FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
FlashAttention-2 is not installed, ignore this if you are not using FlashAttention.
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 1, device: cuda:1, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 4, device: cuda:4, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 3, device: cuda:3, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=1,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=4,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=3,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-48_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 5, device: cuda:5, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=5,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 6, device: cuda:6, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - WARNING - llmtuner.tuner.core.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training.
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
[INFO|training_args.py:1345] 2023-11-12 02:44:48,105 >> Found safetensors installation, but --save_safetensors=False. Safetensors should be a preferred weights saving format due to security and performance reasons. If your model cannot be saved by safetensors please feel free to open an issue at https://github.com/huggingface/safetensors!
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=6,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
[INFO|training_args.py:1798] 2023-11-12 02:44:48,105 >> PyTorch: setting up devices
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 7, device: cuda:7, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=7,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-48_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
/home/data/condaEnv/llama-factory/lib/python3.10/site-packages/transformers/training_args.py:1711: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 2, device: cuda:2, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Process rank: 0, device: cuda:0, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=2,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
11/12/2023 02:44:48 - INFO - llmtuner.tuner.core.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
bf16_full_eval=False,
data_seed=None,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_backend=None,
ddp_broadcast_buffers=None,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=False,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
dispatch_batches=None,
do_eval=False,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_delay=0,
eval_steps=None,
evaluation_strategy=no,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
fsdp=[],
fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},
fsdp_min_num_params=0,
fsdp_transformer_layer_cls_to_wrap=None,
full_determinism=False,
generation_config=None,
generation_max_length=None,
generation_num_beams=None,
gradient_accumulation_steps=8,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_always_push=False,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
include_inputs_for_metrics=False,
include_tokens_per_second=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=5e-05,
length_column_name=length,
load_best_model_at_end=False,
local_rank=0,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/runs/Nov12_02-44-47_k8s-node1,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=3.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
prediction_loss_only=False,
push_to_hub=False,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
ray_scope=last,
remove_unused_columns=True,
report_to=[],
resume_from_checkpoint=None,
run_name=/home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned,
save_on_each_node=False,
save_safetensors=False,
save_steps=200,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=False,
tf32=None,
torch_compile=False,
torch_compile_backend=None,
torch_compile_mode=None,
torchdynamo=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_cpu=False,
use_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
)
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
11/12/2023 02:44:48 - INFO - llmtuner.dsets.loader - Loading dataset /home/hz/projects/LLaMA-Factory/data/wenshu_train.json...
11/12/2023 02:44:48 - WARNING - llmtuner.dsets.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
Using custom data configuration default-6a98bd49aacf7ef0
Loading Dataset Infos from /home/data/condaEnv/llama-factory/lib/python3.10/site-packages/datasets/packaged_modules/json
Overwrite dataset info from restored data version if exists.
Loading Dataset info from /home/hz/.cache/huggingface/datasets/json/default-6a98bd49aacf7ef0/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96
Found cached dataset json (/home/hz/.cache/huggingface/datasets/json/default-6a98bd49aacf7ef0/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96)
Loading Dataset info from /home/hz/.cache/huggingface/datasets/json/default-6a98bd49aacf7ef0/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96
[INFO|tokenization_utils_base.py:2041] 2023-11-12 02:44:49,194 >> loading file tokenizer.model
[INFO|tokenization_utils_base.py:2041] 2023-11-12 02:44:49,194 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:2041] 2023-11-12 02:44:49,194 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:2041] 2023-11-12 02:44:49,194 >> loading file tokenizer_config.json
[INFO|tokenization_utils_base.py:2041] 2023-11-12 02:44:49,194 >> loading file tokenizer.json
[INFO|configuration_utils.py:713] 2023-11-12 02:44:49,377 >> loading configuration file /home/hz/projects/chatglm3-6b-32k/config.json
[INFO|configuration_utils.py:713] 2023-11-12 02:44:49,378 >> loading configuration file /home/hz/projects/chatglm3-6b-32k/config.json
[INFO|configuration_utils.py:775] 2023-11-12 02:44:49,379 >> Model config ChatGLMConfig {
  "_name_or_path": "/home/hz/projects/chatglm3-6b-32k",
  "add_bias_linear": false,
  "add_qkv_bias": true,
  "apply_query_key_layer_scaling": true,
  "apply_residual_connection_post_layernorm": false,
  "architectures": [
    "ChatGLMModel"
  ],
  "attention_dropout": 0.0,
  "attention_softmax_in_fp32": true,
  "auto_map": {
    "AutoConfig": "configuration_chatglm.ChatGLMConfig",
    "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
    "AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification"
  },
  "bias_dropout_fusion": true,
  "classifier_dropout": null,
  "eos_token_id": 2,
  "ffn_hidden_size": 13696,
  "fp32_residual_connection": false,
  "hidden_dropout": 0.0,
  "hidden_size": 4096,
  "kv_channels": 128,
  "layernorm_epsilon": 1e-05,
  "model_type": "chatglm",
  "multi_query_attention": true,
  "multi_query_group_num": 2,
  "num_attention_heads": 32,
  "num_layers": 28,
  "original_rope": true,
  "pad_token_id": 0,
  "padded_vocab_size": 65024,
  "post_layer_norm": true,
  "pre_seq_len": null,
  "prefix_projection": false,
  "quantization_bit": 0,
  "rmsnorm": true,
  "rope_ratio": 50,
  "seq_length": 32768,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.34.0",
  "use_cache": true,
  "vocab_size": 65024
}

[INFO|modeling_utils.py:2990] 2023-11-12 02:44:49,413 >> loading weights file /home/hz/projects/chatglm3-6b-32k/pytorch_model.bin.index.json
[INFO|modeling_utils.py:1220] 2023-11-12 02:44:49,413 >> Instantiating ChatGLMForConditionalGeneration model under default dtype torch.float16.
[INFO|configuration_utils.py:770] 2023-11-12 02:44:49,414 >> Generate config GenerationConfig {
  "eos_token_id": 2,
  "pad_token_id": 0
}


Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:08,  1.40s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.54s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.53s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.58s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.60s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:08,  1.45s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.51s/it]
Loading checkpoint shards:  14%|█▍        | 1/7 [00:01<00:09,  1.53s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:02<00:07,  1.46s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.54s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:02<00:06,  1.40s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.55s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.56s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.57s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.52s/it]
Loading checkpoint shards:  29%|██▊       | 2/7 [00:03<00:07,  1.53s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.41s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.46s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.41s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.48s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:06,  1.52s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:06,  1.52s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.47s/it]
Loading checkpoint shards:  43%|████▎     | 3/7 [00:04<00:05,  1.48s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.37s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.41s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.42s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.40s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:06<00:04,  1.48s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.43s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:06<00:04,  1.48s/it]
Loading checkpoint shards:  57%|█████▋    | 4/7 [00:05<00:04,  1.44s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:06<00:02,  1.37s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:02,  1.39s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:02,  1.39s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:02,  1.44s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:02,  1.46s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:03,  1.50s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:03,  1.51s/it]
Loading checkpoint shards:  71%|███████▏  | 5/7 [00:07<00:02,  1.46s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.34s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.36s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.36s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.42s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.43s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.41s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.48s/it]
Loading checkpoint shards:  86%|████████▌ | 6/7 [00:08<00:01,  1.48s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.17s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.29s/it]
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.21s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.33s/it]
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.21s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.34s/it]
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:58 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.24s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.34s/it]
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.23s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.36s/it]
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.30s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.41s/it]

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.30s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.42s/it]
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
[INFO|modeling_utils.py:3775] 2023-11-12 02:44:59,427 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.

[INFO|modeling_utils.py:3783] 2023-11-12 02:44:59,427 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at /home/hz/projects/chatglm3-6b-32k.
If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
[INFO|modeling_utils.py:3352] 2023-11-12 02:44:59,430 >> Generation config file not found, using a generation config created from the model config.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA

Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.30s/it]
Loading checkpoint shards: 100%|██████████| 7/7 [00:09<00:00,  1.39s/it]
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.utils - Gradient checkpointing enabled.
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.adapter - Fine-tuning method: LoRA
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312
[INFO|tokenization_utils_base.py:952] 2023-11-12 02:44:59,481 >> Assigning ['<|user|>', '<|observation|>'] to the additional_special_tokens key of the tokenizer
11/12/2023 02:44:59 - INFO - llmtuner.tuner.core.loader - trainable params: 1949696 || all params: 6245533696 || trainable%: 0.0312
[INFO|tokenization_utils.py:493] 2023-11-12 02:44:59,596 >> Adding <|user|> to the vocabulary
[INFO|tokenization_utils.py:493] 2023-11-12 02:44:59,596 >> Adding <|observation|> to the vocabulary

Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]Caching processed dataset at /home/hz/.cache/huggingface/datasets/json/default-6a98bd49aacf7ef0/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96/cache-e4be9020819d52ba.arrow

Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<19:59, 317.69 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:05<18:49, 336.51 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:08<18:29, 341.63 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:11<18:04, 348.53 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:14<18:00, 348.77 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:17<18:05, 346.32 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:20<18:00, 347.15 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:23<17:50, 349.31 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:26<17:52, 347.88 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:29<18:03, 343.32 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:31<17:58, 344.12 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:35<18:14, 337.94 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:37<17:59, 341.91 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:40<17:47, 344.78 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:43<17:33, 348.37 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:46<17:34, 347.24 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:49<17:23, 349.68 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:51<17:06, 354.69 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [00:54<17:07, 353.18 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [00:57<17:07, 352.42 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:00<17:05, 351.94 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:03<17:19, 346.27 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:06<17:28, 342.51 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:09<17:18, 344.81 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:12<17:12, 345.92 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:15<17:17, 343.29 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:18<17:12, 343.81 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:20<17:01, 346.71 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:23<17:13, 341.67 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:26<17:14, 340.23 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:29<17:00, 343.92 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:32<17:02, 342.35 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:35<16:48, 346.09 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:38<16:45, 345.96 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:41<16:31, 349.86 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:44<16:32, 348.77 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [01:46<16:23, 350.92 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [01:49<16:28, 348.10 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [01:52<16:29, 346.73 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [01:55<16:34, 344.02 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [01:58<16:33, 343.24 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:01<16:34, 341.88 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:04<16:32, 341.68 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:07<16:27, 342.13 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:10<16:24, 342.29 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:13<16:22, 342.03 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:16<16:22, 340.83 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:19<16:10, 344.12 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:21<15:58, 347.49 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:24<15:52, 348.62 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:27<15:40, 351.76 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:30<15:41, 350.44 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:33<15:37, 350.75 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:35<15:24, 354.93 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:38<15:18, 356.14 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [02:41<15:18, 354.78 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [02:44<15:14, 355.55 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [02:47<15:15, 354.07 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [02:50<15:20, 350.75 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [02:53<15:25, 347.96 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [02:56<15:38, 342.13 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [02:58<15:26, 345.24 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:01<15:16, 348.16 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:04<15:15, 347.17 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:07<15:05, 349.89 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:10<15:03, 349.91 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:13<14:59, 350.38 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:16<15:00, 348.58 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:18<14:55, 349.58 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:21<15:03, 345.32 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:24<14:51, 348.88 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:27<14:45, 350.18 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:30<14:36, 352.60 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:33<14:35, 351.74 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [03:36<14:37, 350.03 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [03:39<14:48, 344.59 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [03:42<14:57, 340.01 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [03:44<14:44, 343.75 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [03:47<14:32, 347.38 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [03:50<14:42, 342.25 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [03:53<14:35, 343.70 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [03:56<14:28, 345.28 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [03:59<14:16, 349.22 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:02<14:06, 352.14 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:04<14:01, 353.00 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:07<13:57, 353.24 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:10<13:53, 353.80 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:13<13:50, 353.91 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:16<13:47, 354.18 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:19<13:51, 351.17 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:21<13:48, 351.20 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:24<13:48, 350.17 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:27<13:46, 349.49 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [04:30<13:55, 344.53 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [04:33<13:58, 342.19 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [04:36<13:58, 340.93 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [04:39<13:52, 342.17 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [04:42<13:47, 343.37 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [04:45<13:37, 345.97 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [04:48<13:34, 346.06 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [04:50<13:26, 348.37 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [04:53<13:26, 347.12 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [04:56<13:25, 346.26 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [04:59<13:20, 347.26 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:02<13:11, 349.77 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:05<13:02, 352.71 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:08<13:28, 340.04 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:11<13:14, 344.70 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:14<13:12, 344.62 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:16<13:01, 348.02 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:19<13:05, 345.20 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [05:22<13:00, 345.86 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [05:25<12:59, 344.89 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [05:28<13:01, 342.81 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [05:31<12:55, 344.11 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [05:34<12:51, 344.96 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [05:37<12:46, 345.90 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [05:40<12:35, 349.27 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [05:42<12:35, 348.13 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [05:45<12:32, 348.10 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [05:48<12:32, 346.72 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [05:51<12:32, 345.66 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [05:54<12:29, 345.63 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [05:57<12:23, 347.09 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:00<12:15, 349.19 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:03<12:13, 348.84 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:05<12:11, 348.64 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:08<12:10, 347.51 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:11<12:16, 343.65 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:14<12:14, 342.97 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [06:17<12:07, 344.81 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [06:20<12:07, 343.59 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [06:23<11:59, 345.91 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [06:26<11:58, 345.27 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [06:29<11:49, 348.31 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [06:31<11:41, 350.44 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [06:34<11:40, 349.53 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [06:37<11:41, 348.05 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [06:40<11:35, 349.57 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [06:43<11:31, 349.90 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [06:46<11:31, 348.34 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [06:49<11:31, 347.05 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [06:52<11:30, 346.12 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [06:54<11:22, 348.63 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [06:57<11:20, 348.17 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:00<11:23, 345.05 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:03<11:15, 347.86 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:06<11:08, 349.83 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [07:09<11:11, 346.78 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [07:12<11:09, 346.36 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [07:15<11:06, 346.81 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [07:18<11:02, 347.19 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [07:20<10:58, 347.95 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [07:23<10:48, 351.64 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [07:26<10:41, 353.82 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [07:29<10:38, 353.94 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [07:31<10:27, 358.69 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [07:34<10:27, 357.03 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [07:37<10:30, 353.57 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [07:40<10:27, 353.84 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [07:43<10:29, 350.90 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [07:46<10:20, 354.80 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [07:49<10:24, 350.82 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [07:51<10:25, 348.68 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [07:55<10:33, 342.55 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [07:57<10:28, 343.87 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:00<10:28, 342.17 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [08:03<10:22, 344.00 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [08:06<10:20, 343.31 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [08:09<10:08, 348.68 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [08:12<10:08, 347.00 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [08:15<10:00, 349.52 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [08:17<09:54, 351.60 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [08:20<09:49, 352.73 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [08:23<09:44, 354.39 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [08:26<09:44, 352.38 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [08:29<09:38, 354.26 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [08:32<09:41, 350.58 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [08:34<09:34, 353.64 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [08:37<09:29, 354.88 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [08:40<09:24, 355.77 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [08:43<09:21, 356.50 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [08:46<09:17, 356.70 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [08:48<09:20, 353.43 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [08:51<09:18, 352.46 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [08:54<09:17, 351.67 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [08:57<09:17, 349.97 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [09:00<09:16, 348.57 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [09:03<09:11, 350.17 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [09:06<09:06, 351.01 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [09:09<09:10, 347.07 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [09:11<09:05, 348.27 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [09:14<09:12, 342.00 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [09:17<09:08, 342.56 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [09:20<09:09, 340.23 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [09:23<09:03, 342.18 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [09:26<09:01, 341.90 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [09:29<08:59, 341.24 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [09:32<08:47, 346.71 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [09:35<08:55, 339.55 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [09:38<08:54, 338.76 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [09:41<08:47, 341.03 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [09:44<08:43, 342.09 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [09:47<08:35, 345.21 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [09:49<08:31, 346.17 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [09:52<08:27, 346.70 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [09:55<08:21, 348.67 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [09:58<08:22, 346.35 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [10:01<08:17, 347.71 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [10:04<08:16, 346.39 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [10:07<08:14, 346.07 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [10:10<08:13, 344.47 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [10:13<08:11, 343.77 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [10:16<08:09, 343.50 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [10:18<08:08, 341.59 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [10:21<08:03, 343.68 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [10:24<07:58, 344.52 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [10:27<07:57, 343.72 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [10:30<07:57, 341.16 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [10:33<07:53, 342.10 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [10:36<07:49, 342.70 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [10:39<07:56, 336.09 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [10:42<07:46, 341.18 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [10:45<07:33, 348.47 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [10:47<07:29, 349.23 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [10:50<07:27, 348.60 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [10:53<07:25, 347.68 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [10:56<07:20, 349.43 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [10:59<07:18, 349.30 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [11:02<07:16, 348.22 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [11:05<07:13, 348.34 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [11:08<07:13, 346.11 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [11:11<07:10, 346.25 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [11:14<07:12, 341.91 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [11:16<07:03, 347.36 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [11:19<06:57, 349.75 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [11:22<06:51, 352.73 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [11:25<06:52, 348.97 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [11:28<06:46, 351.91 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [11:30<06:39, 355.80 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [11:33<06:36, 355.27 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [11:36<06:36, 352.95 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [11:39<06:34, 352.51 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [11:42<06:32, 351.49 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [11:45<06:28, 352.69 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [11:47<06:28, 349.89 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [11:50<06:26, 348.98 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [11:53<06:22, 350.52 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [11:56<06:23, 346.81 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [11:59<06:26, 341.21 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [12:02<06:21, 343.11 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [12:05<06:18, 343.08 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [12:08<06:12, 345.94 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [12:11<06:12, 343.98 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [12:14<06:04, 348.59 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [12:16<06:02, 347.41 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [12:19<06:01, 345.77 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [12:22<05:55, 348.33 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [12:25<05:54, 346.72 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [12:28<05:52, 346.19 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [12:31<05:47, 348.41 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [12:34<05:44, 348.52 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [12:37<05:41, 348.27 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [12:39<05:40, 346.55 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [12:42<05:39, 344.80 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [12:45<05:36, 344.90 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [12:48<05:32, 345.71 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [12:51<05:29, 346.39 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [12:54<05:27, 344.56 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [12:57<05:25, 344.46 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [13:00<05:20, 346.66 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [13:03<05:15, 348.74 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [13:05<05:12, 349.19 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [13:08<05:07, 351.68 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [13:11<05:03, 352.83 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [13:14<05:03, 349.57 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [13:17<05:00, 349.06 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [13:20<04:58, 348.50 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [13:23<04:57, 346.63 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [13:26<04:56, 343.51 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [13:29<04:55, 341.72 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [13:31<04:52, 342.24 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [13:34<04:49, 342.51 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [13:37<04:42, 347.32 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [13:40<04:42, 343.02 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [13:43<04:36, 346.62 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [13:46<04:30, 350.59 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [13:49<04:28, 350.25 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [13:52<04:25, 350.23 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [13:54<04:21, 351.34 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [13:57<04:17, 353.31 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [14:00<04:14, 353.41 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [14:03<04:12, 352.49 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [14:06<04:15, 344.02 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [14:09<04:13, 343.73 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [14:12<04:07, 348.10 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [14:14<04:04, 347.90 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [14:17<04:02, 347.01 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [14:20<03:58, 348.33 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [14:23<03:52, 351.99 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [14:26<03:48, 354.26 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [14:29<03:45, 354.48 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [14:31<03:42, 355.45 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [14:34<03:43, 349.73 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [14:37<03:42, 345.44 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [14:40<03:37, 349.66 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [14:43<03:35, 347.71 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [14:46<03:32, 347.71 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [14:49<03:30, 346.48 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [14:52<03:26, 348.71 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [14:54<03:23, 348.72 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [14:57<03:21, 347.50 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [15:00<03:18, 346.99 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [15:03<03:15, 347.72 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [15:06<03:12, 347.20 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [15:09<03:11, 344.87 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [15:12<03:08, 344.80 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [15:15<03:05, 345.20 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [15:18<03:04, 342.22 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [15:21<03:02, 339.56 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [15:24<03:00, 338.08 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [15:27<02:56, 339.72 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [15:30<02:53, 339.50 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [15:32<02:50, 340.62 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [15:35<02:44, 347.06 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [15:38<02:41, 347.58 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [15:41<02:38, 346.06 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [15:44<02:34, 348.89 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [15:47<02:31, 349.53 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [15:50<02:28, 349.26 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [15:52<02:23, 354.88 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [15:55<02:21, 353.30 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [15:58<02:18, 354.73 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [16:01<02:15, 354.57 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [16:04<02:13, 352.44 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [16:07<02:11, 350.74 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [16:09<02:07, 352.27 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [16:12<02:04, 352.26 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [16:15<02:02, 351.43 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [16:18<02:02, 342.02 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [16:21<01:59, 343.83 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [16:24<01:55, 345.31 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [16:27<01:53, 345.06 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [16:30<01:50, 342.80 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [16:33<01:47, 342.80 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [16:36<01:44, 344.77 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [16:38<01:40, 349.10 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [16:41<01:37, 348.70 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [16:44<01:35, 347.04 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [16:47<01:31, 348.78 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [16:50<01:29, 346.22 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [16:53<01:27, 344.63 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [16:56<01:24, 343.16 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [16:59<01:21, 342.48 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [17:02<01:18, 342.25 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [17:04<01:15, 344.51 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [17:07<01:11, 347.80 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [17:10<01:09, 347.00 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [17:13<01:05, 349.12 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [17:16<01:03, 347.80 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [17:19<01:00, 349.10 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [17:22<00:57, 350.08 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [17:24<00:54, 350.48 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [17:27<00:51, 347.43 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [17:30<00:48, 347.18 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [17:33<00:45, 349.49 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [17:36<00:42, 353.56 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [17:39<00:39, 354.39 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [17:41<00:36, 354.23 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [17:44<00:33, 355.18 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [17:47<00:31, 351.97 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [17:50<00:28, 353.18 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [17:53<00:25, 352.15 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [17:56<00:22, 352.31 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [17:59<00:20, 349.39 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [18:01<00:17, 352.13 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [18:04<00:14, 352.41 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [18:07<00:11, 349.23 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [18:10<00:08, 348.21 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [18:13<00:05, 346.50 examples/s]
Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [18:16<00:02, 343.62 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [18:19<00:00, 341.49 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [18:19<00:00, 347.47 examples/s]
input_ids:
[64790, 64792, 64795, 30910, 13, 30910, 37117, 31915, 32114, 31123, 44531, 31793, 35100, 31915, 32722, 33855, 47805, 54599, 54907, 54556, 36454, 31915, 54593, 39385, 35939, 31951, 31155, 44531, 32116, 48491, 31822, 31915, 41140, 41507, 31709, 31155, 13, 54534, 20727, 54827, 34172, 31915, 32722, 54538, 31771, 30977, 17102, 10618, 30991, 43329, 54544, 33598, 35939, 31951, 31123, 34999, 31793, 30989, 30977, 17102, 13080, 337, 340, 34120, 38233, 34148, 41905, 31300, 54542, 30989, 30977, 17102, 2383, 1010, 34120, 32234, 54530, 41507, 43871, 31807, 31300, 54607, 32136, 32075, 16414, 30991, 34100, 46901, 34490, 31155, 13, 31714, 33485, 40967, 32722, 33163, 31155, 55073, 31793, 30989, 30977, 17102, 13080, 337, 340, 30991, 54542, 30989, 30977, 17102, 2383, 1010, 30991, 54952, 55005, 31201, 31724, 31915, 54599, 54907, 34421, 41947, 31707, 31123, 39234, 54960, 46901, 35939, 31951, 30989, 30977, 17102, 10618, 31681, 13, 30982, 30953, 16414, 5440, 765, 30978, 30970, 30973, 30970, 55493, 55937, 54619, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 35665, 56932, 55270, 34694, 43340, 35841, 35939, 54755, 1589, 765, 30977, 17102, 13080, 337, 340, 5440, 765, 38233, 55493, 55937, 54759, 42503, 32013, 33817, 34899, 31211, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54564, 48189, 36932, 54598, 55660, 54747, 34023, 33507, 56081, 54557, 31155, 32650, 54542, 33652, 31211, 55493, 55937, 32719, 48189, 36932, 54598, 55660, 54747, 34023, 33507, 31123, 39311, 30943, 30940, 30939, 30973, 54540, 30972, 54595, 30939, 30969, 54576, 32871, 45204, 54983, 37884, 33843, 32867, 31155, 54688, 32848, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 33269, 40532, 33507, 54680, 33487, 56568, 54557, 31155, 30974, 30916, 34148, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 55222, 54933, 54553, 46440, 31155, 30974, 30916, 42503, 54608, 36951, 47382, 33163, 32650, 31211, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54712, 33434, 31775, 31155, 30943, 30940, 30939, 30981, 54540, 30970, 54595, 30966, 54576, 31123, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54759, 34773, 36932, 54598, 55752, 56163, 43986, 54807, 32127, 43986, 32258, 35074, 31201, 32258, 38469, 58268, 54542, 33145, 54919, 31123, 32419, 32871, 35715, 36932, 54598, 55660, 54747, 34023, 33507, 31301, 32250, 54541, 30973, 30969, 30930, 30981, 30939, 33350, 31966, 49210, 31123, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54687, 33002, 33507, 39666, 54840, 56462, 54632, 55555, 31123, 37914, 40427, 54541, 30972, 30940, 32095, 31155, 30943, 30940, 30939, 30981, 54540, 30970, 54595, 30972, 54576, 31123, 56707, 55452, 56700, 54759, 54884, 56298, 38323, 30966, 30940, 32095, 31123, 54540, 34473, 30939, 30970, 32197, 30943, 30940, 30939, 30973, 54540, 30972, 54595, 30939, 30973, 54576, 31123, 54884, 56298, 33830, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54619, 55493, 55937, 34231, 36932, 54598, 45204, 54983, 37884, 32500, 31301, 32500, 35908, 54943, 31211, 30943, 30940, 30939, 30973, 30940, 30972, 30939, 30973, 30940, 30943, 30972, 30939, 31798, 54687, 33002, 33507, 35715, 54840, 55493, 55937, 31155, 49210, 31123, 54884, 56298, 54759, 55493, 38068, 54893, 54821, 54907, 32309, 54757, 33685, 54983, 55115, 31155, 30943, 30940, 30939, 30973, 54540, 30972, 54595, 30939, 30969, 54576, 31123, 33002, 33507, 33843, 32867, 54579, 55493, 55937, 54653, 54578, 31155, 30974, 30916, 47607, 54538, 31123, 55493, 55937, 55037, 54674, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 33487, 56568, 54557, 33002, 33507, 31123, 54633, 32843, 54759, 55892, 54731, 35679, 36533, 31123, 33956, 54574, 32843, 36278, 31940, 31123, 54688, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 33487, 38457, 31155, 31672, 31123, 55493, 55937, 32848, 32067, 38114, 31201, 31695, 54960, 33507, 1589, 765, 30977, 17102, 2383, 1010, 5440, 765, 42503, 31813, 31211, 34721, 40410, 36592, 33127, 31201, 36220, 31201, 35928, 54542, 40285, 31123, 54608, 32599, 32867, 31123, 31763, 38890, 31155, 56932, 55270, 54679, 54991, 54746, 31667, 56932, 55270, 54679, 36592, 31123, 33370, 54533, 31628, 35665, 56932, 55270, 31767, 34833, 33191, 31155, 39002, 54538, 31123, 55493, 55937, 54541, 36932, 54598, 55660, 54747, 34023, 33507, 48593, 54533, 31123, 42276, 33507, 36254, 38114, 31201, 31695, 31201, 33603, 31201, 54807, 35514, 33370, 31155, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54716, 54991, 38114, 33002, 33507, 31123, 54633, 31898, 42130, 55493, 55937, 54530, 54679, 54991, 31123, 54720, 56081, 54557, 54960, 33507, 31155, 46483, 31123, 35713, 54611, 33332, 54679, 54991, 54599, 54612, 35086, 54907, 31201, 31865, 43509, 32005, 33332, 35841, 47758, 54612, 38144, 54795, 43059, 54586, 31852, 31123, 35939, 33163, 25109, 64796, 30910, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54579, 54613, 35939, 37632, 54585, 33628, 37555, 56081, 54557, 48189, 36932, 54598, 55660, 54747, 34023, 33507, 31155, 13, 39002, 35783, 55000, 55325, 55205, 37694, 30972, 30940, 54751, 31123, 54781, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 35022, 31155, 13, 54627, 40873, 54613, 35939, 31123, 36575, 35939, 54755, 43922, 38079, 33628, 37555, 31123, 54759, 42503, 45070, 39829, 55191, 31123, 54724, 55194, 32260, 35469, 47633, 32013, 51150, 31123, 39829, 54579, 34773, 36932, 54598, 44179, 31155, 31701, 31123, 31793, 54611, 33817, 32974, 54745, 55428, 32355, 54612, 34900, 31123, 54759, 54960, 54700, 55135, 54745, 39829, 32722, 35783, 55000, 2]
inputs:
[gMASK]sop<|user|> 
 作为一个法律专家，你需要根据提供的法律案件描述和相关法条来生成法律文书的判决结果。你需要充分调用你的法律知识和推理能力。
在JSON格式的法律案件中，“JudgeResult”是需要生成的判决结果，它是根据“JudgeAccusation”（原告被告指控）和“JudgeReason”（法院的推理归纳过程）得出的。“Case”则是案件的标题。
现在给你一个新的案件如下。请根据“JudgeAccusation”和“JudgeReason”字段、相关法律法条和其他有用信息，得出该案件的判决结果“JudgeResult”。
{'Case': '6585杨杰与潘宗锡、时维娟排除妨害纠纷一审民事判决书', 'JudgeAccusation': '原告杨杰向本院提出诉讼请求：潘宗锡、时维娟自坐落于无锡市惠山区的房屋迁出。事实和理由：杨杰购买坐落于无锡市惠山区的房屋，并于2018年4月19日办理存量房买卖转移登记。但至今潘宗锡、时维娟居住在该房屋内拒绝搬出。\n被告潘宗锡、时维娟均未作答辩。\n本院经审理查明如下事实：潘宗锡、时维娟系夫妻关系。2017年5月3日，潘宗锡、时维娟向江苏省无锡市梁溪公证处申请公证居民身份证、居民户口簿和结婚证，用于办理出售无锡市惠山区的房屋（面积为89.71平方米）。同日，潘宗锡、时维娟将上述房屋抵押给邓建峰，债权数额为40万元。2017年5月4日，潘宗锡向王芳借款30万元，年利率15%。2018年4月18日，王芳代理潘宗锡、时维娟与杨杰签订无锡市存量房买卖合同（合同备案号：201804180241），将上述房屋出售给杨杰。同日，王芳向杨杰出具收条明确已收到房款。2018年4月19日，上述房屋转移登记于杨杰名下。\n庭审中，杨杰称因潘宗锡、时维娟拒绝搬出上述房屋，其多次向洛社派出所报警，民警也多次上门处理，但潘宗锡、时维娟拒绝开门。所以，杨杰至今无法占有、使用该房屋', 'JudgeReason': '本院认为：不动产物权的设立、变更、转让和消灭，经依法登记，发生效力。妨害物权或可能妨害物权的，权利人可以排除妨害或者消除危险。本案中，杨杰为无锡市惠山区的房屋所有权人，对该房屋享有占有、使用、收益、处分的权利。潘宗锡、时维娟无权占有上述房屋，其行为侵犯了杨杰的物权，应迁出该房屋。综上，依照《中华人民共和国物权法》第九条、第三十五条、《中华人民共和国民事诉讼法》第一百四十四条之规定，判决如下'}<|assistant|> 潘宗锡、时维娟于本判决生效后十五日内迁出坐落于无锡市惠山区的房屋。
本案受理费减半收取40元，由潘宗锡、时维娟负担。
如不服本判决，可在判决书送达之日起十五日内，向本院递交上诉状，并按对方当事人的人数提出副本，上诉于江苏省无锡市中级人民法院。同时，根据《诉讼费用交纳办法》的规定，向该院预交上诉案件受理费
label_ids:
[-100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 30910, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 54579, 54613, 35939, 37632, 54585, 33628, 37555, 56081, 54557, 48189, 36932, 54598, 55660, 54747, 34023, 33507, 31155, 13, 39002, 35783, 55000, 55325, 55205, 37694, 30972, 30940, 54751, 31123, 54781, 56707, 55452, 56700, 31201, 54554, 55094, 57379, 35022, 31155, 13, 54627, 40873, 54613, 35939, 31123, 36575, 35939, 54755, 43922, 38079, 33628, 37555, 31123, 54759, 42503, 45070, 39829, 55191, 31123, 54724, 55194, 32260, 35469, 47633, 32013, 51150, 31123, 39829, 54579, 34773, 36932, 54598, 44179, 31155, 31701, 31123, 31793, 54611, 33817, 32974, 54745, 55428, 32355, 54612, 34900, 31123, 54759, 54960, 54700, 55135, 54745, 39829, 32722, 35783, 55000, 2]
labels:
潘宗锡、时维娟于本判决生效后十五日内迁出坐落于无锡市惠山区的房屋。
本案受理费减半收取40元，由潘宗锡、时维娟负担。
如不服本判决，可在判决书送达之日起十五日内，向本院递交上诉状，并按对方当事人的人数提出副本，上诉于江苏省无锡市中级人民法院。同时，根据《诉讼费用交纳办法》的规定，向该院预交上诉案件受理费

Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/382000 [00:00<?, ? examples/s][INFO|training_args.py:1345] 2023-11-12 03:03:21,435 >> Found safetensors installation, but --save_safetensors=False. Safetensors should be a preferred weights saving format due to security and performance reasons. If your model cannot be saved by safetensors please feel free to open an issue at https://github.com/huggingface/safetensors!
[INFO|training_args.py:1798] 2023-11-12 03:03:21,435 >> PyTorch: setting up devices
Detected kernel version 4.15.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.

Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<20:13, 314.05 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<20:28, 310.23 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<20:33, 308.80 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<20:35, 308.35 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<21:06, 300.90 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<21:25, 296.40 examples/s]
Running tokenizer on dataset:   0%|          | 1000/382000 [00:03<21:35, 294.16 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:01, 316.18 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:14, 312.83 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:16, 312.49 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:17, 312.20 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:47, 304.51 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<20:59, 301.59 examples/s]
Running tokenizer on dataset:   1%|          | 2000/382000 [00:06<21:07, 299.79 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<19:37, 321.97 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<19:51, 318.16 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<19:52, 317.74 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<20:14, 312.19 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<20:32, 307.49 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<20:44, 304.54 examples/s]
Running tokenizer on dataset:   1%|          | 3000/382000 [00:09<20:50, 303.11 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:12<19:15, 327.18 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:12<19:27, 323.80 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:12<19:30, 322.92 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:12<19:53, 316.85 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:12<20:15, 311.06 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:13<20:32, 306.66 examples/s]
Running tokenizer on dataset:   1%|          | 4000/382000 [00:13<20:37, 305.48 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:15<19:09, 327.98 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:15<19:19, 325.08 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:15<19:29, 322.48 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:15<19:42, 318.78 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:16<20:08, 311.89 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:16<20:36, 305.02 examples/s]
Running tokenizer on dataset:   1%|▏         | 5000/382000 [00:16<20:43, 303.21 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:18<19:28, 321.84 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:18<19:23, 323.12 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:18<19:30, 321.29 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:18<19:45, 317.03 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:19<20:17, 308.77 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:19<20:36, 304.09 examples/s]
Running tokenizer on dataset:   2%|▏         | 6000/382000 [00:19<20:53, 300.02 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:21<19:32, 319.90 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:21<19:35, 319.13 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:21<19:35, 319.06 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:22<19:46, 316.06 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:22<20:13, 308.92 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:23<20:37, 303.07 examples/s]
Running tokenizer on dataset:   2%|▏         | 7000/382000 [00:23<20:41, 302.14 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:24<19:35, 318.05 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:25<19:34, 318.35 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:25<19:30, 319.51 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:25<20:02, 310.96 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:25<19:56, 312.66 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:26<20:25, 305.16 examples/s]
Running tokenizer on dataset:   2%|▏         | 8000/382000 [00:26<20:44, 300.64 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:28<19:39, 316.36 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:28<19:36, 317.01 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:28<19:27, 319.45 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:28<20:44, 299.70 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:29<19:54, 312.31 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:29<20:23, 304.91 examples/s]
Running tokenizer on dataset:   2%|▏         | 9000/382000 [00:29<20:36, 301.57 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:31<19:48, 312.96 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:31<19:50, 312.51 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:31<19:35, 316.35 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:32<20:35, 301.11 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:32<20:03, 308.97 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:33<20:44, 298.97 examples/s]
Running tokenizer on dataset:   3%|▎         | 10000/382000 [00:33<21:02, 294.56 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:34<19:41, 314.00 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:34<19:22, 319.24 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:34<19:52, 311.04 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:35<20:18, 304.57 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:35<19:55, 310.32 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:36<20:38, 299.64 examples/s]
Running tokenizer on dataset:   3%|▎         | 11000/382000 [00:36<20:40, 298.99 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:37<19:40, 313.52 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:37<19:19, 319.14 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:38<19:51, 310.54 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:38<20:07, 306.35 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:38<20:01, 307.97 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:39<20:45, 297.03 examples/s]
Running tokenizer on dataset:   3%|▎         | 12000/382000 [00:39<20:36, 299.25 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:40<19:01, 323.16 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:40<19:23, 317.24 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:41<19:37, 313.41 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:41<19:47, 310.84 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:41<19:42, 311.96 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:42<20:13, 303.99 examples/s]
Running tokenizer on dataset:   3%|▎         | 13000/382000 [00:43<20:19, 302.60 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:43<18:56, 323.89 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:44<19:24, 316.12 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:44<19:47, 309.80 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:45<19:48, 309.67 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:45<19:53, 308.46 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:46<20:01, 306.38 examples/s]
Running tokenizer on dataset:   4%|▎         | 14000/382000 [00:46<20:17, 302.31 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:46<18:52, 324.19 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:47<19:14, 317.92 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:47<19:46, 309.37 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:48<19:34, 312.53 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:48<19:39, 311.09 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:49<19:50, 308.37 examples/s]
Running tokenizer on dataset:   4%|▍         | 15000/382000 [00:49<20:21, 300.54 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:50<18:51, 323.57 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:50<19:04, 319.87 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:50<19:31, 312.52 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:51<19:25, 314.07 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:51<19:26, 313.76 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:52<19:43, 309.15 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:53<18:38, 326.25 examples/s]
Running tokenizer on dataset:   4%|▍         | 16000/382000 [00:53<20:20, 299.96 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:53<18:56, 321.13 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:53<19:25, 313.16 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:54<19:14, 316.03 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:54<19:07, 317.96 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:55<19:36, 310.12 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:56<18:22, 330.03 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:56<18:46, 323.25 examples/s]
Running tokenizer on dataset:   4%|▍         | 17000/382000 [00:56<20:10, 301.54 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:57<19:10, 316.38 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:57<19:04, 318.04 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:57<19:09, 316.55 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:58<19:18, 314.16 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [00:59<18:29, 327.05 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [00:59<18:47, 321.83 examples/s]
Running tokenizer on dataset:   5%|▍         | 18000/382000 [00:59<19:47, 306.54 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [01:00<19:06, 316.51 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [01:00<19:05, 316.86 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [01:00<19:11, 315.23 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [01:02<19:20, 312.74 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:02<18:26, 327.14 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:02<18:45, 321.66 examples/s]
Running tokenizer on dataset:   5%|▍         | 19000/382000 [01:02<19:58, 302.77 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:03<18:59, 317.75 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:03<19:02, 316.82 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:04<19:29, 309.54 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:05<18:24, 326.71 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:05<19:11, 314.41 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:05<18:42, 321.61 examples/s]
Running tokenizer on dataset:   5%|▌         | 20000/382000 [01:06<19:54, 303.11 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:06<19:04, 315.45 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:06<19:00, 316.64 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:07<19:18, 311.60 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:08<18:27, 325.07 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:08<19:25, 309.71 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:08<18:40, 321.16 examples/s]
Running tokenizer on dataset:   5%|▌         | 21000/382000 [01:09<19:38, 306.31 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:09<19:06, 314.06 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:10<19:01, 315.34 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:10<19:39, 305.33 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:11<18:24, 325.18 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:11<19:19, 310.49 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:12<18:47, 318.42 examples/s]
Running tokenizer on dataset:   6%|▌         | 22000/382000 [01:12<19:50, 302.47 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:13<19:10, 311.94 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:13<19:00, 314.75 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:14<19:40, 304.03 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:14<18:17, 326.33 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:15<19:20, 309.36 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:15<18:51, 316.42 examples/s]
Running tokenizer on dataset:   6%|▌         | 23000/382000 [01:16<19:49, 301.88 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:16<19:11, 310.87 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:16<18:56, 314.95 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:17<19:34, 304.88 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:17<18:19, 324.70 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:18<18:41, 318.27 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:18<19:26, 306.78 examples/s]
Running tokenizer on dataset:   6%|▋         | 24000/382000 [01:19<19:39, 303.43 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:19<19:05, 311.79 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:19<18:51, 315.65 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:20<18:28, 321.25 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:20<19:35, 303.81 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:21<18:47, 315.71 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:21<19:27, 305.83 examples/s]
Running tokenizer on dataset:   7%|▋         | 25000/382000 [01:22<19:30, 305.07 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:22<19:09, 309.69 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:23<19:00, 312.07 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:23<18:23, 321.72 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:24<19:36, 302.50 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:24<18:46, 315.18 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:25<19:42, 300.98 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:25<19:00, 311.31 examples/s]
Running tokenizer on dataset:   7%|▋         | 26000/382000 [01:26<19:45, 300.22 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:26<18:52, 313.37 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:26<18:07, 325.38 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:27<19:39, 300.92 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:27<18:32, 318.11 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:28<19:40, 300.84 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:29<18:42, 315.24 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:29<18:33, 318.02 examples/s]
Running tokenizer on dataset:   7%|▋         | 27000/382000 [01:29<19:32, 302.85 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:30<18:14, 322.47 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:30<19:27, 303.22 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:30<18:30, 317.94 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:31<19:23, 304.30 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:32<18:42, 314.43 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:32<18:33, 316.90 examples/s]
Running tokenizer on dataset:   7%|▋         | 28000/382000 [01:32<19:18, 305.54 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:33<18:18, 320.49 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:34<18:29, 317.25 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:34<19:43, 298.37 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:34<19:21, 303.86 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:35<18:37, 315.10 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:35<18:33, 316.21 examples/s]
Running tokenizer on dataset:   8%|▊         | 29000/382000 [01:35<19:30, 301.62 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:36<18:07, 322.86 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:37<18:12, 321.28 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:37<19:43, 297.52 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:38<19:21, 303.17 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:38<18:30, 316.18 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:38<18:18, 319.47 examples/s]
Running tokenizer on dataset:   8%|▊         | 30000/382000 [01:39<19:28, 301.23 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:39<18:06, 322.24 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:40<18:13, 320.16 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:40<19:24, 301.48 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:41<19:07, 305.86 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:41<18:33, 314.27 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:41<18:20, 318.10 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:42<18:02, 322.28 examples/s]
Running tokenizer on dataset:   8%|▊         | 31000/382000 [01:42<19:24, 301.45 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:43<18:06, 321.14 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:44<19:22, 301.15 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:44<18:10, 320.00 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:44<18:30, 314.27 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:44<19:21, 301.27 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:45<17:57, 322.92 examples/s]
Running tokenizer on dataset:   8%|▊         | 32000/382000 [01:45<19:10, 304.25 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:46<18:06, 320.44 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:47<19:12, 302.76 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:47<18:06, 320.20 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:48<18:26, 314.54 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:48<19:15, 301.91 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:48<17:47, 325.16 examples/s]
Running tokenizer on dataset:   9%|▊         | 33000/382000 [01:49<19:00, 305.93 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:49<17:50, 324.05 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:50<19:18, 300.47 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:51<18:01, 320.71 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:51<18:14, 316.92 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:51<19:15, 301.27 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:51<17:42, 325.54 examples/s]
Running tokenizer on dataset:   9%|▉         | 34000/382000 [01:52<19:01, 304.94 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:52<17:49, 323.65 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:53<18:49, 307.18 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:54<18:01, 319.97 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:54<18:09, 317.54 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [01:54<17:33, 327.42 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:54<18:49, 307.18 examples/s]
Running tokenizer on dataset:   9%|▉         | 35000/382000 [01:55<18:53, 306.23 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [01:55<17:44, 324.05 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:57<18:44, 307.69 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [01:57<17:56, 320.58 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [01:57<18:02, 318.68 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [01:57<17:39, 324.59 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:57<18:54, 304.91 examples/s]
Running tokenizer on dataset:   9%|▉         | 36000/382000 [01:58<18:46, 307.06 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [01:58<17:50, 321.35 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [02:00<18:44, 306.71 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [02:00<17:59, 318.65 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [02:00<18:07, 316.40 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:01<17:56, 318.49 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [02:01<18:39, 308.08 examples/s]
Running tokenizer on dataset:  10%|▉         | 37000/382000 [02:02<18:35, 309.23 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:02<17:55, 319.00 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:03<18:05, 315.88 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [02:03<18:56, 302.67 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:03<18:06, 315.81 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:04<18:01, 316.26 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [02:04<18:34, 308.74 examples/s]
Running tokenizer on dataset:  10%|▉         | 38000/382000 [02:05<18:40, 306.95 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:05<18:04, 315.25 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:07<18:15, 312.17 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:07<18:16, 312.00 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:07<19:01, 300.57 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:07<18:12, 312.19 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:07<18:51, 303.08 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:08<17:53, 317.56 examples/s]
Running tokenizer on dataset:  10%|█         | 39000/382000 [02:08<18:36, 307.07 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:10<18:07, 313.57 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:10<18:10, 312.63 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:10<19:10, 297.18 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:10<18:18, 309.54 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:11<19:00, 299.82 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:11<17:55, 316.28 examples/s]
Running tokenizer on dataset:  10%|█         | 40000/382000 [02:11<18:47, 303.27 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:13<18:08, 312.22 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:13<18:18, 309.40 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:13<18:58, 299.64 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:14<18:07, 311.73 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:14<19:03, 298.25 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:14<17:48, 317.23 examples/s]
Running tokenizer on dataset:  11%|█         | 41000/382000 [02:15<18:54, 300.65 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:16<18:08, 311.39 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:16<18:20, 307.92 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:17<19:16, 294.09 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:17<18:27, 305.14 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:17<18:53, 299.85 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:17<17:51, 315.31 examples/s]
Running tokenizer on dataset:  11%|█         | 42000/382000 [02:18<18:45, 302.05 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:19<18:16, 308.33 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:20<18:15, 308.54 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:20<18:05, 310.46 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:20<19:04, 296.19 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:21<17:46, 315.93 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:21<18:48, 300.27 examples/s]
Running tokenizer on dataset:  11%|█▏        | 43000/382000 [02:22<18:52, 299.21 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:23<18:07, 309.74 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:23<18:04, 310.65 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:23<18:11, 307.74 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:24<19:07, 294.66 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:24<17:45, 315.37 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:24<18:35, 302.90 examples/s]
Running tokenizer on dataset:  12%|█▏        | 44000/382000 [02:25<18:46, 300.08 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:26<18:08, 308.55 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:26<18:08, 308.70 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:27<18:15, 305.92 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:27<17:49, 313.32 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:27<18:58, 296.08 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:27<18:27, 304.34 examples/s]
Running tokenizer on dataset:  12%|█▏        | 45000/382000 [02:28<18:27, 304.34 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:29<18:05, 308.75 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:29<18:07, 308.12 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:30<18:06, 307.53 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:30<17:37, 315.91 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:30<18:55, 295.95 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:31<18:31, 302.33 examples/s]
Running tokenizer on dataset:  12%|█▏        | 46000/382000 [02:31<18:34, 301.53 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:32<17:57, 310.10 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:33<18:02, 308.67 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:33<17:53, 310.07 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:33<17:31, 316.82 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:34<18:39, 299.12 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:34<18:28, 302.31 examples/s]
Running tokenizer on dataset:  12%|█▏        | 47000/382000 [02:35<18:31, 301.27 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:36<17:51, 310.91 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:36<17:50, 311.07 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:36<17:55, 308.62 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:37<17:33, 315.24 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:37<18:30, 300.85 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:37<18:21, 303.22 examples/s]
Running tokenizer on dataset:  13%|█▎        | 48000/382000 [02:38<18:15, 304.82 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:39<17:41, 312.90 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:39<17:45, 311.50 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:39<17:37, 312.95 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:40<17:31, 314.89 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:40<18:20, 302.58 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:40<18:04, 306.97 examples/s]
Running tokenizer on dataset:  13%|█▎        | 49000/382000 [02:41<18:00, 308.16 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:42<17:30, 315.14 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:42<17:35, 313.64 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:43<17:24, 315.90 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:43<17:31, 313.90 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:44<18:12, 303.79 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:44<18:01, 306.98 examples/s]
Running tokenizer on dataset:  13%|█▎        | 50000/382000 [02:45<18:15, 303.01 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:45<17:29, 314.33 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:45<17:32, 313.58 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:46<17:18, 316.79 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:46<17:41, 309.95 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:47<17:43, 311.19 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:47<18:16, 301.75 examples/s]
Running tokenizer on dataset:  13%|█▎        | 51000/382000 [02:48<18:00, 306.21 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:48<17:29, 313.37 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:48<17:29, 313.60 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:49<16:58, 321.92 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:49<17:25, 313.79 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:50<17:43, 310.41 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:50<18:13, 301.88 examples/s]
Running tokenizer on dataset:  14%|█▎        | 52000/382000 [02:51<17:59, 305.73 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:51<17:11, 318.05 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:52<17:16, 316.42 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:52<16:52, 323.09 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:53<17:22, 313.66 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:53<17:37, 311.03 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:54<18:17, 299.74 examples/s]
Running tokenizer on dataset:  14%|█▍        | 53000/382000 [02:54<17:54, 306.14 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:54<17:04, 319.27 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:55<17:08, 318.00 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [02:55<16:56, 320.63 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [02:56<17:27, 311.17 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:56<17:21, 314.87 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:57<18:07, 301.61 examples/s]
Running tokenizer on dataset:  14%|█▍        | 54000/382000 [02:57<17:41, 308.92 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [02:58<17:11, 316.01 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [02:58<17:11, 316.01 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [02:58<17:19, 312.51 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [02:59<17:19, 312.70 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [02:59<17:14, 315.97 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [03:00<18:02, 301.99 examples/s]
Running tokenizer on dataset:  14%|█▍        | 55000/382000 [03:01<17:32, 310.73 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [03:01<17:03, 317.42 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [03:01<17:04, 317.09 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:02<17:19, 311.81 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:02<17:13, 313.38 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [03:03<17:14, 315.27 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [03:03<17:51, 304.19 examples/s]
Running tokenizer on dataset:  15%|█▍        | 56000/382000 [03:04<17:33, 309.42 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:04<17:04, 316.27 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:04<17:05, 315.82 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:05<17:08, 314.04 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:05<17:11, 313.24 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [03:06<17:08, 316.02 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [03:07<17:43, 305.53 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:07<16:56, 317.66 examples/s]
Running tokenizer on dataset:  15%|█▍        | 57000/382000 [03:07<17:35, 307.98 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:07<17:04, 315.28 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:08<17:02, 315.05 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:09<17:07, 313.40 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:09<17:05, 315.93 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:10<17:52, 302.09 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:10<16:55, 316.96 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:10<16:56, 316.76 examples/s]
Running tokenizer on dataset:  15%|█▌        | 58000/382000 [03:11<17:41, 305.23 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:11<16:58, 315.28 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:12<17:04, 313.39 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:12<17:00, 316.53 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:13<17:47, 302.46 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:13<16:58, 315.06 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:14<16:55, 316.12 examples/s]
Running tokenizer on dataset:  15%|█▌        | 59000/382000 [03:14<17:31, 307.28 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:14<16:50, 316.55 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:15<16:50, 316.62 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:15<16:54, 317.28 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:17<16:50, 316.63 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:17<16:47, 317.60 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:17<18:00, 298.04 examples/s]
Running tokenizer on dataset:  16%|█▌        | 60000/382000 [03:17<17:26, 307.78 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:17<16:40, 318.85 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:18<16:44, 317.41 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:18<16:53, 316.69 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:20<16:45, 317.25 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:20<16:39, 319.00 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:20<17:56, 298.08 examples/s]
Running tokenizer on dataset:  16%|█▌        | 61000/382000 [03:20<17:34, 304.47 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:20<16:38, 318.53 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:21<16:45, 316.30 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:21<16:40, 319.98 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:23<16:49, 315.01 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:23<16:45, 316.16 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:23<16:31, 319.78 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:23<17:14, 309.18 examples/s]
Running tokenizer on dataset:  16%|█▌        | 62000/382000 [03:23<17:48, 299.53 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:24<16:38, 317.51 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:24<16:35, 320.48 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:26<16:42, 316.36 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:26<16:36, 318.22 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:27<16:30, 319.03 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:27<17:09, 309.98 examples/s]
Running tokenizer on dataset:  16%|█▋        | 63000/382000 [03:27<17:40, 300.88 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:27<16:37, 316.79 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:28<16:39, 318.19 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:29<16:41, 315.52 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:29<16:38, 316.48 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:30<16:27, 318.84 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:30<17:03, 310.55 examples/s]
Running tokenizer on dataset:  17%|█▋        | 64000/382000 [03:30<17:37, 300.69 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:31<16:37, 315.91 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:31<16:33, 319.16 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:32<16:41, 314.42 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:33<16:39, 315.12 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:33<16:24, 319.09 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:33<17:00, 310.60 examples/s]
Running tokenizer on dataset:  17%|█▋        | 65000/382000 [03:33<17:38, 299.37 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:34<16:37, 314.81 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:34<16:32, 318.29 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:36<16:42, 313.34 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:36<16:37, 314.70 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:36<16:18, 319.88 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:36<17:02, 308.92 examples/s]
Running tokenizer on dataset:  17%|█▋        | 66000/382000 [03:37<17:33, 299.87 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:37<16:26, 317.33 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:37<16:35, 316.37 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:39<16:35, 314.34 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:39<16:35, 314.56 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:39<16:51, 308.37 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:40<17:08, 306.18 examples/s]
Running tokenizer on dataset:  18%|█▊        | 67000/382000 [03:40<17:27, 300.73 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:40<16:27, 316.05 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:40<16:35, 315.34 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:42<16:43, 310.78 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:42<16:34, 313.80 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:43<16:46, 308.84 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:43<17:11, 304.51 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:43<16:13, 319.39 examples/s]
Running tokenizer on dataset:  18%|█▊        | 68000/382000 [03:43<17:18, 302.26 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:43<16:27, 316.82 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:45<16:18, 317.70 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:45<16:34, 312.83 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:46<16:40, 309.80 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:46<16:56, 308.05 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:46<16:10, 319.27 examples/s]
Running tokenizer on dataset:  18%|█▊        | 69000/382000 [03:47<17:13, 302.91 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:47<16:31, 314.56 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:48<16:17, 317.22 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:48<16:25, 314.63 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:49<16:32, 311.41 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:49<16:01, 321.51 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:49<16:46, 310.01 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:50<16:17, 318.02 examples/s]
Running tokenizer on dataset:  18%|█▊        | 70000/382000 [03:50<17:00, 305.71 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:52<16:09, 318.73 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:52<16:17, 316.04 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:52<16:35, 309.35 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:52<15:58, 321.41 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:53<16:38, 311.44 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:53<16:11, 319.16 examples/s]
Running tokenizer on dataset:  19%|█▊        | 71000/382000 [03:53<16:50, 307.76 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:55<16:07, 318.42 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:55<16:14, 316.22 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [03:55<16:22, 312.56 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [03:56<15:56, 320.89 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:56<16:32, 312.21 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:56<15:58, 322.41 examples/s]
Running tokenizer on dataset:  19%|█▉        | 72000/382000 [03:56<16:52, 306.18 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [03:58<16:02, 319.11 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [03:58<16:23, 312.24 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [03:59<16:26, 310.12 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [03:59<16:07, 316.27 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [03:59<16:32, 311.45 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [03:59<15:54, 322.65 examples/s]
Running tokenizer on dataset:  19%|█▉        | 73000/382000 [04:00<16:46, 307.12 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [04:01<16:08, 315.87 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [04:01<16:23, 311.06 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:02<16:25, 309.61 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:02<16:11, 313.87 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [04:02<15:52, 322.28 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [04:02<16:33, 310.04 examples/s]
Running tokenizer on dataset:  19%|█▉        | 74000/382000 [04:03<16:43, 307.05 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:04<16:10, 314.22 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:04<16:23, 309.97 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:05<16:12, 312.58 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:05<16:05, 314.86 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [04:05<16:00, 318.61 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [04:05<16:23, 312.09 examples/s]
Running tokenizer on dataset:  20%|█▉        | 75000/382000 [04:06<16:54, 302.65 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:07<16:01, 316.01 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:08<16:11, 312.76 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:08<16:07, 313.19 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:08<15:56, 316.88 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:09<16:03, 316.64 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [04:09<16:33, 308.09 examples/s]
Running tokenizer on dataset:  20%|█▉        | 76000/382000 [04:10<16:55, 301.28 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:10<15:54, 317.49 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:11<16:03, 314.59 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:11<16:04, 313.20 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:12<16:01, 314.24 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:12<16:02, 315.75 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:12<16:22, 310.32 examples/s]
Running tokenizer on dataset:  20%|██        | 77000/382000 [04:13<17:15, 294.55 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:14<15:58, 315.02 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:14<16:08, 311.97 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:15<16:00, 313.45 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:15<16:00, 313.29 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:15<15:53, 317.83 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:15<16:12, 312.58 examples/s]
Running tokenizer on dataset:  20%|██        | 78000/382000 [04:17<17:13, 294.22 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:17<15:54, 315.24 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:17<15:59, 313.82 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:18<16:03, 311.37 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:18<15:56, 313.68 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:18<15:58, 314.91 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:18<16:01, 315.00 examples/s]
Running tokenizer on dataset:  21%|██        | 79000/382000 [04:20<16:55, 298.28 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:20<15:55, 314.11 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:20<15:56, 313.51 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:21<15:44, 316.70 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:21<15:51, 314.13 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:21<16:00, 313.46 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:21<16:12, 310.62 examples/s]
Running tokenizer on dataset:  21%|██        | 80000/382000 [04:23<16:49, 299.10 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:23<15:45, 316.18 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:23<15:43, 317.05 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:24<15:34, 318.96 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:24<15:46, 314.79 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:24<15:50, 315.60 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:25<16:16, 308.14 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:26<15:37, 317.98 examples/s]
Running tokenizer on dataset:  21%|██        | 81000/382000 [04:26<16:42, 300.11 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:27<15:35, 318.50 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:27<15:31, 318.81 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:27<15:41, 315.49 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:28<15:41, 317.41 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:28<16:07, 309.95 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:29<15:34, 317.92 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:30<15:34, 317.81 examples/s]
Running tokenizer on dataset:  21%|██▏       | 82000/382000 [04:30<16:36, 301.15 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:30<15:30, 317.97 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:31<15:40, 316.94 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:31<15:57, 309.04 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:31<15:57, 312.31 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:33<15:36, 316.19 examples/s]
Running tokenizer on dataset:  22%|██▏       | 83000/382000 [04:33<16:17, 305.79 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:33<15:34, 316.78 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:34<15:28, 317.75 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:34<15:37, 316.85 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:34<15:51, 309.98 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:34<16:06, 308.43 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:36<15:45, 312.07 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:36<15:27, 318.16 examples/s]
Running tokenizer on dataset:  22%|██▏       | 84000/382000 [04:36<16:11, 306.90 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:37<15:12, 322.15 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:37<15:31, 315.53 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:37<15:35, 316.36 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:38<16:15, 304.44 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:39<15:31, 315.69 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:39<15:20, 319.47 examples/s]
Running tokenizer on dataset:  22%|██▏       | 85000/382000 [04:39<15:52, 311.82 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:40<15:11, 321.54 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:40<15:34, 315.79 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:40<15:40, 311.64 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:41<16:18, 302.55 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:42<15:31, 314.54 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:42<15:20, 318.45 examples/s]
Running tokenizer on dataset:  23%|██▎       | 86000/382000 [04:42<15:51, 310.93 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:43<15:15, 318.87 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:43<15:28, 316.53 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:44<15:38, 311.27 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:44<16:16, 302.13 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:45<15:29, 314.09 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:46<15:28, 314.57 examples/s]
Running tokenizer on dataset:  23%|██▎       | 87000/382000 [04:46<15:47, 311.22 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:46<15:09, 319.92 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:46<15:23, 317.37 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:47<15:51, 305.87 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:47<15:45, 310.85 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:49<15:21, 315.65 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:49<15:19, 316.56 examples/s]
Running tokenizer on dataset:  23%|██▎       | 88000/382000 [04:49<15:35, 314.21 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:49<15:10, 318.61 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:50<15:28, 314.60 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:50<15:49, 305.57 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:51<15:47, 309.32 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:52<15:19, 315.38 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:52<15:18, 315.57 examples/s]
Running tokenizer on dataset:  23%|██▎       | 89000/382000 [04:52<15:47, 309.32 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:52<15:04, 319.61 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:53<15:18, 316.78 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:53<15:40, 307.31 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:54<15:56, 305.24 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:55<15:09, 317.70 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:55<15:10, 317.49 examples/s]
Running tokenizer on dataset:  24%|██▎       | 90000/382000 [04:55<15:42, 309.70 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [04:55<15:03, 318.78 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [04:56<15:11, 318.16 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [04:57<15:39, 306.52 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:57<15:44, 308.03 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [04:58<15:13, 315.27 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [04:58<15:10, 316.24 examples/s]
Running tokenizer on dataset:  24%|██▍       | 91000/382000 [04:59<15:38, 310.13 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [04:59<15:05, 316.87 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [04:59<15:09, 317.80 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:00<15:32, 307.87 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [05:01<15:53, 304.04 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:01<15:11, 314.77 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:01<15:07, 316.23 examples/s]
Running tokenizer on dataset:  24%|██▍       | 92000/382000 [05:02<15:43, 307.38 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:02<15:08, 314.66 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [05:02<15:10, 316.25 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:03<15:32, 306.79 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [05:04<15:46, 305.20 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:05<15:15, 312.23 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:05<15:14, 312.87 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:05<15:06, 314.40 examples/s]
Running tokenizer on dataset:  24%|██▍       | 93000/382000 [05:05<15:41, 306.90 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:05<15:05, 316.84 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:06<15:20, 309.61 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [05:07<15:47, 303.84 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:08<15:16, 311.08 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:08<15:10, 313.10 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:08<14:58, 316.15 examples/s]
Running tokenizer on dataset:  25%|██▍       | 94000/382000 [05:08<15:45, 304.53 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:09<15:05, 315.71 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:10<15:10, 311.86 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:11<15:43, 304.08 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:11<15:01, 314.88 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:11<15:15, 310.18 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:11<14:52, 317.00 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:12<15:04, 314.95 examples/s]
Running tokenizer on dataset:  25%|██▍       | 95000/382000 [05:12<15:53, 300.90 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:13<15:04, 312.75 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:14<15:45, 302.58 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:14<14:54, 316.33 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:14<15:17, 308.42 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:14<14:45, 318.50 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:15<14:56, 316.68 examples/s]
Running tokenizer on dataset:  25%|██▌       | 96000/382000 [05:15<15:56, 299.10 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:16<15:01, 312.75 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:17<14:48, 317.52 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:17<15:40, 302.91 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:17<15:04, 311.70 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:18<14:37, 320.38 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:18<14:50, 317.73 examples/s]
Running tokenizer on dataset:  25%|██▌       | 97000/382000 [05:19<15:53, 299.02 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:19<14:55, 313.62 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:20<14:45, 317.50 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:20<15:37, 302.88 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:21<14:43, 316.99 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:21<15:13, 307.58 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:21<14:49, 317.19 examples/s]
Running tokenizer on dataset:  26%|██▌       | 98000/382000 [05:22<15:45, 300.29 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:22<14:57, 311.91 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:23<14:46, 315.90 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:24<15:21, 307.14 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:24<14:47, 314.22 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:24<15:10, 307.51 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:24<14:38, 319.85 examples/s]
Running tokenizer on dataset:  26%|██▌       | 99000/382000 [05:25<15:35, 302.39 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:26<14:59, 310.13 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:27<14:54, 312.06 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:27<15:13, 308.56 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:27<14:38, 316.53 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:27<15:11, 305.97 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:27<14:42, 317.38 examples/s]
Running tokenizer on dataset:  26%|██▌       | 100000/382000 [05:28<15:24, 304.97 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:29<14:47, 313.39 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:30<14:49, 312.42 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:30<15:14, 307.20 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:30<14:28, 319.04 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:31<15:06, 306.64 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:31<14:47, 314.38 examples/s]
Running tokenizer on dataset:  26%|██▋       | 101000/382000 [05:32<15:08, 309.20 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:32<14:36, 316.10 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:33<14:41, 314.26 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:33<14:19, 321.10 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:33<15:13, 306.50 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:34<14:57, 308.70 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:34<14:43, 314.71 examples/s]
Running tokenizer on dataset:  27%|██▋       | 102000/382000 [05:35<15:12, 306.68 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:35<14:30, 317.05 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:36<14:28, 317.67 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:37<15:08, 307.14 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:37<14:44, 310.89 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:37<14:50, 309.90 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:37<14:34, 316.80 examples/s]
Running tokenizer on dataset:  27%|██▋       | 103000/382000 [05:38<15:17, 304.07 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:39<15:05, 303.77 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:40<14:52, 308.25 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:40<14:29, 315.18 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:40<14:55, 310.42 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:40<14:25, 318.73 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:40<15:07, 302.99 examples/s]
Running tokenizer on dataset:  27%|██▋       | 104000/382000 [05:41<15:13, 304.23 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:42<14:50, 307.85 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:43<14:47, 308.81 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:43<14:50, 310.94 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:43<14:31, 313.12 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:44<14:54, 306.15 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:44<14:53, 307.81 examples/s]
Running tokenizer on dataset:  27%|██▋       | 105000/382000 [05:45<15:08, 305.03 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:45<14:42, 309.30 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:46<14:17, 317.18 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:46<14:39, 313.78 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:46<14:45, 308.31 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:47<14:37, 312.38 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:47<14:46, 307.90 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:48<14:27, 313.62 examples/s]
Running tokenizer on dataset:  28%|██▊       | 106000/382000 [05:48<15:04, 305.18 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:49<14:27, 313.72 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:49<14:26, 312.69 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:50<14:56, 306.81 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:50<14:28, 314.49 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:50<14:32, 311.91 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:51<14:39, 308.18 examples/s]
Running tokenizer on dataset:  28%|██▊       | 107000/382000 [05:52<15:25, 297.20 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:53<14:33, 310.23 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [05:53<14:19, 314.12 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:53<14:45, 309.40 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:53<14:18, 316.99 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:53<14:33, 310.14 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [05:55<14:41, 306.14 examples/s]
Running tokenizer on dataset:  28%|██▊       | 108000/382000 [05:55<15:07, 301.97 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [05:56<14:24, 312.18 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [05:56<14:15, 314.33 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:56<14:34, 312.01 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [05:56<14:28, 311.96 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [05:56<14:27, 311.17 examples/s]
Running tokenizer on dataset:  29%|██▊       | 109000/382000 [05:58<14:50, 306.52 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [05:58<14:54, 300.59 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [05:59<14:17, 313.72 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [05:59<14:12, 314.33 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [05:59<14:24, 314.70 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [06:00<14:20, 312.78 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [06:00<14:34, 308.70 examples/s]
Running tokenizer on dataset:  29%|██▉       | 110000/382000 [06:01<14:39, 309.11 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:01<14:45, 302.63 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:02<14:06, 315.58 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:02<14:19, 311.89 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [06:02<14:28, 312.14 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:03<14:19, 311.64 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [06:03<14:48, 302.84 examples/s]
Running tokenizer on dataset:  29%|██▉       | 111000/382000 [06:04<14:46, 305.78 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:05<14:46, 301.04 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:05<13:56, 318.09 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:05<14:07, 315.19 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [06:05<14:31, 309.72 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:06<14:10, 313.77 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:06<14:50, 301.01 examples/s]
Running tokenizer on dataset:  29%|██▉       | 112000/382000 [06:08<14:43, 305.63 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:08<14:35, 303.66 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:08<13:53, 317.79 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:08<13:59, 316.82 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [06:09<14:30, 309.05 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:09<14:05, 314.72 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:10<14:36, 304.60 examples/s]
Running tokenizer on dataset:  30%|██▉       | 113000/382000 [06:11<14:42, 304.67 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:11<14:28, 305.10 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:11<13:45, 319.73 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:11<14:01, 314.94 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:12<14:32, 307.12 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:12<14:04, 313.96 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:13<14:29, 305.96 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:14<14:16, 308.30 examples/s]
Running tokenizer on dataset:  30%|██▉       | 114000/382000 [06:14<14:49, 301.23 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:15<13:46, 318.22 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:15<13:53, 316.74 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:15<14:29, 306.98 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:15<13:54, 316.19 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:16<14:28, 304.96 examples/s]
Running tokenizer on dataset:  30%|███       | 115000/382000 [06:18<14:36, 304.69 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:18<13:40, 319.12 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:18<14:15, 307.43 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:18<13:50, 316.76 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:19<14:23, 308.09 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:19<13:54, 314.97 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:19<14:18, 307.41 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:21<13:41, 317.59 examples/s]
Running tokenizer on dataset:  30%|███       | 116000/382000 [06:21<14:27, 306.49 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:21<13:43, 318.00 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:21<14:13, 307.06 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:22<13:50, 315.42 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:22<14:18, 308.55 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:23<14:30, 302.21 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:24<13:42, 316.09 examples/s]
Running tokenizer on dataset:  31%|███       | 117000/382000 [06:24<14:20, 307.88 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:24<13:46, 315.76 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:24<14:12, 306.24 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:25<14:08, 311.28 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:25<13:50, 314.18 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:26<14:17, 305.44 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:27<13:39, 316.23 examples/s]
Running tokenizer on dataset:  31%|███       | 118000/382000 [06:27<14:07, 311.54 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:27<13:44, 315.26 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:28<14:14, 304.10 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:28<14:05, 310.93 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:28<13:48, 313.77 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:29<14:19, 303.50 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:30<13:32, 317.45 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:30<13:42, 314.78 examples/s]
Running tokenizer on dataset:  31%|███       | 119000/382000 [06:31<14:18, 306.27 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:31<14:08, 305.30 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:31<14:01, 311.42 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:31<13:49, 312.40 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:33<14:13, 304.80 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:33<13:28, 318.04 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:34<13:37, 315.51 examples/s]
Running tokenizer on dataset:  31%|███▏      | 120000/382000 [06:34<14:13, 307.00 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:34<14:03, 306.05 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:35<13:43, 313.29 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:35<14:14, 305.32 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:36<14:23, 300.01 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:37<13:23, 318.78 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:37<13:28, 318.03 examples/s]
Running tokenizer on dataset:  32%|███▏      | 121000/382000 [06:37<14:17, 304.38 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:37<13:52, 308.83 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:38<13:35, 315.30 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:38<14:07, 306.68 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:39<14:15, 301.45 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:40<13:20, 318.38 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:40<13:27, 317.05 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:41<13:55, 306.22 examples/s]
Running tokenizer on dataset:  32%|███▏      | 122000/382000 [06:41<14:26, 300.02 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:41<13:31, 315.36 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:41<14:09, 304.89 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:43<14:20, 298.77 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:43<13:20, 317.38 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:43<13:26, 316.36 examples/s]
Running tokenizer on dataset:  32%|███▏      | 123000/382000 [06:44<14:18, 301.59 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:44<14:01, 303.10 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:44<13:31, 314.39 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:44<13:57, 308.18 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:46<14:04, 302.97 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:46<13:18, 318.14 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:46<13:32, 311.57 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:47<13:48, 306.75 examples/s]
Running tokenizer on dataset:  32%|███▏      | 124000/382000 [06:47<14:12, 302.73 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:47<13:22, 316.39 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:48<13:44, 311.65 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:49<14:02, 302.65 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:49<13:16, 317.47 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:49<13:30, 310.81 examples/s]
Running tokenizer on dataset:  33%|███▎      | 125000/382000 [06:50<13:57, 306.98 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:50<13:43, 307.08 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:50<13:24, 314.61 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:51<13:47, 309.41 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:52<13:12, 317.90 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [06:53<13:24, 311.84 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:53<14:09, 298.91 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:54<13:35, 308.88 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:54<13:19, 315.19 examples/s]
Running tokenizer on dataset:  33%|███▎      | 126000/382000 [06:54<13:57, 305.79 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:54<13:41, 310.48 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [06:56<13:07, 318.89 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [06:56<13:23, 311.20 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [06:56<14:02, 300.26 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [06:57<13:23, 312.22 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [06:57<13:10, 317.51 examples/s]
Running tokenizer on dataset:  33%|███▎      | 127000/382000 [06:57<13:56, 304.67 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [06:57<13:43, 308.25 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [06:59<13:06, 317.75 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [06:59<13:09, 315.22 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [06:59<13:50, 303.43 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [07:00<13:12, 315.60 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [07:00<13:25, 310.26 examples/s]
Running tokenizer on dataset:  34%|███▎      | 128000/382000 [07:00<13:43, 308.47 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [07:01<13:48, 305.49 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:02<13:02, 318.09 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:02<13:10, 313.78 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [07:02<13:34, 307.98 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:03<13:12, 314.19 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:03<13:07, 316.21 examples/s]
Running tokenizer on dataset:  34%|███▍      | 129000/382000 [07:03<13:51, 304.21 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [07:04<13:38, 307.81 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:05<13:00, 317.73 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:05<13:03, 315.26 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [07:06<13:46, 302.44 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:06<13:08, 314.50 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:06<13:07, 314.73 examples/s]
Running tokenizer on dataset:  34%|███▍      | 130000/382000 [07:07<13:37, 308.20 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [07:07<13:28, 310.63 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:08<12:54, 319.09 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:08<12:58, 315.97 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:09<13:39, 303.94 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:09<13:02, 315.47 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:09<13:02, 315.56 examples/s]
Running tokenizer on dataset:  34%|███▍      | 131000/382000 [07:10<13:35, 307.80 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [07:10<13:35, 306.72 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:11<12:48, 320.10 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:12<13:05, 312.02 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:12<13:42, 301.52 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:12<12:56, 316.80 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:13<13:02, 314.30 examples/s]
Running tokenizer on dataset:  35%|███▍      | 132000/382000 [07:13<13:42, 303.87 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:14<13:29, 307.75 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:14<12:49, 318.43 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:15<13:03, 311.46 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:16<13:30, 304.89 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:16<12:53, 316.74 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:16<12:58, 314.89 examples/s]
Running tokenizer on dataset:  35%|███▍      | 133000/382000 [07:16<13:32, 306.50 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:17<13:32, 305.15 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:18<12:51, 316.40 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:18<13:05, 309.41 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:19<13:14, 309.58 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:19<12:54, 315.16 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:19<12:59, 313.06 examples/s]
Running tokenizer on dataset:  35%|███▌      | 134000/382000 [07:20<13:40, 302.22 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:20<13:32, 303.82 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:21<12:52, 314.72 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:21<12:56, 311.76 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:22<12:49, 315.88 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:22<12:51, 315.01 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:22<13:26, 303.83 examples/s]
Running tokenizer on dataset:  35%|███▌      | 135000/382000 [07:23<13:35, 302.88 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:24<13:27, 304.48 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:24<12:53, 312.79 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:25<12:56, 310.51 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:25<12:46, 315.88 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:25<12:51, 313.55 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:25<13:27, 302.33 examples/s]
Running tokenizer on dataset:  36%|███▌      | 136000/382000 [07:26<13:20, 307.33 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:27<13:19, 306.39 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:27<12:45, 314.63 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:28<12:47, 312.66 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:28<12:42, 316.23 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:28<12:46, 314.28 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:29<13:28, 300.69 examples/s]
Running tokenizer on dataset:  36%|███▌      | 137000/382000 [07:29<13:12, 309.06 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:30<13:30, 301.04 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:30<12:44, 313.87 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:31<12:54, 308.39 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:32<12:43, 314.35 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:32<12:44, 314.07 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:32<13:09, 306.68 examples/s]
Running tokenizer on dataset:  36%|███▌      | 138000/382000 [07:33<13:21, 304.43 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:33<13:16, 305.09 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:34<12:44, 312.55 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:34<12:45, 311.04 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:35<12:43, 313.05 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:35<12:44, 312.74 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:35<13:14, 303.22 examples/s]
Running tokenizer on dataset:  36%|███▋      | 139000/382000 [07:36<13:08, 308.33 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:37<13:07, 307.30 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:37<12:34, 315.28 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:38<12:42, 310.78 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:38<12:37, 314.34 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:38<12:37, 314.38 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:39<13:10, 303.56 examples/s]
Running tokenizer on dataset:  37%|███▋      | 140000/382000 [07:39<13:02, 309.07 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:40<12:28, 316.56 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:40<13:06, 306.50 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:41<12:44, 308.75 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:41<12:32, 315.10 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:41<12:30, 315.78 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:42<13:23, 297.32 examples/s]
Running tokenizer on dataset:  37%|███▋      | 141000/382000 [07:42<12:52, 312.15 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:43<12:33, 313.13 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:43<13:07, 304.94 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:44<12:37, 310.29 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:44<12:35, 312.43 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:44<12:34, 312.88 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:45<13:14, 299.70 examples/s]
Running tokenizer on dataset:  37%|███▋      | 142000/382000 [07:46<12:46, 313.15 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:46<12:27, 314.57 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:46<13:02, 305.52 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:47<12:24, 314.41 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:48<12:29, 313.48 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:48<12:26, 314.76 examples/s]
Running tokenizer on dataset:  37%|███▋      | 143000/382000 [07:49<12:38, 315.05 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:49<13:16, 297.66 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:49<12:18, 317.02 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:50<12:55, 306.74 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [07:50<12:26, 312.24 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:51<12:22, 315.34 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:51<12:19, 316.34 examples/s]
Running tokenizer on dataset:  38%|███▊      | 144000/382000 [07:52<12:37, 313.99 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:52<13:12, 297.63 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [07:53<12:17, 315.84 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:53<12:59, 304.21 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [07:53<12:16, 315.08 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [07:54<12:21, 314.33 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [07:54<12:25, 312.57 examples/s]
Running tokenizer on dataset:  38%|███▊      | 145000/382000 [07:55<12:38, 312.46 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [07:55<12:59, 301.30 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [07:56<12:14, 316.00 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:57<13:11, 298.24 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [07:57<12:13, 314.80 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [07:57<12:08, 318.45 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [07:57<12:12, 316.58 examples/s]
Running tokenizer on dataset:  38%|███▊      | 146000/382000 [07:58<12:44, 308.81 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [07:59<12:43, 306.31 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [07:59<12:15, 314.20 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [08:00<13:01, 300.80 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:00<12:17, 311.92 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [08:00<12:06, 317.86 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [08:00<12:12, 315.41 examples/s]
Running tokenizer on dataset:  38%|███▊      | 147000/382000 [08:02<12:38, 309.97 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [08:02<12:41, 306.00 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:02<12:14, 313.28 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [08:03<12:50, 303.71 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:03<12:03, 317.70 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:03<12:26, 306.85 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:03<12:09, 315.29 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [08:05<12:23, 311.97 examples/s]
Running tokenizer on dataset:  39%|███▊      | 148000/382000 [08:05<12:35, 309.63 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:05<12:11, 312.88 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:06<12:16, 309.54 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [08:06<12:54, 300.88 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:06<12:04, 315.89 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:07<12:08, 314.30 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [08:08<12:18, 312.93 examples/s]
Running tokenizer on dataset:  39%|███▉      | 149000/382000 [08:08<12:38, 307.33 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:09<12:06, 313.87 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [08:09<12:31, 308.58 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:10<12:03, 313.65 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:10<11:57, 317.66 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:10<11:59, 316.90 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:11<12:18, 311.32 examples/s]
Running tokenizer on dataset:  39%|███▉      | 150000/382000 [08:11<12:32, 308.29 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:12<12:06, 312.28 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:13<11:49, 319.79 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:13<11:59, 314.27 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [08:13<12:27, 308.88 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:13<11:52, 318.50 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:15<12:21, 308.99 examples/s]
Running tokenizer on dataset:  40%|███▉      | 151000/382000 [08:15<12:27, 309.23 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:15<12:07, 310.75 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:16<11:39, 321.68 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:16<11:44, 320.80 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:16<12:15, 312.60 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:16<11:48, 319.05 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:18<12:04, 314.50 examples/s]
Running tokenizer on dataset:  40%|███▉      | 152000/382000 [08:18<12:33, 305.20 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:18<11:53, 315.25 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:19<11:29, 326.24 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:19<11:43, 318.38 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:19<11:34, 323.91 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:19<12:10, 313.65 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:21<12:10, 310.87 examples/s]
Running tokenizer on dataset:  40%|████      | 153000/382000 [08:21<12:24, 307.43 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:21<11:58, 311.78 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:22<11:34, 322.65 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:22<11:39, 320.39 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:22<11:47, 314.99 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:22<12:11, 311.57 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:24<12:02, 312.99 examples/s]
Running tokenizer on dataset:  40%|████      | 154000/382000 [08:25<12:27, 304.89 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:25<11:57, 310.91 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:25<11:37, 319.78 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:25<11:39, 318.61 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:25<11:47, 313.59 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:25<12:00, 315.00 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:27<11:50, 316.87 examples/s]
Running tokenizer on dataset:  41%|████      | 155000/382000 [08:28<12:17, 307.64 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:28<11:47, 313.76 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:28<11:33, 319.97 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:28<11:34, 319.45 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:28<11:55, 315.98 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:29<11:51, 310.81 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:30<11:45, 317.57 examples/s]
Running tokenizer on dataset:  41%|████      | 156000/382000 [08:31<12:09, 309.69 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:31<11:45, 313.45 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:31<11:30, 319.94 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:32<11:34, 318.37 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:32<11:48, 317.38 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:32<11:40, 314.22 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:33<11:46, 315.68 examples/s]
Running tokenizer on dataset:  41%|████      | 157000/382000 [08:34<11:55, 314.64 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:34<11:47, 311.09 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:34<11:20, 323.45 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:35<11:26, 320.57 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:35<11:47, 316.42 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:35<11:47, 309.32 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:37<11:40, 316.90 examples/s]
Running tokenizer on dataset:  41%|████▏     | 158000/382000 [08:37<11:50, 315.46 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:37<11:21, 321.41 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:38<11:57, 305.26 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:38<11:27, 318.52 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:38<11:41, 317.83 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:38<11:40, 311.02 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:40<11:38, 316.29 examples/s]
Running tokenizer on dataset:  42%|████▏     | 159000/382000 [08:40<11:54, 312.15 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:41<11:21, 319.77 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:41<11:26, 317.57 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:41<11:56, 304.10 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:41<11:37, 318.27 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:42<11:45, 307.49 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:43<11:31, 318.30 examples/s]
Running tokenizer on dataset:  42%|████▏     | 160000/382000 [08:44<11:51, 312.09 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:44<11:23, 317.49 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:44<11:24, 316.83 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:44<11:51, 305.20 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:44<11:45, 313.25 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [08:45<11:36, 310.13 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:46<11:36, 314.53 examples/s]
Running tokenizer on dataset:  42%|████▏     | 161000/382000 [08:47<11:39, 316.01 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [08:47<11:13, 320.48 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [08:47<11:17, 318.61 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [08:47<11:35, 310.36 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:47<11:32, 317.74 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:48<11:40, 307.06 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:49<11:39, 311.48 examples/s]
Running tokenizer on dataset:  42%|████▏     | 162000/382000 [08:50<11:28, 319.36 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:50<11:11, 320.06 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:50<11:17, 317.36 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:51<11:30, 311.20 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:51<11:32, 316.05 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [08:51<11:30, 310.01 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:53<11:37, 311.11 examples/s]
Running tokenizer on dataset:  43%|████▎     | 163000/382000 [08:53<11:30, 317.05 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [08:53<11:05, 321.72 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [08:53<11:11, 318.53 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [08:54<11:19, 314.96 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:54<11:26, 317.75 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [08:55<11:33, 307.27 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [08:56<11:26, 314.47 examples/s]
Running tokenizer on dataset:  43%|████▎     | 164000/382000 [08:56<11:43, 310.01 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [08:56<11:12, 316.57 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [08:57<11:23, 311.84 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [08:57<11:26, 310.16 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [08:57<11:33, 313.12 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [08:58<11:24, 309.73 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [08:59<11:32, 310.62 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [08:59<11:04, 319.01 examples/s]
Running tokenizer on dataset:  43%|████▎     | 165000/382000 [09:00<11:38, 310.57 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [09:00<11:14, 314.43 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [09:00<11:21, 316.87 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [09:00<11:20, 311.32 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:01<11:22, 309.31 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [09:02<11:23, 313.00 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:03<11:06, 316.39 examples/s]
Running tokenizer on dataset:  43%|████▎     | 166000/382000 [09:03<11:37, 309.61 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [09:03<11:17, 317.24 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:03<11:16, 311.99 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:03<11:19, 310.37 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:04<11:19, 309.19 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [09:06<11:33, 307.27 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:06<11:04, 316.23 examples/s]
Running tokenizer on dataset:  44%|████▎     | 167000/382000 [09:06<11:40, 307.01 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [09:06<11:10, 319.24 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:07<11:17, 309.96 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:07<11:17, 309.76 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:07<11:15, 309.51 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [09:09<11:24, 309.54 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:09<11:01, 316.18 examples/s]
Running tokenizer on dataset:  44%|████▍     | 168000/382000 [09:09<11:27, 311.19 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [09:10<11:13, 316.45 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:10<11:12, 310.91 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:10<11:16, 309.01 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:11<11:05, 312.45 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:12<11:22, 309.16 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:12<10:58, 315.71 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [09:13<11:00, 321.14 examples/s]
Running tokenizer on dataset:  44%|████▍     | 169000/382000 [09:13<11:40, 303.88 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:13<11:09, 310.66 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:13<11:10, 310.08 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:14<11:03, 312.15 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:15<11:23, 307.15 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:15<10:54, 316.06 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:16<11:01, 319.10 examples/s]
Running tokenizer on dataset:  45%|████▍     | 170000/382000 [09:16<11:31, 306.60 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:16<11:02, 312.44 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:16<11:09, 309.10 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:17<11:08, 308.31 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:19<10:55, 314.33 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:19<11:19, 307.54 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:19<10:56, 320.04 examples/s]
Running tokenizer on dataset:  45%|████▍     | 171000/382000 [09:19<11:37, 302.59 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:19<10:58, 313.00 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:20<11:07, 308.71 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:20<10:54, 313.05 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:22<10:44, 318.03 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:22<11:24, 304.05 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:22<10:54, 319.40 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:22<10:48, 316.10 examples/s]
Running tokenizer on dataset:  45%|████▌     | 172000/382000 [09:23<11:29, 304.35 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:23<11:01, 309.97 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:24<11:01, 308.28 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:25<10:49, 314.28 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:25<10:49, 320.33 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:25<11:25, 301.81 examples/s]
Running tokenizer on dataset:  45%|████▌     | 173000/382000 [09:26<11:18, 307.84 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:26<10:54, 311.92 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:26<11:03, 307.42 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:27<10:52, 311.25 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:28<10:42, 316.06 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:28<10:48, 319.09 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:29<11:19, 303.31 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:29<10:47, 313.70 examples/s]
Running tokenizer on dataset:  46%|████▌     | 174000/382000 [09:29<11:13, 309.05 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:29<10:52, 311.13 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:30<10:45, 313.06 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:31<10:35, 317.95 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:31<10:44, 319.43 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:32<11:13, 304.48 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:32<10:38, 316.58 examples/s]
Running tokenizer on dataset:  46%|████▌     | 175000/382000 [09:32<11:08, 309.88 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:32<10:45, 312.92 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:33<10:35, 316.07 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:34<10:32, 317.75 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:35<10:43, 318.43 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:35<10:34, 317.00 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:35<11:17, 301.13 examples/s]
Running tokenizer on dataset:  46%|████▌     | 176000/382000 [09:35<11:14, 305.61 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:35<10:41, 313.18 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:36<10:36, 314.27 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:37<10:27, 318.97 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:38<10:47, 314.91 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:38<10:28, 318.01 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:38<11:04, 305.49 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:39<10:35, 314.90 examples/s]
Running tokenizer on dataset:  46%|████▋     | 177000/382000 [09:39<11:04, 308.42 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:39<10:32, 314.60 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:40<10:22, 319.66 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:41<10:48, 313.26 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:41<10:26, 317.46 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:42<11:04, 304.11 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:42<10:33, 314.25 examples/s]
Running tokenizer on dataset:  47%|████▋     | 178000/382000 [09:42<11:05, 306.60 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:43<10:32, 313.00 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:44<10:24, 317.11 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:44<10:39, 315.65 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:45<10:26, 315.82 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:45<11:02, 303.57 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:45<10:33, 312.76 examples/s]
Running tokenizer on dataset:  47%|████▋     | 179000/382000 [09:45<10:52, 310.97 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [09:46<10:28, 313.57 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [09:47<10:22, 316.40 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:47<10:32, 317.83 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [09:48<10:21, 316.95 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:48<10:42, 311.09 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [09:48<10:27, 313.87 examples/s]
Running tokenizer on dataset:  47%|████▋     | 180000/382000 [09:48<10:58, 306.60 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [09:49<10:22, 314.64 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [09:50<10:22, 315.02 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:50<10:25, 319.58 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [09:51<10:20, 315.75 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:51<10:35, 313.13 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [09:51<10:24, 313.96 examples/s]
Running tokenizer on dataset:  47%|████▋     | 181000/382000 [09:52<11:04, 302.38 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [09:52<10:15, 316.68 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [09:53<10:20, 314.01 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:53<10:22, 319.71 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [09:54<10:19, 314.90 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:55<10:41, 308.57 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [09:55<10:26, 311.48 examples/s]
Running tokenizer on dataset:  48%|████▊     | 182000/382000 [09:55<10:38, 313.33 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [09:55<10:18, 313.52 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [09:56<10:16, 314.55 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [09:57<10:20, 319.23 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [09:57<10:17, 314.07 examples/s]
Running tokenizer on dataset:  48%|████▊     | 183000/382000 [09:58<10:28, 316.45 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [09:58<10:26, 309.58 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [09:58<10:49, 303.26 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [09:58<10:14, 314.00 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:00<10:10, 316.29 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [10:00<10:14, 320.44 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:00<10:13, 314.80 examples/s]
Running tokenizer on dataset:  48%|████▊     | 184000/382000 [10:01<10:25, 316.67 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:01<10:17, 312.74 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [10:01<10:47, 302.59 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:01<10:04, 317.56 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:03<10:04, 317.48 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [10:03<10:13, 319.46 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:04<10:05, 317.05 examples/s]
Running tokenizer on dataset:  48%|████▊     | 185000/382000 [10:04<10:22, 316.49 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:04<10:12, 313.55 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:04<09:54, 321.34 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [10:05<11:01, 294.84 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:06<10:00, 318.05 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [10:06<10:18, 315.53 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:07<10:00, 318.24 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:07<10:03, 316.55 examples/s]
Running tokenizer on dataset:  49%|████▊     | 186000/382000 [10:07<10:21, 315.56 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:08<09:48, 322.71 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [10:08<10:46, 299.97 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:09<09:55, 319.06 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [10:09<10:13, 316.18 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:10<09:54, 319.47 examples/s]
Running tokenizer on dataset:  49%|████▉     | 187000/382000 [10:10<10:13, 317.74 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:10<10:00, 316.66 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:11<09:44, 323.36 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:12<10:48, 297.65 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:12<09:52, 319.20 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:12<10:07, 317.84 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:13<09:55, 317.54 examples/s]
Running tokenizer on dataset:  49%|████▉     | 188000/382000 [10:14<10:07, 319.34 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:14<09:56, 316.89 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:14<09:38, 325.06 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:15<10:36, 301.44 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:15<09:47, 320.14 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:15<09:59, 320.43 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:16<09:50, 318.61 examples/s]
Running tokenizer on dataset:  49%|████▉     | 189000/382000 [10:17<10:01, 320.83 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:17<09:34, 325.42 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:17<09:52, 317.04 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:18<10:34, 301.04 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:18<09:47, 318.56 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:18<09:54, 321.43 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:19<09:49, 317.08 examples/s]
Running tokenizer on dataset:  50%|████▉     | 190000/382000 [10:20<09:55, 322.19 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:20<09:29, 326.48 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:20<09:54, 314.78 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:21<10:25, 303.98 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:21<09:42, 319.41 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:22<09:48, 323.03 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:22<09:43, 318.51 examples/s]
Running tokenizer on dataset:  50%|█████     | 191000/382000 [10:23<09:49, 324.21 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:23<09:25, 326.98 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:23<09:48, 316.16 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:24<10:10, 309.34 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:25<09:38, 319.77 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:25<09:46, 322.43 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:26<09:39, 319.02 examples/s]
Running tokenizer on dataset:  50%|█████     | 192000/382000 [10:26<09:42, 326.04 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:26<09:27, 324.50 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:26<09:43, 316.85 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:28<10:03, 311.74 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:28<09:41, 323.08 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:28<09:38, 318.25 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:29<09:40, 317.23 examples/s]
Running tokenizer on dataset:  51%|█████     | 193000/382000 [10:29<09:39, 326.04 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:29<09:14, 329.86 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:30<09:44, 314.89 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:31<09:25, 323.61 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:31<10:00, 311.22 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:31<09:38, 323.17 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:32<09:27, 322.40 examples/s]
Running tokenizer on dataset:  51%|█████     | 194000/382000 [10:32<09:40, 323.96 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:32<09:15, 327.40 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:33<09:44, 313.25 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:34<09:27, 320.99 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:34<09:33, 324.11 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:34<10:02, 308.64 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:35<09:28, 320.14 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:35<09:11, 328.38 examples/s]
Running tokenizer on dataset:  51%|█████     | 195000/382000 [10:35<09:39, 322.71 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:36<09:44, 311.59 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:37<09:20, 322.84 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:37<09:30, 324.01 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:37<09:50, 313.20 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:38<09:25, 320.17 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:38<09:08, 328.09 examples/s]
Running tokenizer on dataset:  51%|█████▏    | 196000/382000 [10:38<09:32, 325.13 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:39<09:33, 315.55 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:40<09:19, 321.61 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:40<09:31, 321.71 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:40<09:43, 315.20 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:41<09:06, 327.34 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:41<09:23, 319.61 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 197000/382000 [10:41<09:28, 325.56 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:42<09:29, 316.14 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:43<09:22, 325.30 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:43<09:28, 321.81 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:43<09:18, 320.33 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:44<09:03, 327.24 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 198000/382000 [10:44<09:25, 325.19 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:44<09:21, 318.81 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:45<09:25, 316.67 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:46<09:22, 323.60 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:46<09:14, 321.28 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:46<09:27, 320.95 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 199000/382000 [10:47<09:15, 329.24 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [10:47<08:59, 328.22 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:47<09:17, 319.04 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:49<09:25, 314.69 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:49<09:16, 325.27 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:49<09:19, 323.27 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [10:49<09:11, 321.10 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [10:50<08:56, 328.10 examples/s]
Running tokenizer on dataset:  52%|█████▏    | 200000/382000 [10:50<09:17, 326.18 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [10:51<09:14, 319.28 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [10:52<09:20, 315.56 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:52<09:13, 325.07 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:52<09:14, 324.42 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [10:53<09:13, 318.13 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [10:53<08:54, 327.38 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 201000/382000 [10:53<09:13, 326.89 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [10:54<09:12, 318.38 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [10:55<09:19, 314.52 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:56<09:14, 322.97 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [10:56<09:11, 324.83 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [10:56<09:10, 317.75 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 202000/382000 [10:56<09:10, 326.98 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [10:57<08:59, 322.77 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [10:57<09:09, 318.25 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [10:58<09:13, 316.13 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:59<09:09, 323.81 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [10:59<09:12, 322.22 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [10:59<09:13, 314.26 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 203000/382000 [11:00<09:07, 326.67 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:00<08:55, 322.84 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [11:00<09:11, 315.28 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [11:01<09:14, 313.85 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [11:02<09:06, 323.59 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [11:02<09:09, 322.22 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:02<09:10, 314.31 examples/s]
Running tokenizer on dataset:  53%|█████▎    | 204000/382000 [11:03<09:07, 324.87 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:03<08:55, 321.35 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:03<09:05, 317.24 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:04<09:07, 315.70 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [11:05<09:03, 323.58 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [11:05<09:06, 321.87 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:05<09:08, 313.67 examples/s]
Running tokenizer on dataset:  54%|█████▎    | 205000/382000 [11:06<09:06, 323.91 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:06<08:49, 322.94 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:06<09:03, 316.64 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:08<09:02, 316.89 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [11:08<09:00, 323.49 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [11:08<09:01, 323.14 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:09<09:04, 313.89 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 206000/382000 [11:09<09:05, 322.66 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:09<08:50, 320.75 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:10<09:00, 316.47 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:11<09:01, 316.05 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [11:11<09:03, 319.97 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [11:11<09:02, 320.54 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:12<09:03, 312.54 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 207000/382000 [11:12<09:02, 322.67 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:12<08:46, 321.08 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:13<09:00, 314.46 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:14<09:06, 311.15 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:14<08:59, 320.53 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:14<08:58, 321.32 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:15<09:00, 312.59 examples/s]
Running tokenizer on dataset:  54%|█████▍    | 208000/382000 [11:15<09:01, 321.21 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:15<08:45, 319.86 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:16<08:57, 314.68 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:17<08:59, 313.17 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:17<08:57, 319.82 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:17<08:57, 320.07 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 209000/382000 [11:18<08:57, 321.77 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:18<08:58, 311.79 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:18<08:42, 319.45 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:19<08:54, 314.10 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:20<08:52, 321.07 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:20<08:58, 312.25 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:21<08:55, 319.45 examples/s]
Running tokenizer on dataset:  55%|█████▍    | 210000/382000 [11:21<08:56, 320.80 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:21<08:51, 314.26 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:21<08:36, 321.21 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:22<08:49, 315.61 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:24<08:50, 315.06 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:24<08:50, 320.44 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:24<08:54, 317.76 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 211000/382000 [11:24<08:54, 319.78 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:25<08:47, 314.71 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:25<08:33, 321.50 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:25<08:43, 317.33 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:27<08:44, 316.49 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:27<08:46, 320.75 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:27<08:48, 319.76 examples/s]
Running tokenizer on dataset:  55%|█████▌    | 212000/382000 [11:28<08:54, 318.02 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:28<08:31, 320.76 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:28<08:47, 312.93 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:29<08:40, 317.30 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:30<08:41, 316.63 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:30<08:46, 319.02 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:30<08:47, 318.69 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 213000/382000 [11:31<08:50, 318.84 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:31<08:31, 318.93 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:31<08:47, 310.68 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:32<08:38, 316.42 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:33<08:42, 319.42 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:33<08:38, 316.37 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:33<08:43, 318.91 examples/s]
Running tokenizer on dataset:  56%|█████▌    | 214000/382000 [11:34<08:47, 318.74 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:34<08:23, 321.47 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:34<08:43, 311.16 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:35<08:36, 315.66 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:36<08:36, 321.65 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:36<08:37, 320.69 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:36<08:41, 312.69 examples/s]
Running tokenizer on dataset:  56%|█████▋    | 215000/382000 [11:37<08:43, 318.72 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:37<08:22, 320.17 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:37<08:36, 313.61 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:38<08:29, 318.04 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:39<08:32, 322.24 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:39<08:31, 316.83 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:39<08:37, 319.00 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 216000/382000 [11:40<08:39, 319.25 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:40<08:31, 312.91 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:41<08:34, 312.91 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:41<08:28, 316.81 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:42<08:32, 320.24 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:42<08:33, 319.50 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:43<08:31, 314.66 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 217000/382000 [11:43<08:35, 320.38 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:43<08:17, 319.48 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:44<08:47, 303.42 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:45<08:36, 310.01 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:45<08:33, 317.28 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:46<08:33, 317.67 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:46<08:38, 308.84 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [11:46<08:06, 324.59 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 218000/382000 [11:46<08:33, 319.62 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:47<08:36, 307.80 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:48<08:27, 313.15 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:49<08:27, 319.47 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:49<08:24, 320.81 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:49<08:26, 313.95 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [11:50<08:03, 324.53 examples/s]
Running tokenizer on dataset:  57%|█████▋    | 219000/382000 [11:50<08:31, 318.78 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [11:50<08:20, 315.71 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [11:51<08:14, 319.31 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:52<08:23, 319.62 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:52<08:23, 320.05 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [11:52<08:15, 319.05 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [11:53<08:04, 322.19 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 220000/382000 [11:53<08:25, 320.26 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [11:53<08:17, 315.42 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [11:54<08:12, 318.97 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [11:55<08:10, 320.19 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:55<08:34, 310.80 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:55<08:31, 312.65 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [11:56<07:58, 324.15 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 221000/382000 [11:56<08:22, 320.27 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [11:57<08:13, 315.99 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [11:57<08:11, 317.51 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:58<08:22, 316.68 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [11:58<08:19, 318.07 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [11:58<08:07, 320.25 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [11:59<07:56, 322.90 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 222000/382000 [11:59<08:30, 313.33 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:00<08:12, 314.89 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:00<08:08, 317.03 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [12:01<08:11, 321.34 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [12:01<08:07, 324.00 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:01<08:06, 318.59 examples/s]
Running tokenizer on dataset:  58%|█████▊    | 223000/382000 [12:02<08:18, 319.08 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:02<08:06, 314.63 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:03<08:06, 316.27 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:03<08:02, 319.12 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [12:04<08:07, 321.90 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [12:04<08:04, 323.86 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:04<08:00, 320.59 examples/s]
Running tokenizer on dataset:  59%|█████▊    | 224000/382000 [12:05<08:05, 325.24 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:05<08:04, 313.86 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:06<08:04, 315.89 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:06<07:59, 319.27 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [12:07<08:03, 322.52 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [12:07<08:06, 320.62 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:08<07:56, 321.43 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 225000/382000 [12:08<08:03, 324.73 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:09<07:59, 314.80 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:09<08:03, 314.59 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:10<07:58, 317.74 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:10<07:59, 323.21 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:10<07:59, 322.99 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:11<07:52, 321.62 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 226000/382000 [12:11<08:02, 323.36 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:12<07:55, 315.38 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:12<07:57, 316.55 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:13<07:55, 317.42 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:13<07:52, 325.63 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:14<07:56, 323.39 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:14<07:50, 321.10 examples/s]
Running tokenizer on dataset:  59%|█████▉    | 227000/382000 [12:14<07:59, 322.96 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:15<07:51, 315.74 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:16<07:57, 313.84 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:16<07:55, 315.42 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:17<07:50, 325.04 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:17<07:53, 323.21 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:17<07:48, 320.01 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 228000/382000 [12:17<07:55, 324.19 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:18<07:48, 315.76 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:19<07:55, 313.44 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:19<07:52, 315.39 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:20<07:49, 323.44 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:20<07:50, 323.08 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:20<07:47, 318.86 examples/s]
Running tokenizer on dataset:  60%|█████▉    | 229000/382000 [12:21<07:51, 324.59 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:21<07:38, 320.81 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:22<07:55, 310.97 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:22<07:50, 314.40 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:23<07:49, 321.91 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:23<07:50, 320.91 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:23<07:45, 317.95 examples/s]
Running tokenizer on dataset:  60%|██████    | 230000/382000 [12:24<07:50, 322.97 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:24<07:33, 322.14 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:25<07:47, 314.65 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:25<07:43, 317.46 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:26<07:47, 320.91 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:26<07:48, 320.04 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:26<07:36, 322.14 examples/s]
Running tokenizer on dataset:  60%|██████    | 231000/382000 [12:27<07:49, 321.90 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:27<07:25, 325.36 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:28<07:39, 317.44 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:29<07:39, 317.76 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:29<07:45, 320.43 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:29<07:47, 319.02 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:29<07:31, 323.17 examples/s]
Running tokenizer on dataset:  61%|██████    | 232000/382000 [12:30<07:47, 320.80 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:30<07:25, 323.36 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:31<07:36, 317.29 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:32<07:33, 319.43 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:32<07:42, 320.13 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:32<07:43, 319.36 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:32<07:27, 324.21 examples/s]
Running tokenizer on dataset:  61%|██████    | 233000/382000 [12:33<07:46, 319.60 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:33<07:18, 325.82 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:35<07:34, 316.49 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:35<07:33, 317.52 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:35<07:33, 324.44 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:35<07:33, 324.38 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:36<07:27, 321.77 examples/s]
Running tokenizer on dataset:  61%|██████▏   | 234000/382000 [12:36<07:42, 319.85 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:36<07:12, 328.67 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:38<07:28, 318.62 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:38<07:28, 319.16 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:38<07:29, 324.62 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:38<07:28, 325.55 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:39<07:20, 324.44 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 235000/382000 [12:39<07:32, 324.83 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:39<07:07, 329.98 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:41<07:22, 320.84 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:41<07:21, 321.45 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:41<07:25, 325.30 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:41<07:23, 327.10 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:42<07:13, 327.55 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 236000/382000 [12:42<07:27, 326.02 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:42<07:04, 329.67 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:44<07:17, 321.98 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:44<07:18, 321.80 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:44<07:25, 323.28 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:45<07:25, 323.19 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:45<07:10, 327.64 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 237000/382000 [12:45<07:23, 326.61 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [12:45<07:00, 330.52 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:47<07:13, 322.89 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:47<07:15, 321.34 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:47<07:17, 326.76 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:48<07:18, 325.87 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:48<07:07, 327.19 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [12:48<07:00, 328.02 examples/s]
Running tokenizer on dataset:  62%|██████▏   | 238000/382000 [12:48<07:25, 323.39 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [12:50<07:11, 322.49 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [12:50<07:11, 322.48 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:50<07:11, 329.08 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:51<07:11, 328.76 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [12:51<07:04, 327.53 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [12:51<06:57, 328.07 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 239000/382000 [12:52<07:20, 324.35 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [12:53<07:13, 318.17 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [12:53<07:10, 320.50 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:53<07:08, 329.24 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:54<07:07, 329.64 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [12:54<07:03, 325.66 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 240000/382000 [12:54<07:12, 327.98 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [12:55<06:57, 325.50 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [12:56<07:08, 319.87 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [12:57<07:06, 321.21 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:57<07:07, 327.59 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [12:57<07:06, 328.39 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [12:57<06:59, 326.68 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 241000/382000 [12:57<07:07, 329.61 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [12:58<06:53, 326.49 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [13:00<07:03, 328.23 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:00<07:09, 316.47 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [13:00<07:04, 327.20 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:00<07:08, 317.29 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:00<07:01, 322.65 examples/s]
Running tokenizer on dataset:  63%|██████▎   | 242000/382000 [13:01<07:07, 327.84 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:01<06:47, 329.04 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [13:03<07:02, 326.40 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [13:03<07:04, 324.78 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:03<07:08, 314.98 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:03<07:05, 317.32 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:03<06:58, 322.23 examples/s]
Running tokenizer on dataset:  64%|██████▎   | 243000/382000 [13:04<07:03, 328.29 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:04<06:47, 326.03 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [13:06<06:59, 326.85 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [13:06<06:59, 326.73 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:06<06:59, 319.13 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:06<07:04, 315.30 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:06<06:54, 323.03 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 244000/382000 [13:07<07:05, 324.58 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:07<06:48, 322.88 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:09<06:59, 324.03 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:09<06:58, 324.94 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:09<06:58, 317.68 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:09<07:03, 314.26 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:09<06:54, 321.17 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 245000/382000 [13:10<06:59, 326.82 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:10<06:45, 322.95 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:12<06:58, 322.85 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:12<06:55, 324.94 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:12<07:00, 314.10 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:13<07:03, 311.53 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:13<06:53, 319.06 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 246000/382000 [13:13<07:00, 323.65 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:13<06:43, 322.17 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:15<06:54, 323.61 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:15<06:50, 326.62 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:16<06:57, 314.07 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:16<07:00, 311.23 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:16<06:50, 319.15 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 247000/382000 [13:16<06:57, 323.38 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:16<06:36, 324.98 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:18<06:50, 323.74 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:18<06:48, 325.23 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:19<06:52, 315.43 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:19<06:47, 318.86 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:19<06:59, 309.55 examples/s]
Running tokenizer on dataset:  65%|██████▍   | 248000/382000 [13:19<06:52, 324.94 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:19<06:34, 324.24 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:21<06:50, 321.73 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:21<06:50, 321.92 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:22<06:46, 317.21 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:22<06:43, 319.90 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:22<06:55, 310.39 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 249000/382000 [13:22<06:52, 322.45 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:22<06:29, 325.87 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:24<06:45, 323.10 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:24<06:48, 320.93 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:25<06:44, 316.31 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:25<06:39, 320.15 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:25<06:25, 326.70 examples/s]
Running tokenizer on dataset:  65%|██████▌   | 250000/382000 [13:25<06:50, 321.76 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:25<06:51, 310.95 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:28<06:41, 323.46 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:28<06:46, 319.86 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:28<06:33, 322.68 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:28<06:38, 318.36 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 251000/382000 [13:28<06:46, 322.33 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:28<06:25, 324.15 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:29<06:47, 311.43 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:31<06:36, 325.16 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:31<06:42, 320.33 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:31<06:31, 322.15 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:31<06:37, 317.03 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:31<06:20, 326.15 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 252000/382000 [13:32<06:43, 322.54 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:32<06:43, 311.91 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:34<06:35, 323.74 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:34<06:39, 320.68 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:34<06:30, 319.75 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:35<06:17, 326.10 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:35<06:35, 315.70 examples/s]
Running tokenizer on dataset:  66%|██████▌   | 253000/382000 [13:35<06:39, 322.65 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:36<07:03, 294.99 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:37<06:29, 326.30 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:37<06:33, 323.01 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:38<06:27, 320.34 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:38<06:13, 326.99 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:38<06:30, 317.93 examples/s]
Running tokenizer on dataset:  66%|██████▋   | 254000/382000 [13:38<06:37, 322.36 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:39<06:50, 302.16 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:40<06:24, 327.29 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:40<06:34, 319.70 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:41<06:09, 327.82 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:41<06:23, 320.63 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 255000/382000 [13:41<06:31, 324.05 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:41<06:28, 316.25 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:42<06:42, 305.67 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:43<06:24, 325.00 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:43<06:36, 315.47 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:44<06:08, 325.73 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:44<06:19, 321.62 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 256000/382000 [13:44<06:28, 324.64 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:44<06:25, 316.35 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:45<06:35, 308.78 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:46<06:19, 326.55 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:46<06:30, 317.37 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [13:47<06:06, 325.08 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:47<06:14, 323.10 examples/s]
Running tokenizer on dataset:  67%|██████▋   | 257000/382000 [13:47<06:26, 323.26 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:47<06:20, 318.28 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:48<06:25, 313.67 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:49<06:17, 325.61 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:50<06:26, 318.01 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [13:50<06:03, 324.51 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:50<06:14, 320.58 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 258000/382000 [13:50<06:21, 325.07 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:50<06:17, 317.71 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:52<06:32, 305.45 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:52<06:15, 325.13 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:53<06:23, 317.90 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [13:53<06:03, 322.01 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [13:53<06:11, 320.49 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 259000/382000 [13:53<06:19, 323.90 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [13:53<06:15, 317.14 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:55<06:09, 327.44 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [13:55<06:49, 290.32 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:56<06:18, 319.61 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [13:56<06:01, 320.92 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 260000/382000 [13:56<06:17, 323.22 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [13:56<06:08, 320.46 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [13:57<06:12, 317.04 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:58<06:07, 326.84 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [13:59<06:14, 320.09 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [13:59<05:56, 322.77 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [13:59<06:59, 281.39 examples/s]
Running tokenizer on dataset:  68%|██████▊   | 261000/382000 [13:59<06:12, 324.92 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [13:59<06:07, 317.95 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [14:00<06:13, 313.62 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [14:01<06:03, 327.18 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [14:02<06:14, 317.77 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:02<05:51, 324.07 examples/s]
Running tokenizer on dataset:  69%|██████▊   | 262000/382000 [14:02<06:09, 324.57 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [14:02<06:43, 290.16 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:03<06:05, 317.75 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:03<06:09, 314.00 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [14:04<06:02, 325.45 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [14:05<06:10, 318.72 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:05<05:49, 323.24 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 263000/382000 [14:05<06:07, 324.04 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:06<06:30, 297.41 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:06<06:00, 319.04 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:06<06:04, 315.74 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [14:07<06:03, 321.87 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [14:08<06:07, 318.18 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 264000/382000 [14:09<06:03, 324.56 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:09<05:46, 322.84 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:09<05:55, 321.10 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:09<06:26, 297.84 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:09<05:59, 317.35 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:11<06:00, 321.65 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:12<06:04, 318.57 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:12<05:41, 324.81 examples/s]
Running tokenizer on dataset:  69%|██████▉   | 265000/382000 [14:12<06:04, 321.16 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:12<05:53, 320.02 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:12<06:17, 302.11 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:12<05:57, 316.02 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:14<05:55, 323.58 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:15<05:38, 325.24 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:15<06:01, 318.49 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 266000/382000 [14:15<06:01, 320.73 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:15<05:50, 319.29 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:15<06:13, 302.68 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:16<05:56, 314.09 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:17<05:49, 326.37 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:18<05:35, 325.05 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:18<05:56, 320.15 examples/s]
Running tokenizer on dataset:  70%|██████▉   | 267000/382000 [14:18<05:56, 322.32 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:18<05:46, 320.35 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:19<06:07, 304.75 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:19<05:50, 316.46 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:20<05:47, 324.87 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:21<05:30, 327.08 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:21<05:52, 320.56 examples/s]
Running tokenizer on dataset:  70%|███████   | 268000/382000 [14:21<05:52, 323.16 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:21<05:42, 321.30 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:22<05:46, 317.66 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:22<06:05, 303.62 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:23<05:45, 323.79 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:24<05:24, 329.74 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:24<05:49, 320.04 examples/s]
Running tokenizer on dataset:  70%|███████   | 269000/382000 [14:24<05:51, 321.78 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:24<05:39, 321.23 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:25<05:43, 317.34 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:25<05:56, 308.44 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:26<05:42, 323.73 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:27<05:24, 326.15 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:27<05:44, 321.99 examples/s]
Running tokenizer on dataset:  71%|███████   | 270000/382000 [14:27<05:49, 320.34 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:27<05:33, 323.74 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:28<05:38, 319.50 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:28<05:53, 308.45 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:29<05:37, 326.39 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:30<05:21, 326.18 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:30<05:42, 320.94 examples/s]
Running tokenizer on dataset:  71%|███████   | 271000/382000 [14:30<05:45, 321.40 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:30<05:28, 325.95 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:31<05:33, 321.13 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:31<05:44, 313.87 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:32<05:33, 326.76 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:33<05:19, 325.58 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:33<05:39, 321.26 examples/s]
Running tokenizer on dataset:  71%|███████   | 272000/382000 [14:33<05:41, 321.88 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:34<05:29, 321.54 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:34<05:34, 317.24 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:35<05:38, 316.17 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:35<05:28, 328.46 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:36<05:17, 323.99 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:36<05:35, 322.05 examples/s]
Running tokenizer on dataset:  71%|███████▏  | 273000/382000 [14:37<05:37, 322.61 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:37<05:26, 321.44 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:38<05:30, 317.62 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:38<05:37, 314.15 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:38<05:24, 329.36 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:39<05:17, 320.90 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:39<05:31, 323.04 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 274000/382000 [14:40<05:32, 324.78 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:40<05:23, 321.01 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:41<05:27, 317.19 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:41<05:32, 315.50 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:41<05:25, 325.66 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:42<05:14, 320.77 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 275000/382000 [14:43<05:27, 326.99 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:43<05:32, 318.57 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:43<05:22, 319.84 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:44<05:26, 315.21 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:44<05:28, 316.20 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:44<05:22, 325.14 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [14:46<05:11, 320.63 examples/s]
Running tokenizer on dataset:  72%|███████▏  | 276000/382000 [14:46<05:29, 321.77 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:46<05:30, 317.86 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:46<05:21, 317.70 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:47<05:25, 313.41 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:47<05:26, 315.10 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:47<05:18, 326.21 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [14:49<05:10, 319.28 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 277000/382000 [14:49<05:26, 321.29 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:49<05:25, 319.92 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:49<05:19, 315.77 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:50<05:17, 324.78 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:50<05:24, 314.41 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:50<05:23, 312.26 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [14:52<05:03, 322.81 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 278000/382000 [14:52<05:23, 321.34 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:52<05:23, 318.27 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [14:53<05:16, 316.20 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:54<05:17, 321.76 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [14:54<05:21, 311.38 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:54<05:23, 311.87 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [14:55<05:02, 321.19 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 279000/382000 [14:55<05:21, 320.47 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:55<05:24, 314.26 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [14:56<05:13, 315.39 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:57<05:15, 319.85 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [14:57<05:19, 313.04 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [14:57<05:17, 311.56 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [14:58<04:56, 324.14 examples/s]
Running tokenizer on dataset:  73%|███████▎  | 280000/382000 [14:58<05:19, 318.95 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [14:59<05:23, 311.73 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [14:59<05:06, 319.43 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [15:00<05:12, 320.07 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [15:00<05:11, 315.08 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [15:00<05:16, 312.37 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:01<04:49, 327.94 examples/s]
Running tokenizer on dataset:  74%|███████▎  | 281000/382000 [15:02<05:19, 316.58 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [15:02<05:20, 311.95 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:02<05:04, 318.25 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [15:03<05:08, 321.04 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [15:03<05:09, 316.20 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:03<05:08, 314.08 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:04<04:48, 326.29 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 282000/382000 [15:05<05:15, 317.25 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [15:05<05:14, 314.35 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:05<04:59, 320.77 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [15:06<05:01, 325.35 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:06<05:02, 317.40 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:06<05:10, 312.05 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:07<04:45, 325.93 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 283000/382000 [15:08<05:12, 316.85 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [15:08<05:08, 318.04 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:08<04:53, 323.15 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:09<04:59, 324.28 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:09<04:58, 318.26 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:10<05:06, 312.76 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:10<04:42, 325.54 examples/s]
Running tokenizer on dataset:  74%|███████▍  | 284000/382000 [15:11<05:05, 321.05 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:11<04:50, 323.29 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:11<05:05, 317.52 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:12<04:54, 325.57 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:13<04:55, 318.44 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:13<05:01, 315.02 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:13<04:37, 327.43 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 285000/382000 [15:14<05:02, 320.36 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:14<05:00, 319.85 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:14<04:48, 322.08 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:15<04:50, 327.18 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:16<04:53, 317.29 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:16<04:59, 313.60 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:16<04:34, 327.50 examples/s]
Running tokenizer on dataset:  75%|███████▍  | 286000/382000 [15:17<04:59, 320.79 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:17<04:54, 322.81 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:17<04:44, 322.91 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:18<04:48, 326.30 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:19<04:49, 318.30 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:19<04:32, 326.31 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:19<04:58, 311.25 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 287000/382000 [15:20<04:53, 323.16 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:20<04:40, 324.73 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:20<04:52, 321.63 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:21<04:45, 326.06 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:22<04:44, 320.16 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:22<04:32, 322.88 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:22<04:54, 312.28 examples/s]
Running tokenizer on dataset:  75%|███████▌  | 288000/382000 [15:23<04:52, 321.36 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:24<04:37, 324.34 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:24<04:48, 322.56 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:24<04:42, 325.77 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:25<04:41, 319.93 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:26<04:47, 316.12 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:26<04:30, 321.41 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 289000/382000 [15:26<04:49, 321.14 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:27<04:43, 323.95 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:27<04:35, 322.56 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:27<04:38, 326.96 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:28<04:38, 319.26 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:29<04:25, 324.36 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:29<04:43, 317.53 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 290000/382000 [15:30<04:45, 322.00 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:30<04:39, 325.22 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:30<04:34, 320.42 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:30<04:35, 326.79 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:31<04:37, 316.58 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:32<04:22, 323.52 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:32<04:41, 316.05 examples/s]
Running tokenizer on dataset:  76%|███████▌  | 291000/382000 [15:33<04:41, 323.35 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:33<04:36, 325.18 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:33<04:31, 320.11 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:34<04:33, 325.67 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:35<04:35, 316.05 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:35<04:20, 323.05 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:35<04:38, 316.08 examples/s]
Running tokenizer on dataset:  76%|███████▋  | 292000/382000 [15:36<04:38, 323.42 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:36<04:35, 323.61 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:36<04:26, 322.46 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:37<04:31, 324.16 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:38<04:30, 318.44 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:38<04:16, 323.09 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:38<04:36, 315.20 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 293000/382000 [15:39<04:36, 321.98 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:39<04:33, 322.17 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:39<04:23, 322.02 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:40<04:28, 323.65 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:41<04:27, 317.57 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:41<04:12, 324.70 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:41<04:31, 316.27 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 294000/382000 [15:42<04:35, 319.37 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:42<04:30, 322.13 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:42<04:20, 321.85 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:43<04:23, 326.27 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:44<04:25, 316.97 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [15:44<04:08, 325.65 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:44<04:27, 317.78 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:45<04:24, 325.16 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 295000/382000 [15:45<04:33, 317.94 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:45<04:17, 321.88 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:46<04:21, 324.64 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [15:47<04:04, 327.39 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:47<04:21, 317.61 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:48<04:24, 317.20 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:48<04:22, 324.12 examples/s]
Running tokenizer on dataset:  77%|███████▋  | 296000/382000 [15:48<04:27, 322.02 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:48<04:12, 324.77 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:49<04:19, 324.32 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [15:50<03:59, 330.41 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:50<04:17, 317.93 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:51<04:22, 316.66 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 297000/382000 [15:51<04:24, 321.17 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:51<04:20, 322.05 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [15:51<04:08, 325.34 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:52<04:15, 324.22 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [15:53<03:56, 329.19 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [15:53<04:13, 319.95 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:54<04:17, 319.01 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:54<04:16, 323.10 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 298000/382000 [15:54<04:21, 320.98 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [15:54<04:06, 325.19 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:55<04:10, 327.42 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [15:56<03:54, 327.81 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [15:57<04:11, 318.58 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [15:57<04:11, 321.87 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [15:57<04:12, 325.35 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [15:58<04:01, 326.60 examples/s]
Running tokenizer on dataset:  78%|███████▊  | 299000/382000 [15:58<04:19, 320.25 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [15:58<04:05, 329.79 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [15:59<03:51, 328.14 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [16:00<04:07, 319.77 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [16:00<04:09, 320.40 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [16:00<04:07, 327.06 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:01<03:58, 326.62 examples/s]
Running tokenizer on dataset:  79%|███████▊  | 300000/382000 [16:01<04:14, 322.52 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [16:01<04:03, 328.49 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:02<03:50, 325.10 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:03<04:04, 319.31 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [16:03<04:06, 320.69 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [16:04<04:04, 326.93 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 301000/382000 [16:04<04:08, 325.51 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:04<03:56, 325.72 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [16:04<03:58, 330.58 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:05<03:46, 326.38 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:06<04:02, 317.64 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:06<04:04, 318.65 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [16:07<04:01, 327.57 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:07<03:52, 326.35 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 302000/382000 [16:07<04:06, 324.47 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:07<03:56, 329.28 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:08<03:45, 323.92 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:09<03:58, 318.68 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:10<04:02, 318.16 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:10<03:58, 326.90 examples/s]
Running tokenizer on dataset:  79%|███████▉  | 303000/382000 [16:10<04:01, 326.75 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:10<03:52, 322.32 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:10<03:54, 328.94 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:12<03:41, 324.78 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:12<03:56, 316.70 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:13<03:58, 318.52 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:13<03:55, 326.30 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 304000/382000 [16:13<03:59, 325.57 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:13<03:49, 322.19 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:13<03:49, 330.83 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:15<03:38, 324.50 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:15<03:53, 316.54 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:16<03:51, 327.60 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:16<03:58, 314.89 examples/s]
Running tokenizer on dataset:  80%|███████▉  | 305000/382000 [16:16<03:57, 324.80 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:16<03:47, 320.82 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:16<03:49, 326.21 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:18<03:36, 322.76 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:19<03:51, 315.31 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:19<03:51, 323.80 examples/s]
Running tokenizer on dataset:  80%|████████  | 306000/382000 [16:19<03:53, 326.12 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:19<03:53, 316.47 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:19<03:42, 323.31 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:19<03:47, 325.67 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:21<03:33, 323.27 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:22<03:47, 317.13 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:22<03:48, 323.41 examples/s]
Running tokenizer on dataset:  80%|████████  | 307000/382000 [16:22<03:53, 321.42 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:22<03:51, 314.69 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:22<03:39, 322.73 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:23<03:46, 322.19 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:24<03:29, 324.32 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:25<03:43, 317.92 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:25<03:47, 321.44 examples/s]
Running tokenizer on dataset:  81%|████████  | 308000/382000 [16:25<03:49, 321.80 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:25<03:48, 314.81 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:25<03:37, 321.23 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:26<03:41, 325.28 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:27<03:25, 325.90 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:28<03:40, 317.37 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:28<03:42, 322.93 examples/s]
Running tokenizer on dataset:  81%|████████  | 309000/382000 [16:28<03:48, 320.10 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:29<03:45, 314.80 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:29<03:35, 320.39 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:29<03:37, 325.87 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:30<03:22, 325.82 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:31<03:37, 317.94 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:31<03:39, 323.88 examples/s]
Running tokenizer on dataset:  81%|████████  | 310000/382000 [16:32<03:44, 321.21 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:32<03:30, 322.54 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:32<03:34, 325.92 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:32<03:42, 314.35 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:33<03:20, 324.54 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:34<03:33, 317.99 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:34<03:36, 322.59 examples/s]
Running tokenizer on dataset:  81%|████████▏ | 311000/382000 [16:35<03:40, 321.29 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:35<03:26, 323.99 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:35<03:32, 324.76 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:35<03:38, 315.62 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:36<03:17, 323.73 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:37<03:29, 319.33 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:38<03:34, 322.31 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 312000/382000 [16:38<03:37, 321.13 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:38<03:23, 323.92 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:38<03:29, 325.28 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:38<03:35, 315.75 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:39<03:14, 324.06 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:41<03:25, 320.68 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:41<03:30, 322.74 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 313000/382000 [16:41<03:34, 321.70 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:41<03:21, 323.01 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:41<03:25, 326.70 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:41<03:31, 316.42 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:42<03:11, 324.10 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:44<03:26, 324.88 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:44<03:23, 319.39 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:44<03:21, 326.93 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 314000/382000 [16:44<03:31, 321.15 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:44<03:18, 322.48 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:44<03:27, 318.00 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [16:46<03:09, 321.44 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:47<03:22, 325.22 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:47<03:20, 319.66 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:47<03:19, 326.38 examples/s]
Running tokenizer on dataset:  82%|████████▏ | 315000/382000 [16:47<03:28, 322.00 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:47<03:15, 321.87 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:48<03:26, 314.69 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [16:49<03:07, 320.00 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:50<03:21, 323.32 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:50<03:18, 317.32 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:50<03:16, 326.23 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 316000/382000 [16:50<03:24, 322.06 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:50<03:13, 321.18 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:51<03:23, 314.93 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [16:52<03:03, 320.95 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:53<03:17, 323.95 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:53<03:14, 317.95 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:53<03:14, 324.60 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 317000/382000 [16:53<03:22, 321.47 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [16:53<03:11, 319.34 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:54<03:20, 313.86 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [16:55<03:00, 321.30 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [16:56<03:16, 321.35 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [16:56<03:13, 315.76 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:56<03:11, 323.42 examples/s]
Running tokenizer on dataset:  83%|████████▎ | 318000/382000 [16:56<03:19, 321.46 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [16:57<03:09, 317.28 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:57<03:18, 312.85 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [16:58<02:53, 328.60 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [16:59<03:13, 320.40 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [16:59<03:09, 322.57 examples/s]
Running tokenizer on dataset:  84%|████████▎ | 319000/382000 [17:00<03:16, 320.42 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [17:00<03:11, 313.35 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:00<03:06, 316.91 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [17:00<03:15, 312.25 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:01<02:50, 329.16 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [17:02<03:10, 319.64 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [17:03<03:06, 321.52 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:03<03:07, 314.65 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 320000/382000 [17:03<03:14, 319.33 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:03<03:01, 318.76 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [17:04<03:12, 311.26 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:04<02:48, 326.31 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [17:06<03:08, 318.31 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:06<03:03, 321.79 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:06<02:55, 325.23 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 321000/382000 [17:06<03:11, 318.12 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:06<03:04, 314.09 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:07<03:09, 311.99 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:07<02:44, 328.23 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:09<03:04, 319.55 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:09<02:59, 322.37 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:09<02:57, 321.51 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:09<02:52, 324.80 examples/s]
Running tokenizer on dataset:  84%|████████▍ | 322000/382000 [17:09<03:09, 316.91 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:10<03:04, 314.64 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:10<02:42, 326.08 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:12<02:53, 328.18 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:12<03:01, 318.90 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:12<02:54, 321.18 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:12<02:50, 322.63 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 323000/382000 [17:12<03:05, 317.49 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:13<02:57, 320.64 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:13<02:39, 325.80 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:15<02:55, 324.91 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:15<02:51, 327.29 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:15<02:46, 324.97 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:15<02:51, 320.24 examples/s]
Running tokenizer on dataset:  85%|████████▍ | 324000/382000 [17:15<03:02, 317.82 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:16<02:54, 321.07 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:16<02:35, 328.77 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:18<02:51, 325.77 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:18<02:48, 326.11 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:18<02:47, 321.83 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:18<02:43, 323.70 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 325000/382000 [17:18<02:56, 323.19 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:19<02:52, 319.01 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:19<02:33, 326.67 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:21<02:44, 328.78 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:21<02:49, 325.06 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:21<02:45, 320.33 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:21<02:41, 321.46 examples/s]
Running tokenizer on dataset:  85%|████████▌ | 326000/382000 [17:21<02:53, 323.28 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:22<02:29, 328.75 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:22<02:48, 321.24 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:24<02:45, 326.81 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:24<02:41, 327.59 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:24<02:37, 322.82 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:25<02:42, 319.70 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 327000/382000 [17:25<02:50, 322.65 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:25<02:26, 327.90 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:25<02:45, 320.84 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:27<02:43, 324.47 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:27<02:39, 326.08 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:28<02:38, 322.29 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 328000/382000 [17:28<02:46, 324.26 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:28<02:36, 319.84 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:28<02:24, 325.02 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:29<02:42, 320.72 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:30<02:35, 328.72 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:30<02:40, 323.14 examples/s]
Running tokenizer on dataset:  86%|████████▌ | 329000/382000 [17:31<02:43, 323.99 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:31<02:32, 321.94 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:31<02:35, 320.96 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:32<02:21, 325.16 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:32<02:37, 323.52 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:33<02:32, 327.30 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:33<02:36, 326.01 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:34<02:31, 322.41 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:34<02:29, 320.94 examples/s]
Running tokenizer on dataset:  86%|████████▋ | 330000/382000 [17:34<02:41, 322.20 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:35<02:18, 324.81 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:35<02:35, 321.44 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:36<02:28, 329.56 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:36<02:33, 325.99 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 331000/382000 [17:37<02:36, 325.17 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:37<02:28, 322.43 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:37<02:27, 317.99 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:38<02:14, 326.00 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:38<02:31, 322.61 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:39<02:25, 328.83 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:39<02:29, 328.21 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 332000/382000 [17:40<02:34, 323.86 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:40<02:28, 317.40 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:40<02:25, 315.66 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:41<02:12, 325.08 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:41<02:28, 323.08 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:42<02:23, 326.45 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:42<02:27, 326.35 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 333000/382000 [17:43<02:30, 326.11 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:43<02:25, 315.33 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:43<02:22, 316.84 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:44<02:26, 321.66 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [17:44<02:13, 315.07 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:45<02:21, 325.29 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:45<02:25, 323.85 examples/s]
Running tokenizer on dataset:  87%|████████▋ | 334000/382000 [17:46<02:27, 325.68 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:47<02:22, 315.51 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:47<02:18, 316.65 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:47<02:23, 319.85 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [17:47<02:10, 313.72 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:48<02:17, 326.59 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:49<02:23, 321.04 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 335000/382000 [17:49<02:24, 324.92 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:50<02:15, 316.56 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:50<02:20, 313.54 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:50<02:20, 319.42 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [17:51<02:08, 310.14 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:51<02:13, 328.76 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:52<02:19, 322.42 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 336000/382000 [17:52<02:21, 325.28 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:53<02:16, 313.87 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [17:53<02:16, 307.07 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:54<02:19, 315.19 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [17:54<02:06, 309.06 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:54<02:10, 329.83 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:55<02:16, 322.67 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 337000/382000 [17:55<02:18, 324.72 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [17:56<02:11, 311.48 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [17:56<02:17, 305.31 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:57<02:16, 315.87 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [17:57<02:03, 308.88 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [17:58<02:10, 320.96 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [17:58<02:13, 322.95 examples/s]
Running tokenizer on dataset:  88%|████████▊ | 338000/382000 [17:58<02:15, 324.39 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [17:59<02:07, 313.22 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [18:00<02:12, 310.23 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [18:00<02:16, 307.64 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:00<01:59, 310.02 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [18:01<02:06, 322.96 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [18:01<02:13, 315.27 examples/s]
Running tokenizer on dataset:  89%|████████▊ | 339000/382000 [18:02<02:12, 323.74 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:03<02:03, 314.87 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [18:03<02:08, 311.67 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [18:03<02:10, 313.42 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:03<01:54, 313.12 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [18:04<02:04, 322.14 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [18:04<02:08, 318.48 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 340000/382000 [18:05<02:13, 314.44 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:06<02:04, 313.24 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:06<02:01, 312.49 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [18:06<02:06, 314.97 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:06<01:49, 318.91 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:07<02:00, 323.90 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [18:07<02:04, 320.75 examples/s]
Running tokenizer on dataset:  89%|████████▉ | 341000/382000 [18:08<02:08, 318.42 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:09<01:57, 313.88 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:09<02:01, 312.04 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:10<02:03, 314.53 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:10<01:46, 318.74 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:10<01:58, 321.48 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:11<02:01, 321.33 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 342000/382000 [18:11<02:05, 318.73 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:12<01:54, 313.62 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:12<01:58, 311.77 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:13<01:43, 319.14 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:13<02:01, 312.78 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:13<01:54, 322.62 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:14<01:59, 318.56 examples/s]
Running tokenizer on dataset:  90%|████████▉ | 343000/382000 [18:14<02:01, 320.59 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:15<01:49, 318.57 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:15<01:55, 311.83 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:16<01:39, 320.66 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:16<01:57, 313.98 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:16<01:51, 321.87 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:17<01:56, 317.89 examples/s]
Running tokenizer on dataset:  90%|█████████ | 344000/382000 [18:17<01:59, 317.59 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:18<01:46, 318.45 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:19<01:50, 316.46 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:19<01:36, 321.99 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:19<01:54, 315.07 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:19<01:47, 326.99 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:20<01:53, 317.70 examples/s]
Running tokenizer on dataset:  90%|█████████ | 345000/382000 [18:21<01:56, 317.77 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:22<01:44, 314.86 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:22<01:47, 316.97 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:22<01:33, 321.58 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:22<01:49, 318.63 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:22<01:43, 327.07 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:23<01:48, 322.33 examples/s]
Running tokenizer on dataset:  91%|█████████ | 346000/382000 [18:24<01:53, 317.79 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:25<01:41, 314.52 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:25<01:44, 315.91 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:25<01:29, 323.49 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:25<01:46, 318.33 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:25<01:41, 325.63 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:26<01:45, 323.62 examples/s]
Running tokenizer on dataset:  91%|█████████ | 347000/382000 [18:27<01:48, 321.76 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:28<01:38, 314.52 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:28<01:41, 315.58 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:28<01:26, 322.87 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:28<01:37, 326.72 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:28<01:43, 317.96 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:29<01:42, 322.87 examples/s]
Running tokenizer on dataset:  91%|█████████ | 348000/382000 [18:30<01:45, 322.27 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:31<01:35, 312.76 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:31<01:38, 314.02 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:31<01:23, 322.95 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:32<01:35, 324.72 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:32<01:41, 314.29 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:32<01:39, 321.65 examples/s]
Running tokenizer on dataset:  91%|█████████▏| 349000/382000 [18:33<01:42, 320.72 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:34<01:31, 317.01 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:34<01:19, 325.28 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:34<01:35, 314.77 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:35<01:32, 324.41 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:35<01:38, 313.50 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:35<01:36, 320.52 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 350000/382000 [18:36<01:39, 320.79 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:37<01:16, 324.73 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:37<01:28, 316.33 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:38<01:31, 317.53 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:38<01:28, 325.99 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:38<01:36, 311.61 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:39<01:33, 319.58 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 351000/382000 [18:39<01:37, 318.13 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:41<01:14, 323.10 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:41<01:25, 316.18 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:41<01:28, 316.34 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:41<01:26, 325.17 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:41<01:32, 315.12 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:42<01:29, 323.58 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 352000/382000 [18:42<01:34, 318.05 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [18:44<01:11, 323.45 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:44<01:21, 319.42 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:44<01:22, 325.44 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:44<01:25, 315.88 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:45<01:29, 313.04 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:45<01:26, 323.00 examples/s]
Running tokenizer on dataset:  92%|█████████▏| 353000/382000 [18:45<01:30, 321.93 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [18:47<01:08, 323.43 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:47<01:18, 320.28 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:47<01:19, 327.29 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:47<01:21, 318.16 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:48<01:25, 314.37 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:48<01:23, 322.87 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 354000/382000 [18:49<01:27, 321.01 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [18:50<01:04, 326.27 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:50<01:16, 327.05 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:50<01:15, 316.79 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:50<01:18, 319.37 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:51<01:22, 316.92 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:51<01:20, 324.47 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 355000/382000 [18:52<01:24, 320.08 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [18:53<01:01, 324.97 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:53<01:13, 324.45 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [18:53<01:12, 316.98 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:53<01:15, 316.81 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:54<01:18, 316.61 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:54<01:17, 323.78 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 356000/382000 [18:55<01:20, 321.62 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [18:56<00:58, 324.03 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [18:56<01:10, 324.66 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [18:56<01:09, 316.45 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [18:56<01:12, 317.39 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:57<01:15, 317.16 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [18:57<01:14, 322.10 examples/s]
Running tokenizer on dataset:  93%|█████████▎| 357000/382000 [18:58<01:17, 321.35 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [18:59<00:55, 321.74 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [18:59<01:07, 325.24 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [18:59<01:05, 319.03 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [19:00<01:09, 315.90 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [19:00<01:11, 322.88 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [19:00<01:12, 316.77 examples/s]
Running tokenizer on dataset:  94%|█████████▎| 358000/382000 [19:01<01:14, 320.00 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:02<00:52, 322.69 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [19:02<01:03, 328.28 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:03<01:02, 319.02 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [19:03<01:05, 318.87 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [19:03<01:08, 322.71 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [19:03<01:09, 316.29 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 359000/382000 [19:04<01:11, 321.42 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:05<00:49, 323.93 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:05<01:00, 328.55 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:06<00:59, 318.76 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:06<01:02, 320.44 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [19:06<01:04, 324.70 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [19:06<01:05, 320.88 examples/s]
Running tokenizer on dataset:  94%|█████████▍| 360000/382000 [19:07<01:08, 320.29 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:08<00:46, 324.21 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:08<00:58, 325.63 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:09<00:56, 317.83 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:09<00:59, 318.08 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:09<01:01, 325.59 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:10<01:02, 322.43 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 361000/382000 [19:10<01:05, 322.52 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:11<00:43, 325.07 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:12<00:55, 323.86 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:12<00:53, 318.39 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:12<00:56, 316.01 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:13<00:58, 323.41 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:13<00:59, 321.05 examples/s]
Running tokenizer on dataset:  95%|█████████▍| 362000/382000 [19:13<01:02, 322.38 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:14<00:40, 323.83 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:15<00:52, 324.35 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:15<00:50, 318.17 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:15<00:53, 316.06 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:16<00:55, 322.02 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:16<00:56, 317.81 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 363000/382000 [19:17<00:59, 321.09 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:18<00:36, 325.25 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:18<00:49, 326.13 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:18<00:47, 318.83 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:18<00:50, 317.56 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:19<00:52, 323.93 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:19<00:53, 319.25 examples/s]
Running tokenizer on dataset:  95%|█████████▌| 364000/382000 [19:20<00:56, 319.86 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:21<00:45, 327.26 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:21<00:33, 323.60 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:21<00:44, 317.07 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:22<00:47, 318.60 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:22<00:49, 322.57 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:22<00:50, 319.65 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 365000/382000 [19:23<00:52, 320.91 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:24<00:30, 326.51 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:24<00:42, 327.68 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:25<00:40, 318.21 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:25<00:43, 318.50 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:25<00:46, 323.46 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:25<00:46, 321.80 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 366000/382000 [19:26<00:49, 320.91 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:27<00:39, 326.34 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:27<00:27, 324.48 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:28<00:37, 319.65 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:28<00:40, 318.05 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:28<00:43, 323.42 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:28<00:43, 321.70 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 367000/382000 [19:29<00:46, 321.85 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:30<00:36, 328.06 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:30<00:24, 323.60 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:31<00:34, 317.43 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:31<00:37, 319.22 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:31<00:40, 324.11 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:31<00:40, 320.81 examples/s]
Running tokenizer on dataset:  96%|█████████▋| 368000/382000 [19:32<00:43, 322.68 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:33<00:33, 326.65 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:33<00:21, 321.99 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:34<00:31, 320.66 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:34<00:36, 325.96 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:34<00:34, 318.12 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:34<00:37, 322.00 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 369000/382000 [19:35<00:40, 322.45 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:36<00:30, 327.01 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:36<00:18, 323.61 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:37<00:28, 321.02 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:37<00:33, 324.31 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:37<00:31, 320.26 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:38<00:34, 320.63 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 370000/382000 [19:38<00:37, 323.42 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:39<00:27, 325.50 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:39<00:15, 323.02 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:40<00:24, 321.20 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:40<00:30, 325.58 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:40<00:28, 318.59 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:41<00:31, 320.64 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 371000/382000 [19:41<00:34, 321.31 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:42<00:24, 325.51 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [19:42<00:12, 323.51 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:43<00:21, 318.31 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:43<00:27, 323.04 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:44<00:25, 317.16 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:44<00:28, 319.35 examples/s]
Running tokenizer on dataset:  97%|█████████▋| 372000/382000 [19:44<00:30, 323.06 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:45<00:21, 324.93 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [19:45<00:09, 324.20 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:46<00:18, 319.76 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:47<00:24, 322.84 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:47<00:22, 316.41 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:47<00:25, 319.77 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 373000/382000 [19:48<00:27, 321.73 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:48<00:18, 326.99 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [19:48<00:06, 325.25 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:50<00:15, 320.68 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:50<00:21, 321.74 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:50<00:18, 318.10 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:50<00:22, 315.20 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 374000/382000 [19:51<00:24, 320.82 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:51<00:15, 325.97 examples/s]
Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [19:52<00:03, 323.96 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [19:53<00:12, 319.97 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:53<00:18, 322.73 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:53<00:15, 317.67 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:53<00:18, 315.98 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 375000/382000 [19:54<00:21, 320.02 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [19:54<00:12, 325.60 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [19:55<00:00, 324.05 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [19:55<00:00, 319.48 examples/s]

Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [19:56<00:09, 320.05 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:56<00:15, 322.29 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [19:56<00:12, 316.82 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [19:57<00:15, 315.76 examples/s]
Running tokenizer on dataset:  98%|█████████▊| 376000/382000 [19:57<00:18, 320.73 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [19:58<00:09, 323.59 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [19:59<00:06, 319.11 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [19:59<00:12, 321.21 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [19:59<00:09, 315.25 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [20:00<00:12, 315.20 examples/s]
Running tokenizer on dataset:  99%|█████████▊| 377000/382000 [20:00<00:15, 318.80 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [20:01<00:06, 322.51 examples/s]
Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:02<00:03, 318.79 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [20:02<00:09, 319.28 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [20:03<00:06, 316.52 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [20:03<00:09, 313.86 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 378000/382000 [20:03<00:12, 317.29 examples/s]
Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:04<00:03, 321.38 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:05<00:00, 317.47 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:05<00:00, 316.80 examples/s]

Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [20:05<00:06, 318.20 examples/s]
Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:06<00:03, 315.63 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [20:06<00:06, 313.62 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 379000/382000 [20:07<00:09, 315.23 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:07<00:00, 319.26 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:07<00:00, 316.36 examples/s]

Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:09<00:03, 316.68 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:09<00:00, 312.94 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:09<00:00, 315.82 examples/s]

Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:09<00:03, 313.32 examples/s]
Running tokenizer on dataset:  99%|█████████▉| 380000/382000 [20:10<00:06, 316.21 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:12<00:00, 315.59 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:12<00:00, 315.12 examples/s]

Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:13<00:00, 311.01 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:13<00:00, 314.86 examples/s]

Running tokenizer on dataset: 100%|█████████▉| 381000/382000 [20:13<00:03, 317.28 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:16<00:00, 318.06 examples/s]
Running tokenizer on dataset: 100%|██████████| 382000/382000 [20:16<00:00, 314.01 examples/s]
[INFO|trainer.py:1760] 2023-11-12 03:23:45,386 >> ***** Running training *****
[INFO|trainer.py:1761] 2023-11-12 03:23:45,387 >>   Num examples = 382,000
[INFO|trainer.py:1762] 2023-11-12 03:23:45,387 >>   Num Epochs = 3
[INFO|trainer.py:1763] 2023-11-12 03:23:45,387 >>   Instantaneous batch size per device = 4
[INFO|trainer.py:1766] 2023-11-12 03:23:45,387 >>   Total train batch size (w. parallel, distributed & accumulation) = 256
[INFO|trainer.py:1767] 2023-11-12 03:23:45,387 >>   Gradient Accumulation steps = 8
[INFO|trainer.py:1768] 2023-11-12 03:23:45,387 >>   Total optimization steps = 4,476
[INFO|trainer.py:1769] 2023-11-12 03:23:45,390 >>   Number of trainable parameters = 1,949,696

  0%|          | 0/4476 [00:00<?, ?it/s]
  0%|          | 1/4476 [00:24<30:42:33, 24.70s/it]
  0%|          | 2/4476 [00:47<29:22:31, 23.64s/it]
  0%|          | 3/4476 [01:10<28:50:04, 23.21s/it]
  0%|          | 4/4476 [01:33<28:45:20, 23.15s/it]
  0%|          | 5/4476 [01:56<28:38:31, 23.06s/it]
  0%|          | 6/4476 [02:18<28:26:52, 22.91s/it]
  0%|          | 7/4476 [02:42<28:37:53, 23.06s/it]
  0%|          | 8/4476 [03:05<28:37:44, 23.07s/it]
  0%|          | 9/4476 [03:28<28:41:32, 23.12s/it]
  0%|          | 10/4476 [03:51<28:42:46, 23.15s/it]
                                                    
{'loss': 1.2211, 'learning_rate': 4.99993842168232e-05, 'epoch': 0.01}

  0%|          | 10/4476 [03:51<28:42:46, 23.15s/it]
  0%|          | 11/4476 [04:14<28:43:31, 23.16s/it]
  0%|          | 12/4476 [04:38<28:43:44, 23.17s/it]
  0%|          | 13/4476 [05:01<28:52:17, 23.29s/it]
  0%|          | 14/4476 [05:25<28:55:36, 23.34s/it]
  0%|          | 15/4476 [05:48<28:54:14, 23.33s/it]
  0%|          | 16/4476 [06:11<28:52:57, 23.31s/it]
  0%|          | 17/4476 [06:35<28:53:49, 23.33s/it]
  0%|          | 18/4476 [06:58<28:55:06, 23.35s/it]
  0%|          | 19/4476 [07:21<28:55:34, 23.36s/it]
  0%|          | 20/4476 [07:45<28:51:07, 23.31s/it]
                                                    
{'loss': 1.0276, 'learning_rate': 4.9997536897627915e-05, 'epoch': 0.01}

  0%|          | 20/4476 [07:45<28:51:07, 23.31s/it]
  0%|          | 21/4476 [08:08<28:49:11, 23.29s/it]
  0%|          | 22/4476 [08:31<28:47:29, 23.27s/it]
  1%|          | 23/4476 [08:54<28:47:10, 23.27s/it]
  1%|          | 24/4476 [09:18<28:46:41, 23.27s/it]
  1%|          | 25/4476 [09:41<28:42:47, 23.22s/it]
  1%|          | 26/4476 [10:04<28:40:17, 23.19s/it]
  1%|          | 27/4476 [10:27<28:47:15, 23.29s/it]
  1%|          | 28/4476 [10:51<28:46:54, 23.29s/it]
  1%|          | 29/4476 [11:14<28:42:33, 23.24s/it]
  1%|          | 30/4476 [11:37<28:38:55, 23.20s/it]
                                                    
{'loss': 0.8587, 'learning_rate': 4.9994458133418e-05, 'epoch': 0.02}

  1%|          | 30/4476 [11:37<28:38:55, 23.20s/it]
  1%|          | 31/4476 [12:00<28:37:56, 23.19s/it]
  1%|          | 32/4476 [12:23<28:40:11, 23.22s/it]
  1%|          | 33/4476 [12:47<28:41:52, 23.25s/it]
  1%|          | 34/4476 [13:10<28:42:46, 23.27s/it]
  1%|          | 35/4476 [13:33<28:46:26, 23.33s/it]
  1%|          | 36/4476 [13:57<28:41:46, 23.27s/it]
  1%|          | 37/4476 [14:20<28:40:42, 23.26s/it]
  1%|          | 38/4476 [14:43<28:47:39, 23.36s/it]
  1%|          | 39/4476 [15:07<28:48:57, 23.38s/it]
  1%|          | 40/4476 [15:30<28:51:49, 23.42s/it]
                                                    
{'loss': 0.7431, 'learning_rate': 4.999014807586154e-05, 'epoch': 0.03}

  1%|          | 40/4476 [15:30<28:51:49, 23.42s/it]
  1%|          | 41/4476 [15:54<28:52:50, 23.44s/it]
  1%|          | 42/4476 [16:18<28:57:44, 23.51s/it]
  1%|          | 43/4476 [16:41<28:49:58, 23.42s/it]
  1%|          | 44/4476 [17:04<28:38:58, 23.27s/it]
  1%|          | 45/4476 [17:27<28:42:45, 23.33s/it]
  1%|          | 46/4476 [17:50<28:34:23, 23.22s/it]
  1%|          | 47/4476 [18:13<28:36:20, 23.25s/it]
  1%|          | 48/4476 [18:37<28:43:32, 23.35s/it]
  1%|          | 49/4476 [19:00<28:40:50, 23.32s/it]
  1%|          | 50/4476 [19:23<28:31:17, 23.20s/it]
                                                    
{'loss': 0.6841, 'learning_rate': 4.9984606937283405e-05, 'epoch': 0.03}

  1%|          | 50/4476 [19:23<28:31:17, 23.20s/it]
  1%|          | 51/4476 [19:47<28:37:00, 23.28s/it]
  1%|          | 52/4476 [20:10<28:40:26, 23.33s/it]
  1%|          | 53/4476 [20:33<28:41:32, 23.35s/it]
  1%|          | 54/4476 [20:57<28:47:23, 23.44s/it]
  1%|          | 55/4476 [21:20<28:41:53, 23.37s/it]
  1%|▏         | 56/4476 [21:44<28:39:50, 23.35s/it]
  1%|▏         | 57/4476 [22:07<28:42:00, 23.38s/it]
  1%|▏         | 58/4476 [22:30<28:38:36, 23.34s/it]
  1%|▏         | 59/4476 [22:54<28:42:14, 23.39s/it]
  1%|▏         | 60/4476 [23:17<28:36:28, 23.32s/it]
                                                    
{'loss': 0.6452, 'learning_rate': 4.9977834990654804e-05, 'epoch': 0.04}

  1%|▏         | 60/4476 [23:17<28:36:28, 23.32s/it]
  1%|▏         | 61/4476 [23:41<28:43:31, 23.42s/it]
  1%|▏         | 62/4476 [24:04<28:44:24, 23.44s/it]
  1%|▏         | 63/4476 [24:28<28:42:55, 23.43s/it]
  1%|▏         | 64/4476 [24:51<28:48:07, 23.50s/it]
  1%|▏         | 65/4476 [25:15<28:46:21, 23.48s/it]
  1%|▏         | 66/4476 [25:38<28:50:54, 23.55s/it]
  1%|▏         | 67/4476 [26:02<28:51:32, 23.56s/it]
  2%|▏         | 68/4476 [26:25<28:41:40, 23.43s/it]
  2%|▏         | 69/4476 [26:48<28:37:37, 23.38s/it]
  2%|▏         | 70/4476 [27:12<28:34:51, 23.35s/it]
                                                    
{'loss': 0.6347, 'learning_rate': 4.99698325695798e-05, 'epoch': 0.05}

  2%|▏         | 70/4476 [27:12<28:34:51, 23.35s/it]
  2%|▏         | 71/4476 [27:35<28:39:59, 23.43s/it]
  2%|▏         | 72/4476 [27:58<28:31:09, 23.31s/it]
  2%|▏         | 73/4476 [28:22<28:30:10, 23.30s/it]
  2%|▏         | 74/4476 [28:45<28:36:56, 23.40s/it]
  2%|▏         | 75/4476 [29:09<28:35:04, 23.38s/it]
  2%|▏         | 76/4476 [29:32<28:42:41, 23.49s/it]
  2%|▏         | 77/4476 [29:55<28:35:54, 23.40s/it]
  2%|▏         | 78/4476 [30:19<28:35:12, 23.40s/it]
  2%|▏         | 79/4476 [30:42<28:36:21, 23.42s/it]
  2%|▏         | 80/4476 [31:06<28:40:36, 23.48s/it]
                                                    
{'loss': 0.6109, 'learning_rate': 4.9960600068278876e-05, 'epoch': 0.05}

  2%|▏         | 80/4476 [31:06<28:40:36, 23.48s/it]
  2%|▏         | 81/4476 [31:29<28:36:06, 23.43s/it]
  2%|▏         | 82/4476 [31:53<28:41:30, 23.51s/it]
  2%|▏         | 83/4476 [32:16<28:30:03, 23.36s/it]
  2%|▏         | 84/4476 [32:39<28:28:57, 23.35s/it]
  2%|▏         | 85/4476 [33:03<28:28:04, 23.34s/it]
  2%|▏         | 86/4476 [33:26<28:27:08, 23.33s/it]
  2%|▏         | 87/4476 [33:49<28:25:11, 23.31s/it]
  2%|▏         | 88/4476 [34:13<28:27:07, 23.34s/it]
  2%|▏         | 89/4476 [34:36<28:29:06, 23.37s/it]
  2%|▏         | 90/4476 [34:59<28:26:34, 23.35s/it]
                                                    
{'loss': 0.5911, 'learning_rate': 4.995013794156957e-05, 'epoch': 0.06}

  2%|▏         | 90/4476 [34:59<28:26:34, 23.35s/it]
  2%|▏         | 91/4476 [35:23<28:32:49, 23.44s/it]
  2%|▏         | 92/4476 [35:46<28:24:01, 23.32s/it]
  2%|▏         | 93/4476 [36:10<28:27:16, 23.37s/it]
  2%|▏         | 94/4476 [36:33<28:21:42, 23.30s/it]
  2%|▏         | 95/4476 [36:56<28:27:09, 23.38s/it]
  2%|▏         | 96/4476 [37:20<28:29:36, 23.42s/it]
  2%|▏         | 97/4476 [37:43<28:26:06, 23.38s/it]
  2%|▏         | 98/4476 [38:06<28:24:24, 23.36s/it]
  2%|▏         | 99/4476 [38:30<28:20:13, 23.31s/it]
  2%|▏         | 100/4476 [38:53<28:27:44, 23.42s/it]
                                                     
{'loss': 0.5803, 'learning_rate': 4.993844670484401e-05, 'epoch': 0.07}

  2%|▏         | 100/4476 [38:53<28:27:44, 23.42s/it]
  2%|▏         | 101/4476 [39:17<28:27:20, 23.41s/it]
  2%|▏         | 102/4476 [39:40<28:17:00, 23.28s/it]
  2%|▏         | 103/4476 [40:03<28:22:59, 23.37s/it]
  2%|▏         | 104/4476 [40:26<28:17:19, 23.29s/it]
  2%|▏         | 105/4476 [40:50<28:16:24, 23.29s/it]
  2%|▏         | 106/4476 [41:13<28:14:10, 23.26s/it]
  2%|▏         | 107/4476 [41:36<28:15:04, 23.28s/it]
  2%|▏         | 108/4476 [41:59<28:15:20, 23.29s/it]
  2%|▏         | 109/4476 [42:23<28:19:48, 23.35s/it]
  2%|▏         | 110/4476 [42:46<28:23:03, 23.40s/it]
                                                     
{'loss': 0.5902, 'learning_rate': 4.992552693404354e-05, 'epoch': 0.07}

  2%|▏         | 110/4476 [42:46<28:23:03, 23.40s/it]
  2%|▏         | 111/4476 [43:09<28:15:07, 23.30s/it]
  3%|▎         | 112/4476 [43:33<28:10:51, 23.25s/it]
  3%|▎         | 113/4476 [43:56<28:18:03, 23.35s/it]
  3%|▎         | 114/4476 [44:20<28:18:53, 23.37s/it]
  3%|▎         | 115/4476 [44:43<28:17:50, 23.36s/it]
  3%|▎         | 116/4476 [45:06<28:14:20, 23.32s/it]
  3%|▎         | 117/4476 [45:30<28:15:28, 23.34s/it]
  3%|▎         | 118/4476 [45:53<28:14:36, 23.33s/it]
  3%|▎         | 119/4476 [46:16<28:12:29, 23.31s/it]
  3%|▎         | 120/4476 [46:39<28:04:39, 23.20s/it]
                                                     
{'loss': 0.5745, 'learning_rate': 4.991137926563036e-05, 'epoch': 0.08}

  3%|▎         | 120/4476 [46:39<28:04:39, 23.20s/it]
  3%|▎         | 121/4476 [47:02<28:04:12, 23.20s/it]
  3%|▎         | 122/4476 [47:26<28:13:42, 23.34s/it]
  3%|▎         | 123/4476 [47:50<28:20:11, 23.43s/it]
  3%|▎         | 124/4476 [48:13<28:18:02, 23.41s/it]
  3%|▎         | 125/4476 [48:36<28:21:08, 23.46s/it]
  3%|▎         | 126/4476 [49:00<28:17:50, 23.42s/it]
  3%|▎         | 127/4476 [49:23<28:14:30, 23.38s/it]
  3%|▎         | 128/4476 [49:46<28:11:27, 23.34s/it]
  3%|▎         | 129/4476 [50:10<28:08:17, 23.30s/it]
  3%|▎         | 130/4476 [50:33<28:10:43, 23.34s/it]
                                                     
{'loss': 0.5538, 'learning_rate': 4.9896004396556176e-05, 'epoch': 0.09}

  3%|▎         | 130/4476 [50:33<28:10:43, 23.34s/it]
  3%|▎         | 131/4476 [50:56<28:09:30, 23.33s/it]
  3%|▎         | 132/4476 [51:20<28:12:19, 23.37s/it]
  3%|▎         | 133/4476 [51:43<28:17:23, 23.45s/it]
  3%|▎         | 134/4476 [52:06<28:08:47, 23.34s/it]
  3%|▎         | 135/4476 [52:30<28:06:00, 23.30s/it]
  3%|▎         | 136/4476 [52:53<28:01:26, 23.25s/it]
  3%|▎         | 137/4476 [53:16<28:07:01, 23.33s/it]
  3%|▎         | 138/4476 [53:40<28:11:44, 23.40s/it]
  3%|▎         | 139/4476 [54:03<28:07:38, 23.35s/it]
  3%|▎         | 140/4476 [54:27<28:09:13, 23.37s/it]
                                                     
{'loss': 0.5495, 'learning_rate': 4.987940308422783e-05, 'epoch': 0.09}

  3%|▎         | 140/4476 [54:27<28:09:13, 23.37s/it]
  3%|▎         | 141/4476 [54:50<28:03:51, 23.31s/it]
  3%|▎         | 142/4476 [55:13<28:09:54, 23.40s/it]
  3%|▎         | 143/4476 [55:37<28:06:07, 23.35s/it]
  3%|▎         | 144/4476 [56:00<28:08:09, 23.38s/it]
  3%|▎         | 145/4476 [56:23<27:58:46, 23.26s/it]
  3%|▎         | 146/4476 [56:46<28:02:05, 23.31s/it]
  3%|▎         | 147/4476 [57:10<28:01:43, 23.31s/it]
  3%|▎         | 148/4476 [57:33<28:02:54, 23.33s/it]
  3%|▎         | 149/4476 [57:57<28:09:54, 23.43s/it]
  3%|▎         | 150/4476 [58:20<28:04:58, 23.37s/it]
                                                     
{'loss': 0.5433, 'learning_rate': 4.986157614647005e-05, 'epoch': 0.1}

  3%|▎         | 150/4476 [58:20<28:04:58, 23.37s/it]
  3%|▎         | 151/4476 [58:44<28:11:40, 23.47s/it]
  3%|▎         | 152/4476 [59:07<28:05:34, 23.39s/it]
  3%|▎         | 153/4476 [59:31<28:11:11, 23.47s/it]
  3%|▎         | 154/4476 [59:54<28:11:55, 23.49s/it]
  3%|▎         | 155/4476 [1:00:17<28:07:23, 23.43s/it]
  3%|▎         | 156/4476 [1:00:41<28:14:01, 23.53s/it]
  4%|▎         | 157/4476 [1:01:04<28:07:13, 23.44s/it]
  4%|▎         | 158/4476 [1:01:28<28:01:30, 23.37s/it]
  4%|▎         | 159/4476 [1:01:51<28:05:52, 23.43s/it]
  4%|▎         | 160/4476 [1:02:14<28:02:06, 23.38s/it]
                                                       
{'loss': 0.548, 'learning_rate': 4.984252446148508e-05, 'epoch': 0.11}

  4%|▎         | 160/4476 [1:02:14<28:02:06, 23.38s/it]
  4%|▎         | 161/4476 [1:02:37<27:54:43, 23.29s/it]
  4%|▎         | 162/4476 [1:03:01<27:52:53, 23.27s/it]
  4%|▎         | 163/4476 [1:03:24<27:52:29, 23.27s/it]
  4%|▎         | 164/4476 [1:03:47<27:53:08, 23.28s/it]
  4%|▎         | 165/4476 [1:04:10<27:50:56, 23.26s/it]
  4%|▎         | 166/4476 [1:04:34<27:55:03, 23.32s/it]
  4%|▎         | 167/4476 [1:04:58<28:04:32, 23.46s/it]
  4%|▍         | 168/4476 [1:05:21<28:10:56, 23.55s/it]
  4%|▍         | 169/4476 [1:05:44<27:57:48, 23.37s/it]
  4%|▍         | 170/4476 [1:06:08<27:55:40, 23.35s/it]
                                                       
{'loss': 0.5361, 'learning_rate': 4.98222489678095e-05, 'epoch': 0.11}

  4%|▍         | 170/4476 [1:06:08<27:55:40, 23.35s/it]
  4%|▍         | 171/4476 [1:06:31<27:54:24, 23.34s/it]
  4%|▍         | 172/4476 [1:06:54<27:53:33, 23.33s/it]
  4%|▍         | 173/4476 [1:07:18<27:51:42, 23.31s/it]
  4%|▍         | 174/4476 [1:07:41<27:54:52, 23.36s/it]
  4%|▍         | 175/4476 [1:08:04<27:46:39, 23.25s/it]
  4%|▍         | 176/4476 [1:08:27<27:45:07, 23.23s/it]
  4%|▍         | 177/4476 [1:08:51<27:49:26, 23.30s/it]
  4%|▍         | 178/4476 [1:09:14<27:48:55, 23.30s/it]
  4%|▍         | 179/4476 [1:09:38<27:52:08, 23.35s/it]
  4%|▍         | 180/4476 [1:10:01<27:48:11, 23.30s/it]
                                                       
{'loss': 0.5331, 'learning_rate': 4.980075066426796e-05, 'epoch': 0.12}

  4%|▍         | 180/4476 [1:10:01<27:48:11, 23.30s/it]
  4%|▍         | 181/4476 [1:10:24<27:46:26, 23.28s/it]
  4%|▍         | 182/4476 [1:10:47<27:46:55, 23.29s/it]
  4%|▍         | 183/4476 [1:11:11<27:50:09, 23.34s/it]
  4%|▍         | 184/4476 [1:11:34<27:48:37, 23.33s/it]
  4%|▍         | 185/4476 [1:11:58<27:55:57, 23.43s/it]
  4%|▍         | 186/4476 [1:12:21<27:57:02, 23.46s/it]
  4%|▍         | 187/4476 [1:12:44<27:48:54, 23.35s/it]
  4%|▍         | 188/4476 [1:13:07<27:40:13, 23.23s/it]
  4%|▍         | 189/4476 [1:13:30<27:31:30, 23.11s/it]
  4%|▍         | 190/4476 [1:13:54<27:46:26, 23.33s/it]
                                                       
{'loss': 0.53, 'learning_rate': 4.977803060992393e-05, 'epoch': 0.13}

  4%|▍         | 190/4476 [1:13:54<27:46:26, 23.33s/it]
  4%|▍         | 191/4476 [1:14:17<27:40:23, 23.25s/it]
  4%|▍         | 192/4476 [1:14:40<27:40:54, 23.26s/it]
  4%|▍         | 193/4476 [1:15:04<27:40:50, 23.27s/it]
  4%|▍         | 194/4476 [1:15:27<27:34:39, 23.19s/it]
  4%|▍         | 195/4476 [1:15:50<27:38:29, 23.24s/it]
  4%|▍         | 196/4476 [1:16:13<27:37:24, 23.23s/it]
  4%|▍         | 197/4476 [1:16:37<27:42:05, 23.31s/it]
  4%|▍         | 198/4476 [1:17:00<27:44:00, 23.34s/it]
  4%|▍         | 199/4476 [1:17:23<27:39:34, 23.28s/it]
  4%|▍         | 200/4476 [1:17:47<27:42:30, 23.33s/it]
                                                       
{'loss': 0.5135, 'learning_rate': 4.97540899240276e-05, 'epoch': 0.13}

  4%|▍         | 200/4476 [1:17:47<27:42:30, 23.33s/it][INFO|trainer.py:2939] 2023-11-12 04:41:34,337 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-200
[INFO|tokenization_utils_base.py:2437] 2023-11-12 04:41:34,371 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-200/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 04:41:34,372 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-200/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 04:41:34,372 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-200/added_tokens.json

  4%|▍         | 201/4476 [1:18:10<27:42:08, 23.33s/it]
  5%|▍         | 202/4476 [1:18:33<27:46:04, 23.39s/it]
  5%|▍         | 203/4476 [1:18:57<27:39:00, 23.30s/it]
  5%|▍         | 204/4476 [1:19:20<27:34:29, 23.24s/it]
  5%|▍         | 205/4476 [1:19:43<27:37:53, 23.29s/it]
  5%|▍         | 206/4476 [1:20:06<27:37:16, 23.29s/it]
  5%|▍         | 207/4476 [1:20:30<27:45:27, 23.41s/it]
  5%|▍         | 208/4476 [1:20:53<27:39:53, 23.33s/it]
  5%|▍         | 209/4476 [1:21:16<27:34:03, 23.26s/it]
  5%|▍         | 210/4476 [1:21:40<27:38:03, 23.32s/it]
                                                       
{'loss': 0.5101, 'learning_rate': 4.972892978596069e-05, 'epoch': 0.14}

  5%|▍         | 210/4476 [1:21:40<27:38:03, 23.32s/it]
  5%|▍         | 211/4476 [1:22:03<27:36:13, 23.30s/it]
  5%|▍         | 212/4476 [1:22:26<27:36:22, 23.31s/it]
  5%|▍         | 213/4476 [1:22:50<27:35:08, 23.30s/it]
  5%|▍         | 214/4476 [1:23:13<27:30:38, 23.24s/it]
  5%|▍         | 215/4476 [1:23:36<27:33:36, 23.28s/it]
  5%|▍         | 216/4476 [1:23:59<27:34:54, 23.31s/it]
  5%|▍         | 217/4476 [1:24:23<27:44:44, 23.45s/it]
  5%|▍         | 218/4476 [1:24:46<27:37:32, 23.36s/it]
  5%|▍         | 219/4476 [1:25:09<27:29:48, 23.25s/it]
  5%|▍         | 220/4476 [1:25:33<27:34:44, 23.33s/it]
                                                       
{'loss': 0.5125, 'learning_rate': 4.970255143517838e-05, 'epoch': 0.15}

  5%|▍         | 220/4476 [1:25:33<27:34:44, 23.33s/it]
  5%|▍         | 221/4476 [1:25:56<27:39:15, 23.40s/it]
  5%|▍         | 222/4476 [1:26:20<27:43:12, 23.46s/it]
  5%|▍         | 223/4476 [1:26:43<27:39:00, 23.40s/it]
  5%|▌         | 224/4476 [1:27:07<27:34:53, 23.35s/it]
  5%|▌         | 225/4476 [1:27:29<27:26:02, 23.23s/it]
  5%|▌         | 226/4476 [1:27:53<27:21:25, 23.17s/it]
  5%|▌         | 227/4476 [1:28:16<27:23:04, 23.20s/it]
  5%|▌         | 228/4476 [1:28:39<27:21:39, 23.19s/it]
  5%|▌         | 229/4476 [1:29:02<27:26:17, 23.26s/it]
  5%|▌         | 230/4476 [1:29:26<27:29:14, 23.31s/it]
                                                       
{'loss': 0.4928, 'learning_rate': 4.967495617114826e-05, 'epoch': 0.15}

  5%|▌         | 230/4476 [1:29:26<27:29:14, 23.31s/it]
  5%|▌         | 231/4476 [1:29:49<27:24:13, 23.24s/it]
  5%|▌         | 232/4476 [1:30:13<27:33:14, 23.37s/it]
  5%|▌         | 233/4476 [1:30:36<27:38:20, 23.45s/it]
  5%|▌         | 234/4476 [1:31:00<27:39:44, 23.48s/it]
  5%|▌         | 235/4476 [1:31:23<27:29:54, 23.34s/it]
  5%|▌         | 236/4476 [1:31:46<27:27:43, 23.32s/it]
  5%|▌         | 237/4476 [1:32:09<27:26:27, 23.30s/it]
  5%|▌         | 238/4476 [1:32:32<27:21:40, 23.24s/it]
  5%|▌         | 239/4476 [1:32:56<27:28:34, 23.35s/it]
  5%|▌         | 240/4476 [1:33:19<27:26:20, 23.32s/it]
                                                       
{'loss': 0.4878, 'learning_rate': 4.964614535328626e-05, 'epoch': 0.16}

  5%|▌         | 240/4476 [1:33:19<27:26:20, 23.32s/it]
  5%|▌         | 241/4476 [1:33:43<27:33:03, 23.42s/it]
  5%|▌         | 242/4476 [1:34:06<27:19:22, 23.23s/it]
  5%|▌         | 243/4476 [1:34:29<27:20:32, 23.25s/it]
  5%|▌         | 244/4476 [1:34:52<27:25:15, 23.33s/it]
  5%|▌         | 245/4476 [1:35:16<27:24:01, 23.31s/it]
  5%|▌         | 246/4476 [1:35:40<27:33:40, 23.46s/it]
  6%|▌         | 247/4476 [1:36:03<27:28:35, 23.39s/it]
  6%|▌         | 248/4476 [1:36:26<27:27:10, 23.38s/it]
  6%|▌         | 249/4476 [1:36:49<27:19:06, 23.27s/it]
  6%|▌         | 250/4476 [1:37:13<27:21:35, 23.31s/it]
                                                       
{'loss': 0.5017, 'learning_rate': 4.961612040088973e-05, 'epoch': 0.17}

  6%|▌         | 250/4476 [1:37:13<27:21:35, 23.31s/it]
  6%|▌         | 251/4476 [1:37:36<27:22:20, 23.32s/it]
  6%|▌         | 252/4476 [1:37:59<27:25:39, 23.38s/it]
  6%|▌         | 253/4476 [1:38:23<27:27:44, 23.41s/it]
  6%|▌         | 254/4476 [1:38:46<27:31:09, 23.47s/it]
  6%|▌         | 255/4476 [1:39:10<27:31:13, 23.47s/it]
  6%|▌         | 256/4476 [1:39:33<27:30:32, 23.47s/it]
  6%|▌         | 257/4476 [1:39:57<27:30:55, 23.48s/it]
  6%|▌         | 258/4476 [1:40:20<27:20:51, 23.34s/it]
  6%|▌         | 259/4476 [1:40:43<27:23:39, 23.39s/it]
  6%|▌         | 260/4476 [1:41:07<27:20:34, 23.35s/it]
                                                       
{'loss': 0.4863, 'learning_rate': 4.9584882793067534e-05, 'epoch': 0.17}

  6%|▌         | 260/4476 [1:41:07<27:20:34, 23.35s/it]
  6%|▌         | 261/4476 [1:41:30<27:21:10, 23.36s/it]
  6%|▌         | 262/4476 [1:41:53<27:13:54, 23.26s/it]
  6%|▌         | 263/4476 [1:42:17<27:21:29, 23.38s/it]
  6%|▌         | 264/4476 [1:42:40<27:23:36, 23.41s/it]
  6%|▌         | 265/4476 [1:43:03<27:18:17, 23.34s/it]
  6%|▌         | 266/4476 [1:43:27<27:17:35, 23.34s/it]
  6%|▌         | 267/4476 [1:43:50<27:15:28, 23.31s/it]
  6%|▌         | 268/4476 [1:44:13<26:59:10, 23.09s/it]
  6%|▌         | 269/4476 [1:44:36<27:04:36, 23.17s/it]
  6%|▌         | 270/4476 [1:45:00<27:14:49, 23.32s/it]
                                                       
{'loss': 0.4847, 'learning_rate': 4.955243406866713e-05, 'epoch': 0.18}

  6%|▌         | 270/4476 [1:45:00<27:14:49, 23.32s/it]
  6%|▌         | 271/4476 [1:45:23<27:12:42, 23.30s/it]
  6%|▌         | 272/4476 [1:45:46<27:08:23, 23.24s/it]
  6%|▌         | 273/4476 [1:46:09<27:12:06, 23.30s/it]
  6%|▌         | 274/4476 [1:46:33<27:07:55, 23.25s/it]
  6%|▌         | 275/4476 [1:46:56<27:10:15, 23.28s/it]
  6%|▌         | 276/4476 [1:47:19<27:12:05, 23.32s/it]
  6%|▌         | 277/4476 [1:47:43<27:12:00, 23.32s/it]
  6%|▌         | 278/4476 [1:48:06<27:14:37, 23.36s/it]
  6%|▌         | 279/4476 [1:48:30<27:16:21, 23.39s/it]
  6%|▋         | 280/4476 [1:48:53<27:17:44, 23.42s/it]
                                                       
{'loss': 0.4868, 'learning_rate': 4.951877582619881e-05, 'epoch': 0.19}

  6%|▋         | 280/4476 [1:48:53<27:17:44, 23.42s/it]
  6%|▋         | 281/4476 [1:49:16<27:10:24, 23.32s/it]
  6%|▋         | 282/4476 [1:49:39<27:05:33, 23.26s/it]
  6%|▋         | 283/4476 [1:50:03<27:16:06, 23.41s/it]
  6%|▋         | 284/4476 [1:50:26<27:10:03, 23.33s/it]
  6%|▋         | 285/4476 [1:50:49<27:09:33, 23.33s/it]
  6%|▋         | 286/4476 [1:51:13<27:10:11, 23.34s/it]
  6%|▋         | 287/4476 [1:51:37<27:19:56, 23.49s/it]
  6%|▋         | 288/4476 [1:52:00<27:14:07, 23.41s/it]
  6%|▋         | 289/4476 [1:52:23<27:09:17, 23.35s/it]
  6%|▋         | 290/4476 [1:52:46<27:05:40, 23.30s/it]
                                                       
{'loss': 0.4748, 'learning_rate': 4.948390972375694e-05, 'epoch': 0.19}

  6%|▋         | 290/4476 [1:52:46<27:05:40, 23.30s/it]
  7%|▋         | 291/4476 [1:53:10<27:06:50, 23.32s/it]
  7%|▋         | 292/4476 [1:53:33<27:06:48, 23.33s/it]
  7%|▋         | 293/4476 [1:53:57<27:10:04, 23.38s/it]
  7%|▋         | 294/4476 [1:54:20<27:16:30, 23.48s/it]
  7%|▋         | 295/4476 [1:54:44<27:17:19, 23.50s/it]
  7%|▋         | 296/4476 [1:55:07<27:10:16, 23.40s/it]
  7%|▋         | 297/4476 [1:55:30<27:10:20, 23.41s/it]
  7%|▋         | 298/4476 [1:55:54<27:07:44, 23.38s/it]
  7%|▋         | 299/4476 [1:56:17<27:11:11, 23.43s/it]
  7%|▋         | 300/4476 [1:56:41<27:09:05, 23.41s/it]
                                                       
{'loss': 0.4764, 'learning_rate': 4.944783747893825e-05, 'epoch': 0.2}

  7%|▋         | 300/4476 [1:56:41<27:09:05, 23.41s/it]
  7%|▋         | 301/4476 [1:57:04<27:15:01, 23.50s/it]
  7%|▋         | 302/4476 [1:57:28<27:10:46, 23.44s/it]
  7%|▋         | 303/4476 [1:57:51<27:09:00, 23.42s/it]
  7%|▋         | 304/4476 [1:58:15<27:11:38, 23.47s/it]
  7%|▋         | 305/4476 [1:58:38<27:09:45, 23.44s/it]
  7%|▋         | 306/4476 [1:59:01<27:04:55, 23.38s/it]
  7%|▋         | 307/4476 [1:59:25<27:03:53, 23.37s/it]
  7%|▋         | 308/4476 [1:59:48<27:07:49, 23.43s/it]
  7%|▋         | 309/4476 [2:00:12<27:11:55, 23.50s/it]
  7%|▋         | 310/4476 [2:00:35<27:15:34, 23.56s/it]
                                                       
{'loss': 0.4712, 'learning_rate': 4.941056086875727e-05, 'epoch': 0.21}

  7%|▋         | 310/4476 [2:00:35<27:15:34, 23.56s/it]
  7%|▋         | 311/4476 [2:00:59<27:14:08, 23.54s/it]
  7%|▋         | 312/4476 [2:01:22<27:12:13, 23.52s/it]
  7%|▋         | 313/4476 [2:01:46<27:10:08, 23.49s/it]
  7%|▋         | 314/4476 [2:02:09<27:05:44, 23.44s/it]
  7%|▋         | 315/4476 [2:02:32<27:02:00, 23.39s/it]
  7%|▋         | 316/4476 [2:02:56<26:58:13, 23.34s/it]
  7%|▋         | 317/4476 [2:03:19<26:57:05, 23.33s/it]
  7%|▋         | 318/4476 [2:03:42<26:53:40, 23.29s/it]
  7%|▋         | 319/4476 [2:04:06<26:56:18, 23.33s/it]
  7%|▋         | 320/4476 [2:04:29<26:58:26, 23.37s/it]
                                                       
{'loss': 0.4642, 'learning_rate': 4.937208172955876e-05, 'epoch': 0.21}

  7%|▋         | 320/4476 [2:04:29<26:58:26, 23.37s/it]
  7%|▋         | 321/4476 [2:04:52<26:56:37, 23.34s/it]
  7%|▋         | 322/4476 [2:05:16<26:58:27, 23.38s/it]
  7%|▋         | 323/4476 [2:05:39<27:01:34, 23.43s/it]
  7%|▋         | 324/4476 [2:06:02<26:52:40, 23.30s/it]
  7%|▋         | 325/4476 [2:06:25<26:48:59, 23.26s/it]
  7%|▋         | 326/4476 [2:06:49<26:58:25, 23.40s/it]
  7%|▋         | 327/4476 [2:07:13<26:59:01, 23.41s/it]
  7%|▋         | 328/4476 [2:07:36<26:59:02, 23.42s/it]
  7%|▋         | 329/4476 [2:08:00<27:00:47, 23.45s/it]
  7%|▋         | 330/4476 [2:08:23<26:57:05, 23.40s/it]
                                                       
{'loss': 0.4642, 'learning_rate': 4.9332401956927224e-05, 'epoch': 0.22}

  7%|▋         | 330/4476 [2:08:23<26:57:05, 23.40s/it]
  7%|▋         | 331/4476 [2:08:47<27:05:27, 23.53s/it]
  7%|▋         | 332/4476 [2:09:10<27:05:41, 23.54s/it]
  7%|▋         | 333/4476 [2:09:34<27:03:08, 23.51s/it]
  7%|▋         | 334/4476 [2:09:57<26:57:46, 23.43s/it]
  7%|▋         | 335/4476 [2:10:20<26:54:21, 23.39s/it]
  8%|▊         | 336/4476 [2:10:43<26:47:32, 23.30s/it]
  8%|▊         | 337/4476 [2:11:06<26:42:13, 23.23s/it]
  8%|▊         | 338/4476 [2:11:29<26:33:29, 23.11s/it]
  8%|▊         | 339/4476 [2:11:53<26:36:59, 23.16s/it]
  8%|▊         | 340/4476 [2:12:16<26:35:51, 23.15s/it]
                                                       
{'loss': 0.4709, 'learning_rate': 4.9291523505593604e-05, 'epoch': 0.23}

  8%|▊         | 340/4476 [2:12:16<26:35:51, 23.15s/it]
  8%|▊         | 341/4476 [2:12:39<26:30:23, 23.08s/it]
  8%|▊         | 342/4476 [2:13:02<26:37:03, 23.18s/it]
  8%|▊         | 343/4476 [2:13:25<26:40:22, 23.23s/it]
  8%|▊         | 344/4476 [2:13:49<26:43:16, 23.28s/it]
  8%|▊         | 345/4476 [2:14:12<26:41:38, 23.26s/it]
  8%|▊         | 346/4476 [2:14:36<26:49:08, 23.38s/it]
  8%|▊         | 347/4476 [2:14:59<26:51:18, 23.41s/it]
  8%|▊         | 348/4476 [2:15:22<26:47:53, 23.37s/it]
  8%|▊         | 349/4476 [2:15:46<26:51:16, 23.43s/it]
  8%|▊         | 350/4476 [2:16:09<26:49:53, 23.41s/it]
                                                       
{'loss': 0.461, 'learning_rate': 4.9249448389338905e-05, 'epoch': 0.23}

  8%|▊         | 350/4476 [2:16:09<26:49:53, 23.41s/it]
  8%|▊         | 351/4476 [2:16:33<26:53:21, 23.47s/it]
  8%|▊         | 352/4476 [2:16:56<26:47:37, 23.39s/it]
  8%|▊         | 353/4476 [2:17:19<26:39:45, 23.28s/it]
  8%|▊         | 354/4476 [2:17:42<26:39:51, 23.29s/it]
  8%|▊         | 355/4476 [2:18:06<26:48:17, 23.42s/it]
  8%|▊         | 356/4476 [2:18:30<26:55:32, 23.53s/it]
  8%|▊         | 357/4476 [2:18:53<26:49:35, 23.45s/it]
  8%|▊         | 358/4476 [2:19:17<26:49:15, 23.45s/it]
  8%|▊         | 359/4476 [2:19:40<26:55:47, 23.55s/it]
  8%|▊         | 360/4476 [2:20:04<26:49:49, 23.47s/it]
                                                       
{'loss': 0.4677, 'learning_rate': 4.920617868089501e-05, 'epoch': 0.24}

  8%|▊         | 360/4476 [2:20:04<26:49:49, 23.47s/it]
  8%|▊         | 361/4476 [2:20:27<26:43:51, 23.39s/it]
  8%|▊         | 362/4476 [2:20:51<26:51:24, 23.50s/it]
  8%|▊         | 363/4476 [2:21:14<26:48:45, 23.47s/it]
  8%|▊         | 364/4476 [2:21:38<26:53:59, 23.55s/it]
  8%|▊         | 365/4476 [2:22:01<26:52:37, 23.54s/it]
  8%|▊         | 366/4476 [2:22:25<26:56:21, 23.60s/it]
  8%|▊         | 367/4476 [2:22:48<26:50:19, 23.51s/it]
  8%|▊         | 368/4476 [2:23:12<26:54:20, 23.58s/it]
  8%|▊         | 369/4476 [2:23:35<26:47:39, 23.49s/it]
  8%|▊         | 370/4476 [2:23:59<26:44:43, 23.45s/it]
                                                       
{'loss': 0.4564, 'learning_rate': 4.9161716511842614e-05, 'epoch': 0.25}

  8%|▊         | 370/4476 [2:23:59<26:44:43, 23.45s/it]
  8%|▊         | 371/4476 [2:24:22<26:41:54, 23.41s/it]
  8%|▊         | 372/4476 [2:24:45<26:39:41, 23.39s/it]
  8%|▊         | 373/4476 [2:25:09<26:40:55, 23.41s/it]
  8%|▊         | 374/4476 [2:25:32<26:41:10, 23.42s/it]
  8%|▊         | 375/4476 [2:25:56<26:36:47, 23.36s/it]
  8%|▊         | 376/4476 [2:26:19<26:37:03, 23.37s/it]
  8%|▊         | 377/4476 [2:26:42<26:37:12, 23.38s/it]
  8%|▊         | 378/4476 [2:27:06<26:38:07, 23.40s/it]
  8%|▊         | 379/4476 [2:27:29<26:34:24, 23.35s/it]
  8%|▊         | 380/4476 [2:27:52<26:27:58, 23.26s/it]
                                                       
{'loss': 0.4663, 'learning_rate': 4.911606407250617e-05, 'epoch': 0.25}

  8%|▊         | 380/4476 [2:27:52<26:27:58, 23.26s/it]
  9%|▊         | 381/4476 [2:28:16<26:34:55, 23.37s/it]
  9%|▊         | 382/4476 [2:28:39<26:33:09, 23.35s/it]
  9%|▊         | 383/4476 [2:29:02<26:27:08, 23.27s/it]
  9%|▊         | 384/4476 [2:29:25<26:25:53, 23.25s/it]
  9%|▊         | 385/4476 [2:29:48<26:18:18, 23.15s/it]
  9%|▊         | 386/4476 [2:30:11<26:19:40, 23.17s/it]
  9%|▊         | 387/4476 [2:30:35<26:26:28, 23.28s/it]
  9%|▊         | 388/4476 [2:30:58<26:31:41, 23.36s/it]
  9%|▊         | 389/4476 [2:31:22<26:29:45, 23.34s/it]
  9%|▊         | 390/4476 [2:31:45<26:31:58, 23.38s/it]
                                                       
{'loss': 0.4682, 'learning_rate': 4.9069223611846014e-05, 'epoch': 0.26}

  9%|▊         | 390/4476 [2:31:45<26:31:58, 23.38s/it]
  9%|▊         | 391/4476 [2:32:08<26:25:41, 23.29s/it]
  9%|▉         | 392/4476 [2:32:31<26:20:01, 23.21s/it]
  9%|▉         | 393/4476 [2:32:55<26:24:30, 23.28s/it]
  9%|▉         | 394/4476 [2:33:19<26:33:00, 23.42s/it]
  9%|▉         | 395/4476 [2:33:42<26:30:29, 23.38s/it]
  9%|▉         | 396/4476 [2:34:05<26:23:48, 23.29s/it]
  9%|▉         | 397/4476 [2:34:28<26:20:25, 23.25s/it]
  9%|▉         | 398/4476 [2:34:52<26:24:59, 23.32s/it]
  9%|▉         | 399/4476 [2:35:15<26:21:54, 23.28s/it]
  9%|▉         | 400/4476 [2:35:38<26:15:48, 23.20s/it]
                                                       
{'loss': 0.4636, 'learning_rate': 4.9021197437347555e-05, 'epoch': 0.27}

  9%|▉         | 400/4476 [2:35:38<26:15:48, 23.20s/it][INFO|trainer.py:2939] 2023-11-12 05:59:25,475 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-400
[INFO|tokenization_utils_base.py:2437] 2023-11-12 05:59:25,507 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-400/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 05:59:25,507 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-400/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 05:59:25,507 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-400/added_tokens.json

  9%|▉         | 401/4476 [2:36:01<26:26:55, 23.37s/it]
  9%|▉         | 402/4476 [2:36:25<26:33:27, 23.47s/it]
  9%|▉         | 403/4476 [2:36:48<26:23:21, 23.32s/it]
  9%|▉         | 404/4476 [2:37:12<26:23:54, 23.34s/it]
  9%|▉         | 405/4476 [2:37:35<26:25:03, 23.36s/it]
  9%|▉         | 406/4476 [2:37:58<26:24:19, 23.36s/it]
  9%|▉         | 407/4476 [2:38:22<26:25:51, 23.38s/it]
  9%|▉         | 408/4476 [2:38:45<26:26:19, 23.40s/it]
  9%|▉         | 409/4476 [2:39:08<26:22:15, 23.34s/it]
  9%|▉         | 410/4476 [2:39:32<26:24:35, 23.38s/it]
                                                       
{'loss': 0.4569, 'learning_rate': 4.897198791490762e-05, 'epoch': 0.27}

  9%|▉         | 410/4476 [2:39:32<26:24:35, 23.38s/it]
  9%|▉         | 411/4476 [2:39:56<26:30:09, 23.47s/it]
  9%|▉         | 412/4476 [2:40:19<26:27:15, 23.43s/it]
  9%|▉         | 413/4476 [2:40:42<26:17:33, 23.30s/it]
  9%|▉         | 414/4476 [2:41:05<26:17:09, 23.30s/it]
  9%|▉         | 415/4476 [2:41:28<26:14:59, 23.27s/it]
  9%|▉         | 416/4476 [2:41:52<26:22:44, 23.39s/it]
  9%|▉         | 417/4476 [2:42:16<26:23:50, 23.41s/it]
  9%|▉         | 418/4476 [2:42:39<26:21:46, 23.39s/it]
  9%|▉         | 419/4476 [2:43:02<26:18:28, 23.34s/it]
  9%|▉         | 420/4476 [2:43:25<26:17:45, 23.34s/it]
                                                       
{'loss': 0.462, 'learning_rate': 4.8921597468717887e-05, 'epoch': 0.28}

  9%|▉         | 420/4476 [2:43:25<26:17:45, 23.34s/it]
  9%|▉         | 421/4476 [2:43:49<26:24:13, 23.44s/it]
  9%|▉         | 422/4476 [2:44:13<26:26:08, 23.48s/it]
  9%|▉         | 423/4476 [2:44:36<26:25:53, 23.48s/it]
  9%|▉         | 424/4476 [2:44:59<26:16:20, 23.34s/it]
  9%|▉         | 425/4476 [2:45:23<26:20:44, 23.41s/it]
 10%|▉         | 426/4476 [2:45:47<26:28:46, 23.54s/it]
 10%|▉         | 427/4476 [2:46:10<26:28:21, 23.54s/it]
 10%|▉         | 428/4476 [2:46:33<26:23:10, 23.47s/it]
 10%|▉         | 429/4476 [2:46:57<26:24:36, 23.49s/it]
 10%|▉         | 430/4476 [2:47:20<26:21:42, 23.46s/it]
                                                       
{'loss': 0.4563, 'learning_rate': 4.887002858114548e-05, 'epoch': 0.29}

 10%|▉         | 430/4476 [2:47:20<26:21:42, 23.46s/it]
 10%|▉         | 431/4476 [2:47:44<26:19:14, 23.43s/it]
 10%|▉         | 432/4476 [2:48:07<26:15:47, 23.38s/it]
 10%|▉         | 433/4476 [2:48:30<26:14:31, 23.37s/it]
 10%|▉         | 434/4476 [2:48:54<26:20:53, 23.47s/it]
 10%|▉         | 435/4476 [2:49:18<26:22:14, 23.49s/it]
 10%|▉         | 436/4476 [2:49:41<26:17:25, 23.43s/it]
 10%|▉         | 437/4476 [2:50:04<26:17:22, 23.43s/it]
 10%|▉         | 438/4476 [2:50:27<26:10:14, 23.33s/it]
 10%|▉         | 439/4476 [2:50:51<26:09:03, 23.32s/it]
 10%|▉         | 440/4476 [2:51:14<26:07:38, 23.30s/it]
                                                       
{'loss': 0.4563, 'learning_rate': 4.881728379261068e-05, 'epoch': 0.29}

 10%|▉         | 440/4476 [2:51:14<26:07:38, 23.30s/it]
 10%|▉         | 441/4476 [2:51:38<26:14:48, 23.42s/it]
 10%|▉         | 442/4476 [2:52:01<26:08:35, 23.33s/it]
 10%|▉         | 443/4476 [2:52:24<26:12:55, 23.40s/it]
 10%|▉         | 444/4476 [2:52:47<26:06:35, 23.31s/it]
 10%|▉         | 445/4476 [2:53:11<26:05:04, 23.30s/it]
 10%|▉         | 446/4476 [2:53:34<26:09:43, 23.37s/it]
 10%|▉         | 447/4476 [2:53:57<26:05:19, 23.31s/it]
 10%|█         | 448/4476 [2:54:21<26:08:10, 23.36s/it]
 10%|█         | 449/4476 [2:54:44<26:03:41, 23.30s/it]
 10%|█         | 450/4476 [2:55:07<25:59:08, 23.24s/it]
                                                       
{'loss': 0.4468, 'learning_rate': 4.876336570146175e-05, 'epoch': 0.3}

 10%|█         | 450/4476 [2:55:07<25:59:08, 23.24s/it]
 10%|█         | 451/4476 [2:55:30<26:01:37, 23.28s/it]
 10%|█         | 452/4476 [2:55:54<26:00:57, 23.27s/it]
 10%|█         | 453/4476 [2:56:17<25:59:06, 23.25s/it]
 10%|█         | 454/4476 [2:56:40<26:00:53, 23.29s/it]
 10%|█         | 455/4476 [2:57:04<26:01:33, 23.30s/it]
 10%|█         | 456/4476 [2:57:27<26:03:03, 23.33s/it]
 10%|█         | 457/4476 [2:57:50<26:05:00, 23.36s/it]
 10%|█         | 458/4476 [2:58:14<25:58:24, 23.27s/it]
 10%|█         | 459/4476 [2:58:37<26:01:40, 23.33s/it]
 10%|█         | 460/4476 [2:59:01<26:05:50, 23.39s/it]
                                                       
{'loss': 0.4508, 'learning_rate': 4.870827696384698e-05, 'epoch': 0.31}

 10%|█         | 460/4476 [2:59:01<26:05:50, 23.39s/it]
 10%|█         | 461/4476 [2:59:24<26:06:49, 23.41s/it]
 10%|█         | 462/4476 [2:59:47<26:05:21, 23.40s/it]
 10%|█         | 463/4476 [3:00:11<26:06:00, 23.41s/it]
 10%|█         | 464/4476 [3:00:34<26:05:41, 23.42s/it]
 10%|█         | 465/4476 [3:00:58<26:07:45, 23.45s/it]
 10%|█         | 466/4476 [3:01:21<26:06:23, 23.44s/it]
 10%|█         | 467/4476 [3:01:44<25:58:01, 23.32s/it]
 10%|█         | 468/4476 [3:02:07<25:55:52, 23.29s/it]
 10%|█         | 469/4476 [3:02:31<25:57:51, 23.33s/it]
 11%|█         | 470/4476 [3:02:54<25:56:18, 23.31s/it]
                                                       
{'loss': 0.4507, 'learning_rate': 4.865202029358379e-05, 'epoch': 0.31}

 11%|█         | 470/4476 [3:02:54<25:56:18, 23.31s/it]
 11%|█         | 471/4476 [3:03:17<25:54:52, 23.29s/it]
 11%|█         | 472/4476 [3:03:41<26:04:02, 23.44s/it]
 11%|█         | 473/4476 [3:04:05<26:09:01, 23.52s/it]
 11%|█         | 474/4476 [3:04:29<26:11:00, 23.55s/it]
 11%|█         | 475/4476 [3:04:52<26:15:28, 23.63s/it]
 11%|█         | 476/4476 [3:05:15<26:04:12, 23.46s/it]
 11%|█         | 477/4476 [3:05:39<25:57:51, 23.37s/it]
 11%|█         | 478/4476 [3:06:02<25:53:35, 23.32s/it]
 11%|█         | 479/4476 [3:06:25<25:58:03, 23.39s/it]
 11%|█         | 480/4476 [3:06:49<25:56:46, 23.37s/it]
                                                       
{'loss': 0.4486, 'learning_rate': 4.859459846202507e-05, 'epoch': 0.32}

 11%|█         | 480/4476 [3:06:49<25:56:46, 23.37s/it]
 11%|█         | 481/4476 [3:07:12<25:51:03, 23.29s/it]
 11%|█         | 482/4476 [3:07:35<25:47:57, 23.25s/it]
 11%|█         | 483/4476 [3:07:58<25:50:13, 23.29s/it]
 11%|█         | 484/4476 [3:08:22<25:51:52, 23.32s/it]
 11%|█         | 485/4476 [3:08:45<25:47:48, 23.27s/it]
 11%|█         | 486/4476 [3:09:08<25:45:46, 23.24s/it]
 11%|█         | 487/4476 [3:09:31<25:44:49, 23.24s/it]
 11%|█         | 488/4476 [3:09:55<25:48:01, 23.29s/it]
 11%|█         | 489/4476 [3:10:18<25:39:50, 23.17s/it]
 11%|█         | 490/4476 [3:10:41<25:41:42, 23.21s/it]
                                                       
{'loss': 0.4423, 'learning_rate': 4.853601429792265e-05, 'epoch': 0.33}

 11%|█         | 490/4476 [3:10:41<25:41:42, 23.21s/it]
 11%|█         | 491/4476 [3:11:04<25:39:06, 23.17s/it]
 11%|█         | 492/4476 [3:11:27<25:38:49, 23.18s/it]
 11%|█         | 493/4476 [3:11:50<25:39:06, 23.19s/it]
 11%|█         | 494/4476 [3:12:14<25:39:58, 23.20s/it]
 11%|█         | 495/4476 [3:12:37<25:40:33, 23.22s/it]
 11%|█         | 496/4476 [3:13:00<25:35:46, 23.15s/it]
 11%|█         | 497/4476 [3:13:23<25:40:41, 23.23s/it]
 11%|█         | 498/4476 [3:13:47<25:45:52, 23.32s/it]
 11%|█         | 499/4476 [3:14:10<25:45:16, 23.31s/it]
 11%|█         | 500/4476 [3:14:34<25:49:12, 23.38s/it]
                                                       
{'loss': 0.4369, 'learning_rate': 4.847627068728795e-05, 'epoch': 0.34}

 11%|█         | 500/4476 [3:14:34<25:49:12, 23.38s/it]
 11%|█         | 501/4476 [3:14:57<25:47:56, 23.37s/it]
 11%|█         | 502/4476 [3:15:20<25:50:26, 23.41s/it]
 11%|█         | 503/4476 [3:15:44<25:47:51, 23.38s/it]
 11%|█▏        | 504/4476 [3:16:07<25:52:02, 23.44s/it]
 11%|█▏        | 505/4476 [3:16:31<25:47:29, 23.38s/it]
 11%|█▏        | 506/4476 [3:16:54<25:42:12, 23.31s/it]
 11%|█▏        | 507/4476 [3:17:17<25:48:23, 23.41s/it]
 11%|█▏        | 508/4476 [3:17:41<25:47:37, 23.40s/it]
 11%|█▏        | 509/4476 [3:18:04<25:45:38, 23.38s/it]
 11%|█▏        | 510/4476 [3:18:28<25:47:43, 23.41s/it]
                                                       
{'loss': 0.4429, 'learning_rate': 4.841537057324979e-05, 'epoch': 0.34}

 11%|█▏        | 510/4476 [3:18:28<25:47:43, 23.41s/it]
 11%|█▏        | 511/4476 [3:18:51<25:47:31, 23.42s/it]
 11%|█▏        | 512/4476 [3:19:14<25:44:47, 23.38s/it]
 11%|█▏        | 513/4476 [3:19:37<25:37:15, 23.27s/it]
 11%|█▏        | 514/4476 [3:20:01<25:40:41, 23.33s/it]
 12%|█▏        | 515/4476 [3:20:24<25:38:47, 23.31s/it]
 12%|█▏        | 516/4476 [3:20:48<25:43:33, 23.39s/it]
 12%|█▏        | 517/4476 [3:21:11<25:37:56, 23.31s/it]
 12%|█▏        | 518/4476 [3:21:34<25:36:13, 23.29s/it]
 12%|█▏        | 519/4476 [3:21:57<25:40:21, 23.36s/it]
 12%|█▏        | 520/4476 [3:22:21<25:48:45, 23.49s/it]
                                                       
{'loss': 0.4389, 'learning_rate': 4.835331695590943e-05, 'epoch': 0.35}

 12%|█▏        | 520/4476 [3:22:21<25:48:45, 23.49s/it]
 12%|█▏        | 521/4476 [3:22:45<25:43:36, 23.42s/it]
 12%|█▏        | 522/4476 [3:23:08<25:40:35, 23.38s/it]
 12%|█▏        | 523/4476 [3:23:31<25:42:01, 23.41s/it]
 12%|█▏        | 524/4476 [3:23:54<25:36:01, 23.32s/it]
 12%|█▏        | 525/4476 [3:24:18<25:35:16, 23.31s/it]
 12%|█▏        | 526/4476 [3:24:41<25:35:49, 23.33s/it]
 12%|█▏        | 527/4476 [3:25:04<25:36:52, 23.35s/it]
 12%|█▏        | 528/4476 [3:25:28<25:45:20, 23.49s/it]
 12%|█▏        | 529/4476 [3:25:51<25:39:52, 23.41s/it]
 12%|█▏        | 530/4476 [3:26:15<25:41:15, 23.44s/it]
                                                       
{'loss': 0.44, 'learning_rate': 4.829011289219276e-05, 'epoch': 0.36}

 12%|█▏        | 530/4476 [3:26:15<25:41:15, 23.44s/it]
 12%|█▏        | 531/4476 [3:26:38<25:39:05, 23.41s/it]
 12%|█▏        | 532/4476 [3:27:02<25:42:47, 23.47s/it]
 12%|█▏        | 533/4476 [3:27:25<25:43:46, 23.49s/it]
 12%|█▏        | 534/4476 [3:27:49<25:40:31, 23.45s/it]
 12%|█▏        | 535/4476 [3:28:13<25:44:26, 23.51s/it]
 12%|█▏        | 536/4476 [3:28:36<25:36:38, 23.40s/it]
 12%|█▏        | 537/4476 [3:28:59<25:43:35, 23.51s/it]
 12%|█▏        | 538/4476 [3:29:23<25:38:58, 23.45s/it]
 12%|█▏        | 539/4476 [3:29:46<25:30:54, 23.33s/it]
 12%|█▏        | 540/4476 [3:30:09<25:30:56, 23.34s/it]
                                                       
{'loss': 0.4476, 'learning_rate': 4.82257614956997e-05, 'epoch': 0.36}

 12%|█▏        | 540/4476 [3:30:09<25:30:56, 23.34s/it]
 12%|█▏        | 541/4476 [3:30:32<25:30:35, 23.34s/it]
 12%|█▏        | 542/4476 [3:30:56<25:31:01, 23.35s/it]
 12%|█▏        | 543/4476 [3:31:19<25:28:05, 23.31s/it]
 12%|█▏        | 544/4476 [3:31:42<25:28:05, 23.32s/it]
 12%|█▏        | 545/4476 [3:32:06<25:26:07, 23.29s/it]
 12%|█▏        | 546/4476 [3:32:29<25:29:25, 23.35s/it]
 12%|█▏        | 547/4476 [3:32:52<25:28:42, 23.35s/it]
 12%|█▏        | 548/4476 [3:33:16<25:36:18, 23.47s/it]
 12%|█▏        | 549/4476 [3:33:39<25:32:01, 23.41s/it]
 12%|█▏        | 550/4476 [3:34:03<25:33:50, 23.44s/it]
                                                       
{'loss': 0.4367, 'learning_rate': 4.816026593655085e-05, 'epoch': 0.37}

 12%|█▏        | 550/4476 [3:34:03<25:33:50, 23.44s/it]
 12%|█▏        | 551/4476 [3:34:26<25:29:15, 23.38s/it]
 12%|█▏        | 552/4476 [3:34:49<25:23:32, 23.30s/it]
 12%|█▏        | 553/4476 [3:35:12<25:17:49, 23.21s/it]
 12%|█▏        | 554/4476 [3:35:36<25:19:31, 23.25s/it]
 12%|█▏        | 555/4476 [3:35:59<25:22:41, 23.30s/it]
 12%|█▏        | 556/4476 [3:36:22<25:19:48, 23.26s/it]
 12%|█▏        | 557/4476 [3:36:46<25:24:10, 23.34s/it]
 12%|█▏        | 558/4476 [3:37:09<25:17:35, 23.24s/it]
 12%|█▏        | 559/4476 [3:37:32<25:23:11, 23.33s/it]
 13%|█▎        | 560/4476 [3:37:56<25:29:18, 23.43s/it]
                                                       
{'loss': 0.4357, 'learning_rate': 4.809362944123129e-05, 'epoch': 0.38}

 13%|█▎        | 560/4476 [3:37:56<25:29:18, 23.43s/it]
 13%|█▎        | 561/4476 [3:38:20<25:36:37, 23.55s/it]
 13%|█▎        | 562/4476 [3:38:43<25:35:09, 23.53s/it]
 13%|█▎        | 563/4476 [3:39:07<25:30:46, 23.47s/it]
 13%|█▎        | 564/4476 [3:39:30<25:36:51, 23.57s/it]
 13%|█▎        | 565/4476 [3:39:53<25:25:00, 23.40s/it]
 13%|█▎        | 566/4476 [3:40:17<25:19:57, 23.32s/it]
 13%|█▎        | 567/4476 [3:40:40<25:26:35, 23.43s/it]
 13%|█▎        | 568/4476 [3:41:04<25:26:38, 23.44s/it]
 13%|█▎        | 569/4476 [3:41:27<25:23:24, 23.40s/it]
 13%|█▎        | 570/4476 [3:41:50<25:19:26, 23.34s/it]
                                                       
{'loss': 0.4492, 'learning_rate': 4.802585529243164e-05, 'epoch': 0.38}

 13%|█▎        | 570/4476 [3:41:50<25:19:26, 23.34s/it]
 13%|█▎        | 571/4476 [3:42:13<25:12:18, 23.24s/it]
 13%|█▎        | 572/4476 [3:42:37<25:16:56, 23.31s/it]
 13%|█▎        | 573/4476 [3:43:00<25:20:01, 23.37s/it]
 13%|█▎        | 574/4476 [3:43:24<25:20:38, 23.38s/it]
 13%|█▎        | 575/4476 [3:43:47<25:12:25, 23.26s/it]
 13%|█▎        | 576/4476 [3:44:10<25:16:32, 23.33s/it]
 13%|█▎        | 577/4476 [3:44:34<25:20:20, 23.40s/it]
 13%|█▎        | 578/4476 [3:44:57<25:18:18, 23.37s/it]
 13%|█▎        | 579/4476 [3:45:20<25:17:23, 23.36s/it]
 13%|█▎        | 580/4476 [3:45:44<25:25:06, 23.49s/it]
                                                       
{'loss': 0.4403, 'learning_rate': 4.795694682888635e-05, 'epoch': 0.39}

 13%|█▎        | 580/4476 [3:45:44<25:25:06, 23.49s/it]
 13%|█▎        | 581/4476 [3:46:07<25:17:40, 23.38s/it]
 13%|█▎        | 582/4476 [3:46:30<25:14:42, 23.34s/it]
 13%|█▎        | 583/4476 [3:46:54<25:14:01, 23.33s/it]
 13%|█▎        | 584/4476 [3:47:17<25:16:55, 23.39s/it]
 13%|█▎        | 585/4476 [3:47:41<25:14:00, 23.35s/it]
 13%|█▎        | 586/4476 [3:48:04<25:14:43, 23.36s/it]
 13%|█▎        | 587/4476 [3:48:27<25:17:44, 23.42s/it]
 13%|█▎        | 588/4476 [3:48:51<25:14:53, 23.38s/it]
 13%|█▎        | 589/4476 [3:49:14<25:10:18, 23.31s/it]
 13%|█▎        | 590/4476 [3:49:37<25:02:22, 23.20s/it]
                                                       
{'loss': 0.4406, 'learning_rate': 4.7886907445209234e-05, 'epoch': 0.4}

 13%|█▎        | 590/4476 [3:49:37<25:02:22, 23.20s/it]
 13%|█▎        | 591/4476 [3:50:00<24:59:08, 23.15s/it]
 13%|█▎        | 592/4476 [3:50:24<25:10:16, 23.33s/it]
 13%|█▎        | 593/4476 [3:50:47<25:09:36, 23.33s/it]
 13%|█▎        | 594/4476 [3:51:10<25:06:08, 23.28s/it]
 13%|█▎        | 595/4476 [3:51:33<25:05:18, 23.27s/it]
 13%|█▎        | 596/4476 [3:51:57<25:08:21, 23.33s/it]
 13%|█▎        | 597/4476 [3:52:20<25:12:58, 23.40s/it]
 13%|█▎        | 598/4476 [3:52:44<25:15:28, 23.45s/it]
 13%|█▎        | 599/4476 [3:53:07<25:06:21, 23.31s/it]
 13%|█▎        | 600/4476 [3:53:31<25:10:21, 23.38s/it]
                                                       
{'loss': 0.4317, 'learning_rate': 4.781574059172621e-05, 'epoch': 0.4}

 13%|█▎        | 600/4476 [3:53:31<25:10:21, 23.38s/it][INFO|trainer.py:2939] 2023-11-12 07:17:18,259 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-600
[INFO|tokenization_utils_base.py:2437] 2023-11-12 07:17:18,290 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-600/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 07:17:18,290 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-600/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 07:17:18,290 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-600/added_tokens.json

 13%|█▎        | 601/4476 [3:53:54<25:03:31, 23.28s/it]
 13%|█▎        | 602/4476 [3:54:17<25:13:27, 23.44s/it]
 13%|█▎        | 603/4476 [3:54:41<25:08:54, 23.38s/it]
 13%|█▎        | 604/4476 [3:55:04<25:11:36, 23.42s/it]
 14%|█▎        | 605/4476 [3:55:27<24:59:09, 23.24s/it]
 14%|█▎        | 606/4476 [3:55:50<24:50:56, 23.12s/it]
 14%|█▎        | 607/4476 [3:56:13<24:53:51, 23.17s/it]
 14%|█▎        | 608/4476 [3:56:37<24:58:54, 23.25s/it]
 14%|█▎        | 609/4476 [3:57:00<25:08:42, 23.41s/it]
 14%|█▎        | 610/4476 [3:57:24<25:15:34, 23.52s/it]
                                                       
{'loss': 0.4379, 'learning_rate': 4.7743449774305386e-05, 'epoch': 0.41}

 14%|█▎        | 610/4476 [3:57:24<25:15:34, 23.52s/it]
 14%|█▎        | 611/4476 [3:57:47<25:11:11, 23.46s/it]
 14%|█▎        | 612/4476 [3:58:11<25:14:14, 23.51s/it]
 14%|█▎        | 613/4476 [3:58:34<25:06:30, 23.40s/it]
 14%|█▎        | 614/4476 [3:58:58<25:05:47, 23.39s/it]
 14%|█▎        | 615/4476 [3:59:21<25:12:13, 23.50s/it]
 14%|█▍        | 616/4476 [3:59:45<25:14:59, 23.55s/it]
 14%|█▍        | 617/4476 [4:00:08<25:11:27, 23.50s/it]
 14%|█▍        | 618/4476 [4:00:32<25:11:34, 23.51s/it]
 14%|█▍        | 619/4476 [4:00:55<25:07:15, 23.45s/it]
 14%|█▍        | 620/4476 [4:01:19<25:09:01, 23.48s/it]
                                                       
{'loss': 0.4324, 'learning_rate': 4.7670038554184296e-05, 'epoch': 0.42}

 14%|█▍        | 620/4476 [4:01:19<25:09:01, 23.48s/it]
 14%|█▍        | 621/4476 [4:01:42<25:06:03, 23.44s/it]
 14%|█▍        | 622/4476 [4:02:05<25:02:50, 23.40s/it]
 14%|█▍        | 623/4476 [4:02:29<25:00:39, 23.37s/it]
 14%|█▍        | 624/4476 [4:02:52<24:58:55, 23.35s/it]
 14%|█▍        | 625/4476 [4:03:15<24:57:38, 23.33s/it]
 14%|█▍        | 626/4476 [4:03:39<25:01:36, 23.40s/it]
 14%|█▍        | 627/4476 [4:04:02<24:59:48, 23.38s/it]
 14%|█▍        | 628/4476 [4:04:25<24:54:01, 23.30s/it]
 14%|█▍        | 629/4476 [4:04:49<24:55:03, 23.32s/it]
 14%|█▍        | 630/4476 [4:05:12<25:00:06, 23.40s/it]
                                                       
{'loss': 0.4329, 'learning_rate': 4.7595510547794465e-05, 'epoch': 0.42}

 14%|█▍        | 630/4476 [4:05:12<25:00:06, 23.40s/it]
 14%|█▍        | 631/4476 [4:05:36<25:06:44, 23.51s/it]
 14%|█▍        | 632/4476 [4:05:59<24:58:23, 23.39s/it]
 14%|█▍        | 633/4476 [4:06:22<24:55:13, 23.34s/it]
 14%|█▍        | 634/4476 [4:06:46<24:53:04, 23.32s/it]
 14%|█▍        | 635/4476 [4:07:09<24:54:10, 23.34s/it]
 14%|█▍        | 636/4476 [4:07:32<24:55:39, 23.37s/it]
 14%|█▍        | 637/4476 [4:07:56<25:03:07, 23.49s/it]
 14%|█▍        | 638/4476 [4:08:20<25:02:48, 23.49s/it]
 14%|█▍        | 639/4476 [4:08:43<25:06:39, 23.56s/it]
 14%|█▍        | 640/4476 [4:09:07<24:58:40, 23.44s/it]
                                                       
{'loss': 0.4259, 'learning_rate': 4.751986942658332e-05, 'epoch': 0.43}

 14%|█▍        | 640/4476 [4:09:07<24:58:40, 23.44s/it]
 14%|█▍        | 641/4476 [4:09:30<24:55:36, 23.40s/it]
 14%|█▍        | 642/4476 [4:09:53<24:50:38, 23.33s/it]
 14%|█▍        | 643/4476 [4:10:17<24:58:35, 23.46s/it]
 14%|█▍        | 644/4476 [4:10:40<24:59:15, 23.47s/it]
 14%|█▍        | 645/4476 [4:11:04<24:59:32, 23.49s/it]
 14%|█▍        | 646/4476 [4:11:27<24:57:56, 23.47s/it]
 14%|█▍        | 647/4476 [4:11:50<24:53:14, 23.40s/it]
 14%|█▍        | 648/4476 [4:12:14<24:58:51, 23.49s/it]
 14%|█▍        | 649/4476 [4:12:38<24:56:12, 23.46s/it]
 15%|█▍        | 650/4476 [4:13:01<24:52:22, 23.40s/it]
                                                       
{'loss': 0.4256, 'learning_rate': 4.744311891683325e-05, 'epoch': 0.44}

 15%|█▍        | 650/4476 [4:13:01<24:52:22, 23.40s/it]
 15%|█▍        | 651/4476 [4:13:24<24:46:33, 23.32s/it]
 15%|█▍        | 652/4476 [4:13:47<24:48:00, 23.35s/it]
 15%|█▍        | 653/4476 [4:14:11<24:53:05, 23.43s/it]
 15%|█▍        | 654/4476 [4:14:34<24:46:38, 23.34s/it]
 15%|█▍        | 655/4476 [4:14:58<24:49:57, 23.40s/it]
 15%|█▍        | 656/4476 [4:15:21<24:43:00, 23.29s/it]
 15%|█▍        | 657/4476 [4:15:44<24:44:13, 23.32s/it]
 15%|█▍        | 658/4476 [4:16:08<24:45:32, 23.35s/it]
 15%|█▍        | 659/4476 [4:16:31<24:48:55, 23.40s/it]
 15%|█▍        | 660/4476 [4:16:55<24:54:23, 23.50s/it]
                                                       
{'loss': 0.4289, 'learning_rate': 4.736526279947807e-05, 'epoch': 0.44}

 15%|█▍        | 660/4476 [4:16:55<24:54:23, 23.50s/it]
 15%|█▍        | 661/4476 [4:17:18<24:58:11, 23.56s/it]
 15%|█▍        | 662/4476 [4:17:42<24:49:17, 23.43s/it]
 15%|█▍        | 663/4476 [4:18:05<24:44:59, 23.37s/it]
 15%|█▍        | 664/4476 [4:18:28<24:40:37, 23.30s/it]
 15%|█▍        | 665/4476 [4:18:51<24:35:03, 23.22s/it]
 15%|█▍        | 666/4476 [4:19:14<24:38:16, 23.28s/it]
 15%|█▍        | 667/4476 [4:19:38<24:37:00, 23.27s/it]
 15%|█▍        | 668/4476 [4:20:01<24:39:42, 23.31s/it]
 15%|█▍        | 669/4476 [4:20:25<24:44:24, 23.39s/it]
 15%|█▍        | 670/4476 [4:20:48<24:41:13, 23.35s/it]
                                                       
{'loss': 0.4353, 'learning_rate': 4.728630490991676e-05, 'epoch': 0.45}

 15%|█▍        | 670/4476 [4:20:48<24:41:13, 23.35s/it]
 15%|█▍        | 671/4476 [4:21:11<24:41:34, 23.36s/it]
 15%|█▌        | 672/4476 [4:21:35<24:39:08, 23.33s/it]
 15%|█▌        | 673/4476 [4:21:58<24:42:32, 23.39s/it]
 15%|█▌        | 674/4476 [4:22:21<24:40:54, 23.37s/it]
 15%|█▌        | 675/4476 [4:22:44<24:32:56, 23.25s/it]
 15%|█▌        | 676/4476 [4:23:08<24:38:26, 23.34s/it]
 15%|█▌        | 677/4476 [4:23:31<24:37:54, 23.34s/it]
 15%|█▌        | 678/4476 [4:23:55<24:36:27, 23.32s/it]
 15%|█▌        | 679/4476 [4:24:18<24:34:10, 23.29s/it]
 15%|█▌        | 680/4476 [4:24:41<24:36:39, 23.34s/it]
                                                       
{'loss': 0.4413, 'learning_rate': 4.7206249137824535e-05, 'epoch': 0.46}

 15%|█▌        | 680/4476 [4:24:41<24:36:39, 23.34s/it]
 15%|█▌        | 681/4476 [4:25:05<24:39:54, 23.40s/it]
 15%|█▌        | 682/4476 [4:25:28<24:36:13, 23.35s/it]
 15%|█▌        | 683/4476 [4:25:51<24:35:14, 23.34s/it]
 15%|█▌        | 684/4476 [4:26:15<24:32:45, 23.30s/it]
 15%|█▌        | 685/4476 [4:26:38<24:28:17, 23.24s/it]
 15%|█▌        | 686/4476 [4:27:01<24:38:22, 23.40s/it]
 15%|█▌        | 687/4476 [4:27:25<24:46:09, 23.53s/it]
 15%|█▌        | 688/4476 [4:27:49<24:40:21, 23.45s/it]
 15%|█▌        | 689/4476 [4:28:12<24:35:53, 23.38s/it]
 15%|█▌        | 690/4476 [4:28:35<24:35:00, 23.38s/it]
                                                       
{'loss': 0.4302, 'learning_rate': 4.7125099426961185e-05, 'epoch': 0.46}

 15%|█▌        | 690/4476 [4:28:35<24:35:00, 23.38s/it]
 15%|█▌        | 691/4476 [4:28:58<24:27:22, 23.26s/it]
 15%|█▌        | 692/4476 [4:29:21<24:25:53, 23.24s/it]
 15%|█▌        | 693/4476 [4:29:44<24:24:50, 23.23s/it]
 16%|█▌        | 694/4476 [4:30:08<24:33:17, 23.37s/it]
 16%|█▌        | 695/4476 [4:30:31<24:30:14, 23.33s/it]
 16%|█▌        | 696/4476 [4:30:55<24:28:12, 23.30s/it]
 16%|█▌        | 697/4476 [4:31:18<24:36:31, 23.44s/it]
 16%|█▌        | 698/4476 [4:31:42<24:34:48, 23.42s/it]
 16%|█▌        | 699/4476 [4:32:05<24:29:14, 23.34s/it]
 16%|█▌        | 700/4476 [4:32:28<24:28:20, 23.33s/it]
                                                       
{'loss': 0.4365, 'learning_rate': 4.704285977497687e-05, 'epoch': 0.47}

 16%|█▌        | 700/4476 [4:32:28<24:28:20, 23.33s/it]
 16%|█▌        | 701/4476 [4:32:52<24:28:08, 23.33s/it]
 16%|█▌        | 702/4476 [4:33:15<24:21:10, 23.23s/it]
 16%|█▌        | 703/4476 [4:33:38<24:25:59, 23.31s/it]
 16%|█▌        | 704/4476 [4:34:02<24:33:32, 23.44s/it]
 16%|█▌        | 705/4476 [4:34:25<24:35:59, 23.48s/it]
 16%|█▌        | 706/4476 [4:34:49<24:31:04, 23.41s/it]
 16%|█▌        | 707/4476 [4:35:12<24:24:56, 23.32s/it]
 16%|█▌        | 708/4476 [4:35:35<24:22:08, 23.28s/it]
 16%|█▌        | 709/4476 [4:35:58<24:22:38, 23.30s/it]
 16%|█▌        | 710/4476 [4:36:22<24:25:17, 23.35s/it]
                                                       
{'loss': 0.4238, 'learning_rate': 4.6959534233215116e-05, 'epoch': 0.48}

 16%|█▌        | 710/4476 [4:36:22<24:25:17, 23.35s/it]
 16%|█▌        | 711/4476 [4:36:45<24:32:12, 23.46s/it]
 16%|█▌        | 712/4476 [4:37:09<24:32:08, 23.47s/it]
 16%|█▌        | 713/4476 [4:37:33<24:33:52, 23.50s/it]
 16%|█▌        | 714/4476 [4:37:56<24:31:30, 23.47s/it]
 16%|█▌        | 715/4476 [4:38:19<24:26:57, 23.40s/it]
 16%|█▌        | 716/4476 [4:38:43<24:27:39, 23.42s/it]
 16%|█▌        | 717/4476 [4:39:06<24:24:44, 23.38s/it]
 16%|█▌        | 718/4476 [4:39:29<24:21:04, 23.33s/it]
 16%|█▌        | 719/4476 [4:39:52<24:16:13, 23.26s/it]
 16%|█▌        | 720/4476 [4:40:16<24:19:56, 23.32s/it]
                                                       
{'loss': 0.4284, 'learning_rate': 4.687512690651328e-05, 'epoch': 0.48}

 16%|█▌        | 720/4476 [4:40:16<24:19:56, 23.32s/it]
 16%|█▌        | 721/4476 [4:40:39<24:18:47, 23.31s/it]
 16%|█▌        | 722/4476 [4:41:02<24:18:08, 23.31s/it]
 16%|█▌        | 723/4476 [4:41:26<24:18:05, 23.31s/it]
 16%|█▌        | 724/4476 [4:41:49<24:18:23, 23.32s/it]
 16%|█▌        | 725/4476 [4:42:12<24:17:28, 23.31s/it]
 16%|█▌        | 726/4476 [4:42:36<24:17:36, 23.32s/it]
 16%|█▌        | 727/4476 [4:42:59<24:10:37, 23.22s/it]
 16%|█▋        | 728/4476 [4:43:22<24:10:57, 23.23s/it]
 16%|█▋        | 729/4476 [4:43:45<24:10:16, 23.22s/it]
 16%|█▋        | 730/4476 [4:44:09<24:15:21, 23.31s/it]
                                                       
{'loss': 0.4193, 'learning_rate': 4.678964195300028e-05, 'epoch': 0.49}

 16%|█▋        | 730/4476 [4:44:09<24:15:21, 23.31s/it]
 16%|█▋        | 731/4476 [4:44:32<24:09:53, 23.23s/it]
 16%|█▋        | 732/4476 [4:44:55<24:17:00, 23.35s/it]
 16%|█▋        | 733/4476 [4:45:18<24:12:03, 23.28s/it]
 16%|█▋        | 734/4476 [4:45:42<24:13:12, 23.30s/it]
 16%|█▋        | 735/4476 [4:46:05<24:12:50, 23.30s/it]
 16%|█▋        | 736/4476 [4:46:28<24:11:18, 23.28s/it]
 16%|█▋        | 737/4476 [4:46:51<24:10:36, 23.28s/it]
 16%|█▋        | 738/4476 [4:47:15<24:14:52, 23.35s/it]
 17%|█▋        | 739/4476 [4:47:38<24:10:31, 23.29s/it]
 17%|█▋        | 740/4476 [4:48:01<24:08:33, 23.26s/it]
                                                       
{'loss': 0.4256, 'learning_rate': 4.670308358389184e-05, 'epoch': 0.5}

 17%|█▋        | 740/4476 [4:48:01<24:08:33, 23.26s/it]
 17%|█▋        | 741/4476 [4:48:24<24:02:12, 23.17s/it]
 17%|█▋        | 742/4476 [4:48:48<24:05:21, 23.22s/it]
 17%|█▋        | 743/4476 [4:49:11<24:03:50, 23.21s/it]
 17%|█▋        | 744/4476 [4:49:34<24:08:37, 23.29s/it]
 17%|█▋        | 745/4476 [4:49:57<24:03:37, 23.22s/it]
 17%|█▋        | 746/4476 [4:50:21<24:07:04, 23.28s/it]
 17%|█▋        | 747/4476 [4:50:44<24:06:07, 23.27s/it]
 17%|█▋        | 748/4476 [4:51:07<24:02:26, 23.22s/it]
 17%|█▋        | 749/4476 [4:51:30<23:56:58, 23.13s/it]
 17%|█▋        | 750/4476 [4:51:53<23:57:39, 23.15s/it]
                                                       
{'loss': 0.4288, 'learning_rate': 4.6615456063282944e-05, 'epoch': 0.5}

 17%|█▋        | 750/4476 [4:51:53<23:57:39, 23.15s/it]
 17%|█▋        | 751/4476 [4:52:16<23:56:50, 23.14s/it]
 17%|█▋        | 752/4476 [4:52:40<23:58:16, 23.17s/it]
 17%|█▋        | 753/4476 [4:53:03<23:59:33, 23.20s/it]
 17%|█▋        | 754/4476 [4:53:26<24:02:12, 23.25s/it]
 17%|█▋        | 755/4476 [4:53:49<24:01:14, 23.24s/it]
 17%|█▋        | 756/4476 [4:54:13<24:10:30, 23.40s/it]
 17%|█▋        | 757/4476 [4:54:37<24:15:06, 23.48s/it]
 17%|█▋        | 758/4476 [4:55:01<24:19:44, 23.56s/it]
 17%|█▋        | 759/4476 [4:55:24<24:15:07, 23.49s/it]
 17%|█▋        | 760/4476 [4:55:47<24:08:20, 23.39s/it]
                                                       
{'loss': 0.4335, 'learning_rate': 4.652676370793784e-05, 'epoch': 0.51}

 17%|█▋        | 760/4476 [4:55:47<24:08:20, 23.39s/it]
 17%|█▋        | 761/4476 [4:56:11<24:11:45, 23.45s/it]
 17%|█▋        | 762/4476 [4:56:34<24:01:55, 23.29s/it]
 17%|█▋        | 763/4476 [4:56:57<24:10:19, 23.44s/it]
 17%|█▋        | 764/4476 [4:57:21<24:11:20, 23.46s/it]
 17%|█▋        | 765/4476 [4:57:44<24:12:40, 23.49s/it]
 17%|█▋        | 766/4476 [4:58:08<24:17:29, 23.57s/it]
 17%|█▋        | 767/4476 [4:58:32<24:17:35, 23.58s/it]
 17%|█▋        | 768/4476 [4:58:56<24:21:25, 23.65s/it]
 17%|█▋        | 769/4476 [4:59:19<24:18:46, 23.61s/it]
 17%|█▋        | 770/4476 [4:59:42<24:09:59, 23.48s/it]
                                                       
{'loss': 0.4271, 'learning_rate': 4.643701088707736e-05, 'epoch': 0.52}

 17%|█▋        | 770/4476 [4:59:42<24:09:59, 23.48s/it]
 17%|█▋        | 771/4476 [5:00:06<24:09:55, 23.48s/it]
 17%|█▋        | 772/4476 [5:00:29<24:10:09, 23.49s/it]
 17%|█▋        | 773/4476 [5:00:53<24:09:58, 23.49s/it]
 17%|█▋        | 774/4476 [5:01:16<24:08:28, 23.48s/it]
 17%|█▋        | 775/4476 [5:01:40<24:05:22, 23.43s/it]
 17%|█▋        | 776/4476 [5:02:03<24:05:09, 23.43s/it]
 17%|█▋        | 777/4476 [5:02:27<24:07:07, 23.47s/it]
 17%|█▋        | 778/4476 [5:02:50<24:01:08, 23.38s/it]
 17%|█▋        | 779/4476 [5:03:14<24:08:09, 23.50s/it]
 17%|█▋        | 780/4476 [5:03:37<24:02:39, 23.42s/it]
                                                       
{'loss': 0.4304, 'learning_rate': 4.634620202216366e-05, 'epoch': 0.52}

 17%|█▋        | 780/4476 [5:03:37<24:02:39, 23.42s/it]
 17%|█▋        | 781/4476 [5:04:00<23:59:43, 23.38s/it]
 17%|█▋        | 782/4476 [5:04:23<23:58:39, 23.37s/it]
 17%|█▋        | 783/4476 [5:04:47<23:58:37, 23.37s/it]
 18%|█▊        | 784/4476 [5:05:10<23:53:21, 23.29s/it]
 18%|█▊        | 785/4476 [5:05:33<23:53:27, 23.30s/it]
 18%|█▊        | 786/4476 [5:05:57<23:53:44, 23.31s/it]
 18%|█▊        | 787/4476 [5:06:20<23:57:55, 23.39s/it]
 18%|█▊        | 788/4476 [5:06:43<23:53:06, 23.32s/it]
 18%|█▊        | 789/4476 [5:07:07<23:51:29, 23.30s/it]
 18%|█▊        | 790/4476 [5:07:30<24:00:43, 23.45s/it]
                                                       
{'loss': 0.4249, 'learning_rate': 4.625434158668246e-05, 'epoch': 0.53}

 18%|█▊        | 790/4476 [5:07:30<24:00:43, 23.45s/it]
 18%|█▊        | 791/4476 [5:07:54<23:58:44, 23.43s/it]
 18%|█▊        | 792/4476 [5:08:17<24:00:35, 23.46s/it]
 18%|█▊        | 793/4476 [5:08:40<23:54:43, 23.37s/it]
 18%|█▊        | 794/4476 [5:09:04<23:56:42, 23.41s/it]
 18%|█▊        | 795/4476 [5:09:27<23:51:06, 23.33s/it]
 18%|█▊        | 796/4476 [5:09:50<23:47:05, 23.27s/it]
 18%|█▊        | 797/4476 [5:10:13<23:46:27, 23.26s/it]
 18%|█▊        | 798/4476 [5:10:37<23:47:39, 23.29s/it]
 18%|█▊        | 799/4476 [5:11:00<23:51:58, 23.37s/it]
 18%|█▊        | 800/4476 [5:11:24<23:51:32, 23.37s/it]
                                                       
{'loss': 0.4322, 'learning_rate': 4.6161434105922616e-05, 'epoch': 0.54}

 18%|█▊        | 800/4476 [5:11:24<23:51:32, 23.37s/it][INFO|trainer.py:2939] 2023-11-12 08:35:11,424 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-800
[INFO|tokenization_utils_base.py:2437] 2023-11-12 08:35:11,456 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-800/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 08:35:11,456 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-800/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 08:35:11,456 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-800/added_tokens.json

 18%|█▊        | 801/4476 [5:11:47<23:53:34, 23.41s/it]
 18%|█▊        | 802/4476 [5:12:11<23:55:52, 23.45s/it]
 18%|█▊        | 803/4476 [5:12:35<24:01:29, 23.55s/it]
 18%|█▊        | 804/4476 [5:12:58<23:59:45, 23.53s/it]
 18%|█▊        | 805/4476 [5:13:21<23:53:14, 23.43s/it]
 18%|█▊        | 806/4476 [5:13:45<23:54:56, 23.46s/it]
 18%|█▊        | 807/4476 [5:14:08<23:53:10, 23.44s/it]
 18%|█▊        | 808/4476 [5:14:31<23:45:59, 23.33s/it]
 18%|█▊        | 809/4476 [5:14:54<23:39:20, 23.22s/it]
 18%|█▊        | 810/4476 [5:15:17<23:33:04, 23.13s/it]
                                                       
{'loss': 0.4229, 'learning_rate': 4.6067484156753234e-05, 'epoch': 0.54}

 18%|█▊        | 810/4476 [5:15:17<23:33:04, 23.13s/it]
 18%|█▊        | 811/4476 [5:15:40<23:36:43, 23.19s/it]
 18%|█▊        | 812/4476 [5:16:04<23:36:52, 23.20s/it]
 18%|█▊        | 813/4476 [5:16:27<23:38:49, 23.24s/it]
 18%|█▊        | 814/4476 [5:16:50<23:42:26, 23.31s/it]
 18%|█▊        | 815/4476 [5:17:14<23:44:09, 23.34s/it]
 18%|█▊        | 816/4476 [5:17:37<23:34:17, 23.19s/it]
 18%|█▊        | 817/4476 [5:18:00<23:33:52, 23.18s/it]
 18%|█▊        | 818/4476 [5:18:23<23:39:37, 23.29s/it]
 18%|█▊        | 819/4476 [5:18:46<23:35:11, 23.22s/it]
 18%|█▊        | 820/4476 [5:19:10<23:33:54, 23.20s/it]
                                                       
{'loss': 0.4252, 'learning_rate': 4.597249636739815e-05, 'epoch': 0.55}

 18%|█▊        | 820/4476 [5:19:10<23:33:54, 23.20s/it]
 18%|█▊        | 821/4476 [5:19:33<23:34:02, 23.21s/it]
 18%|█▊        | 822/4476 [5:19:56<23:33:45, 23.21s/it]
 18%|█▊        | 823/4476 [5:20:20<23:43:44, 23.38s/it]
 18%|█▊        | 824/4476 [5:20:43<23:44:35, 23.41s/it]
 18%|█▊        | 825/4476 [5:21:07<23:43:06, 23.39s/it]
 18%|█▊        | 826/4476 [5:21:30<23:41:39, 23.37s/it]
 18%|█▊        | 827/4476 [5:21:53<23:35:20, 23.27s/it]
 18%|█▊        | 828/4476 [5:22:16<23:35:35, 23.28s/it]
 19%|█▊        | 829/4476 [5:22:39<23:33:24, 23.25s/it]
 19%|█▊        | 830/4476 [5:23:03<23:40:40, 23.38s/it]
                                                       
{'loss': 0.413, 'learning_rate': 4.5876475417207974e-05, 'epoch': 0.56}

 19%|█▊        | 830/4476 [5:23:03<23:40:40, 23.38s/it]
 19%|█▊        | 831/4476 [5:23:27<23:45:57, 23.47s/it]
 19%|█▊        | 832/4476 [5:23:50<23:41:22, 23.40s/it]
 19%|█▊        | 833/4476 [5:24:14<23:47:54, 23.52s/it]
 19%|█▊        | 834/4476 [5:24:37<23:44:21, 23.47s/it]
 19%|█▊        | 835/4476 [5:25:01<23:40:50, 23.41s/it]
 19%|█▊        | 836/4476 [5:25:24<23:43:16, 23.46s/it]
 19%|█▊        | 837/4476 [5:25:48<23:48:17, 23.55s/it]
 19%|█▊        | 838/4476 [5:26:11<23:42:51, 23.47s/it]
 19%|█▊        | 839/4476 [5:26:34<23:35:28, 23.35s/it]
 19%|█▉        | 840/4476 [5:26:58<23:42:29, 23.47s/it]
                                                       
{'loss': 0.4186, 'learning_rate': 4.577942603642959e-05, 'epoch': 0.56}

 19%|█▉        | 840/4476 [5:26:58<23:42:29, 23.47s/it]
 19%|█▉        | 841/4476 [5:27:21<23:35:55, 23.37s/it]
 19%|█▉        | 842/4476 [5:27:44<23:33:57, 23.35s/it]
 19%|█▉        | 843/4476 [5:28:08<23:36:37, 23.40s/it]
 19%|█▉        | 844/4476 [5:28:31<23:37:16, 23.41s/it]
 19%|█▉        | 845/4476 [5:28:55<23:34:38, 23.38s/it]
 19%|█▉        | 846/4476 [5:29:18<23:32:59, 23.36s/it]
 19%|█▉        | 847/4476 [5:29:41<23:31:59, 23.35s/it]
 19%|█▉        | 848/4476 [5:30:05<23:31:33, 23.34s/it]
 19%|█▉        | 849/4476 [5:30:28<23:28:34, 23.30s/it]
 19%|█▉        | 850/4476 [5:30:51<23:32:42, 23.38s/it]
                                                       
{'loss': 0.4233, 'learning_rate': 4.568135300597306e-05, 'epoch': 0.57}

 19%|█▉        | 850/4476 [5:30:51<23:32:42, 23.38s/it]
 19%|█▉        | 851/4476 [5:31:15<23:35:31, 23.43s/it]
 19%|█▉        | 852/4476 [5:31:38<23:32:34, 23.39s/it]
 19%|█▉        | 853/4476 [5:32:01<23:30:19, 23.36s/it]
 19%|█▉        | 854/4476 [5:32:24<23:22:14, 23.23s/it]
 19%|█▉        | 855/4476 [5:32:48<23:23:13, 23.25s/it]
 19%|█▉        | 856/4476 [5:33:11<23:32:17, 23.41s/it]
 19%|█▉        | 857/4476 [5:33:35<23:28:35, 23.35s/it]
 19%|█▉        | 858/4476 [5:33:58<23:26:12, 23.32s/it]
 19%|█▉        | 859/4476 [5:34:21<23:23:06, 23.28s/it]
 19%|█▉        | 860/4476 [5:34:45<23:27:43, 23.36s/it]
                                                       
{'loss': 0.4177, 'learning_rate': 4.5582261157176164e-05, 'epoch': 0.58}

 19%|█▉        | 860/4476 [5:34:45<23:27:43, 23.36s/it]
 19%|█▉        | 861/4476 [5:35:08<23:27:13, 23.36s/it]
 19%|█▉        | 862/4476 [5:35:31<23:24:50, 23.32s/it]
 19%|█▉        | 863/4476 [5:35:55<23:32:59, 23.47s/it]
 19%|█▉        | 864/4476 [5:36:18<23:23:56, 23.32s/it]
 19%|█▉        | 865/4476 [5:36:41<23:25:21, 23.35s/it]
 19%|█▉        | 866/4476 [5:37:05<23:24:38, 23.35s/it]
 19%|█▉        | 867/4476 [5:37:29<23:30:57, 23.46s/it]
 19%|█▉        | 868/4476 [5:37:52<23:28:21, 23.42s/it]
 19%|█▉        | 869/4476 [5:38:15<23:26:30, 23.40s/it]
 19%|█▉        | 870/4476 [5:38:39<23:24:21, 23.37s/it]
                                                       
{'loss': 0.4236, 'learning_rate': 4.5482155371566384e-05, 'epoch': 0.58}

 19%|█▉        | 870/4476 [5:38:39<23:24:21, 23.37s/it]
 19%|█▉        | 871/4476 [5:39:02<23:31:35, 23.49s/it]
 19%|█▉        | 872/4476 [5:39:26<23:26:41, 23.42s/it]
 20%|█▉        | 873/4476 [5:39:49<23:25:24, 23.40s/it]
 20%|█▉        | 874/4476 [5:40:12<23:22:19, 23.36s/it]
 20%|█▉        | 875/4476 [5:40:35<23:19:39, 23.32s/it]
 20%|█▉        | 876/4476 [5:40:59<23:24:34, 23.41s/it]
 20%|█▉        | 877/4476 [5:41:22<23:18:28, 23.31s/it]
 20%|█▉        | 878/4476 [5:41:45<23:17:29, 23.30s/it]
 20%|█▉        | 879/4476 [5:42:09<23:19:04, 23.34s/it]
 20%|█▉        | 880/4476 [5:42:32<23:21:57, 23.39s/it]
                                                       
{'loss': 0.4228, 'learning_rate': 4.538104058062042e-05, 'epoch': 0.59}

 20%|█▉        | 880/4476 [5:42:32<23:21:57, 23.39s/it]
 20%|█▉        | 881/4476 [5:42:56<23:19:38, 23.36s/it]
 20%|█▉        | 882/4476 [5:43:19<23:23:57, 23.44s/it]
 20%|█▉        | 883/4476 [5:43:43<23:21:07, 23.40s/it]
 20%|█▉        | 884/4476 [5:44:06<23:24:08, 23.45s/it]
 20%|█▉        | 885/4476 [5:44:30<23:23:44, 23.45s/it]
 20%|█▉        | 886/4476 [5:44:53<23:23:07, 23.45s/it]
 20%|█▉        | 887/4476 [5:45:16<23:15:57, 23.34s/it]
 20%|█▉        | 888/4476 [5:45:39<23:14:36, 23.32s/it]
 20%|█▉        | 889/4476 [5:46:02<23:09:12, 23.24s/it]
 20%|█▉        | 890/4476 [5:46:26<23:12:44, 23.30s/it]
                                                       
{'loss': 0.4181, 'learning_rate': 4.5278921765521234e-05, 'epoch': 0.6}

 20%|█▉        | 890/4476 [5:46:26<23:12:44, 23.30s/it]
 20%|█▉        | 891/4476 [5:46:49<23:10:39, 23.27s/it]
 20%|█▉        | 892/4476 [5:47:12<23:11:36, 23.30s/it]
 20%|█▉        | 893/4476 [5:47:36<23:13:12, 23.33s/it]
 20%|█▉        | 894/4476 [5:47:59<23:09:59, 23.28s/it]
 20%|█▉        | 895/4476 [5:48:22<23:07:06, 23.24s/it]
 20%|██        | 896/4476 [5:48:46<23:09:09, 23.28s/it]
 20%|██        | 897/4476 [5:49:09<23:05:21, 23.22s/it]
 20%|██        | 898/4476 [5:49:32<23:07:07, 23.26s/it]
 20%|██        | 899/4476 [5:49:55<23:02:46, 23.19s/it]
 20%|██        | 900/4476 [5:50:18<23:03:32, 23.21s/it]
                                                       
{'loss': 0.4261, 'learning_rate': 4.51758039569127e-05, 'epoch': 0.6}

 20%|██        | 900/4476 [5:50:18<23:03:32, 23.21s/it]
 20%|██        | 901/4476 [5:50:41<23:01:04, 23.18s/it]
 20%|██        | 902/4476 [5:51:05<23:01:26, 23.19s/it]
 20%|██        | 903/4476 [5:51:28<23:07:03, 23.29s/it]
 20%|██        | 904/4476 [5:51:52<23:09:25, 23.34s/it]
 20%|██        | 905/4476 [5:52:15<23:11:06, 23.37s/it]
 20%|██        | 906/4476 [5:52:38<23:08:54, 23.34s/it]
 20%|██        | 907/4476 [5:53:01<23:06:10, 23.30s/it]
 20%|██        | 908/4476 [5:53:25<23:12:41, 23.42s/it]
 20%|██        | 909/4476 [5:53:48<23:05:21, 23.30s/it]
 20%|██        | 910/4476 [5:54:12<23:07:57, 23.35s/it]
                                                       
{'loss': 0.4217, 'learning_rate': 4.5071692234651764e-05, 'epoch': 0.61}

 20%|██        | 910/4476 [5:54:12<23:07:57, 23.35s/it]
 20%|██        | 911/4476 [5:54:35<23:06:28, 23.33s/it]
 20%|██        | 912/4476 [5:54:58<23:03:57, 23.30s/it]
 20%|██        | 913/4476 [5:55:21<23:00:57, 23.26s/it]
 20%|██        | 914/4476 [5:55:45<23:02:37, 23.29s/it]
 20%|██        | 915/4476 [5:56:08<23:10:34, 23.43s/it]
 20%|██        | 916/4476 [5:56:32<23:08:23, 23.40s/it]
 20%|██        | 917/4476 [5:56:55<23:10:01, 23.43s/it]
 21%|██        | 918/4476 [5:57:19<23:12:23, 23.48s/it]
 21%|██        | 919/4476 [5:57:43<23:16:15, 23.55s/it]
 21%|██        | 920/4476 [5:58:06<23:09:42, 23.45s/it]
                                                       
{'loss': 0.4191, 'learning_rate': 4.4966591727558184e-05, 'epoch': 0.62}

 21%|██        | 920/4476 [5:58:06<23:09:42, 23.45s/it]
 21%|██        | 921/4476 [5:58:30<23:15:18, 23.55s/it]
 21%|██        | 922/4476 [5:58:53<23:10:19, 23.47s/it]
 21%|██        | 923/4476 [5:59:17<23:16:07, 23.58s/it]
 21%|██        | 924/4476 [5:59:40<23:05:30, 23.40s/it]
 21%|██        | 925/4476 [6:00:03<23:01:37, 23.34s/it]
 21%|██        | 926/4476 [6:00:27<23:09:42, 23.49s/it]
 21%|██        | 927/4476 [6:00:50<23:04:26, 23.41s/it]
 21%|██        | 928/4476 [6:01:13<22:56:42, 23.28s/it]
 21%|██        | 929/4476 [6:01:37<23:01:34, 23.37s/it]
 21%|██        | 930/4476 [6:02:00<23:04:20, 23.42s/it]
                                                       
{'loss': 0.4247, 'learning_rate': 4.48605076131619e-05, 'epoch': 0.62}

 21%|██        | 930/4476 [6:02:00<23:04:20, 23.42s/it]
 21%|██        | 931/4476 [6:02:23<23:01:03, 23.37s/it]
 21%|██        | 932/4476 [6:02:46<22:55:40, 23.29s/it]
 21%|██        | 933/4476 [6:03:10<22:53:02, 23.25s/it]
 21%|██        | 934/4476 [6:03:33<22:53:37, 23.27s/it]
 21%|██        | 935/4476 [6:03:57<22:59:34, 23.38s/it]
 21%|██        | 936/4476 [6:04:20<22:59:10, 23.38s/it]
 21%|██        | 937/4476 [6:04:44<23:04:44, 23.48s/it]
 21%|██        | 938/4476 [6:05:07<22:59:08, 23.39s/it]
 21%|██        | 939/4476 [6:05:30<22:57:46, 23.37s/it]
 21%|██        | 940/4476 [6:05:53<22:55:43, 23.34s/it]
                                                       
{'loss': 0.4236, 'learning_rate': 4.475344511744794e-05, 'epoch': 0.63}

 21%|██        | 940/4476 [6:05:53<22:55:43, 23.34s/it]
 21%|██        | 941/4476 [6:06:17<22:59:22, 23.41s/it]
 21%|██        | 942/4476 [6:06:41<23:02:43, 23.48s/it]
 21%|██        | 943/4476 [6:07:04<23:00:54, 23.45s/it]
 21%|██        | 944/4476 [6:07:27<23:00:34, 23.45s/it]
 21%|██        | 945/4476 [6:07:50<22:50:47, 23.29s/it]
 21%|██        | 946/4476 [6:08:14<22:55:36, 23.38s/it]
 21%|██        | 947/4476 [6:08:37<22:54:55, 23.38s/it]
 21%|██        | 948/4476 [6:09:01<22:55:35, 23.39s/it]
 21%|██        | 949/4476 [6:09:24<22:50:36, 23.32s/it]
 21%|██        | 950/4476 [6:09:47<22:49:41, 23.31s/it]
                                                       
{'loss': 0.4172, 'learning_rate': 4.464540951459902e-05, 'epoch': 0.64}

 21%|██        | 950/4476 [6:09:47<22:49:41, 23.31s/it]
 21%|██        | 951/4476 [6:10:11<22:49:56, 23.32s/it]
 21%|██▏       | 952/4476 [6:10:34<22:58:44, 23.47s/it]
 21%|██▏       | 953/4476 [6:10:58<22:59:25, 23.49s/it]
 21%|██▏       | 954/4476 [6:11:22<23:03:03, 23.56s/it]
 21%|██▏       | 955/4476 [6:11:45<22:53:55, 23.41s/it]
 21%|██▏       | 956/4476 [6:12:08<22:51:48, 23.38s/it]
 21%|██▏       | 957/4476 [6:12:31<22:49:47, 23.36s/it]
 21%|██▏       | 958/4476 [6:12:55<22:54:24, 23.44s/it]
 21%|██▏       | 959/4476 [6:13:18<22:55:26, 23.46s/it]
 21%|██▏       | 960/4476 [6:13:42<22:52:09, 23.42s/it]
                                                       
{'loss': 0.4209, 'learning_rate': 4.4536406126735664e-05, 'epoch': 0.64}

 21%|██▏       | 960/4476 [6:13:42<22:52:09, 23.42s/it]
 21%|██▏       | 961/4476 [6:14:05<22:45:26, 23.31s/it]
 21%|██▏       | 962/4476 [6:14:28<22:48:54, 23.37s/it]
 22%|██▏       | 963/4476 [6:14:52<22:51:17, 23.42s/it]
 22%|██▏       | 964/4476 [6:15:15<22:50:12, 23.41s/it]
 22%|██▏       | 965/4476 [6:15:38<22:45:46, 23.34s/it]
 22%|██▏       | 966/4476 [6:16:02<22:44:52, 23.33s/it]
 22%|██▏       | 967/4476 [6:16:25<22:44:05, 23.32s/it]
 22%|██▏       | 968/4476 [6:16:48<22:43:10, 23.32s/it]
 22%|██▏       | 969/4476 [6:17:12<22:45:40, 23.36s/it]
 22%|██▏       | 970/4476 [6:17:36<22:50:52, 23.46s/it]
                                                       
{'loss': 0.4179, 'learning_rate': 4.442644032365407e-05, 'epoch': 0.65}

 22%|██▏       | 970/4476 [6:17:36<22:50:52, 23.46s/it]
 22%|██▏       | 971/4476 [6:17:59<22:50:55, 23.47s/it]
 22%|██▏       | 972/4476 [6:18:22<22:48:42, 23.44s/it]
 22%|██▏       | 973/4476 [6:18:46<22:49:03, 23.45s/it]
 22%|██▏       | 974/4476 [6:19:09<22:49:07, 23.46s/it]
 22%|██▏       | 975/4476 [6:19:33<22:49:15, 23.47s/it]
 22%|██▏       | 976/4476 [6:19:57<22:52:59, 23.54s/it]
 22%|██▏       | 977/4476 [6:20:20<22:52:54, 23.54s/it]
 22%|██▏       | 978/4476 [6:20:44<22:57:17, 23.62s/it]
 22%|██▏       | 979/4476 [6:21:07<22:51:45, 23.54s/it]
 22%|██▏       | 980/4476 [6:21:30<22:44:14, 23.41s/it]
                                                       
{'loss': 0.4166, 'learning_rate': 4.431551752256155e-05, 'epoch': 0.66}

 22%|██▏       | 980/4476 [6:21:30<22:44:14, 23.41s/it]
 22%|██▏       | 981/4476 [6:21:54<22:48:58, 23.50s/it]
 22%|██▏       | 982/4476 [6:22:18<22:47:40, 23.49s/it]
 22%|██▏       | 983/4476 [6:22:41<22:48:43, 23.51s/it]
 22%|██▏       | 984/4476 [6:23:04<22:45:09, 23.46s/it]
 22%|██▏       | 985/4476 [6:23:28<22:43:48, 23.44s/it]
 22%|██▏       | 986/4476 [6:23:51<22:43:51, 23.45s/it]
 22%|██▏       | 987/4476 [6:24:15<22:42:47, 23.44s/it]
 22%|██▏       | 988/4476 [6:24:38<22:44:02, 23.46s/it]
 22%|██▏       | 989/4476 [6:25:02<22:48:57, 23.56s/it]
 22%|██▏       | 990/4476 [6:25:25<22:42:46, 23.46s/it]
                                                       
{'loss': 0.4173, 'learning_rate': 4.420364318780973e-05, 'epoch': 0.66}

 22%|██▏       | 990/4476 [6:25:25<22:42:46, 23.46s/it]
 22%|██▏       | 991/4476 [6:25:49<22:46:02, 23.52s/it]
 22%|██▏       | 992/4476 [6:26:12<22:45:02, 23.51s/it]
 22%|██▏       | 993/4476 [6:26:36<22:44:45, 23.51s/it]
 22%|██▏       | 994/4476 [6:26:59<22:44:35, 23.51s/it]
 22%|██▏       | 995/4476 [6:27:22<22:35:31, 23.36s/it]
 22%|██▏       | 996/4476 [6:27:46<22:36:31, 23.39s/it]
 22%|██▏       | 997/4476 [6:28:09<22:39:27, 23.45s/it]
 22%|██▏       | 998/4476 [6:28:33<22:38:45, 23.44s/it]
 22%|██▏       | 999/4476 [6:28:56<22:37:21, 23.42s/it]
 22%|██▏       | 1000/4476 [6:29:20<22:36:52, 23.42s/it]
                                                        
{'loss': 0.4166, 'learning_rate': 4.4090822830625236e-05, 'epoch': 0.67}

 22%|██▏       | 1000/4476 [6:29:20<22:36:52, 23.42s/it][INFO|trainer.py:2939] 2023-11-12 09:53:07,385 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1000
[INFO|tokenization_utils_base.py:2437] 2023-11-12 09:53:07,416 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1000/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 09:53:07,416 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1000/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 09:53:07,416 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1000/added_tokens.json

 22%|██▏       | 1001/4476 [6:29:43<22:35:09, 23.40s/it]
 22%|██▏       | 1002/4476 [6:30:06<22:35:47, 23.42s/it]
 22%|██▏       | 1003/4476 [6:30:30<22:37:07, 23.45s/it]
 22%|██▏       | 1004/4476 [6:30:53<22:33:39, 23.39s/it]
 22%|██▏       | 1005/4476 [6:31:17<22:36:11, 23.44s/it]
 22%|██▏       | 1006/4476 [6:31:40<22:39:14, 23.50s/it]
 22%|██▏       | 1007/4476 [6:32:04<22:32:29, 23.39s/it]
 23%|██▎       | 1008/4476 [6:32:27<22:28:13, 23.33s/it]
 23%|██▎       | 1009/4476 [6:32:50<22:28:03, 23.33s/it]
 23%|██▎       | 1010/4476 [6:33:14<22:30:45, 23.38s/it]
                                                        
{'loss': 0.4173, 'learning_rate': 4.3977062008838307e-05, 'epoch': 0.68}

 23%|██▎       | 1010/4476 [6:33:14<22:30:45, 23.38s/it]
 23%|██▎       | 1011/4476 [6:33:37<22:25:34, 23.30s/it]
 23%|██▎       | 1012/4476 [6:34:00<22:20:06, 23.21s/it]
 23%|██▎       | 1013/4476 [6:34:23<22:21:46, 23.25s/it]
 23%|██▎       | 1014/4476 [6:34:46<22:20:05, 23.23s/it]
 23%|██▎       | 1015/4476 [6:35:09<22:19:05, 23.21s/it]
 23%|██▎       | 1016/4476 [6:35:33<22:19:46, 23.23s/it]
 23%|██▎       | 1017/4476 [6:35:56<22:21:19, 23.27s/it]
 23%|██▎       | 1018/4476 [6:36:19<22:22:31, 23.29s/it]
 23%|██▎       | 1019/4476 [6:36:43<22:21:08, 23.28s/it]
 23%|██▎       | 1020/4476 [6:37:06<22:24:22, 23.34s/it]
                                                        
{'loss': 0.4049, 'learning_rate': 4.3862366326608975e-05, 'epoch': 0.68}

 23%|██▎       | 1020/4476 [6:37:06<22:24:22, 23.34s/it]
 23%|██▎       | 1021/4476 [6:37:29<22:23:04, 23.32s/it]
 23%|██▎       | 1022/4476 [6:37:53<22:25:55, 23.38s/it]
 23%|██▎       | 1023/4476 [6:38:16<22:20:01, 23.28s/it]
 23%|██▎       | 1024/4476 [6:38:39<22:21:00, 23.31s/it]
 23%|██▎       | 1025/4476 [6:39:03<22:19:00, 23.28s/it]
 23%|██▎       | 1026/4476 [6:39:26<22:17:58, 23.27s/it]
 23%|██▎       | 1027/4476 [6:39:49<22:23:18, 23.37s/it]
 23%|██▎       | 1028/4476 [6:40:13<22:26:16, 23.43s/it]
 23%|██▎       | 1029/4476 [6:40:36<22:24:15, 23.40s/it]
 23%|██▎       | 1030/4476 [6:41:00<22:25:08, 23.42s/it]
                                                        
{'loss': 0.4143, 'learning_rate': 4.374674143415096e-05, 'epoch': 0.69}

 23%|██▎       | 1030/4476 [6:41:00<22:25:08, 23.42s/it]
 23%|██▎       | 1031/4476 [6:41:23<22:18:14, 23.31s/it]
 23%|██▎       | 1032/4476 [6:41:46<22:16:07, 23.28s/it]
 23%|██▎       | 1033/4476 [6:42:09<22:14:31, 23.26s/it]
 23%|██▎       | 1034/4476 [6:42:33<22:17:26, 23.31s/it]
 23%|██▎       | 1035/4476 [6:42:56<22:21:24, 23.39s/it]
 23%|██▎       | 1036/4476 [6:43:19<22:13:19, 23.26s/it]
 23%|██▎       | 1037/4476 [6:43:42<22:12:16, 23.24s/it]
 23%|██▎       | 1038/4476 [6:44:05<22:09:18, 23.20s/it]
 23%|██▎       | 1039/4476 [6:44:29<22:13:14, 23.27s/it]
 23%|██▎       | 1040/4476 [6:44:52<22:14:52, 23.31s/it]
                                                        
{'loss': 0.4219, 'learning_rate': 4.363019302745334e-05, 'epoch': 0.7}

 23%|██▎       | 1040/4476 [6:44:52<22:14:52, 23.31s/it]
 23%|██▎       | 1041/4476 [6:45:16<22:14:08, 23.30s/it]
 23%|██▎       | 1042/4476 [6:45:39<22:15:59, 23.34s/it]
 23%|██▎       | 1043/4476 [6:46:03<22:17:45, 23.38s/it]
 23%|██▎       | 1044/4476 [6:46:26<22:20:02, 23.43s/it]
 23%|██▎       | 1045/4476 [6:46:50<22:21:06, 23.45s/it]
 23%|██▎       | 1046/4476 [6:47:13<22:18:30, 23.41s/it]
 23%|██▎       | 1047/4476 [6:47:36<22:17:03, 23.40s/it]
 23%|██▎       | 1048/4476 [6:48:00<22:15:10, 23.37s/it]
 23%|██▎       | 1049/4476 [6:48:23<22:14:08, 23.36s/it]
 23%|██▎       | 1050/4476 [6:48:47<22:21:26, 23.49s/it]
                                                        
{'loss': 0.4152, 'learning_rate': 4.3512726847999987e-05, 'epoch': 0.7}

 23%|██▎       | 1050/4476 [6:48:47<22:21:26, 23.49s/it]
 23%|██▎       | 1051/4476 [6:49:10<22:16:36, 23.42s/it]
 24%|██▎       | 1052/4476 [6:49:33<22:08:53, 23.29s/it]
 24%|██▎       | 1053/4476 [6:49:56<22:08:20, 23.28s/it]
 24%|██▎       | 1054/4476 [6:50:19<22:08:04, 23.29s/it]
 24%|██▎       | 1055/4476 [6:50:43<22:12:57, 23.38s/it]
 24%|██▎       | 1056/4476 [6:51:07<22:15:49, 23.44s/it]
 24%|██▎       | 1057/4476 [6:51:30<22:16:53, 23.46s/it]
 24%|██▎       | 1058/4476 [6:51:53<22:13:27, 23.41s/it]
 24%|██▎       | 1059/4476 [6:52:17<22:09:19, 23.34s/it]
 24%|██▎       | 1060/4476 [6:52:40<22:06:53, 23.31s/it]
                                                        
{'loss': 0.4153, 'learning_rate': 4.339434868248665e-05, 'epoch': 0.71}

 24%|██▎       | 1060/4476 [6:52:40<22:06:53, 23.31s/it]
 24%|██▎       | 1061/4476 [6:53:03<22:07:10, 23.32s/it]
 24%|██▎       | 1062/4476 [6:53:27<22:12:46, 23.42s/it]
 24%|██▎       | 1063/4476 [6:53:50<22:06:59, 23.33s/it]
 24%|██▍       | 1064/4476 [6:54:13<22:09:36, 23.38s/it]
 24%|██▍       | 1065/4476 [6:54:37<22:11:22, 23.42s/it]
 24%|██▍       | 1066/4476 [6:55:00<22:02:18, 23.27s/it]
 24%|██▍       | 1067/4476 [6:55:23<22:03:35, 23.30s/it]
 24%|██▍       | 1068/4476 [6:55:46<21:58:05, 23.21s/it]
 24%|██▍       | 1069/4476 [6:56:10<21:59:23, 23.24s/it]
 24%|██▍       | 1070/4476 [6:56:33<21:59:02, 23.24s/it]
                                                        
{'loss': 0.4148, 'learning_rate': 4.3275064362535966e-05, 'epoch': 0.72}

 24%|██▍       | 1070/4476 [6:56:33<21:59:02, 23.24s/it]
 24%|██▍       | 1071/4476 [6:56:56<22:00:52, 23.28s/it]
 24%|██▍       | 1072/4476 [6:57:19<21:54:27, 23.17s/it]
 24%|██▍       | 1073/4476 [6:57:42<21:52:27, 23.14s/it]
 24%|██▍       | 1074/4476 [6:58:05<21:55:00, 23.19s/it]
 24%|██▍       | 1075/4476 [6:58:29<21:55:45, 23.21s/it]
 24%|██▍       | 1076/4476 [6:58:52<21:52:58, 23.17s/it]
 24%|██▍       | 1077/4476 [6:59:15<21:53:37, 23.19s/it]
 24%|██▍       | 1078/4476 [6:59:38<21:57:13, 23.26s/it]
 24%|██▍       | 1079/4476 [7:00:01<21:52:14, 23.18s/it]
 24%|██▍       | 1080/4476 [7:00:25<21:54:46, 23.23s/it]
                                                        
{'loss': 0.4147, 'learning_rate': 4.315487976441014e-05, 'epoch': 0.72}

 24%|██▍       | 1080/4476 [7:00:25<21:54:46, 23.23s/it]
 24%|██▍       | 1081/4476 [7:00:48<21:59:35, 23.32s/it]
 24%|██▍       | 1082/4476 [7:01:12<21:59:21, 23.32s/it]
 24%|██▍       | 1083/4476 [7:01:35<22:02:01, 23.38s/it]
 24%|██▍       | 1084/4476 [7:01:58<21:59:08, 23.33s/it]
 24%|██▍       | 1085/4476 [7:02:22<22:05:48, 23.46s/it]
 24%|██▍       | 1086/4476 [7:02:46<22:04:44, 23.45s/it]
 24%|██▍       | 1087/4476 [7:03:09<22:08:06, 23.51s/it]
 24%|██▍       | 1088/4476 [7:03:33<22:03:59, 23.45s/it]
 24%|██▍       | 1089/4476 [7:03:56<22:01:03, 23.40s/it]
 24%|██▍       | 1090/4476 [7:04:20<22:06:17, 23.50s/it]
                                                        
{'loss': 0.41, 'learning_rate': 4.303380080872145e-05, 'epoch': 0.73}

 24%|██▍       | 1090/4476 [7:04:20<22:06:17, 23.50s/it]
 24%|██▍       | 1091/4476 [7:04:43<22:01:44, 23.43s/it]
 24%|██▍       | 1092/4476 [7:05:06<21:59:36, 23.40s/it]
 24%|██▍       | 1093/4476 [7:05:29<21:55:51, 23.34s/it]
 24%|██▍       | 1094/4476 [7:05:53<21:55:43, 23.34s/it]
 24%|██▍       | 1095/4476 [7:06:16<21:50:23, 23.25s/it]
 24%|██▍       | 1096/4476 [7:06:39<21:57:00, 23.38s/it]
 25%|██▍       | 1097/4476 [7:07:03<22:03:38, 23.50s/it]
 25%|██▍       | 1098/4476 [7:07:26<21:54:45, 23.35s/it]
 25%|██▍       | 1099/4476 [7:07:49<21:51:39, 23.30s/it]
 25%|██▍       | 1100/4476 [7:08:13<21:54:30, 23.36s/it]
                                                        
{'loss': 0.4119, 'learning_rate': 4.291183346014063e-05, 'epoch': 0.74}

 25%|██▍       | 1100/4476 [7:08:13<21:54:30, 23.36s/it]
 25%|██▍       | 1101/4476 [7:08:36<21:51:56, 23.32s/it]
 25%|██▍       | 1102/4476 [7:08:59<21:50:11, 23.30s/it]
 25%|██▍       | 1103/4476 [7:09:23<21:50:39, 23.31s/it]
 25%|██▍       | 1104/4476 [7:09:46<21:49:55, 23.31s/it]
 25%|██▍       | 1105/4476 [7:10:10<21:57:27, 23.45s/it]
 25%|██▍       | 1106/4476 [7:10:34<22:02:36, 23.55s/it]
 25%|██▍       | 1107/4476 [7:10:57<22:01:20, 23.53s/it]
 25%|██▍       | 1108/4476 [7:11:21<22:04:21, 23.59s/it]
 25%|██▍       | 1109/4476 [7:11:44<21:59:59, 23.52s/it]
 25%|██▍       | 1110/4476 [7:12:08<21:58:19, 23.50s/it]
                                                        
{'loss': 0.4173, 'learning_rate': 4.278898372710296e-05, 'epoch': 0.74}

 25%|██▍       | 1110/4476 [7:12:08<21:58:19, 23.50s/it]
 25%|██▍       | 1111/4476 [7:12:31<21:55:29, 23.46s/it]
 25%|██▍       | 1112/4476 [7:12:54<21:55:15, 23.46s/it]
 25%|██▍       | 1113/4476 [7:13:18<21:52:27, 23.42s/it]
 25%|██▍       | 1114/4476 [7:13:41<21:47:48, 23.34s/it]
 25%|██▍       | 1115/4476 [7:14:04<21:49:43, 23.38s/it]
 25%|██▍       | 1116/4476 [7:14:28<21:51:55, 23.43s/it]
 25%|██▍       | 1117/4476 [7:14:51<21:49:12, 23.39s/it]
 25%|██▍       | 1118/4476 [7:15:15<21:51:35, 23.44s/it]
 25%|██▌       | 1119/4476 [7:15:38<21:50:06, 23.42s/it]
 25%|██▌       | 1120/4476 [7:16:01<21:47:52, 23.38s/it]
                                                        
{'loss': 0.4119, 'learning_rate': 4.266525766151238e-05, 'epoch': 0.75}

 25%|██▌       | 1120/4476 [7:16:01<21:47:52, 23.38s/it]
 25%|██▌       | 1121/4476 [7:16:25<21:43:34, 23.31s/it]
 25%|██▌       | 1122/4476 [7:16:48<21:44:03, 23.33s/it]
 25%|██▌       | 1123/4476 [7:17:11<21:43:57, 23.33s/it]
 25%|██▌       | 1124/4476 [7:17:35<21:42:05, 23.31s/it]
 25%|██▌       | 1125/4476 [7:17:58<21:37:31, 23.23s/it]
 25%|██▌       | 1126/4476 [7:18:21<21:43:52, 23.35s/it]
 25%|██▌       | 1127/4476 [7:18:44<21:40:57, 23.31s/it]
 25%|██▌       | 1128/4476 [7:19:08<21:40:04, 23.30s/it]
 25%|██▌       | 1129/4476 [7:19:31<21:39:29, 23.30s/it]
 25%|██▌       | 1130/4476 [7:19:54<21:41:39, 23.34s/it]
                                                        
{'loss': 0.4163, 'learning_rate': 4.254066135844326e-05, 'epoch': 0.76}

 25%|██▌       | 1130/4476 [7:19:54<21:41:39, 23.34s/it]
 25%|██▌       | 1131/4476 [7:20:18<21:37:47, 23.28s/it]
 25%|██▌       | 1132/4476 [7:20:41<21:34:41, 23.23s/it]
 25%|██▌       | 1133/4476 [7:21:04<21:36:15, 23.27s/it]
 25%|██▌       | 1134/4476 [7:21:27<21:36:26, 23.28s/it]
 25%|██▌       | 1135/4476 [7:21:51<21:37:03, 23.29s/it]
 25%|██▌       | 1136/4476 [7:22:14<21:40:19, 23.36s/it]
 25%|██▌       | 1137/4476 [7:22:38<21:41:23, 23.39s/it]
 25%|██▌       | 1138/4476 [7:23:01<21:35:44, 23.29s/it]
 25%|██▌       | 1139/4476 [7:23:24<21:34:47, 23.28s/it]
 25%|██▌       | 1140/4476 [7:23:47<21:38:19, 23.35s/it]
                                                        
{'loss': 0.4104, 'learning_rate': 4.2415200955840184e-05, 'epoch': 0.76}

 25%|██▌       | 1140/4476 [7:23:47<21:38:19, 23.35s/it]
 25%|██▌       | 1141/4476 [7:24:11<21:39:44, 23.38s/it]
 26%|██▌       | 1142/4476 [7:24:34<21:33:29, 23.28s/it]
 26%|██▌       | 1143/4476 [7:24:57<21:30:10, 23.23s/it]
 26%|██▌       | 1144/4476 [7:25:20<21:29:46, 23.23s/it]
 26%|██▌       | 1145/4476 [7:25:43<21:28:58, 23.22s/it]
 26%|██▌       | 1146/4476 [7:26:07<21:28:52, 23.22s/it]
 26%|██▌       | 1147/4476 [7:26:30<21:35:59, 23.36s/it]
 26%|██▌       | 1148/4476 [7:26:54<21:35:46, 23.36s/it]
 26%|██▌       | 1149/4476 [7:27:17<21:35:01, 23.35s/it]
 26%|██▌       | 1150/4476 [7:27:40<21:34:44, 23.36s/it]
                                                        
{'loss': 0.4045, 'learning_rate': 4.228888263421557e-05, 'epoch': 0.77}

 26%|██▌       | 1150/4476 [7:27:40<21:34:44, 23.36s/it]
 26%|██▌       | 1151/4476 [7:28:04<21:32:39, 23.33s/it]
 26%|██▌       | 1152/4476 [7:28:27<21:28:54, 23.27s/it]
 26%|██▌       | 1153/4476 [7:28:50<21:26:44, 23.23s/it]
 26%|██▌       | 1154/4476 [7:29:13<21:29:12, 23.28s/it]
 26%|██▌       | 1155/4476 [7:29:37<21:36:33, 23.42s/it]
 26%|██▌       | 1156/4476 [7:30:00<21:25:05, 23.22s/it]
 26%|██▌       | 1157/4476 [7:30:23<21:28:52, 23.30s/it]
 26%|██▌       | 1158/4476 [7:30:47<21:31:27, 23.35s/it]
 26%|██▌       | 1159/4476 [7:31:10<21:27:26, 23.29s/it]
 26%|██▌       | 1160/4476 [7:31:33<21:27:47, 23.30s/it]
                                                        
{'loss': 0.413, 'learning_rate': 4.216171261634521e-05, 'epoch': 0.78}

 26%|██▌       | 1160/4476 [7:31:33<21:27:47, 23.30s/it]
 26%|██▌       | 1161/4476 [7:31:57<21:27:09, 23.30s/it]
 26%|██▌       | 1162/4476 [7:32:20<21:30:33, 23.37s/it]
 26%|██▌       | 1163/4476 [7:32:43<21:26:44, 23.30s/it]
 26%|██▌       | 1164/4476 [7:33:07<21:26:53, 23.31s/it]
 26%|██▌       | 1165/4476 [7:33:30<21:25:58, 23.30s/it]
 26%|██▌       | 1166/4476 [7:33:53<21:21:10, 23.22s/it]
 26%|██▌       | 1167/4476 [7:34:16<21:22:01, 23.25s/it]
 26%|██▌       | 1168/4476 [7:34:40<21:29:44, 23.39s/it]
 26%|██▌       | 1169/4476 [7:35:03<21:28:20, 23.37s/it]
 26%|██▌       | 1170/4476 [7:35:27<21:29:06, 23.40s/it]
                                                        
{'loss': 0.4112, 'learning_rate': 4.2033697166961716e-05, 'epoch': 0.78}

 26%|██▌       | 1170/4476 [7:35:27<21:29:06, 23.40s/it]
 26%|██▌       | 1171/4476 [7:35:51<21:35:32, 23.52s/it]
 26%|██▌       | 1172/4476 [7:36:14<21:33:08, 23.48s/it]
 26%|██▌       | 1173/4476 [7:36:37<21:27:29, 23.39s/it]
 26%|██▌       | 1174/4476 [7:37:00<21:21:40, 23.29s/it]
 26%|██▋       | 1175/4476 [7:37:24<21:28:28, 23.42s/it]
 26%|██▋       | 1176/4476 [7:37:48<21:34:05, 23.53s/it]
 26%|██▋       | 1177/4476 [7:38:11<21:33:13, 23.52s/it]
 26%|██▋       | 1178/4476 [7:38:34<21:27:03, 23.42s/it]
 26%|██▋       | 1179/4476 [7:38:58<21:23:59, 23.37s/it]
 26%|██▋       | 1180/4476 [7:39:21<21:25:14, 23.40s/it]
                                                        
{'loss': 0.4018, 'learning_rate': 4.1904842592445906e-05, 'epoch': 0.79}

 26%|██▋       | 1180/4476 [7:39:21<21:25:14, 23.40s/it]
 26%|██▋       | 1181/4476 [7:39:44<21:24:33, 23.39s/it]
 26%|██▋       | 1182/4476 [7:40:08<21:23:25, 23.38s/it]
 26%|██▋       | 1183/4476 [7:40:31<21:26:08, 23.43s/it]
 26%|██▋       | 1184/4476 [7:40:55<21:29:25, 23.50s/it]
 26%|██▋       | 1185/4476 [7:41:18<21:24:22, 23.42s/it]
 26%|██▋       | 1186/4476 [7:41:42<21:21:01, 23.36s/it]
 27%|██▋       | 1187/4476 [7:42:05<21:17:30, 23.31s/it]
 27%|██▋       | 1188/4476 [7:42:28<21:16:09, 23.29s/it]
 27%|██▋       | 1189/4476 [7:42:51<21:15:26, 23.28s/it]
 27%|██▋       | 1190/4476 [7:43:15<21:15:41, 23.29s/it]
                                                        
{'loss': 0.4068, 'learning_rate': 4.177515524051609e-05, 'epoch': 0.8}

 27%|██▋       | 1190/4476 [7:43:15<21:15:41, 23.29s/it]
 27%|██▋       | 1191/4476 [7:43:38<21:15:59, 23.31s/it]
 27%|██▋       | 1192/4476 [7:44:01<21:13:21, 23.26s/it]
 27%|██▋       | 1193/4476 [7:44:25<21:22:05, 23.43s/it]
 27%|██▋       | 1194/4476 [7:44:48<21:19:01, 23.38s/it]
 27%|██▋       | 1195/4476 [7:45:11<21:12:02, 23.26s/it]
 27%|██▋       | 1196/4476 [7:45:34<21:12:27, 23.28s/it]
 27%|██▋       | 1197/4476 [7:45:57<21:04:51, 23.14s/it]
 27%|██▋       | 1198/4476 [7:46:20<21:02:30, 23.11s/it]
 27%|██▋       | 1199/4476 [7:46:44<21:05:44, 23.17s/it]
 27%|██▋       | 1200/4476 [7:47:07<21:07:22, 23.21s/it]
                                                        
{'loss': 0.4029, 'learning_rate': 4.1644641499915454e-05, 'epoch': 0.8}

 27%|██▋       | 1200/4476 [7:47:07<21:07:22, 23.21s/it][INFO|trainer.py:2939] 2023-11-12 11:10:54,633 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1200
[INFO|tokenization_utils_base.py:2437] 2023-11-12 11:10:54,664 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1200/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 11:10:54,664 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1200/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 11:10:54,665 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1200/added_tokens.json

 27%|██▋       | 1201/4476 [7:47:30<21:13:07, 23.32s/it]
 27%|██▋       | 1202/4476 [7:47:54<21:15:24, 23.37s/it]
 27%|██▋       | 1203/4476 [7:48:18<21:20:39, 23.48s/it]
 27%|██▋       | 1204/4476 [7:48:41<21:22:07, 23.51s/it]
 27%|██▋       | 1205/4476 [7:49:05<21:24:57, 23.57s/it]
 27%|██▋       | 1206/4476 [7:49:28<21:18:21, 23.46s/it]
 27%|██▋       | 1207/4476 [7:49:51<21:14:46, 23.40s/it]
 27%|██▋       | 1208/4476 [7:50:15<21:18:41, 23.48s/it]
 27%|██▋       | 1209/4476 [7:50:39<21:23:43, 23.58s/it]
 27%|██▋       | 1210/4476 [7:51:02<21:20:02, 23.52s/it]
                                                        
{'loss': 0.4009, 'learning_rate': 4.151330780009726e-05, 'epoch': 0.81}

 27%|██▋       | 1210/4476 [7:51:02<21:20:02, 23.52s/it]
 27%|██▋       | 1211/4476 [7:51:26<21:18:58, 23.50s/it]
 27%|██▋       | 1212/4476 [7:51:49<21:10:50, 23.36s/it]
 27%|██▋       | 1213/4476 [7:52:12<21:14:17, 23.43s/it]
 27%|██▋       | 1214/4476 [7:52:36<21:19:42, 23.54s/it]
 27%|██▋       | 1215/4476 [7:52:59<21:15:21, 23.47s/it]
 27%|██▋       | 1216/4476 [7:53:23<21:09:57, 23.37s/it]
 27%|██▋       | 1217/4476 [7:53:46<21:05:28, 23.30s/it]
 27%|██▋       | 1218/4476 [7:54:09<21:08:34, 23.36s/it]
 27%|██▋       | 1219/4476 [7:54:33<21:10:10, 23.40s/it]
 27%|██▋       | 1220/4476 [7:54:56<21:08:25, 23.37s/it]
                                                        
{'loss': 0.4073, 'learning_rate': 4.1381160610908134e-05, 'epoch': 0.82}

 27%|██▋       | 1220/4476 [7:54:56<21:08:25, 23.37s/it]
 27%|██▋       | 1221/4476 [7:55:19<21:07:29, 23.36s/it]
 27%|██▋       | 1222/4476 [7:55:43<21:06:51, 23.36s/it]
 27%|██▋       | 1223/4476 [7:56:06<21:04:29, 23.32s/it]
 27%|██▋       | 1224/4476 [7:56:30<21:10:20, 23.44s/it]
 27%|██▋       | 1225/4476 [7:56:53<21:07:42, 23.40s/it]
 27%|██▋       | 1226/4476 [7:57:16<21:06:11, 23.38s/it]
 27%|██▋       | 1227/4476 [7:57:40<21:10:04, 23.45s/it]
 27%|██▋       | 1228/4476 [7:58:03<21:07:11, 23.41s/it]
 27%|██▋       | 1229/4476 [7:58:27<21:06:30, 23.40s/it]
 27%|██▋       | 1230/4476 [7:58:50<21:05:14, 23.39s/it]
                                                        
{'loss': 0.4138, 'learning_rate': 4.124820644226936e-05, 'epoch': 0.82}

 27%|██▋       | 1230/4476 [7:58:50<21:05:14, 23.39s/it]
 28%|██▊       | 1231/4476 [7:59:13<21:04:27, 23.38s/it]
 28%|██▊       | 1232/4476 [7:59:37<21:10:35, 23.50s/it]
 28%|██▊       | 1233/4476 [8:00:01<21:11:01, 23.52s/it]
 28%|██▊       | 1234/4476 [8:00:24<21:10:56, 23.52s/it]
 28%|██▊       | 1235/4476 [8:00:48<21:07:53, 23.47s/it]
 28%|██▊       | 1236/4476 [8:01:11<21:04:54, 23.42s/it]
 28%|██▊       | 1237/4476 [8:01:34<21:01:56, 23.38s/it]
 28%|██▊       | 1238/4476 [8:01:57<20:56:13, 23.28s/it]
 28%|██▊       | 1239/4476 [8:02:21<20:56:45, 23.30s/it]
 28%|██▊       | 1240/4476 [8:02:44<20:56:53, 23.30s/it]
                                                        
{'loss': 0.4139, 'learning_rate': 4.111445184385616e-05, 'epoch': 0.83}

 28%|██▊       | 1240/4476 [8:02:44<20:56:53, 23.30s/it]
 28%|██▊       | 1241/4476 [8:03:08<21:03:25, 23.43s/it]
 28%|██▊       | 1242/4476 [8:03:31<21:00:09, 23.38s/it]
 28%|██▊       | 1243/4476 [8:03:54<20:59:39, 23.38s/it]
 28%|██▊       | 1244/4476 [8:04:18<20:57:40, 23.35s/it]
 28%|██▊       | 1245/4476 [8:04:41<20:59:33, 23.39s/it]
 28%|██▊       | 1246/4476 [8:05:04<20:54:43, 23.31s/it]
 28%|██▊       | 1247/4476 [8:05:27<20:53:22, 23.29s/it]
 28%|██▊       | 1248/4476 [8:05:51<20:51:26, 23.26s/it]
 28%|██▊       | 1249/4476 [8:06:14<20:55:28, 23.34s/it]
 28%|██▊       | 1250/4476 [8:06:37<20:54:34, 23.33s/it]
                                                        
{'loss': 0.4062, 'learning_rate': 4.097990340477507e-05, 'epoch': 0.84}

 28%|██▊       | 1250/4476 [8:06:37<20:54:34, 23.33s/it]
 28%|██▊       | 1251/4476 [8:07:01<20:50:52, 23.27s/it]
 28%|██▊       | 1252/4476 [8:07:24<20:55:32, 23.37s/it]
 28%|██▊       | 1253/4476 [8:07:48<20:58:07, 23.42s/it]
 28%|██▊       | 1254/4476 [8:08:11<20:55:48, 23.39s/it]
 28%|██▊       | 1255/4476 [8:08:34<20:57:08, 23.42s/it]
 28%|██▊       | 1256/4476 [8:08:58<20:58:33, 23.45s/it]
 28%|██▊       | 1257/4476 [8:09:21<20:52:08, 23.34s/it]
 28%|██▊       | 1258/4476 [8:09:44<20:50:14, 23.31s/it]
 28%|██▊       | 1259/4476 [8:10:08<20:49:38, 23.31s/it]
 28%|██▊       | 1260/4476 [8:10:31<20:51:34, 23.35s/it]
                                                        
{'loss': 0.4044, 'learning_rate': 4.0844567753239276e-05, 'epoch': 0.84}

 28%|██▊       | 1260/4476 [8:10:31<20:51:34, 23.35s/it]
 28%|██▊       | 1261/4476 [8:10:55<20:56:06, 23.44s/it]
 28%|██▊       | 1262/4476 [8:11:18<20:53:40, 23.40s/it]
 28%|██▊       | 1263/4476 [8:11:41<20:50:13, 23.35s/it]
 28%|██▊       | 1264/4476 [8:12:05<20:55:17, 23.45s/it]
 28%|██▊       | 1265/4476 [8:12:28<20:47:00, 23.30s/it]
 28%|██▊       | 1266/4476 [8:12:51<20:39:51, 23.17s/it]
 28%|██▊       | 1267/4476 [8:13:14<20:34:21, 23.08s/it]
 28%|██▊       | 1268/4476 [8:13:37<20:37:20, 23.14s/it]
 28%|██▊       | 1269/4476 [8:14:00<20:39:40, 23.19s/it]
 28%|██▊       | 1270/4476 [8:14:23<20:36:10, 23.14s/it]
                                                        
{'loss': 0.3978, 'learning_rate': 4.070845155624221e-05, 'epoch': 0.85}

 28%|██▊       | 1270/4476 [8:14:23<20:36:10, 23.14s/it]
 28%|██▊       | 1271/4476 [8:14:47<20:38:46, 23.19s/it]
 28%|██▊       | 1272/4476 [8:15:10<20:43:40, 23.29s/it]
 28%|██▊       | 1273/4476 [8:15:34<20:47:29, 23.37s/it]
 28%|██▊       | 1274/4476 [8:15:57<20:46:34, 23.36s/it]
 28%|██▊       | 1275/4476 [8:16:20<20:43:45, 23.31s/it]
 29%|██▊       | 1276/4476 [8:16:44<20:46:05, 23.36s/it]
 29%|██▊       | 1277/4476 [8:17:07<20:48:07, 23.41s/it]
 29%|██▊       | 1278/4476 [8:17:30<20:45:26, 23.37s/it]
 29%|██▊       | 1279/4476 [8:17:54<20:43:04, 23.33s/it]
 29%|██▊       | 1280/4476 [8:18:17<20:42:23, 23.32s/it]
                                                        
{'loss': 0.4102, 'learning_rate': 4.0571561519228984e-05, 'epoch': 0.86}

 29%|██▊       | 1280/4476 [8:18:17<20:42:23, 23.32s/it]
 29%|██▊       | 1281/4476 [8:18:41<20:47:40, 23.43s/it]
 29%|██▊       | 1282/4476 [8:19:04<20:43:16, 23.36s/it]
 29%|██▊       | 1283/4476 [8:19:27<20:40:53, 23.32s/it]
 29%|██▊       | 1284/4476 [8:19:51<20:42:51, 23.36s/it]
 29%|██▊       | 1285/4476 [8:20:14<20:37:38, 23.27s/it]
 29%|██▊       | 1286/4476 [8:20:37<20:33:17, 23.20s/it]
 29%|██▉       | 1287/4476 [8:21:00<20:36:50, 23.27s/it]
 29%|██▉       | 1288/4476 [8:21:24<20:39:40, 23.33s/it]
 29%|██▉       | 1289/4476 [8:21:47<20:35:18, 23.26s/it]
 29%|██▉       | 1290/4476 [8:22:10<20:38:37, 23.33s/it]
                                                        
{'loss': 0.4052, 'learning_rate': 4.043390438576616e-05, 'epoch': 0.86}

 29%|██▉       | 1290/4476 [8:22:10<20:38:37, 23.33s/it]
 29%|██▉       | 1291/4476 [8:22:34<20:42:32, 23.41s/it]
 29%|██▉       | 1292/4476 [8:22:57<20:41:16, 23.39s/it]
 29%|██▉       | 1293/4476 [8:23:20<20:40:55, 23.39s/it]
 29%|██▉       | 1294/4476 [8:23:44<20:39:24, 23.37s/it]
 29%|██▉       | 1295/4476 [8:24:08<20:45:45, 23.50s/it]
 29%|██▉       | 1296/4476 [8:24:31<20:41:39, 23.43s/it]
 29%|██▉       | 1297/4476 [8:24:54<20:42:27, 23.45s/it]
 29%|██▉       | 1298/4476 [8:25:18<20:40:13, 23.42s/it]
 29%|██▉       | 1299/4476 [8:25:41<20:32:32, 23.28s/it]
 29%|██▉       | 1300/4476 [8:26:04<20:31:30, 23.27s/it]
                                                        
{'loss': 0.4048, 'learning_rate': 4.029548693720949e-05, 'epoch': 0.87}

 29%|██▉       | 1300/4476 [8:26:04<20:31:30, 23.27s/it]
 29%|██▉       | 1301/4476 [8:26:27<20:29:06, 23.23s/it]
 29%|██▉       | 1302/4476 [8:26:50<20:29:34, 23.24s/it]
 29%|██▉       | 1303/4476 [8:27:14<20:29:03, 23.24s/it]
 29%|██▉       | 1304/4476 [8:27:37<20:30:20, 23.27s/it]
 29%|██▉       | 1305/4476 [8:28:00<20:32:32, 23.32s/it]
 29%|██▉       | 1306/4476 [8:28:24<20:35:24, 23.38s/it]
 29%|██▉       | 1307/4476 [8:28:47<20:33:16, 23.35s/it]
 29%|██▉       | 1308/4476 [8:29:10<20:33:13, 23.36s/it]
 29%|██▉       | 1309/4476 [8:29:34<20:30:28, 23.31s/it]
 29%|██▉       | 1310/4476 [8:29:57<20:35:25, 23.41s/it]
                                                        
{'loss': 0.4008, 'learning_rate': 4.0156315992369864e-05, 'epoch': 0.88}

 29%|██▉       | 1310/4476 [8:29:57<20:35:25, 23.41s/it]
 29%|██▉       | 1311/4476 [8:30:21<20:32:55, 23.37s/it]
 29%|██▉       | 1312/4476 [8:30:44<20:29:31, 23.32s/it]
 29%|██▉       | 1313/4476 [8:31:07<20:33:44, 23.40s/it]
 29%|██▉       | 1314/4476 [8:31:31<20:31:37, 23.37s/it]
 29%|██▉       | 1315/4476 [8:31:54<20:25:26, 23.26s/it]
 29%|██▉       | 1316/4476 [8:32:17<20:29:53, 23.35s/it]
 29%|██▉       | 1317/4476 [8:32:40<20:23:52, 23.25s/it]
 29%|██▉       | 1318/4476 [8:33:03<20:22:29, 23.23s/it]
 29%|██▉       | 1319/4476 [8:33:27<20:25:31, 23.29s/it]
 29%|██▉       | 1320/4476 [8:33:50<20:28:01, 23.35s/it]
                                                        
{'loss': 0.4038, 'learning_rate': 4.001639840717741e-05, 'epoch': 0.88}

 29%|██▉       | 1320/4476 [8:33:50<20:28:01, 23.35s/it]
 30%|██▉       | 1321/4476 [8:34:14<20:33:31, 23.46s/it]
 30%|██▉       | 1322/4476 [8:34:38<20:34:23, 23.48s/it]
 30%|██▉       | 1323/4476 [8:35:01<20:34:42, 23.50s/it]
 30%|██▉       | 1324/4476 [8:35:25<20:33:36, 23.48s/it]
 30%|██▉       | 1325/4476 [8:35:48<20:33:24, 23.49s/it]
 30%|██▉       | 1326/4476 [8:36:11<20:29:46, 23.42s/it]
 30%|██▉       | 1327/4476 [8:36:35<20:28:34, 23.41s/it]
 30%|██▉       | 1328/4476 [8:36:59<20:33:48, 23.52s/it]
 30%|██▉       | 1329/4476 [8:37:22<20:32:03, 23.49s/it]
 30%|██▉       | 1330/4476 [8:37:45<20:29:22, 23.45s/it]
                                                        
{'loss': 0.408, 'learning_rate': 3.9875741074343744e-05, 'epoch': 0.89}

 30%|██▉       | 1330/4476 [8:37:45<20:29:22, 23.45s/it]
 30%|██▉       | 1331/4476 [8:38:09<20:26:01, 23.39s/it]
 30%|██▉       | 1332/4476 [8:38:32<20:22:46, 23.34s/it]
 30%|██▉       | 1333/4476 [8:38:55<20:21:44, 23.32s/it]
 30%|██▉       | 1334/4476 [8:39:18<20:18:34, 23.27s/it]
 30%|██▉       | 1335/4476 [8:39:42<20:21:17, 23.33s/it]
 30%|██▉       | 1336/4476 [8:40:05<20:23:21, 23.38s/it]
 30%|██▉       | 1337/4476 [8:40:29<20:24:29, 23.41s/it]
 30%|██▉       | 1338/4476 [8:40:52<20:20:14, 23.33s/it]
 30%|██▉       | 1339/4476 [8:41:15<20:22:47, 23.39s/it]
 30%|██▉       | 1340/4476 [8:41:39<20:28:59, 23.51s/it]
                                                        
{'loss': 0.406, 'learning_rate': 3.973435092302239e-05, 'epoch': 0.9}

 30%|██▉       | 1340/4476 [8:41:39<20:28:59, 23.51s/it]
 30%|██▉       | 1341/4476 [8:42:03<20:27:47, 23.50s/it]
 30%|██▉       | 1342/4476 [8:42:26<20:29:41, 23.54s/it]
 30%|███       | 1343/4476 [8:42:49<20:21:58, 23.40s/it]
 30%|███       | 1344/4476 [8:43:12<20:18:28, 23.34s/it]
 30%|███       | 1345/4476 [8:43:35<20:10:13, 23.19s/it]
 30%|███       | 1346/4476 [8:43:59<20:11:52, 23.23s/it]
 30%|███       | 1347/4476 [8:44:22<20:07:54, 23.16s/it]
 30%|███       | 1348/4476 [8:44:45<20:11:49, 23.24s/it]
 30%|███       | 1349/4476 [8:45:08<20:07:42, 23.17s/it]
 30%|███       | 1350/4476 [8:45:31<20:05:47, 23.14s/it]
                                                        
{'loss': 0.3991, 'learning_rate': 3.959223491846749e-05, 'epoch': 0.9}

 30%|███       | 1350/4476 [8:45:31<20:05:47, 23.14s/it]
 30%|███       | 1351/4476 [8:45:55<20:13:50, 23.31s/it]
 30%|███       | 1352/4476 [8:46:18<20:13:18, 23.30s/it]
 30%|███       | 1353/4476 [8:46:41<20:12:24, 23.29s/it]
 30%|███       | 1354/4476 [8:47:05<20:12:56, 23.31s/it]
 30%|███       | 1355/4476 [8:47:28<20:10:35, 23.27s/it]
 30%|███       | 1356/4476 [8:47:51<20:04:22, 23.16s/it]
 30%|███       | 1357/4476 [8:48:14<20:02:24, 23.13s/it]
 30%|███       | 1358/4476 [8:48:37<20:03:55, 23.17s/it]
 30%|███       | 1359/4476 [8:49:01<20:06:26, 23.22s/it]
 30%|███       | 1360/4476 [8:49:24<20:08:38, 23.27s/it]
                                                        
{'loss': 0.4091, 'learning_rate': 3.94494000616906e-05, 'epoch': 0.91}

 30%|███       | 1360/4476 [8:49:24<20:08:38, 23.27s/it]
 30%|███       | 1361/4476 [8:49:47<20:03:36, 23.18s/it]
 30%|███       | 1362/4476 [8:50:10<20:05:04, 23.22s/it]
 30%|███       | 1363/4476 [8:50:33<20:05:20, 23.23s/it]
 30%|███       | 1364/4476 [8:50:57<20:03:34, 23.21s/it]
 30%|███       | 1365/4476 [8:51:20<20:00:05, 23.15s/it]
 31%|███       | 1366/4476 [8:51:43<20:00:21, 23.16s/it]
 31%|███       | 1367/4476 [8:52:06<19:57:33, 23.11s/it]
 31%|███       | 1368/4476 [8:52:29<20:02:49, 23.22s/it]
 31%|███       | 1369/4476 [8:52:53<20:06:51, 23.31s/it]
 31%|███       | 1370/4476 [8:53:16<20:07:22, 23.32s/it]
                                                        
{'loss': 0.4, 'learning_rate': 3.93058533891159e-05, 'epoch': 0.92}

 31%|███       | 1370/4476 [8:53:16<20:07:22, 23.32s/it]
 31%|███       | 1371/4476 [8:53:39<20:04:12, 23.27s/it]
 31%|███       | 1372/4476 [8:54:02<20:02:52, 23.25s/it]
 31%|███       | 1373/4476 [8:54:26<20:08:45, 23.37s/it]
 31%|███       | 1374/4476 [8:54:50<20:10:18, 23.41s/it]
 31%|███       | 1375/4476 [8:55:13<20:07:29, 23.36s/it]
 31%|███       | 1376/4476 [8:55:36<20:05:36, 23.33s/it]
 31%|███       | 1377/4476 [8:55:59<20:01:59, 23.27s/it]
 31%|███       | 1378/4476 [8:56:23<20:00:58, 23.26s/it]
 31%|███       | 1379/4476 [8:56:46<20:01:44, 23.28s/it]
 31%|███       | 1380/4476 [8:57:09<20:01:42, 23.29s/it]
                                                        
{'loss': 0.4112, 'learning_rate': 3.916160197223344e-05, 'epoch': 0.92}

 31%|███       | 1380/4476 [8:57:09<20:01:42, 23.29s/it]
 31%|███       | 1381/4476 [8:57:33<20:05:17, 23.37s/it]
 31%|███       | 1382/4476 [8:57:56<20:05:03, 23.37s/it]
 31%|███       | 1383/4476 [8:58:20<20:05:46, 23.39s/it]
 31%|███       | 1384/4476 [8:58:43<20:04:15, 23.37s/it]
 31%|███       | 1385/4476 [8:59:06<20:04:18, 23.38s/it]
 31%|███       | 1386/4476 [8:59:30<20:09:35, 23.49s/it]
 31%|███       | 1387/4476 [8:59:53<20:03:26, 23.38s/it]
 31%|███       | 1388/4476 [9:00:16<19:58:41, 23.29s/it]
 31%|███       | 1389/4476 [9:00:39<19:58:07, 23.29s/it]
 31%|███       | 1390/4476 [9:01:03<19:55:11, 23.24s/it]
                                                        
{'loss': 0.4024, 'learning_rate': 3.901665291725091e-05, 'epoch': 0.93}

 31%|███       | 1390/4476 [9:01:03<19:55:11, 23.24s/it]
 31%|███       | 1391/4476 [9:01:26<19:55:38, 23.25s/it]
 31%|███       | 1392/4476 [9:01:49<19:58:57, 23.33s/it]
 31%|███       | 1393/4476 [9:02:13<19:59:20, 23.34s/it]
 31%|███       | 1394/4476 [9:02:36<19:57:55, 23.32s/it]
 31%|███       | 1395/4476 [9:02:59<19:56:35, 23.30s/it]
 31%|███       | 1396/4476 [9:03:22<19:52:24, 23.23s/it]
 31%|███       | 1397/4476 [9:03:46<19:54:12, 23.27s/it]
 31%|███       | 1398/4476 [9:04:09<19:52:56, 23.25s/it]
 31%|███▏      | 1399/4476 [9:04:32<19:53:05, 23.26s/it]
 31%|███▏      | 1400/4476 [9:04:56<19:57:05, 23.35s/it]
                                                        
{'loss': 0.4048, 'learning_rate': 3.887101336474346e-05, 'epoch': 0.94}

 31%|███▏      | 1400/4476 [9:04:56<19:57:05, 23.35s/it][INFO|trainer.py:2939] 2023-11-12 12:28:43,500 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1400
[INFO|tokenization_utils_base.py:2437] 2023-11-12 12:28:43,531 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1400/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 12:28:43,531 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1400/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 12:28:43,531 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1400/added_tokens.json

 31%|███▏      | 1401/4476 [9:05:19<19:56:51, 23.35s/it]
 31%|███▏      | 1402/4476 [9:05:43<19:59:40, 23.42s/it]
 31%|███▏      | 1403/4476 [9:06:06<19:57:49, 23.39s/it]
 31%|███▏      | 1404/4476 [9:06:29<19:56:44, 23.37s/it]
 31%|███▏      | 1405/4476 [9:06:53<19:54:56, 23.35s/it]
 31%|███▏      | 1406/4476 [9:07:16<19:52:10, 23.30s/it]
 31%|███▏      | 1407/4476 [9:07:40<19:57:56, 23.42s/it]
 31%|███▏      | 1408/4476 [9:08:03<19:59:11, 23.45s/it]
 31%|███▏      | 1409/4476 [9:08:27<20:01:16, 23.50s/it]
 32%|███▏      | 1410/4476 [9:08:50<19:58:11, 23.45s/it]
                                                        
{'loss': 0.4112, 'learning_rate': 3.8724690489302004e-05, 'epoch': 0.94}

 32%|███▏      | 1410/4476 [9:08:50<19:58:11, 23.45s/it]
 32%|███▏      | 1411/4476 [9:09:13<19:53:05, 23.36s/it]
 32%|███▏      | 1412/4476 [9:09:36<19:44:00, 23.19s/it]
 32%|███▏      | 1413/4476 [9:09:59<19:47:17, 23.26s/it]
 32%|███▏      | 1414/4476 [9:10:23<19:49:52, 23.32s/it]
 32%|███▏      | 1415/4476 [9:10:46<19:43:11, 23.19s/it]
 32%|███▏      | 1416/4476 [9:11:09<19:45:31, 23.25s/it]
 32%|███▏      | 1417/4476 [9:11:33<19:50:48, 23.36s/it]
 32%|███▏      | 1418/4476 [9:11:56<19:50:21, 23.36s/it]
 32%|███▏      | 1419/4476 [9:12:20<19:51:43, 23.39s/it]
 32%|███▏      | 1420/4476 [9:12:43<19:49:34, 23.36s/it]
                                                        
{'loss': 0.3947, 'learning_rate': 3.857769149917973e-05, 'epoch': 0.95}

 32%|███▏      | 1420/4476 [9:12:43<19:49:34, 23.36s/it]
 32%|███▏      | 1421/4476 [9:13:06<19:48:06, 23.33s/it]
 32%|███▏      | 1422/4476 [9:13:30<19:52:56, 23.44s/it]
 32%|███▏      | 1423/4476 [9:13:53<19:50:21, 23.39s/it]
 32%|███▏      | 1424/4476 [9:14:16<19:42:33, 23.25s/it]
 32%|███▏      | 1425/4476 [9:14:39<19:43:52, 23.28s/it]
 32%|███▏      | 1426/4476 [9:15:03<19:47:35, 23.36s/it]
 32%|███▏      | 1427/4476 [9:15:26<19:47:24, 23.37s/it]
 32%|███▏      | 1428/4476 [9:15:49<19:43:39, 23.30s/it]
 32%|███▏      | 1429/4476 [9:16:12<19:39:15, 23.22s/it]
 32%|███▏      | 1430/4476 [9:16:36<19:38:55, 23.22s/it]
                                                        
{'loss': 0.4005, 'learning_rate': 3.843002363593707e-05, 'epoch': 0.96}

 32%|███▏      | 1430/4476 [9:16:36<19:38:55, 23.22s/it]
 32%|███▏      | 1431/4476 [9:16:59<19:40:16, 23.26s/it]
 32%|███▏      | 1432/4476 [9:17:22<19:36:47, 23.20s/it]
 32%|███▏      | 1433/4476 [9:17:46<19:44:08, 23.35s/it]
 32%|███▏      | 1434/4476 [9:18:09<19:42:40, 23.33s/it]
 32%|███▏      | 1435/4476 [9:18:32<19:36:10, 23.21s/it]
 32%|███▏      | 1436/4476 [9:18:55<19:37:18, 23.24s/it]
 32%|███▏      | 1437/4476 [9:19:19<19:37:32, 23.25s/it]
 32%|███▏      | 1438/4476 [9:19:42<19:44:19, 23.39s/it]
 32%|███▏      | 1439/4476 [9:20:06<19:49:21, 23.50s/it]
 32%|███▏      | 1440/4476 [9:20:29<19:44:17, 23.40s/it]
                                                        
{'loss': 0.3976, 'learning_rate': 3.828169417408488e-05, 'epoch': 0.96}

 32%|███▏      | 1440/4476 [9:20:29<19:44:17, 23.40s/it]
 32%|███▏      | 1441/4476 [9:20:52<19:41:52, 23.36s/it]
 32%|███▏      | 1442/4476 [9:21:16<19:41:24, 23.36s/it]
 32%|███▏      | 1443/4476 [9:21:39<19:40:07, 23.35s/it]
 32%|███▏      | 1444/4476 [9:22:03<19:41:12, 23.37s/it]
 32%|███▏      | 1445/4476 [9:22:26<19:42:02, 23.40s/it]
 32%|███▏      | 1446/4476 [9:22:49<19:39:08, 23.35s/it]
 32%|███▏      | 1447/4476 [9:23:13<19:43:15, 23.44s/it]
 32%|███▏      | 1448/4476 [9:23:37<19:48:16, 23.55s/it]
 32%|███▏      | 1449/4476 [9:24:00<19:46:58, 23.53s/it]
 32%|███▏      | 1450/4476 [9:24:24<19:45:32, 23.51s/it]
                                                        
{'loss': 0.4006, 'learning_rate': 3.8132710420726146e-05, 'epoch': 0.97}

 32%|███▏      | 1450/4476 [9:24:24<19:45:32, 23.51s/it]
 32%|███▏      | 1451/4476 [9:24:47<19:42:13, 23.45s/it]
 32%|███▏      | 1452/4476 [9:25:10<19:41:42, 23.45s/it]
 32%|███▏      | 1453/4476 [9:25:34<19:40:02, 23.42s/it]
 32%|███▏      | 1454/4476 [9:25:57<19:38:11, 23.39s/it]
 33%|███▎      | 1455/4476 [9:26:21<19:41:26, 23.46s/it]
 33%|███▎      | 1456/4476 [9:26:44<19:45:34, 23.55s/it]
 33%|███▎      | 1457/4476 [9:27:08<19:44:19, 23.54s/it]
 33%|███▎      | 1458/4476 [9:27:32<19:45:50, 23.58s/it]
 33%|███▎      | 1459/4476 [9:27:55<19:40:21, 23.47s/it]
 33%|███▎      | 1460/4476 [9:28:18<19:37:09, 23.42s/it]
                                                        
{'loss': 0.398, 'learning_rate': 3.7983079715195984e-05, 'epoch': 0.98}

 33%|███▎      | 1460/4476 [9:28:18<19:37:09, 23.42s/it]
 33%|███▎      | 1461/4476 [9:28:41<19:34:44, 23.38s/it]
 33%|███▎      | 1462/4476 [9:29:05<19:33:14, 23.36s/it]
 33%|███▎      | 1463/4476 [9:29:28<19:38:12, 23.46s/it]
 33%|███▎      | 1464/4476 [9:29:52<19:37:18, 23.45s/it]
 33%|███▎      | 1465/4476 [9:30:15<19:31:44, 23.35s/it]
 33%|███▎      | 1466/4476 [9:30:38<19:32:59, 23.38s/it]
 33%|███▎      | 1467/4476 [9:31:02<19:29:45, 23.33s/it]
 33%|███▎      | 1468/4476 [9:31:25<19:35:49, 23.45s/it]
 33%|███▎      | 1469/4476 [9:31:49<19:31:40, 23.38s/it]
 33%|███▎      | 1470/4476 [9:32:12<19:29:23, 23.34s/it]
                                                        
{'loss': 0.3987, 'learning_rate': 3.78328094287001e-05, 'epoch': 0.99}

 33%|███▎      | 1470/4476 [9:32:12<19:29:23, 23.34s/it]
 33%|███▎      | 1471/4476 [9:32:35<19:27:43, 23.32s/it]
 33%|███▎      | 1472/4476 [9:32:59<19:29:53, 23.37s/it]
 33%|███▎      | 1473/4476 [9:33:22<19:31:46, 23.41s/it]
 33%|███▎      | 1474/4476 [9:33:46<19:31:12, 23.41s/it]
 33%|███▎      | 1475/4476 [9:34:09<19:32:40, 23.45s/it]
 33%|███▎      | 1476/4476 [9:34:33<19:33:37, 23.47s/it]
 33%|███▎      | 1477/4476 [9:34:56<19:34:20, 23.49s/it]
 33%|███▎      | 1478/4476 [9:35:20<19:35:33, 23.53s/it]
 33%|███▎      | 1479/4476 [9:35:43<19:30:49, 23.44s/it]
 33%|███▎      | 1480/4476 [9:36:06<19:28:09, 23.39s/it]
                                                        
{'loss': 0.4013, 'learning_rate': 3.768190696395162e-05, 'epoch': 0.99}

 33%|███▎      | 1480/4476 [9:36:06<19:28:09, 23.39s/it]
 33%|███▎      | 1481/4476 [9:36:30<19:30:21, 23.45s/it]
 33%|███▎      | 1482/4476 [9:36:53<19:30:12, 23.45s/it]
 33%|███▎      | 1483/4476 [9:37:16<19:26:09, 23.38s/it]
 33%|███▎      | 1484/4476 [9:37:39<19:17:04, 23.20s/it]
 33%|███▎      | 1485/4476 [9:38:02<19:15:38, 23.18s/it]
 33%|███▎      | 1486/4476 [9:38:26<19:15:38, 23.19s/it]
 33%|███▎      | 1487/4476 [9:38:49<19:15:50, 23.20s/it]
 33%|███▎      | 1488/4476 [9:39:12<19:19:11, 23.28s/it]
 33%|███▎      | 1489/4476 [9:39:36<19:19:53, 23.30s/it]
 33%|███▎      | 1490/4476 [9:39:59<19:16:33, 23.24s/it]
                                                        
{'loss': 0.4028, 'learning_rate': 3.7530379754806494e-05, 'epoch': 1.0}

 33%|███▎      | 1490/4476 [9:39:59<19:16:33, 23.24s/it]
 33%|███▎      | 1491/4476 [9:40:22<19:16:59, 23.26s/it]
 33%|███▎      | 1492/4476 [9:40:46<19:20:40, 23.34s/it]
 33%|███▎      | 1493/4476 [9:41:09<19:21:24, 23.36s/it]
 33%|███▎      | 1494/4476 [9:41:32<19:20:19, 23.35s/it]
 33%|███▎      | 1495/4476 [9:41:56<19:19:59, 23.35s/it]
 33%|███▎      | 1496/4476 [9:42:19<19:21:08, 23.38s/it]
 33%|███▎      | 1497/4476 [9:42:43<19:22:06, 23.41s/it]
 33%|███▎      | 1498/4476 [9:43:06<19:20:12, 23.38s/it]
 33%|███▎      | 1499/4476 [9:43:29<19:19:28, 23.37s/it]
 34%|███▎      | 1500/4476 [9:43:53<19:18:10, 23.35s/it]
                                                        
{'loss': 0.4036, 'learning_rate': 3.737823526589722e-05, 'epoch': 1.01}

 34%|███▎      | 1500/4476 [9:43:53<19:18:10, 23.35s/it]
 34%|███▎      | 1501/4476 [9:44:16<19:18:27, 23.36s/it]
 34%|███▎      | 1502/4476 [9:44:39<19:17:42, 23.36s/it]
 34%|███▎      | 1503/4476 [9:45:03<19:17:38, 23.36s/it]
 34%|███▎      | 1504/4476 [9:45:26<19:16:17, 23.34s/it]
 34%|███▎      | 1505/4476 [9:45:49<19:14:35, 23.32s/it]
 34%|███▎      | 1506/4476 [9:46:13<19:16:07, 23.36s/it]
 34%|███▎      | 1507/4476 [9:46:36<19:15:55, 23.36s/it]
 34%|███▎      | 1508/4476 [9:46:59<19:14:38, 23.34s/it]
 34%|███▎      | 1509/4476 [9:47:23<19:16:16, 23.38s/it]
 34%|███▎      | 1510/4476 [9:47:47<19:22:06, 23.51s/it]
                                                        
{'loss': 0.3937, 'learning_rate': 3.7225480992265125e-05, 'epoch': 1.01}

 34%|███▎      | 1510/4476 [9:47:47<19:22:06, 23.51s/it]
 34%|███▍      | 1511/4476 [9:48:10<19:17:54, 23.43s/it]
 34%|███▍      | 1512/4476 [9:48:33<19:15:26, 23.39s/it]
 34%|███▍      | 1513/4476 [9:48:56<19:11:25, 23.32s/it]
 34%|███▍      | 1514/4476 [9:49:20<19:12:48, 23.35s/it]
 34%|███▍      | 1515/4476 [9:49:43<19:11:33, 23.33s/it]
 34%|███▍      | 1516/4476 [9:50:06<19:11:05, 23.33s/it]
 34%|███▍      | 1517/4476 [9:50:30<19:09:04, 23.30s/it]
 34%|███▍      | 1518/4476 [9:50:53<19:07:28, 23.28s/it]
 34%|███▍      | 1519/4476 [9:51:16<19:11:18, 23.36s/it]
 34%|███▍      | 1520/4476 [9:51:40<19:09:48, 23.34s/it]
                                                        
{'loss': 0.4007, 'learning_rate': 3.707212445899116e-05, 'epoch': 1.02}

 34%|███▍      | 1520/4476 [9:51:40<19:09:48, 23.34s/it]
 34%|███▍      | 1521/4476 [9:52:03<19:07:59, 23.31s/it]
 34%|███▍      | 1522/4476 [9:52:26<19:07:25, 23.31s/it]
 34%|███▍      | 1523/4476 [9:52:49<19:05:31, 23.28s/it]
 34%|███▍      | 1524/4476 [9:53:13<19:07:39, 23.33s/it]
 34%|███▍      | 1525/4476 [9:53:36<19:09:04, 23.36s/it]
 34%|███▍      | 1526/4476 [9:54:00<19:10:18, 23.40s/it]
 34%|███▍      | 1527/4476 [9:54:23<19:07:30, 23.35s/it]
 34%|███▍      | 1528/4476 [9:54:46<19:07:30, 23.35s/it]
 34%|███▍      | 1529/4476 [9:55:10<19:05:19, 23.32s/it]
 34%|███▍      | 1530/4476 [9:55:33<19:10:56, 23.44s/it]
                                                        
{'loss': 0.4004, 'learning_rate': 3.6918173220825204e-05, 'epoch': 1.03}

 34%|███▍      | 1530/4476 [9:55:33<19:10:56, 23.44s/it]
 34%|███▍      | 1531/4476 [9:55:57<19:14:44, 23.53s/it]
 34%|███▍      | 1532/4476 [9:56:21<19:14:20, 23.53s/it]
 34%|███▍      | 1533/4476 [9:56:44<19:17:41, 23.60s/it]
 34%|███▍      | 1534/4476 [9:57:08<19:17:36, 23.61s/it]
 34%|███▍      | 1535/4476 [9:57:31<19:10:37, 23.47s/it]
 34%|███▍      | 1536/4476 [9:57:54<19:06:12, 23.39s/it]
 34%|███▍      | 1537/4476 [9:58:18<19:07:51, 23.43s/it]
 34%|███▍      | 1538/4476 [9:58:41<19:05:53, 23.40s/it]
 34%|███▍      | 1539/4476 [9:59:04<19:02:25, 23.34s/it]
 34%|███▍      | 1540/4476 [9:59:28<19:05:02, 23.40s/it]
                                                        
{'loss': 0.4004, 'learning_rate': 3.6763634861813836e-05, 'epoch': 1.03}

 34%|███▍      | 1540/4476 [9:59:28<19:05:02, 23.40s/it]
 34%|███▍      | 1541/4476 [9:59:52<19:09:10, 23.49s/it]
 34%|███▍      | 1542/4476 [10:00:15<19:00:03, 23.31s/it]
 34%|███▍      | 1543/4476 [10:00:38<19:02:18, 23.37s/it]
 34%|███▍      | 1544/4476 [10:01:01<19:02:27, 23.38s/it]
 35%|███▍      | 1545/4476 [10:01:25<19:07:18, 23.49s/it]
 35%|███▍      | 1546/4476 [10:01:49<19:06:22, 23.48s/it]
 35%|███▍      | 1547/4476 [10:02:12<19:04:56, 23.45s/it]
 35%|███▍      | 1548/4476 [10:02:35<19:01:54, 23.40s/it]
 35%|███▍      | 1549/4476 [10:02:59<19:00:13, 23.37s/it]
 35%|███▍      | 1550/4476 [10:03:22<18:56:28, 23.30s/it]
                                                         
{'loss': 0.3991, 'learning_rate': 3.660851699492679e-05, 'epoch': 1.04}

 35%|███▍      | 1550/4476 [10:03:22<18:56:28, 23.30s/it]
 35%|███▍      | 1551/4476 [10:03:45<19:00:15, 23.39s/it]
 35%|███▍      | 1552/4476 [10:04:09<18:56:24, 23.32s/it]
 35%|███▍      | 1553/4476 [10:04:32<19:04:22, 23.49s/it]
 35%|███▍      | 1554/4476 [10:04:56<19:04:29, 23.50s/it]
 35%|███▍      | 1555/4476 [10:05:19<19:03:33, 23.49s/it]
 35%|███▍      | 1556/4476 [10:05:43<19:07:20, 23.58s/it]
 35%|███▍      | 1557/4476 [10:06:07<19:09:15, 23.62s/it]
 35%|███▍      | 1558/4476 [10:06:30<19:07:54, 23.60s/it]
 35%|███▍      | 1559/4476 [10:06:54<19:08:40, 23.63s/it]
 35%|███▍      | 1560/4476 [10:07:17<19:03:11, 23.52s/it]
                                                         
{'loss': 0.4042, 'learning_rate': 3.645282726168191e-05, 'epoch': 1.05}

 35%|███▍      | 1560/4476 [10:07:17<19:03:11, 23.52s/it]
 35%|███▍      | 1561/4476 [10:07:41<18:59:13, 23.45s/it]
 35%|███▍      | 1562/4476 [10:08:04<18:54:07, 23.35s/it]
 35%|███▍      | 1563/4476 [10:08:28<18:59:27, 23.47s/it]
 35%|███▍      | 1564/4476 [10:08:51<18:56:37, 23.42s/it]
 35%|███▍      | 1565/4476 [10:09:14<18:51:58, 23.33s/it]
 35%|███▍      | 1566/4476 [10:09:38<18:57:59, 23.46s/it]
 35%|███▌      | 1567/4476 [10:10:01<18:56:05, 23.43s/it]
 35%|███▌      | 1568/4476 [10:10:25<19:01:05, 23.54s/it]
 35%|███▌      | 1569/4476 [10:10:48<18:56:22, 23.45s/it]
 35%|███▌      | 1570/4476 [10:11:12<18:56:15, 23.46s/it]
                                                         
{'loss': 0.4043, 'learning_rate': 3.6296573331768664e-05, 'epoch': 1.05}

 35%|███▌      | 1570/4476 [10:11:12<18:56:15, 23.46s/it]
 35%|███▌      | 1571/4476 [10:11:35<18:57:23, 23.49s/it]
 35%|███▌      | 1572/4476 [10:11:59<19:00:00, 23.55s/it]
 35%|███▌      | 1573/4476 [10:12:22<18:55:54, 23.48s/it]
 35%|███▌      | 1574/4476 [10:12:46<18:59:38, 23.56s/it]
 35%|███▌      | 1575/4476 [10:13:09<18:55:35, 23.49s/it]
 35%|███▌      | 1576/4476 [10:13:33<18:53:51, 23.46s/it]
 35%|███▌      | 1577/4476 [10:13:56<18:50:11, 23.39s/it]
 35%|███▌      | 1578/4476 [10:14:19<18:50:57, 23.42s/it]
 35%|███▌      | 1579/4476 [10:14:42<18:45:11, 23.30s/it]
 35%|███▌      | 1580/4476 [10:15:06<18:43:20, 23.27s/it]
                                                         
{'loss': 0.3948, 'learning_rate': 3.613976290267036e-05, 'epoch': 1.06}

 35%|███▌      | 1580/4476 [10:15:06<18:43:20, 23.27s/it]
 35%|███▌      | 1581/4476 [10:15:29<18:45:31, 23.33s/it]
 35%|███▌      | 1582/4476 [10:15:53<18:47:07, 23.37s/it]
 35%|███▌      | 1583/4476 [10:16:16<18:48:39, 23.41s/it]
 35%|███▌      | 1584/4476 [10:16:39<18:45:54, 23.36s/it]
 35%|███▌      | 1585/4476 [10:17:03<18:46:06, 23.37s/it]
 35%|███▌      | 1586/4476 [10:17:26<18:45:49, 23.37s/it]
 35%|███▌      | 1587/4476 [10:17:50<18:49:35, 23.46s/it]
 35%|███▌      | 1588/4476 [10:18:13<18:49:46, 23.47s/it]
 36%|███▌      | 1589/4476 [10:18:37<18:47:07, 23.42s/it]
 36%|███▌      | 1590/4476 [10:19:00<18:45:27, 23.40s/it]
                                                         
{'loss': 0.3952, 'learning_rate': 3.598240369928494e-05, 'epoch': 1.07}

 36%|███▌      | 1590/4476 [10:19:00<18:45:27, 23.40s/it]
 36%|███▌      | 1591/4476 [10:19:23<18:42:41, 23.35s/it]
 36%|███▌      | 1592/4476 [10:19:47<18:48:03, 23.47s/it]
 36%|███▌      | 1593/4476 [10:20:10<18:47:37, 23.47s/it]
 36%|███▌      | 1594/4476 [10:20:33<18:40:37, 23.33s/it]
 36%|███▌      | 1595/4476 [10:20:57<18:40:58, 23.35s/it]
 36%|███▌      | 1596/4476 [10:21:20<18:36:48, 23.27s/it]
 36%|███▌      | 1597/4476 [10:21:43<18:38:11, 23.30s/it]
 36%|███▌      | 1598/4476 [10:22:06<18:36:11, 23.27s/it]
 36%|███▌      | 1599/4476 [10:22:30<18:36:41, 23.29s/it]
 36%|███▌      | 1600/4476 [10:22:53<18:36:55, 23.30s/it]
                                                         
{'loss': 0.4002, 'learning_rate': 3.5824503473544405e-05, 'epoch': 1.07}

 36%|███▌      | 1600/4476 [10:22:53<18:36:55, 23.30s/it][INFO|trainer.py:2939] 2023-11-12 13:46:40,777 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1600
[INFO|tokenization_utils_base.py:2437] 2023-11-12 13:46:40,815 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1600/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 13:46:40,816 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1600/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 13:46:40,816 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1600/added_tokens.json

 36%|███▌      | 1601/4476 [10:23:17<18:40:10, 23.38s/it]
 36%|███▌      | 1602/4476 [10:23:40<18:41:29, 23.41s/it]
 36%|███▌      | 1603/4476 [10:24:03<18:38:34, 23.36s/it]
 36%|███▌      | 1604/4476 [10:24:26<18:33:35, 23.26s/it]
 36%|███▌      | 1605/4476 [10:24:50<18:33:46, 23.28s/it]
 36%|███▌      | 1606/4476 [10:25:13<18:36:14, 23.34s/it]
 36%|███▌      | 1607/4476 [10:25:36<18:34:17, 23.30s/it]
 36%|███▌      | 1608/4476 [10:26:00<18:33:12, 23.29s/it]
 36%|███▌      | 1609/4476 [10:26:23<18:35:22, 23.34s/it]
 36%|███▌      | 1610/4476 [10:26:46<18:32:22, 23.29s/it]
                                                         
{'loss': 0.4079, 'learning_rate': 3.566607000403298e-05, 'epoch': 1.08}

 36%|███▌      | 1610/4476 [10:26:46<18:32:22, 23.29s/it]
 36%|███▌      | 1611/4476 [10:27:10<18:31:42, 23.28s/it]
 36%|███▌      | 1612/4476 [10:27:32<18:26:50, 23.19s/it]
 36%|███▌      | 1613/4476 [10:27:56<18:31:03, 23.28s/it]
 36%|███▌      | 1614/4476 [10:28:19<18:30:54, 23.29s/it]
 36%|███▌      | 1615/4476 [10:28:43<18:36:30, 23.42s/it]
 36%|███▌      | 1616/4476 [10:29:06<18:35:53, 23.41s/it]
 36%|███▌      | 1617/4476 [10:29:30<18:36:56, 23.44s/it]
 36%|███▌      | 1618/4476 [10:29:53<18:35:17, 23.41s/it]
 36%|███▌      | 1619/4476 [10:30:17<18:32:47, 23.37s/it]
 36%|███▌      | 1620/4476 [10:30:40<18:29:18, 23.30s/it]
                                                         
{'loss': 0.3942, 'learning_rate': 3.5523030408223166e-05, 'epoch': 1.09}

 36%|███▌      | 1620/4476 [10:30:40<18:29:18, 23.30s/it]
 36%|███▌      | 1621/4476 [10:31:03<18:24:36, 23.21s/it]
 36%|███▌      | 1622/4476 [10:31:26<18:27:19, 23.28s/it]
 36%|███▋      | 1623/4476 [10:31:50<18:32:36, 23.40s/it]
 36%|███▋      | 1624/4476 [10:32:13<18:30:15, 23.36s/it]
 36%|███▋      | 1625/4476 [10:32:37<18:33:42, 23.44s/it]
 36%|███▋      | 1626/4476 [10:33:00<18:33:39, 23.45s/it]
 36%|███▋      | 1627/4476 [10:33:23<18:23:53, 23.25s/it]
 36%|███▋      | 1628/4476 [10:33:46<18:21:21, 23.20s/it]
 36%|███▋      | 1629/4476 [10:34:09<18:24:32, 23.28s/it]
 36%|███▋      | 1630/4476 [10:34:33<18:24:21, 23.28s/it]
                                                         
{'loss': 0.3935, 'learning_rate': 3.5363605299319165e-05, 'epoch': 1.09}

 36%|███▋      | 1630/4476 [10:34:33<18:24:21, 23.28s/it]
 36%|███▋      | 1631/4476 [10:34:56<18:23:37, 23.28s/it]
 36%|███▋      | 1632/4476 [10:35:20<18:29:09, 23.40s/it]
 36%|███▋      | 1633/4476 [10:35:43<18:25:15, 23.33s/it]
 37%|███▋      | 1634/4476 [10:36:06<18:27:32, 23.38s/it]
 37%|███▋      | 1635/4476 [10:36:30<18:25:34, 23.35s/it]
 37%|███▋      | 1636/4476 [10:36:53<18:26:44, 23.38s/it]
 37%|███▋      | 1637/4476 [10:37:16<18:19:19, 23.23s/it]
 37%|███▋      | 1638/4476 [10:37:39<18:21:19, 23.28s/it]
 37%|███▋      | 1639/4476 [10:38:03<18:21:19, 23.29s/it]
 37%|███▋      | 1640/4476 [10:38:26<18:21:04, 23.29s/it]
                                                         
{'loss': 0.3898, 'learning_rate': 3.520366965171161e-05, 'epoch': 1.1}

 37%|███▋      | 1640/4476 [10:38:26<18:21:04, 23.29s/it]
 37%|███▋      | 1641/4476 [10:38:50<18:27:00, 23.43s/it]
 37%|███▋      | 1642/4476 [10:39:13<18:22:51, 23.35s/it]
 37%|███▋      | 1643/4476 [10:39:36<18:25:02, 23.40s/it]
 37%|███▋      | 1644/4476 [10:40:00<18:22:31, 23.36s/it]
 37%|███▋      | 1645/4476 [10:40:23<18:25:35, 23.43s/it]
 37%|███▋      | 1646/4476 [10:40:47<18:26:32, 23.46s/it]
 37%|███▋      | 1647/4476 [10:41:10<18:27:58, 23.50s/it]
 37%|███▋      | 1648/4476 [10:41:34<18:29:05, 23.53s/it]
 37%|███▋      | 1649/4476 [10:41:57<18:24:46, 23.45s/it]
 37%|███▋      | 1650/4476 [10:42:21<18:24:17, 23.45s/it]
                                                         
{'loss': 0.4006, 'learning_rate': 3.504323134425501e-05, 'epoch': 1.11}

 37%|███▋      | 1650/4476 [10:42:21<18:24:17, 23.45s/it]
 37%|███▋      | 1651/4476 [10:42:44<18:21:26, 23.39s/it]
 37%|███▋      | 1652/4476 [10:43:08<18:22:38, 23.43s/it]
 37%|███▋      | 1653/4476 [10:43:31<18:21:10, 23.40s/it]
 37%|███▋      | 1654/4476 [10:43:54<18:19:24, 23.38s/it]
 37%|███▋      | 1655/4476 [10:44:18<18:23:29, 23.47s/it]
 37%|███▋      | 1656/4476 [10:44:41<18:20:00, 23.40s/it]
 37%|███▋      | 1657/4476 [10:45:04<18:15:34, 23.32s/it]
 37%|███▋      | 1658/4476 [10:45:28<18:20:32, 23.43s/it]
 37%|███▋      | 1659/4476 [10:45:51<18:19:17, 23.41s/it]
 37%|███▋      | 1660/4476 [10:46:15<18:23:22, 23.51s/it]
                                                         
{'loss': 0.4089, 'learning_rate': 3.48822982805662e-05, 'epoch': 1.11}

 37%|███▋      | 1660/4476 [10:46:15<18:23:22, 23.51s/it]
 37%|███▋      | 1661/4476 [10:46:38<18:20:28, 23.46s/it]
 37%|███▋      | 1662/4476 [10:47:02<18:17:50, 23.41s/it]
 37%|███▋      | 1663/4476 [10:47:25<18:16:06, 23.38s/it]
 37%|███▋      | 1664/4476 [10:47:48<18:12:37, 23.31s/it]
 37%|███▋      | 1665/4476 [10:48:11<18:09:29, 23.25s/it]
 37%|███▋      | 1666/4476 [10:48:35<18:11:28, 23.31s/it]
 37%|███▋      | 1667/4476 [10:48:58<18:06:53, 23.22s/it]
 37%|███▋      | 1668/4476 [10:49:21<18:07:21, 23.23s/it]
 37%|███▋      | 1669/4476 [10:49:44<18:08:29, 23.27s/it]
 37%|███▋      | 1670/4476 [10:50:08<18:12:01, 23.35s/it]
                                                         
{'loss': 0.3982, 'learning_rate': 3.472087838863505e-05, 'epoch': 1.12}

 37%|███▋      | 1670/4476 [10:50:08<18:12:01, 23.35s/it]
 37%|███▋      | 1671/4476 [10:50:31<18:12:39, 23.37s/it]
 37%|███▋      | 1672/4476 [10:50:55<18:12:45, 23.38s/it]
 37%|███▋      | 1673/4476 [10:51:18<18:10:22, 23.34s/it]
 37%|███▋      | 1674/4476 [10:51:41<18:12:44, 23.40s/it]
 37%|███▋      | 1675/4476 [10:52:05<18:13:06, 23.42s/it]
 37%|███▋      | 1676/4476 [10:52:28<18:11:31, 23.39s/it]
 37%|███▋      | 1677/4476 [10:52:51<18:06:47, 23.30s/it]
 37%|███▋      | 1678/4476 [10:53:15<18:09:24, 23.36s/it]
 38%|███▊      | 1679/4476 [10:53:38<18:08:40, 23.35s/it]
 38%|███▊      | 1680/4476 [10:54:01<18:02:21, 23.23s/it]
                                                         
{'loss': 0.399, 'learning_rate': 3.455897962043387e-05, 'epoch': 1.13}

 38%|███▊      | 1680/4476 [10:54:01<18:02:21, 23.23s/it]
 38%|███▊      | 1681/4476 [10:54:24<18:00:21, 23.19s/it]
 38%|███▊      | 1682/4476 [10:54:48<18:08:41, 23.38s/it]
 38%|███▊      | 1683/4476 [10:55:11<18:09:07, 23.40s/it]
 38%|███▊      | 1684/4476 [10:55:35<18:10:00, 23.42s/it]
 38%|███▊      | 1685/4476 [10:55:58<18:06:46, 23.36s/it]
 38%|███▊      | 1686/4476 [10:56:21<17:59:35, 23.22s/it]
 38%|███▊      | 1687/4476 [10:56:44<18:00:31, 23.25s/it]
 38%|███▊      | 1688/4476 [10:57:08<17:59:46, 23.24s/it]
 38%|███▊      | 1689/4476 [10:57:31<18:00:33, 23.26s/it]
 38%|███▊      | 1690/4476 [10:57:54<18:01:35, 23.29s/it]
                                                         
{'loss': 0.3964, 'learning_rate': 3.4396609951525676e-05, 'epoch': 1.13}

 38%|███▊      | 1690/4476 [10:57:54<18:01:35, 23.29s/it]
 38%|███▊      | 1691/4476 [10:58:17<17:59:29, 23.26s/it]
 38%|███▊      | 1692/4476 [10:58:41<18:00:08, 23.28s/it]
 38%|███▊      | 1693/4476 [10:59:04<17:59:20, 23.27s/it]
 38%|███▊      | 1694/4476 [10:59:27<18:01:23, 23.32s/it]
 38%|███▊      | 1695/4476 [10:59:51<18:03:34, 23.38s/it]
 38%|███▊      | 1696/4476 [11:00:14<18:01:46, 23.35s/it]
 38%|███▊      | 1697/4476 [11:00:38<18:02:31, 23.37s/it]
 38%|███▊      | 1698/4476 [11:01:00<17:54:11, 23.20s/it]
 38%|███▊      | 1699/4476 [11:01:24<17:57:36, 23.28s/it]
 38%|███▊      | 1700/4476 [11:01:47<17:56:40, 23.27s/it]
                                                         
{'loss': 0.3909, 'learning_rate': 3.423377738067132e-05, 'epoch': 1.14}

 38%|███▊      | 1700/4476 [11:01:47<17:56:40, 23.27s/it]
 38%|███▊      | 1701/4476 [11:02:10<17:52:23, 23.19s/it]
 38%|███▊      | 1702/4476 [11:02:34<17:55:38, 23.27s/it]
 38%|███▊      | 1703/4476 [11:02:57<17:57:29, 23.31s/it]
 38%|███▊      | 1704/4476 [11:03:20<17:56:40, 23.30s/it]
 38%|███▊      | 1705/4476 [11:03:44<17:56:37, 23.31s/it]
 38%|███▊      | 1706/4476 [11:04:07<17:56:50, 23.33s/it]
 38%|███▊      | 1707/4476 [11:04:30<17:56:51, 23.33s/it]
 38%|███▊      | 1708/4476 [11:04:54<17:57:59, 23.37s/it]
 38%|███▊      | 1709/4476 [11:05:18<18:02:52, 23.48s/it]
 38%|███▊      | 1710/4476 [11:05:41<18:01:31, 23.46s/it]
                                                         
{'loss': 0.4015, 'learning_rate': 3.407048992943541e-05, 'epoch': 1.15}

 38%|███▊      | 1710/4476 [11:05:41<18:01:31, 23.46s/it]
 38%|███▊      | 1711/4476 [11:06:04<17:54:44, 23.32s/it]
 38%|███▊      | 1712/4476 [11:06:28<17:57:02, 23.38s/it]
 38%|███▊      | 1713/4476 [11:06:51<17:58:53, 23.43s/it]
 38%|███▊      | 1714/4476 [11:07:15<18:00:41, 23.48s/it]
 38%|███▊      | 1715/4476 [11:07:38<18:01:00, 23.49s/it]
 38%|███▊      | 1716/4476 [11:08:01<17:57:31, 23.42s/it]
 38%|███▊      | 1717/4476 [11:08:25<17:52:56, 23.33s/it]
 38%|███▊      | 1718/4476 [11:08:48<17:51:18, 23.31s/it]
 38%|███▊      | 1719/4476 [11:09:11<17:48:55, 23.26s/it]
 38%|███▊      | 1720/4476 [11:09:34<17:46:21, 23.22s/it]
                                                         
{'loss': 0.3915, 'learning_rate': 3.39067556417912e-05, 'epoch': 1.15}

 38%|███▊      | 1720/4476 [11:09:34<17:46:21, 23.22s/it]
 38%|███▊      | 1721/4476 [11:09:58<17:49:21, 23.29s/it]
 38%|███▊      | 1722/4476 [11:10:21<17:51:23, 23.34s/it]
 38%|███▊      | 1723/4476 [11:10:44<17:52:43, 23.38s/it]
 39%|███▊      | 1724/4476 [11:11:08<17:51:06, 23.35s/it]
 39%|███▊      | 1725/4476 [11:11:31<17:52:40, 23.40s/it]
 39%|███▊      | 1726/4476 [11:11:54<17:48:36, 23.32s/it]
 39%|███▊      | 1727/4476 [11:12:18<17:46:14, 23.27s/it]
 39%|███▊      | 1728/4476 [11:12:41<17:46:01, 23.28s/it]
 39%|███▊      | 1729/4476 [11:13:04<17:44:31, 23.25s/it]
 39%|███▊      | 1730/4476 [11:13:27<17:46:13, 23.30s/it]
                                                         
{'loss': 0.3845, 'learning_rate': 3.374258258372426e-05, 'epoch': 1.16}

 39%|███▊      | 1730/4476 [11:13:27<17:46:13, 23.30s/it]
 39%|███▊      | 1731/4476 [11:13:51<17:47:43, 23.34s/it]
 39%|███▊      | 1732/4476 [11:14:14<17:47:19, 23.34s/it]
 39%|███▊      | 1733/4476 [11:14:38<17:49:40, 23.40s/it]
 39%|███▊      | 1734/4476 [11:15:01<17:44:03, 23.28s/it]
 39%|███▉      | 1735/4476 [11:15:24<17:40:01, 23.20s/it]
 39%|███▉      | 1736/4476 [11:15:47<17:39:56, 23.21s/it]
 39%|███▉      | 1737/4476 [11:16:10<17:41:29, 23.25s/it]
 39%|███▉      | 1738/4476 [11:16:34<17:47:39, 23.40s/it]
 39%|███▉      | 1739/4476 [11:16:57<17:45:46, 23.36s/it]
 39%|███▉      | 1740/4476 [11:17:21<17:48:27, 23.43s/it]
                                                         
{'loss': 0.4018, 'learning_rate': 3.357797884283517e-05, 'epoch': 1.17}

 39%|███▉      | 1740/4476 [11:17:21<17:48:27, 23.43s/it]
 39%|███▉      | 1741/4476 [11:17:44<17:42:02, 23.30s/it]
 39%|███▉      | 1742/4476 [11:18:07<17:42:48, 23.32s/it]
 39%|███▉      | 1743/4476 [11:18:31<17:43:13, 23.34s/it]
 39%|███▉      | 1744/4476 [11:18:54<17:43:26, 23.36s/it]
 39%|███▉      | 1745/4476 [11:19:18<17:45:09, 23.40s/it]
 39%|███▉      | 1746/4476 [11:19:41<17:39:42, 23.29s/it]
 39%|███▉      | 1747/4476 [11:20:04<17:42:15, 23.35s/it]
 39%|███▉      | 1748/4476 [11:20:28<17:43:35, 23.39s/it]
 39%|███▉      | 1749/4476 [11:20:51<17:45:15, 23.44s/it]
 39%|███▉      | 1750/4476 [11:21:14<17:40:10, 23.33s/it]
                                                         
{'loss': 0.3914, 'learning_rate': 3.3412952527941096e-05, 'epoch': 1.17}

 39%|███▉      | 1750/4476 [11:21:14<17:40:10, 23.33s/it]
 39%|███▉      | 1751/4476 [11:21:38<17:40:19, 23.35s/it]
 39%|███▉      | 1752/4476 [11:22:01<17:38:10, 23.31s/it]
 39%|███▉      | 1753/4476 [11:22:24<17:39:49, 23.35s/it]
 39%|███▉      | 1754/4476 [11:22:48<17:38:27, 23.33s/it]
 39%|███▉      | 1755/4476 [11:23:11<17:41:30, 23.41s/it]
 39%|███▉      | 1756/4476 [11:23:34<17:39:29, 23.37s/it]
 39%|███▉      | 1757/4476 [11:23:58<17:40:20, 23.40s/it]
 39%|███▉      | 1758/4476 [11:24:21<17:36:53, 23.33s/it]
 39%|███▉      | 1759/4476 [11:24:44<17:35:26, 23.31s/it]
 39%|███▉      | 1760/4476 [11:25:07<17:28:15, 23.16s/it]
                                                         
{'loss': 0.3909, 'learning_rate': 3.32475117686763e-05, 'epoch': 1.18}

 39%|███▉      | 1760/4476 [11:25:07<17:28:15, 23.16s/it]
 39%|███▉      | 1761/4476 [11:25:31<17:30:23, 23.21s/it]
 39%|███▉      | 1762/4476 [11:25:53<17:26:17, 23.13s/it]
 39%|███▉      | 1763/4476 [11:26:17<17:27:27, 23.17s/it]
 39%|███▉      | 1764/4476 [11:26:40<17:27:40, 23.18s/it]
 39%|███▉      | 1765/4476 [11:27:03<17:29:29, 23.23s/it]
 39%|███▉      | 1766/4476 [11:27:26<17:29:24, 23.23s/it]
 39%|███▉      | 1767/4476 [11:27:50<17:34:13, 23.35s/it]
 39%|███▉      | 1768/4476 [11:28:13<17:32:21, 23.32s/it]
 40%|███▉      | 1769/4476 [11:28:36<17:29:18, 23.26s/it]
 40%|███▉      | 1770/4476 [11:29:00<17:27:44, 23.23s/it]
                                                         
{'loss': 0.3993, 'learning_rate': 3.308166471509171e-05, 'epoch': 1.19}

 40%|███▉      | 1770/4476 [11:29:00<17:27:44, 23.23s/it]
 40%|███▉      | 1771/4476 [11:29:23<17:29:27, 23.28s/it]
 40%|███▉      | 1772/4476 [11:29:46<17:30:33, 23.31s/it]
 40%|███▉      | 1773/4476 [11:30:10<17:29:11, 23.29s/it]
 40%|███▉      | 1774/4476 [11:30:33<17:25:45, 23.22s/it]
 40%|███▉      | 1775/4476 [11:30:56<17:31:47, 23.36s/it]
 40%|███▉      | 1776/4476 [11:31:20<17:30:52, 23.35s/it]
 40%|███▉      | 1777/4476 [11:31:43<17:27:33, 23.29s/it]
 40%|███▉      | 1778/4476 [11:32:06<17:28:37, 23.32s/it]
 40%|███▉      | 1779/4476 [11:32:30<17:33:24, 23.44s/it]
 40%|███▉      | 1780/4476 [11:32:53<17:30:10, 23.37s/it]
                                                         
{'loss': 0.3906, 'learning_rate': 3.2915419537253346e-05, 'epoch': 1.19}

 40%|███▉      | 1780/4476 [11:32:53<17:30:10, 23.37s/it]
 40%|███▉      | 1781/4476 [11:33:16<17:26:04, 23.29s/it]
 40%|███▉      | 1782/4476 [11:33:40<17:28:23, 23.35s/it]
 40%|███▉      | 1783/4476 [11:34:03<17:27:11, 23.33s/it]
 40%|███▉      | 1784/4476 [11:34:26<17:26:38, 23.33s/it]
 40%|███▉      | 1785/4476 [11:34:50<17:30:44, 23.43s/it]
 40%|███▉      | 1786/4476 [11:35:14<17:31:04, 23.44s/it]
 40%|███▉      | 1787/4476 [11:35:37<17:32:07, 23.48s/it]
 40%|███▉      | 1788/4476 [11:36:01<17:31:02, 23.46s/it]
 40%|███▉      | 1789/4476 [11:36:24<17:28:21, 23.41s/it]
 40%|███▉      | 1790/4476 [11:36:47<17:26:13, 23.37s/it]
                                                         
{'loss': 0.3897, 'learning_rate': 3.274878442483991e-05, 'epoch': 1.2}

 40%|███▉      | 1790/4476 [11:36:47<17:26:13, 23.37s/it]
 40%|████      | 1791/4476 [11:37:11<17:27:32, 23.41s/it]
 40%|████      | 1792/4476 [11:37:34<17:24:34, 23.35s/it]
 40%|████      | 1793/4476 [11:37:58<17:28:58, 23.46s/it]
 40%|████      | 1794/4476 [11:38:21<17:31:29, 23.52s/it]
 40%|████      | 1795/4476 [11:38:44<17:27:29, 23.44s/it]
 40%|████      | 1796/4476 [11:39:08<17:27:24, 23.45s/it]
 40%|████      | 1797/4476 [11:39:31<17:24:26, 23.39s/it]
 40%|████      | 1798/4476 [11:39:54<17:22:25, 23.36s/it]
 40%|████      | 1799/4476 [11:40:18<17:22:08, 23.36s/it]
 40%|████      | 1800/4476 [11:40:41<17:20:11, 23.32s/it]
                                                         
{'loss': 0.3954, 'learning_rate': 3.258176758673932e-05, 'epoch': 1.21}

 40%|████      | 1800/4476 [11:40:41<17:20:11, 23.32s/it][INFO|trainer.py:2939] 2023-11-12 15:04:28,764 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1800
[INFO|tokenization_utils_base.py:2437] 2023-11-12 15:04:28,795 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1800/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 15:04:28,795 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1800/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 15:04:28,795 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-1800/added_tokens.json

 40%|████      | 1801/4476 [11:41:05<17:25:13, 23.44s/it]
 40%|████      | 1802/4476 [11:41:29<17:29:02, 23.54s/it]
 40%|████      | 1803/4476 [11:41:52<17:30:23, 23.58s/it]
 40%|████      | 1804/4476 [11:42:16<17:26:29, 23.50s/it]
 40%|████      | 1805/4476 [11:42:39<17:26:11, 23.50s/it]
 40%|████      | 1806/4476 [11:43:02<17:21:55, 23.41s/it]
 40%|████      | 1807/4476 [11:43:25<17:14:44, 23.26s/it]
 40%|████      | 1808/4476 [11:43:48<17:13:13, 23.24s/it]
 40%|████      | 1809/4476 [11:44:12<17:14:13, 23.27s/it]
 40%|████      | 1810/4476 [11:44:35<17:09:14, 23.16s/it]
                                                         
{'loss': 0.3835, 'learning_rate': 3.241437725064431e-05, 'epoch': 1.21}

 40%|████      | 1810/4476 [11:44:35<17:09:14, 23.16s/it]
 40%|████      | 1811/4476 [11:44:58<17:13:46, 23.27s/it]
 40%|████      | 1812/4476 [11:45:22<17:17:24, 23.37s/it]
 41%|████      | 1813/4476 [11:45:45<17:15:31, 23.33s/it]
 41%|████      | 1814/4476 [11:46:08<17:17:38, 23.39s/it]
 41%|████      | 1815/4476 [11:46:32<17:18:40, 23.42s/it]
 41%|████      | 1816/4476 [11:46:55<17:16:29, 23.38s/it]
 41%|████      | 1817/4476 [11:47:18<17:13:21, 23.32s/it]
 41%|████      | 1818/4476 [11:47:42<17:17:58, 23.43s/it]
 41%|████      | 1819/4476 [11:48:06<17:18:10, 23.44s/it]
 41%|████      | 1820/4476 [11:48:29<17:16:22, 23.41s/it]
                                                         
{'loss': 0.3854, 'learning_rate': 3.224662166264711e-05, 'epoch': 1.22}

 41%|████      | 1820/4476 [11:48:29<17:16:22, 23.41s/it]
 41%|████      | 1821/4476 [11:48:52<17:17:06, 23.44s/it]
 41%|████      | 1822/4476 [11:49:16<17:18:47, 23.48s/it]
 41%|████      | 1823/4476 [11:49:39<17:15:32, 23.42s/it]
 41%|████      | 1824/4476 [11:50:03<17:12:44, 23.37s/it]
 41%|████      | 1825/4476 [11:50:26<17:16:14, 23.45s/it]
 41%|████      | 1826/4476 [11:50:49<17:12:34, 23.38s/it]
 41%|████      | 1827/4476 [11:51:13<17:10:51, 23.35s/it]
 41%|████      | 1828/4476 [11:51:36<17:11:43, 23.38s/it]
 41%|████      | 1829/4476 [11:51:59<17:08:36, 23.32s/it]
 41%|████      | 1830/4476 [11:52:22<17:02:01, 23.17s/it]
                                                         
{'loss': 0.3924, 'learning_rate': 3.207850908683322e-05, 'epoch': 1.23}

 41%|████      | 1830/4476 [11:52:22<17:02:01, 23.17s/it]
 41%|████      | 1831/4476 [11:52:45<17:03:01, 23.21s/it]
 41%|████      | 1832/4476 [11:53:09<17:10:50, 23.39s/it]
 41%|████      | 1833/4476 [11:53:32<17:04:46, 23.26s/it]
 41%|████      | 1834/4476 [11:53:56<17:10:19, 23.40s/it]
 41%|████      | 1835/4476 [11:54:19<17:07:13, 23.34s/it]
 41%|████      | 1836/4476 [11:54:42<17:07:17, 23.35s/it]
 41%|████      | 1837/4476 [11:55:06<17:04:34, 23.29s/it]
 41%|████      | 1838/4476 [11:55:29<17:08:34, 23.39s/it]
 41%|████      | 1839/4476 [11:55:53<17:10:11, 23.44s/it]
 41%|████      | 1840/4476 [11:56:16<17:07:14, 23.38s/it]
                                                         
{'loss': 0.3888, 'learning_rate': 3.191004780487434e-05, 'epoch': 1.23}

 41%|████      | 1840/4476 [11:56:16<17:07:14, 23.38s/it]
 41%|████      | 1841/4476 [11:56:39<17:07:22, 23.39s/it]
 41%|████      | 1842/4476 [11:57:03<17:05:53, 23.37s/it]
 41%|████      | 1843/4476 [11:57:26<17:08:17, 23.43s/it]
 41%|████      | 1844/4476 [11:57:50<17:05:41, 23.38s/it]
 41%|████      | 1845/4476 [11:58:13<17:00:24, 23.27s/it]
 41%|████      | 1846/4476 [11:58:36<16:59:39, 23.26s/it]
 41%|████▏     | 1847/4476 [11:59:00<17:06:09, 23.42s/it]
 41%|████▏     | 1848/4476 [11:59:23<17:10:08, 23.52s/it]
 41%|████▏     | 1849/4476 [11:59:47<17:08:57, 23.50s/it]
 41%|████▏     | 1850/4476 [12:00:10<17:02:59, 23.37s/it]
                                                         
{'loss': 0.3914, 'learning_rate': 3.1741246115620336e-05, 'epoch': 1.24}

 41%|████▏     | 1850/4476 [12:00:10<17:02:59, 23.37s/it]
 41%|████▏     | 1851/4476 [12:00:34<17:08:17, 23.50s/it]
 41%|████▏     | 1852/4476 [12:00:57<17:06:33, 23.47s/it]
 41%|████▏     | 1853/4476 [12:01:21<17:06:23, 23.48s/it]
 41%|████▏     | 1854/4476 [12:01:44<17:06:58, 23.50s/it]
 41%|████▏     | 1855/4476 [12:02:08<17:04:03, 23.44s/it]
 41%|████▏     | 1856/4476 [12:02:31<17:01:08, 23.38s/it]
 41%|████▏     | 1857/4476 [12:02:54<17:01:46, 23.41s/it]
 42%|████▏     | 1858/4476 [12:03:18<17:02:38, 23.44s/it]
 42%|████▏     | 1859/4476 [12:03:41<17:03:51, 23.47s/it]
 42%|████▏     | 1860/4476 [12:04:05<17:01:48, 23.44s/it]
                                                         
{'loss': 0.391, 'learning_rate': 3.157211233469042e-05, 'epoch': 1.25}

 42%|████▏     | 1860/4476 [12:04:05<17:01:48, 23.44s/it]
 42%|████▏     | 1861/4476 [12:04:28<17:02:43, 23.47s/it]
 42%|████▏     | 1862/4476 [12:04:52<17:00:47, 23.43s/it]
 42%|████▏     | 1863/4476 [12:05:15<16:59:22, 23.41s/it]
 42%|████▏     | 1864/4476 [12:05:38<16:57:29, 23.37s/it]
 42%|████▏     | 1865/4476 [12:06:02<16:58:58, 23.42s/it]
 42%|████▏     | 1866/4476 [12:06:25<16:53:12, 23.29s/it]
 42%|████▏     | 1867/4476 [12:06:48<16:50:36, 23.24s/it]
 42%|████▏     | 1868/4476 [12:07:11<16:52:04, 23.28s/it]
 42%|████▏     | 1869/4476 [12:07:34<16:50:41, 23.26s/it]
 42%|████▏     | 1870/4476 [12:07:58<16:56:24, 23.40s/it]
                                                         
{'loss': 0.3916, 'learning_rate': 3.140265479406358e-05, 'epoch': 1.25}

 42%|████▏     | 1870/4476 [12:07:58<16:56:24, 23.40s/it]
 42%|████▏     | 1871/4476 [12:08:21<16:52:50, 23.33s/it]
 42%|████▏     | 1872/4476 [12:08:44<16:50:47, 23.29s/it]
 42%|████▏     | 1873/4476 [12:09:08<16:48:55, 23.26s/it]
 42%|████▏     | 1874/4476 [12:09:31<16:48:30, 23.26s/it]
 42%|████▏     | 1875/4476 [12:09:54<16:49:21, 23.28s/it]
 42%|████▏     | 1876/4476 [12:10:17<16:46:24, 23.22s/it]
 42%|████▏     | 1877/4476 [12:10:40<16:44:57, 23.20s/it]
 42%|████▏     | 1878/4476 [12:11:04<16:45:17, 23.22s/it]
 42%|████▏     | 1879/4476 [12:11:27<16:46:59, 23.27s/it]
 42%|████▏     | 1880/4476 [12:11:51<16:50:17, 23.35s/it]
                                                         
{'loss': 0.4012, 'learning_rate': 3.1232881841668015e-05, 'epoch': 1.26}

 42%|████▏     | 1880/4476 [12:11:51<16:50:17, 23.35s/it]
 42%|████▏     | 1881/4476 [12:12:14<16:50:13, 23.36s/it]
 42%|████▏     | 1882/4476 [12:12:37<16:48:03, 23.32s/it]
 42%|████▏     | 1883/4476 [12:13:00<16:44:24, 23.24s/it]
 42%|████▏     | 1884/4476 [12:13:23<16:40:26, 23.16s/it]
 42%|████▏     | 1885/4476 [12:13:47<16:43:53, 23.25s/it]
 42%|████▏     | 1886/4476 [12:14:10<16:48:28, 23.36s/it]
 42%|████▏     | 1887/4476 [12:14:34<16:47:24, 23.35s/it]
 42%|████▏     | 1888/4476 [12:14:57<16:43:47, 23.27s/it]
 42%|████▏     | 1889/4476 [12:15:20<16:39:49, 23.19s/it]
 42%|████▏     | 1890/4476 [12:15:43<16:40:41, 23.22s/it]
                                                         
{'loss': 0.3934, 'learning_rate': 3.106280184096996e-05, 'epoch': 1.27}

 42%|████▏     | 1890/4476 [12:15:43<16:40:41, 23.22s/it]
 42%|████▏     | 1891/4476 [12:16:06<16:40:28, 23.22s/it]
 42%|████▏     | 1892/4476 [12:16:29<16:38:21, 23.18s/it]
 42%|████▏     | 1893/4476 [12:16:53<16:45:18, 23.35s/it]
 42%|████▏     | 1894/4476 [12:17:17<16:50:00, 23.47s/it]
 42%|████▏     | 1895/4476 [12:17:40<16:44:39, 23.36s/it]
 42%|████▏     | 1896/4476 [12:18:03<16:44:30, 23.36s/it]
 42%|████▏     | 1897/4476 [12:18:27<16:44:40, 23.37s/it]
 42%|████▏     | 1898/4476 [12:18:50<16:44:17, 23.37s/it]
 42%|████▏     | 1899/4476 [12:19:14<16:44:24, 23.39s/it]
 42%|████▏     | 1900/4476 [12:19:37<16:45:15, 23.41s/it]
                                                         
{'loss': 0.3908, 'learning_rate': 3.089242317056168e-05, 'epoch': 1.27}

 42%|████▏     | 1900/4476 [12:19:37<16:45:15, 23.41s/it]
 42%|████▏     | 1901/4476 [12:20:00<16:41:18, 23.33s/it]
 42%|████▏     | 1902/4476 [12:20:23<16:40:56, 23.33s/it]
 43%|████▎     | 1903/4476 [12:20:47<16:44:54, 23.43s/it]
 43%|████▎     | 1904/4476 [12:21:10<16:37:35, 23.27s/it]
 43%|████▎     | 1905/4476 [12:21:33<16:36:34, 23.26s/it]
 43%|████▎     | 1906/4476 [12:21:57<16:36:43, 23.27s/it]
 43%|████▎     | 1907/4476 [12:22:20<16:36:50, 23.28s/it]
 43%|████▎     | 1908/4476 [12:22:44<16:42:38, 23.43s/it]
 43%|████▎     | 1909/4476 [12:23:07<16:42:35, 23.43s/it]
 43%|████▎     | 1910/4476 [12:23:30<16:41:18, 23.41s/it]
                                                         
{'loss': 0.3972, 'learning_rate': 3.072175422374867e-05, 'epoch': 1.28}

 43%|████▎     | 1910/4476 [12:23:30<16:41:18, 23.41s/it]
 43%|████▎     | 1911/4476 [12:23:54<16:38:28, 23.36s/it]
 43%|████▎     | 1912/4476 [12:24:17<16:40:01, 23.40s/it]
 43%|████▎     | 1913/4476 [12:24:41<16:41:15, 23.44s/it]
 43%|████▎     | 1914/4476 [12:25:04<16:43:23, 23.50s/it]
 43%|████▎     | 1915/4476 [12:25:28<16:42:32, 23.49s/it]
 43%|████▎     | 1916/4476 [12:25:51<16:42:15, 23.49s/it]
 43%|████▎     | 1917/4476 [12:26:15<16:38:18, 23.41s/it]
 43%|████▎     | 1918/4476 [12:26:38<16:43:03, 23.53s/it]
 43%|████▎     | 1919/4476 [12:27:02<16:43:30, 23.55s/it]
 43%|████▎     | 1920/4476 [12:27:25<16:40:27, 23.48s/it]
                                                         
{'loss': 0.3963, 'learning_rate': 3.055080340813623e-05, 'epoch': 1.29}

 43%|████▎     | 1920/4476 [12:27:25<16:40:27, 23.48s/it]
 43%|████▎     | 1921/4476 [12:27:49<16:41:10, 23.51s/it]
 43%|████▎     | 1922/4476 [12:28:12<16:38:31, 23.46s/it]
 43%|████▎     | 1923/4476 [12:28:36<16:36:16, 23.41s/it]
 43%|████▎     | 1924/4476 [12:28:59<16:33:49, 23.37s/it]
 43%|████▎     | 1925/4476 [12:29:22<16:30:01, 23.29s/it]
 43%|████▎     | 1926/4476 [12:29:45<16:34:12, 23.39s/it]
 43%|████▎     | 1927/4476 [12:30:09<16:37:24, 23.48s/it]
 43%|████▎     | 1928/4476 [12:30:32<16:33:47, 23.40s/it]
 43%|████▎     | 1929/4476 [12:30:56<16:37:09, 23.49s/it]
 43%|████▎     | 1930/4476 [12:31:20<16:36:48, 23.49s/it]
                                                         
{'loss': 0.3941, 'learning_rate': 3.0379579145215287e-05, 'epoch': 1.29}

 43%|████▎     | 1930/4476 [12:31:20<16:36:48, 23.49s/it]
 43%|████▎     | 1931/4476 [12:31:43<16:31:15, 23.37s/it]
 43%|████▎     | 1932/4476 [12:32:06<16:29:30, 23.34s/it]
 43%|████▎     | 1933/4476 [12:32:29<16:31:40, 23.40s/it]
 43%|████▎     | 1934/4476 [12:32:53<16:32:22, 23.42s/it]
 43%|████▎     | 1935/4476 [12:33:16<16:29:03, 23.35s/it]
 43%|████▎     | 1936/4476 [12:33:39<16:26:39, 23.31s/it]
 43%|████▎     | 1937/4476 [12:34:03<16:24:57, 23.28s/it]
 43%|████▎     | 1938/4476 [12:34:26<16:27:45, 23.35s/it]
 43%|████▎     | 1939/4476 [12:34:50<16:28:27, 23.38s/it]
 43%|████▎     | 1940/4476 [12:35:13<16:27:42, 23.37s/it]
                                                         
{'loss': 0.3887, 'learning_rate': 3.0208089869947475e-05, 'epoch': 1.3}

 43%|████▎     | 1940/4476 [12:35:13<16:27:42, 23.37s/it]
 43%|████▎     | 1941/4476 [12:35:36<16:26:01, 23.34s/it]
 43%|████▎     | 1942/4476 [12:35:59<16:23:01, 23.28s/it]
 43%|████▎     | 1943/4476 [12:36:22<16:20:02, 23.21s/it]
 43%|████▎     | 1944/4476 [12:36:46<16:26:13, 23.37s/it]
 43%|████▎     | 1945/4476 [12:37:09<16:25:04, 23.35s/it]
 43%|████▎     | 1946/4476 [12:37:33<16:24:07, 23.34s/it]
 43%|████▎     | 1947/4476 [12:37:56<16:23:34, 23.34s/it]
 44%|████▎     | 1948/4476 [12:38:19<16:23:56, 23.35s/it]
 44%|████▎     | 1949/4476 [12:38:43<16:23:04, 23.34s/it]
 44%|████▎     | 1950/4476 [12:39:06<16:22:29, 23.34s/it]
                                                         
{'loss': 0.3879, 'learning_rate': 3.0036344030349644e-05, 'epoch': 1.31}

 44%|████▎     | 1950/4476 [12:39:06<16:22:29, 23.34s/it]
 44%|████▎     | 1951/4476 [12:39:29<16:22:48, 23.35s/it]
 44%|████▎     | 1952/4476 [12:39:53<16:21:36, 23.33s/it]
 44%|████▎     | 1953/4476 [12:40:16<16:19:24, 23.29s/it]
 44%|████▎     | 1954/4476 [12:40:39<16:18:50, 23.29s/it]
 44%|████▎     | 1955/4476 [12:41:03<16:23:19, 23.40s/it]
 44%|████▎     | 1956/4476 [12:41:26<16:20:28, 23.34s/it]
 44%|████▎     | 1957/4476 [12:41:49<16:20:07, 23.35s/it]
 44%|████▎     | 1958/4476 [12:42:13<16:19:11, 23.33s/it]
 44%|████▍     | 1959/4476 [12:42:36<16:15:38, 23.26s/it]
 44%|████▍     | 1960/4476 [12:42:59<16:16:05, 23.28s/it]
                                                         
{'loss': 0.3945, 'learning_rate': 2.9864350087077702e-05, 'epoch': 1.31}

 44%|████▍     | 1960/4476 [12:42:59<16:16:05, 23.28s/it]
 44%|████▍     | 1961/4476 [12:43:23<16:17:26, 23.32s/it]
 44%|████▍     | 1962/4476 [12:43:46<16:16:25, 23.30s/it]
 44%|████▍     | 1963/4476 [12:44:09<16:15:37, 23.29s/it]
 44%|████▍     | 1964/4476 [12:44:33<16:20:05, 23.41s/it]
 44%|████▍     | 1965/4476 [12:44:56<16:22:53, 23.49s/it]
 44%|████▍     | 1966/4476 [12:45:20<16:25:42, 23.56s/it]
 44%|████▍     | 1967/4476 [12:45:44<16:27:13, 23.61s/it]
 44%|████▍     | 1968/4476 [12:46:07<16:24:35, 23.55s/it]
 44%|████▍     | 1969/4476 [12:46:31<16:21:14, 23.48s/it]
 44%|████▍     | 1970/4476 [12:46:54<16:14:55, 23.34s/it]
                                                         
{'loss': 0.3909, 'learning_rate': 2.969211651300978e-05, 'epoch': 1.32}

 44%|████▍     | 1970/4476 [12:46:54<16:14:55, 23.34s/it]
 44%|████▍     | 1971/4476 [12:47:17<16:17:05, 23.40s/it]
 44%|████▍     | 1972/4476 [12:47:41<16:17:17, 23.42s/it]
 44%|████▍     | 1973/4476 [12:48:04<16:14:27, 23.36s/it]
 44%|████▍     | 1974/4476 [12:48:27<16:10:22, 23.27s/it]
 44%|████▍     | 1975/4476 [12:48:50<16:10:56, 23.29s/it]
 44%|████▍     | 1976/4476 [12:49:14<16:16:01, 23.42s/it]
 44%|████▍     | 1977/4476 [12:49:37<16:09:27, 23.28s/it]
 44%|████▍     | 1978/4476 [12:50:00<16:10:09, 23.30s/it]
 44%|████▍     | 1979/4476 [12:50:24<16:11:05, 23.33s/it]
 44%|████▍     | 1980/4476 [12:50:47<16:09:19, 23.30s/it]
                                                         
{'loss': 0.3871, 'learning_rate': 2.9519651792828877e-05, 'epoch': 1.33}

 44%|████▍     | 1980/4476 [12:50:47<16:09:19, 23.30s/it]
 44%|████▍     | 1981/4476 [12:51:10<16:05:09, 23.21s/it]
 44%|████▍     | 1982/4476 [12:51:33<16:06:02, 23.24s/it]
 44%|████▍     | 1983/4476 [12:51:57<16:08:13, 23.30s/it]
 44%|████▍     | 1984/4476 [12:52:20<16:05:31, 23.25s/it]
 44%|████▍     | 1985/4476 [12:52:43<16:05:07, 23.25s/it]
 44%|████▍     | 1986/4476 [12:53:06<16:05:17, 23.26s/it]
 44%|████▍     | 1987/4476 [12:53:30<16:04:48, 23.26s/it]
 44%|████▍     | 1988/4476 [12:53:53<16:01:22, 23.18s/it]
 44%|████▍     | 1989/4476 [12:54:16<16:02:27, 23.22s/it]
 44%|████▍     | 1990/4476 [12:54:39<16:05:04, 23.29s/it]
                                                         
{'loss': 0.3803, 'learning_rate': 2.9346964422604846e-05, 'epoch': 1.33}

 44%|████▍     | 1990/4476 [12:54:39<16:05:04, 23.29s/it]
 44%|████▍     | 1991/4476 [12:55:03<16:04:48, 23.30s/it]
 45%|████▍     | 1992/4476 [12:55:26<16:07:00, 23.36s/it]
 45%|████▍     | 1993/4476 [12:55:50<16:09:58, 23.44s/it]
 45%|████▍     | 1994/4476 [12:56:13<16:09:05, 23.43s/it]
 45%|████▍     | 1995/4476 [12:56:36<16:05:10, 23.34s/it]
 45%|████▍     | 1996/4476 [12:57:00<16:03:10, 23.30s/it]
 45%|████▍     | 1997/4476 [12:57:23<16:00:20, 23.24s/it]
 45%|████▍     | 1998/4476 [12:57:46<15:59:52, 23.24s/it]
 45%|████▍     | 1999/4476 [12:58:09<16:01:24, 23.29s/it]
 45%|████▍     | 2000/4476 [12:58:33<16:03:03, 23.34s/it]
                                                         
{'loss': 0.3868, 'learning_rate': 2.9174062909375892e-05, 'epoch': 1.34}

 45%|████▍     | 2000/4476 [12:58:33<16:03:03, 23.34s/it][INFO|trainer.py:2939] 2023-11-12 16:22:20,495 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2000
[INFO|tokenization_utils_base.py:2437] 2023-11-12 16:22:20,536 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2000/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 16:22:20,536 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2000/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 16:22:20,536 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2000/added_tokens.json

 45%|████▍     | 2001/4476 [12:58:56<16:05:16, 23.40s/it]
 45%|████▍     | 2002/4476 [12:59:20<16:05:51, 23.42s/it]
 45%|████▍     | 2003/4476 [12:59:43<16:02:29, 23.35s/it]
 45%|████▍     | 2004/4476 [13:00:06<16:01:12, 23.33s/it]
 45%|████▍     | 2005/4476 [13:00:29<15:56:44, 23.23s/it]
 45%|████▍     | 2006/4476 [13:00:53<15:57:58, 23.27s/it]
 45%|████▍     | 2007/4476 [13:01:16<15:56:17, 23.24s/it]
 45%|████▍     | 2008/4476 [13:01:39<15:56:22, 23.25s/it]
 45%|████▍     | 2009/4476 [13:02:02<15:55:23, 23.24s/it]
 45%|████▍     | 2010/4476 [13:02:26<15:55:48, 23.26s/it]
                                                         
{'loss': 0.385, 'learning_rate': 2.9000955770729464e-05, 'epoch': 1.35}

 45%|████▍     | 2010/4476 [13:02:26<15:55:48, 23.26s/it]
 45%|████▍     | 2011/4476 [13:02:49<15:57:14, 23.30s/it]
 45%|████▍     | 2012/4476 [13:03:13<16:01:42, 23.42s/it]
 45%|████▍     | 2013/4476 [13:03:36<15:58:35, 23.35s/it]
 45%|████▍     | 2014/4476 [13:03:59<15:58:34, 23.36s/it]
 45%|████▌     | 2015/4476 [13:04:22<15:56:45, 23.33s/it]
 45%|████▌     | 2016/4476 [13:04:46<15:57:05, 23.34s/it]
 45%|████▌     | 2017/4476 [13:05:09<15:59:08, 23.40s/it]
 45%|████▌     | 2018/4476 [13:05:33<15:57:16, 23.37s/it]
 45%|████▌     | 2019/4476 [13:05:56<15:57:00, 23.37s/it]
 45%|████▌     | 2020/4476 [13:06:19<15:56:14, 23.36s/it]
                                                         
{'loss': 0.3871, 'learning_rate': 2.8827651534382655e-05, 'epoch': 1.35}

 45%|████▌     | 2020/4476 [13:06:19<15:56:14, 23.36s/it]
 45%|████▌     | 2021/4476 [13:06:43<15:54:11, 23.32s/it]
 45%|████▌     | 2022/4476 [13:07:06<15:56:00, 23.37s/it]
 45%|████▌     | 2023/4476 [13:07:29<15:52:59, 23.31s/it]
 45%|████▌     | 2024/4476 [13:07:53<15:52:47, 23.31s/it]
 45%|████▌     | 2025/4476 [13:08:16<15:54:04, 23.36s/it]
 45%|████▌     | 2026/4476 [13:08:39<15:53:11, 23.34s/it]
 45%|████▌     | 2027/4476 [13:09:03<15:55:47, 23.42s/it]
 45%|████▌     | 2028/4476 [13:09:26<15:54:45, 23.40s/it]
 45%|████▌     | 2029/4476 [13:09:50<15:54:47, 23.41s/it]
 45%|████▌     | 2030/4476 [13:10:13<15:53:39, 23.39s/it]
                                                         
{'loss': 0.3956, 'learning_rate': 2.8654158737762122e-05, 'epoch': 1.36}

 45%|████▌     | 2030/4476 [13:10:13<15:53:39, 23.39s/it]
 45%|████▌     | 2031/4476 [13:10:36<15:52:44, 23.38s/it]
 45%|████▌     | 2032/4476 [13:11:00<15:52:04, 23.37s/it]
 45%|████▌     | 2033/4476 [13:11:23<15:51:54, 23.38s/it]
 45%|████▌     | 2034/4476 [13:11:47<15:50:41, 23.36s/it]
 45%|████▌     | 2035/4476 [13:12:10<15:48:28, 23.31s/it]
 45%|████▌     | 2036/4476 [13:12:33<15:47:37, 23.30s/it]
 46%|████▌     | 2037/4476 [13:12:56<15:41:25, 23.16s/it]
 46%|████▌     | 2038/4476 [13:13:19<15:42:54, 23.21s/it]
 46%|████▌     | 2039/4476 [13:13:42<15:43:59, 23.24s/it]
 46%|████▌     | 2040/4476 [13:14:06<15:41:09, 23.18s/it]
                                                         
{'loss': 0.3884, 'learning_rate': 2.8480485927583506e-05, 'epoch': 1.37}

 46%|████▌     | 2040/4476 [13:14:06<15:41:09, 23.18s/it]
 46%|████▌     | 2041/4476 [13:14:29<15:44:09, 23.26s/it]
 46%|████▌     | 2042/4476 [13:14:52<15:46:22, 23.33s/it]
 46%|████▌     | 2043/4476 [13:15:16<15:44:51, 23.30s/it]
 46%|████▌     | 2044/4476 [13:15:39<15:48:58, 23.41s/it]
 46%|████▌     | 2045/4476 [13:16:02<15:44:01, 23.30s/it]
 46%|████▌     | 2046/4476 [13:16:26<15:43:52, 23.31s/it]
 46%|████▌     | 2047/4476 [13:16:49<15:43:37, 23.31s/it]
 46%|████▌     | 2048/4476 [13:17:12<15:42:30, 23.29s/it]
 46%|████▌     | 2049/4476 [13:17:36<15:44:32, 23.35s/it]
 46%|████▌     | 2050/4476 [13:17:59<15:43:26, 23.33s/it]
                                                         
{'loss': 0.3829, 'learning_rate': 2.8306641659430382e-05, 'epoch': 1.37}

 46%|████▌     | 2050/4476 [13:17:59<15:43:26, 23.33s/it]
 46%|████▌     | 2051/4476 [13:18:23<15:47:10, 23.44s/it]
 46%|████▌     | 2052/4476 [13:18:46<15:47:11, 23.45s/it]
 46%|████▌     | 2053/4476 [13:19:10<15:49:49, 23.52s/it]
 46%|████▌     | 2054/4476 [13:19:33<15:49:38, 23.53s/it]
 46%|████▌     | 2055/4476 [13:19:57<15:49:24, 23.53s/it]
 46%|████▌     | 2056/4476 [13:20:21<15:49:29, 23.54s/it]
 46%|████▌     | 2057/4476 [13:20:43<15:41:28, 23.35s/it]
 46%|████▌     | 2058/4476 [13:21:07<15:41:16, 23.36s/it]
 46%|████▌     | 2059/4476 [13:21:30<15:41:37, 23.38s/it]
 46%|████▌     | 2060/4476 [13:21:54<15:45:54, 23.49s/it]
                                                         
{'loss': 0.3916, 'learning_rate': 2.8132634497332815e-05, 'epoch': 1.38}

 46%|████▌     | 2060/4476 [13:21:54<15:45:54, 23.49s/it]
 46%|████▌     | 2061/4476 [13:22:17<15:40:38, 23.37s/it]
 46%|████▌     | 2062/4476 [13:22:40<15:39:27, 23.35s/it]
 46%|████▌     | 2063/4476 [13:23:04<15:37:05, 23.30s/it]
 46%|████▌     | 2064/4476 [13:23:27<15:38:55, 23.36s/it]
 46%|████▌     | 2065/4476 [13:23:51<15:40:54, 23.42s/it]
 46%|████▌     | 2066/4476 [13:24:14<15:40:54, 23.43s/it]
 46%|████▌     | 2067/4476 [13:24:37<15:35:01, 23.29s/it]
 46%|████▌     | 2068/4476 [13:25:01<15:37:13, 23.35s/it]
 46%|████▌     | 2069/4476 [13:25:24<15:34:26, 23.29s/it]
 46%|████▌     | 2070/4476 [13:25:47<15:34:08, 23.30s/it]
                                                         
{'loss': 0.3924, 'learning_rate': 2.7958473013345447e-05, 'epoch': 1.39}

 46%|████▌     | 2070/4476 [13:25:47<15:34:08, 23.30s/it]
 46%|████▋     | 2071/4476 [13:26:10<15:34:38, 23.32s/it]
 46%|████▋     | 2072/4476 [13:26:34<15:34:43, 23.33s/it]
 46%|████▋     | 2073/4476 [13:26:57<15:29:05, 23.20s/it]
 46%|████▋     | 2074/4476 [13:27:20<15:25:14, 23.11s/it]
 46%|████▋     | 2075/4476 [13:27:43<15:28:57, 23.21s/it]
 46%|████▋     | 2076/4476 [13:28:06<15:29:13, 23.23s/it]
 46%|████▋     | 2077/4476 [13:28:30<15:29:38, 23.25s/it]
 46%|████▋     | 2078/4476 [13:28:53<15:26:26, 23.18s/it]
 46%|████▋     | 2079/4476 [13:29:16<15:30:33, 23.29s/it]
 46%|████▋     | 2080/4476 [13:29:39<15:27:58, 23.24s/it]
                                                         
{'loss': 0.3906, 'learning_rate': 2.7784165787125226e-05, 'epoch': 1.39}

 46%|████▋     | 2080/4476 [13:29:39<15:27:58, 23.24s/it]
 46%|████▋     | 2081/4476 [13:30:02<15:25:35, 23.19s/it]
 47%|████▋     | 2082/4476 [13:30:25<15:23:06, 23.14s/it]
 47%|████▋     | 2083/4476 [13:30:48<15:21:30, 23.11s/it]
 47%|████▋     | 2084/4476 [13:31:12<15:28:37, 23.29s/it]
 47%|████▋     | 2085/4476 [13:31:35<15:27:12, 23.27s/it]
 47%|████▋     | 2086/4476 [13:31:59<15:28:54, 23.32s/it]
 47%|████▋     | 2087/4476 [13:32:22<15:27:28, 23.29s/it]
 47%|████▋     | 2088/4476 [13:32:45<15:28:37, 23.33s/it]
 47%|████▋     | 2089/4476 [13:33:09<15:30:58, 23.40s/it]
 47%|████▋     | 2090/4476 [13:33:32<15:31:40, 23.43s/it]
                                                         
{'loss': 0.383, 'learning_rate': 2.7609721405508758e-05, 'epoch': 1.4}

 47%|████▋     | 2090/4476 [13:33:32<15:31:40, 23.43s/it]
 47%|████▋     | 2091/4476 [13:33:56<15:27:20, 23.33s/it]
 47%|████▋     | 2092/4476 [13:34:19<15:29:30, 23.39s/it]
 47%|████▋     | 2093/4476 [13:34:43<15:32:44, 23.48s/it]
 47%|████▋     | 2094/4476 [13:35:06<15:30:32, 23.44s/it]
 47%|████▋     | 2095/4476 [13:35:29<15:27:43, 23.38s/it]
 47%|████▋     | 2096/4476 [13:35:53<15:31:32, 23.48s/it]
 47%|████▋     | 2097/4476 [13:36:16<15:27:58, 23.40s/it]
 47%|████▋     | 2098/4476 [13:36:39<15:18:48, 23.18s/it]
 47%|████▋     | 2099/4476 [13:37:02<15:16:58, 23.15s/it]
 47%|████▋     | 2100/4476 [13:37:25<15:19:39, 23.22s/it]
                                                         
{'loss': 0.3892, 'learning_rate': 2.7435148462089282e-05, 'epoch': 1.41}

 47%|████▋     | 2100/4476 [13:37:25<15:19:39, 23.22s/it]
 47%|████▋     | 2101/4476 [13:37:49<15:22:07, 23.30s/it]
 47%|████▋     | 2102/4476 [13:38:13<15:27:16, 23.44s/it]
 47%|████▋     | 2103/4476 [13:38:36<15:23:56, 23.36s/it]
 47%|████▋     | 2104/4476 [13:38:59<15:22:58, 23.35s/it]
 47%|████▋     | 2105/4476 [13:39:23<15:22:49, 23.35s/it]
 47%|████▋     | 2106/4476 [13:39:46<15:22:39, 23.36s/it]
 47%|████▋     | 2107/4476 [13:40:09<15:24:34, 23.42s/it]
 47%|████▋     | 2108/4476 [13:40:33<15:27:03, 23.49s/it]
 47%|████▋     | 2109/4476 [13:40:56<15:25:16, 23.45s/it]
 47%|████▋     | 2110/4476 [13:41:20<15:20:53, 23.35s/it]
                                                         
{'loss': 0.3866, 'learning_rate': 2.7260455556793325e-05, 'epoch': 1.41}

 47%|████▋     | 2110/4476 [13:41:20<15:20:53, 23.35s/it]
 47%|████▋     | 2111/4476 [13:41:43<15:19:23, 23.32s/it]
 47%|████▋     | 2112/4476 [13:42:06<15:18:56, 23.32s/it]
 47%|████▋     | 2113/4476 [13:42:30<15:20:38, 23.38s/it]
 47%|████▋     | 2114/4476 [13:42:53<15:18:21, 23.33s/it]
 47%|████▋     | 2115/4476 [13:43:16<15:17:31, 23.32s/it]
 47%|████▋     | 2116/4476 [13:43:39<15:17:03, 23.32s/it]
 47%|████▋     | 2117/4476 [13:44:03<15:16:39, 23.31s/it]
 47%|████▋     | 2118/4476 [13:44:27<15:21:32, 23.45s/it]
 47%|████▋     | 2119/4476 [13:44:50<15:19:16, 23.40s/it]
 47%|████▋     | 2120/4476 [13:45:13<15:15:54, 23.33s/it]
                                                         
{'loss': 0.382, 'learning_rate': 2.708565129545706e-05, 'epoch': 1.42}

 47%|████▋     | 2120/4476 [13:45:13<15:15:54, 23.33s/it]
 47%|████▋     | 2121/4476 [13:45:36<15:16:00, 23.34s/it]
 47%|████▋     | 2122/4476 [13:46:00<15:17:04, 23.37s/it]
 47%|████▋     | 2123/4476 [13:46:23<15:18:50, 23.43s/it]
 47%|████▋     | 2124/4476 [13:46:46<15:12:24, 23.28s/it]
 47%|████▋     | 2125/4476 [13:47:09<15:09:31, 23.21s/it]
 47%|████▋     | 2126/4476 [13:47:33<15:11:10, 23.26s/it]
 48%|████▊     | 2127/4476 [13:47:56<15:11:49, 23.29s/it]
 48%|████▊     | 2128/4476 [13:48:19<15:10:57, 23.28s/it]
 48%|████▊     | 2129/4476 [13:48:43<15:13:27, 23.35s/it]
 48%|████▊     | 2130/4476 [13:49:06<15:11:58, 23.32s/it]
                                                         
{'loss': 0.3825, 'learning_rate': 2.691074428940237e-05, 'epoch': 1.43}

 48%|████▊     | 2130/4476 [13:49:06<15:11:58, 23.32s/it]
 48%|████▊     | 2131/4476 [13:49:30<15:15:57, 23.44s/it]
 48%|████▊     | 2132/4476 [13:49:53<15:12:01, 23.35s/it]
 48%|████▊     | 2133/4476 [13:50:16<15:11:01, 23.33s/it]
 48%|████▊     | 2134/4476 [13:50:39<15:08:43, 23.28s/it]
 48%|████▊     | 2135/4476 [13:51:03<15:08:11, 23.28s/it]
 48%|████▊     | 2136/4476 [13:51:26<15:08:53, 23.30s/it]
 48%|████▊     | 2137/4476 [13:51:49<15:08:06, 23.29s/it]
 48%|████▊     | 2138/4476 [13:52:13<15:09:01, 23.33s/it]
 48%|████▊     | 2139/4476 [13:52:36<15:08:43, 23.33s/it]
 48%|████▊     | 2140/4476 [13:53:00<15:12:53, 23.45s/it]
                                                         
{'loss': 0.3828, 'learning_rate': 2.673574315501259e-05, 'epoch': 1.43}

 48%|████▊     | 2140/4476 [13:53:00<15:12:53, 23.45s/it]
 48%|████▊     | 2141/4476 [13:53:23<15:12:39, 23.45s/it]
 48%|████▊     | 2142/4476 [13:53:47<15:11:15, 23.43s/it]
 48%|████▊     | 2143/4476 [13:54:09<15:03:46, 23.24s/it]
 48%|████▊     | 2144/4476 [13:54:33<15:03:19, 23.24s/it]
 48%|████▊     | 2145/4476 [13:54:56<15:02:10, 23.22s/it]
 48%|████▊     | 2146/4476 [13:55:19<15:00:59, 23.20s/it]
 48%|████▊     | 2147/4476 [13:55:43<15:03:54, 23.29s/it]
 48%|████▊     | 2148/4476 [13:56:06<15:03:30, 23.29s/it]
 48%|████▊     | 2149/4476 [13:56:29<15:03:54, 23.31s/it]
 48%|████▊     | 2150/4476 [13:56:53<15:05:19, 23.35s/it]
                                                         
{'loss': 0.3845, 'learning_rate': 2.656065651330808e-05, 'epoch': 1.44}

 48%|████▊     | 2150/4476 [13:56:53<15:05:19, 23.35s/it]
 48%|████▊     | 2151/4476 [13:57:16<15:04:27, 23.34s/it]
 48%|████▊     | 2152/4476 [13:57:39<15:05:05, 23.37s/it]
 48%|████▊     | 2153/4476 [13:58:03<15:07:50, 23.45s/it]
 48%|████▊     | 2154/4476 [13:58:26<15:03:55, 23.36s/it]
 48%|████▊     | 2155/4476 [13:58:49<15:01:45, 23.31s/it]
 48%|████▊     | 2156/4476 [13:59:13<15:02:05, 23.33s/it]
 48%|████▊     | 2157/4476 [13:59:36<14:56:10, 23.19s/it]
 48%|████▊     | 2158/4476 [13:59:59<14:58:26, 23.26s/it]
 48%|████▊     | 2159/4476 [14:00:22<14:58:05, 23.26s/it]
 48%|████▊     | 2160/4476 [14:00:46<15:03:54, 23.42s/it]
                                                         
{'loss': 0.3904, 'learning_rate': 2.6385492989521522e-05, 'epoch': 1.45}

 48%|████▊     | 2160/4476 [14:00:46<15:03:54, 23.42s/it]
 48%|████▊     | 2161/4476 [14:01:09<15:03:24, 23.41s/it]
 48%|████▊     | 2162/4476 [14:01:32<14:58:37, 23.30s/it]
 48%|████▊     | 2163/4476 [14:01:56<14:59:08, 23.32s/it]
 48%|████▊     | 2164/4476 [14:02:19<14:57:47, 23.30s/it]
 48%|████▊     | 2165/4476 [14:02:42<14:57:56, 23.31s/it]
 48%|████▊     | 2166/4476 [14:03:06<14:57:36, 23.31s/it]
 48%|████▊     | 2167/4476 [14:03:29<14:55:21, 23.27s/it]
 48%|████▊     | 2168/4476 [14:03:52<14:54:52, 23.26s/it]
 48%|████▊     | 2169/4476 [14:04:16<14:57:47, 23.35s/it]
 48%|████▊     | 2170/4476 [14:04:39<14:56:41, 23.33s/it]
                                                         
{'loss': 0.3934, 'learning_rate': 2.6210261212673004e-05, 'epoch': 1.45}

 48%|████▊     | 2170/4476 [14:04:39<14:56:41, 23.33s/it]
 49%|████▊     | 2171/4476 [14:05:02<14:55:05, 23.30s/it]
 49%|████▊     | 2172/4476 [14:05:26<14:56:31, 23.35s/it]
 49%|████▊     | 2173/4476 [14:05:49<14:58:30, 23.41s/it]
 49%|████▊     | 2174/4476 [14:06:12<14:53:06, 23.28s/it]
 49%|████▊     | 2175/4476 [14:06:36<14:57:59, 23.42s/it]
 49%|████▊     | 2176/4476 [14:06:59<14:55:19, 23.36s/it]
 49%|████▊     | 2177/4476 [14:07:23<14:57:30, 23.42s/it]
 49%|████▊     | 2178/4476 [14:07:46<15:00:51, 23.52s/it]
 49%|████▊     | 2179/4476 [14:08:10<15:03:23, 23.60s/it]
 49%|████▊     | 2180/4476 [14:08:33<14:58:42, 23.49s/it]
                                                         
{'loss': 0.3893, 'learning_rate': 2.6034969815144938e-05, 'epoch': 1.46}

 49%|████▊     | 2180/4476 [14:08:33<14:58:42, 23.49s/it]
 49%|████▊     | 2181/4476 [14:08:57<14:53:33, 23.36s/it]
 49%|████▊     | 2182/4476 [14:09:20<14:55:18, 23.42s/it]
 49%|████▉     | 2183/4476 [14:09:43<14:46:42, 23.20s/it]
 49%|████▉     | 2184/4476 [14:10:06<14:46:12, 23.20s/it]
 49%|████▉     | 2185/4476 [14:10:29<14:43:26, 23.14s/it]
 49%|████▉     | 2186/4476 [14:10:53<14:49:16, 23.30s/it]
 49%|████▉     | 2187/4476 [14:11:16<14:48:20, 23.29s/it]
 49%|████▉     | 2188/4476 [14:11:39<14:48:00, 23.29s/it]
 49%|████▉     | 2189/4476 [14:12:03<14:52:59, 23.43s/it]
 49%|████▉     | 2190/4476 [14:12:26<14:52:04, 23.41s/it]
                                                         
{'loss': 0.3965, 'learning_rate': 2.5859627432256816e-05, 'epoch': 1.47}

 49%|████▉     | 2190/4476 [14:12:26<14:52:04, 23.41s/it]
 49%|████▉     | 2191/4476 [14:12:50<14:49:46, 23.36s/it]
 49%|████▉     | 2192/4476 [14:13:13<14:47:52, 23.32s/it]
 49%|████▉     | 2193/4476 [14:13:37<14:52:19, 23.45s/it]
 49%|████▉     | 2194/4476 [14:14:00<14:48:30, 23.36s/it]
 49%|████▉     | 2195/4476 [14:14:23<14:47:29, 23.34s/it]
 49%|████▉     | 2196/4476 [14:14:47<14:50:33, 23.44s/it]
 49%|████▉     | 2197/4476 [14:15:10<14:51:41, 23.48s/it]
 49%|████▉     | 2198/4476 [14:15:33<14:48:21, 23.40s/it]
 49%|████▉     | 2199/4476 [14:15:57<14:44:30, 23.31s/it]
 49%|████▉     | 2200/4476 [14:16:20<14:44:04, 23.31s/it]
                                                         
{'loss': 0.3833, 'learning_rate': 2.568424270183981e-05, 'epoch': 1.47}

 49%|████▉     | 2200/4476 [14:16:20<14:44:04, 23.31s/it][INFO|trainer.py:2939] 2023-11-12 17:40:07,612 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2200
[INFO|tokenization_utils_base.py:2437] 2023-11-12 17:40:07,643 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2200/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 17:40:07,643 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2200/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 17:40:07,643 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2200/added_tokens.json

 49%|████▉     | 2201/4476 [14:16:43<14:42:15, 23.27s/it]
 49%|████▉     | 2202/4476 [14:17:06<14:42:02, 23.27s/it]
 49%|████▉     | 2203/4476 [14:17:30<14:43:40, 23.33s/it]
 49%|████▉     | 2204/4476 [14:17:53<14:42:31, 23.31s/it]
 49%|████▉     | 2205/4476 [14:18:17<14:44:34, 23.37s/it]
 49%|████▉     | 2206/4476 [14:18:40<14:44:41, 23.38s/it]
 49%|████▉     | 2207/4476 [14:19:03<14:42:53, 23.35s/it]
 49%|████▉     | 2208/4476 [14:19:27<14:43:51, 23.38s/it]
 49%|████▉     | 2209/4476 [14:19:50<14:43:12, 23.38s/it]
 49%|████▉     | 2210/4476 [14:20:13<14:41:16, 23.33s/it]
                                                         
{'loss': 0.3822, 'learning_rate': 2.550882426381123e-05, 'epoch': 1.48}

 49%|████▉     | 2210/4476 [14:20:13<14:41:16, 23.33s/it]
 49%|████▉     | 2211/4476 [14:20:36<14:37:47, 23.25s/it]
 49%|████▉     | 2212/4476 [14:21:00<14:38:05, 23.27s/it]
 49%|████▉     | 2213/4476 [14:21:23<14:37:28, 23.26s/it]
 49%|████▉     | 2214/4476 [14:21:46<14:38:18, 23.30s/it]
 49%|████▉     | 2215/4476 [14:22:10<14:41:04, 23.38s/it]
 50%|████▉     | 2216/4476 [14:22:33<14:40:08, 23.37s/it]
 50%|████▉     | 2217/4476 [14:22:57<14:38:50, 23.34s/it]
 50%|████▉     | 2218/4476 [14:23:20<14:35:18, 23.26s/it]
 50%|████▉     | 2219/4476 [14:23:43<14:32:32, 23.20s/it]
 50%|████▉     | 2220/4476 [14:24:06<14:32:16, 23.20s/it]
                                                         
{'loss': 0.3823, 'learning_rate': 2.5333380759748925e-05, 'epoch': 1.49}

 50%|████▉     | 2220/4476 [14:24:06<14:32:16, 23.20s/it]
 50%|████▉     | 2221/4476 [14:24:29<14:31:32, 23.19s/it]
 50%|████▉     | 2222/4476 [14:24:52<14:34:19, 23.27s/it]
 50%|████▉     | 2223/4476 [14:25:16<14:36:16, 23.34s/it]
 50%|████▉     | 2224/4476 [14:25:40<14:39:10, 23.42s/it]
 50%|████▉     | 2225/4476 [14:26:03<14:34:39, 23.31s/it]
 50%|████▉     | 2226/4476 [14:26:26<14:35:57, 23.36s/it]
 50%|████▉     | 2227/4476 [14:26:49<14:33:20, 23.30s/it]
 50%|████▉     | 2228/4476 [14:27:12<14:31:55, 23.27s/it]
 50%|████▉     | 2229/4476 [14:27:36<14:31:56, 23.28s/it]
 50%|████▉     | 2230/4476 [14:27:59<14:32:45, 23.31s/it]
                                                         
{'loss': 0.3861, 'learning_rate': 2.515792083246556e-05, 'epoch': 1.49}

 50%|████▉     | 2230/4476 [14:27:59<14:32:45, 23.31s/it]
 50%|████▉     | 2231/4476 [14:28:22<14:30:00, 23.25s/it]
 50%|████▉     | 2232/4476 [14:28:46<14:30:45, 23.28s/it]
 50%|████▉     | 2233/4476 [14:29:09<14:27:33, 23.21s/it]
 50%|████▉     | 2234/4476 [14:29:32<14:29:53, 23.28s/it]
 50%|████▉     | 2235/4476 [14:29:55<14:28:45, 23.26s/it]
 50%|████▉     | 2236/4476 [14:30:19<14:29:59, 23.30s/it]
 50%|████▉     | 2237/4476 [14:30:42<14:26:54, 23.23s/it]
 50%|█████     | 2238/4476 [14:31:05<14:27:24, 23.25s/it]
 50%|█████     | 2239/4476 [14:31:28<14:26:50, 23.25s/it]
 50%|█████     | 2240/4476 [14:31:51<14:24:21, 23.19s/it]
                                                         
{'loss': 0.3861, 'learning_rate': 2.4982453125582834e-05, 'epoch': 1.5}

 50%|█████     | 2240/4476 [14:31:51<14:24:21, 23.19s/it]
 50%|█████     | 2241/4476 [14:32:14<14:20:59, 23.11s/it]
 50%|█████     | 2242/4476 [14:32:38<14:23:42, 23.20s/it]
 50%|█████     | 2243/4476 [14:33:01<14:21:02, 23.14s/it]
 50%|█████     | 2244/4476 [14:33:24<14:24:11, 23.23s/it]
 50%|█████     | 2245/4476 [14:33:47<14:23:10, 23.21s/it]
 50%|█████     | 2246/4476 [14:34:11<14:24:40, 23.26s/it]
 50%|█████     | 2247/4476 [14:34:34<14:27:35, 23.35s/it]
 50%|█████     | 2248/4476 [14:34:58<14:27:44, 23.37s/it]
 50%|█████     | 2249/4476 [14:35:21<14:31:39, 23.48s/it]
 50%|█████     | 2250/4476 [14:35:45<14:32:56, 23.53s/it]
                                                         
{'loss': 0.3909, 'learning_rate': 2.4806986283105712e-05, 'epoch': 1.51}

 50%|█████     | 2250/4476 [14:35:45<14:32:56, 23.53s/it]
 50%|█████     | 2251/4476 [14:36:08<14:30:02, 23.46s/it]
 50%|█████     | 2252/4476 [14:36:32<14:29:59, 23.47s/it]
 50%|█████     | 2253/4476 [14:36:55<14:25:55, 23.37s/it]
 50%|█████     | 2254/4476 [14:37:18<14:20:03, 23.22s/it]
 50%|█████     | 2255/4476 [14:37:41<14:20:21, 23.24s/it]
 50%|█████     | 2256/4476 [14:38:05<14:23:29, 23.34s/it]
 50%|█████     | 2257/4476 [14:38:28<14:24:31, 23.38s/it]
 50%|█████     | 2258/4476 [14:38:52<14:28:46, 23.50s/it]
 50%|█████     | 2259/4476 [14:39:16<14:29:21, 23.53s/it]
 50%|█████     | 2260/4476 [14:39:39<14:32:01, 23.61s/it]
                                                         
{'loss': 0.3854, 'learning_rate': 2.463152894899658e-05, 'epoch': 1.51}

 50%|█████     | 2260/4476 [14:39:39<14:32:01, 23.61s/it]
 51%|█████     | 2261/4476 [14:40:03<14:30:08, 23.57s/it]
 51%|█████     | 2262/4476 [14:40:26<14:26:00, 23.47s/it]
 51%|█████     | 2263/4476 [14:40:50<14:28:53, 23.56s/it]
 51%|█████     | 2264/4476 [14:41:13<14:25:26, 23.47s/it]
 51%|█████     | 2265/4476 [14:41:36<14:23:02, 23.42s/it]
 51%|█████     | 2266/4476 [14:42:00<14:18:55, 23.32s/it]
 51%|█████     | 2267/4476 [14:42:23<14:19:16, 23.34s/it]
 51%|█████     | 2268/4476 [14:42:46<14:18:33, 23.33s/it]
 51%|█████     | 2269/4476 [14:43:10<14:20:13, 23.39s/it]
 51%|█████     | 2270/4476 [14:43:33<14:16:37, 23.30s/it]
                                                         
{'loss': 0.394, 'learning_rate': 2.445608976674939e-05, 'epoch': 1.52}

 51%|█████     | 2270/4476 [14:43:33<14:16:37, 23.30s/it]
 51%|█████     | 2271/4476 [14:43:57<14:20:35, 23.42s/it]
 51%|█████     | 2272/4476 [14:44:20<14:21:45, 23.46s/it]
 51%|█████     | 2273/4476 [14:44:43<14:19:24, 23.41s/it]
 51%|█████     | 2274/4476 [14:45:07<14:18:40, 23.40s/it]
 51%|█████     | 2275/4476 [14:45:30<14:18:37, 23.41s/it]
 51%|█████     | 2276/4476 [14:45:53<14:15:46, 23.34s/it]
 51%|█████     | 2277/4476 [14:46:17<14:15:40, 23.35s/it]
 51%|█████     | 2278/4476 [14:46:40<14:13:53, 23.31s/it]
 51%|█████     | 2279/4476 [14:47:03<14:12:47, 23.29s/it]
 51%|█████     | 2280/4476 [14:47:26<14:12:21, 23.29s/it]
                                                         
{'loss': 0.3866, 'learning_rate': 2.4280677378963906e-05, 'epoch': 1.53}

 51%|█████     | 2280/4476 [14:47:26<14:12:21, 23.29s/it]
 51%|█████     | 2281/4476 [14:47:50<14:12:00, 23.29s/it]
 51%|█████     | 2282/4476 [14:48:13<14:11:09, 23.28s/it]
 51%|█████     | 2283/4476 [14:48:36<14:10:56, 23.28s/it]
 51%|█████     | 2284/4476 [14:49:00<14:12:24, 23.33s/it]
 51%|█████     | 2285/4476 [14:49:23<14:11:28, 23.32s/it]
 51%|█████     | 2286/4476 [14:49:46<14:12:28, 23.36s/it]
 51%|█████     | 2287/4476 [14:50:10<14:11:21, 23.34s/it]
 51%|█████     | 2288/4476 [14:50:33<14:07:44, 23.25s/it]
 51%|█████     | 2289/4476 [14:50:56<14:07:01, 23.24s/it]
 51%|█████     | 2290/4476 [14:51:20<14:11:52, 23.38s/it]
                                                         
{'loss': 0.3879, 'learning_rate': 2.410530042691992e-05, 'epoch': 1.53}

 51%|█████     | 2290/4476 [14:51:20<14:11:52, 23.38s/it]
 51%|█████     | 2291/4476 [14:51:43<14:08:39, 23.30s/it]
 51%|█████     | 2292/4476 [14:52:06<14:09:17, 23.33s/it]
 51%|█████     | 2293/4476 [14:52:30<14:08:40, 23.33s/it]
 51%|█████▏    | 2294/4476 [14:52:53<14:07:17, 23.30s/it]
 51%|█████▏    | 2295/4476 [14:53:16<14:08:30, 23.34s/it]
 51%|█████▏    | 2296/4476 [14:53:40<14:08:38, 23.36s/it]
 51%|█████▏    | 2297/4476 [14:54:03<14:07:09, 23.33s/it]
 51%|█████▏    | 2298/4476 [14:54:26<14:04:35, 23.27s/it]
 51%|█████▏    | 2299/4476 [14:54:49<14:04:06, 23.26s/it]
 51%|█████▏    | 2300/4476 [14:55:12<14:02:49, 23.24s/it]
                                                         
{'loss': 0.388, 'learning_rate': 2.3929967550151568e-05, 'epoch': 1.54}

 51%|█████▏    | 2300/4476 [14:55:12<14:02:49, 23.24s/it]
 51%|█████▏    | 2301/4476 [14:55:36<14:02:32, 23.24s/it]
 51%|█████▏    | 2302/4476 [14:55:59<13:58:31, 23.14s/it]
 51%|█████▏    | 2303/4476 [14:56:22<14:00:33, 23.21s/it]
 51%|█████▏    | 2304/4476 [14:56:46<14:03:32, 23.30s/it]
 51%|█████▏    | 2305/4476 [14:57:09<14:02:38, 23.29s/it]
 52%|█████▏    | 2306/4476 [14:57:32<14:03:40, 23.33s/it]
 52%|█████▏    | 2307/4476 [14:57:56<14:04:29, 23.36s/it]
 52%|█████▏    | 2308/4476 [14:58:18<13:57:18, 23.17s/it]
 52%|█████▏    | 2309/4476 [14:58:41<13:56:00, 23.15s/it]
 52%|█████▏    | 2310/4476 [14:59:05<14:01:02, 23.30s/it]
                                                         
{'loss': 0.3868, 'learning_rate': 2.375468738602171e-05, 'epoch': 1.55}

 52%|█████▏    | 2310/4476 [14:59:05<14:01:02, 23.30s/it]
 52%|█████▏    | 2311/4476 [14:59:28<14:00:25, 23.29s/it]
 52%|█████▏    | 2312/4476 [14:59:52<14:02:39, 23.36s/it]
 52%|█████▏    | 2313/4476 [15:00:15<13:58:27, 23.26s/it]
 52%|█████▏    | 2314/4476 [15:00:38<13:57:14, 23.24s/it]
 52%|█████▏    | 2315/4476 [15:01:02<13:59:28, 23.31s/it]
 52%|█████▏    | 2316/4476 [15:01:25<14:00:25, 23.35s/it]
 52%|█████▏    | 2317/4476 [15:01:48<13:59:40, 23.34s/it]
 52%|█████▏    | 2318/4476 [15:02:12<13:59:14, 23.33s/it]
 52%|█████▏    | 2319/4476 [15:02:35<13:55:46, 23.25s/it]
 52%|█████▏    | 2320/4476 [15:02:58<13:56:00, 23.27s/it]
                                                         
{'loss': 0.3769, 'learning_rate': 2.3579468569296464e-05, 'epoch': 1.55}

 52%|█████▏    | 2320/4476 [15:02:58<13:56:00, 23.27s/it]
 52%|█████▏    | 2321/4476 [15:03:21<13:55:17, 23.26s/it]
 52%|█████▏    | 2322/4476 [15:03:45<13:58:07, 23.35s/it]
 52%|█████▏    | 2323/4476 [15:04:08<14:00:41, 23.43s/it]
 52%|█████▏    | 2324/4476 [15:04:32<13:59:15, 23.40s/it]
 52%|█████▏    | 2325/4476 [15:04:55<14:02:46, 23.51s/it]
 52%|█████▏    | 2326/4476 [15:05:19<14:00:02, 23.44s/it]
 52%|█████▏    | 2327/4476 [15:05:42<13:57:58, 23.40s/it]
 52%|█████▏    | 2328/4476 [15:06:06<13:59:16, 23.44s/it]
 52%|█████▏    | 2329/4476 [15:06:29<13:57:30, 23.40s/it]
 52%|█████▏    | 2330/4476 [15:06:52<13:57:27, 23.41s/it]
                                                         
{'loss': 0.3811, 'learning_rate': 2.340431973171978e-05, 'epoch': 1.56}

 52%|█████▏    | 2330/4476 [15:06:52<13:57:27, 23.41s/it]
 52%|█████▏    | 2331/4476 [15:07:16<13:55:38, 23.37s/it]
 52%|█████▏    | 2332/4476 [15:07:39<13:54:31, 23.35s/it]
 52%|█████▏    | 2333/4476 [15:08:02<13:53:46, 23.34s/it]
 52%|█████▏    | 2334/4476 [15:08:25<13:49:39, 23.24s/it]
 52%|█████▏    | 2335/4476 [15:08:49<13:52:14, 23.32s/it]
 52%|█████▏    | 2336/4476 [15:09:12<13:52:34, 23.34s/it]
 52%|█████▏    | 2337/4476 [15:09:35<13:51:41, 23.33s/it]
 52%|█████▏    | 2338/4476 [15:09:59<13:51:05, 23.32s/it]
 52%|█████▏    | 2339/4476 [15:10:22<13:50:22, 23.31s/it]
 52%|█████▏    | 2340/4476 [15:10:45<13:49:56, 23.31s/it]
                                                         
{'loss': 0.3805, 'learning_rate': 2.3229249501588278e-05, 'epoch': 1.57}

 52%|█████▏    | 2340/4476 [15:10:45<13:49:56, 23.31s/it]
 52%|█████▏    | 2341/4476 [15:11:09<13:51:22, 23.36s/it]
 52%|█████▏    | 2342/4476 [15:11:33<13:54:34, 23.46s/it]
 52%|█████▏    | 2343/4476 [15:11:56<13:54:49, 23.48s/it]
 52%|█████▏    | 2344/4476 [15:12:20<13:54:04, 23.47s/it]
 52%|█████▏    | 2345/4476 [15:12:43<13:51:39, 23.42s/it]
 52%|█████▏    | 2346/4476 [15:13:06<13:45:32, 23.25s/it]
 52%|█████▏    | 2347/4476 [15:13:29<13:45:18, 23.26s/it]
 52%|█████▏    | 2348/4476 [15:13:52<13:46:06, 23.29s/it]
 52%|█████▏    | 2349/4476 [15:14:16<13:44:56, 23.27s/it]
 53%|█████▎    | 2350/4476 [15:14:39<13:46:06, 23.31s/it]
                                                         
{'loss': 0.3822, 'learning_rate': 2.3054266503326165e-05, 'epoch': 1.57}

 53%|█████▎    | 2350/4476 [15:14:39<13:46:06, 23.31s/it]
 53%|█████▎    | 2351/4476 [15:15:02<13:46:02, 23.32s/it]
 53%|█████▎    | 2352/4476 [15:15:26<13:47:24, 23.37s/it]
 53%|█████▎    | 2353/4476 [15:15:49<13:46:51, 23.37s/it]
 53%|█████▎    | 2354/4476 [15:16:12<13:43:00, 23.27s/it]
 53%|█████▎    | 2355/4476 [15:16:36<13:43:53, 23.31s/it]
 53%|█████▎    | 2356/4476 [15:16:59<13:42:35, 23.28s/it]
 53%|█████▎    | 2357/4476 [15:17:22<13:43:18, 23.31s/it]
 53%|█████▎    | 2358/4476 [15:17:46<13:42:58, 23.31s/it]
 53%|█████▎    | 2359/4476 [15:18:09<13:42:53, 23.32s/it]
 53%|█████▎    | 2360/4476 [15:18:32<13:44:29, 23.38s/it]
                                                         
{'loss': 0.3875, 'learning_rate': 2.2879379357060345e-05, 'epoch': 1.58}

 53%|█████▎    | 2360/4476 [15:18:32<13:44:29, 23.38s/it]
 53%|█████▎    | 2361/4476 [15:18:56<13:43:29, 23.36s/it]
 53%|█████▎    | 2362/4476 [15:19:19<13:38:55, 23.24s/it]
 53%|█████▎    | 2363/4476 [15:19:42<13:41:34, 23.33s/it]
 53%|█████▎    | 2364/4476 [15:20:06<13:41:54, 23.35s/it]
 53%|█████▎    | 2365/4476 [15:20:29<13:41:18, 23.34s/it]
 53%|█████▎    | 2366/4476 [15:20:53<13:45:04, 23.46s/it]
 53%|█████▎    | 2367/4476 [15:21:16<13:42:11, 23.39s/it]
 53%|█████▎    | 2368/4476 [15:21:39<13:37:13, 23.26s/it]
 53%|█████▎    | 2369/4476 [15:22:02<13:38:09, 23.30s/it]
 53%|█████▎    | 2370/4476 [15:22:26<13:38:00, 23.31s/it]
                                                         
{'loss': 0.3884, 'learning_rate': 2.2704596678195827e-05, 'epoch': 1.59}

 53%|█████▎    | 2370/4476 [15:22:26<13:38:00, 23.31s/it]
 53%|█████▎    | 2371/4476 [15:22:49<13:41:45, 23.42s/it]
 53%|█████▎    | 2372/4476 [15:23:13<13:40:04, 23.39s/it]
 53%|█████▎    | 2373/4476 [15:23:36<13:41:13, 23.43s/it]
 53%|█████▎    | 2374/4476 [15:24:00<13:42:08, 23.47s/it]
 53%|█████▎    | 2375/4476 [15:24:23<13:42:14, 23.48s/it]
 53%|█████▎    | 2376/4476 [15:24:46<13:39:10, 23.41s/it]
 53%|█████▎    | 2377/4476 [15:25:10<13:38:28, 23.40s/it]
 53%|█████▎    | 2378/4476 [15:25:33<13:39:24, 23.43s/it]
 53%|█████▎    | 2379/4476 [15:25:56<13:34:04, 23.29s/it]
 53%|█████▎    | 2380/4476 [15:26:20<13:33:07, 23.28s/it]
                                                         
{'loss': 0.381, 'learning_rate': 2.2529927076991283e-05, 'epoch': 1.59}

 53%|█████▎    | 2380/4476 [15:26:20<13:33:07, 23.28s/it]
 53%|█████▎    | 2381/4476 [15:26:43<13:33:13, 23.29s/it]
 53%|█████▎    | 2382/4476 [15:27:06<13:29:29, 23.19s/it]
 53%|█████▎    | 2383/4476 [15:27:29<13:29:24, 23.20s/it]
 53%|█████▎    | 2384/4476 [15:27:52<13:28:33, 23.19s/it]
 53%|█████▎    | 2385/4476 [15:28:16<13:29:34, 23.23s/it]
 53%|█████▎    | 2386/4476 [15:28:39<13:27:59, 23.20s/it]
 53%|█████▎    | 2387/4476 [15:29:02<13:29:05, 23.24s/it]
 53%|█████▎    | 2388/4476 [15:29:25<13:30:16, 23.28s/it]
 53%|█████▎    | 2389/4476 [15:29:48<13:27:24, 23.21s/it]
 53%|█████▎    | 2390/4476 [15:30:12<13:28:22, 23.25s/it]
                                                         
{'loss': 0.3933, 'learning_rate': 2.2355379158134843e-05, 'epoch': 1.6}

 53%|█████▎    | 2390/4476 [15:30:12<13:28:22, 23.25s/it]
 53%|█████▎    | 2391/4476 [15:30:35<13:25:17, 23.17s/it]
 53%|█████▎    | 2392/4476 [15:30:58<13:20:58, 23.06s/it]
 53%|█████▎    | 2393/4476 [15:31:21<13:21:39, 23.09s/it]
 53%|█████▎    | 2394/4476 [15:31:44<13:22:47, 23.14s/it]
 54%|█████▎    | 2395/4476 [15:32:07<13:22:40, 23.14s/it]
 54%|█████▎    | 2396/4476 [15:32:30<13:24:54, 23.22s/it]
 54%|█████▎    | 2397/4476 [15:32:54<13:26:36, 23.28s/it]
 54%|█████▎    | 2398/4476 [15:33:17<13:28:09, 23.33s/it]
 54%|█████▎    | 2399/4476 [15:33:41<13:27:24, 23.32s/it]
 54%|█████▎    | 2400/4476 [15:34:04<13:27:13, 23.33s/it]
                                                         
{'loss': 0.3853, 'learning_rate': 2.2180961520320278e-05, 'epoch': 1.61}

 54%|█████▎    | 2400/4476 [15:34:04<13:27:13, 23.33s/it][INFO|trainer.py:2939] 2023-11-12 18:57:51,743 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2400
[INFO|tokenization_utils_base.py:2437] 2023-11-12 18:57:51,774 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2400/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 18:57:51,774 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2400/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 18:57:51,774 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2400/added_tokens.json

 54%|█████▎    | 2401/4476 [15:34:27<13:24:41, 23.27s/it]
 54%|█████▎    | 2402/4476 [15:34:51<13:25:54, 23.31s/it]
 54%|█████▎    | 2403/4476 [15:35:14<13:27:17, 23.37s/it]
 54%|█████▎    | 2404/4476 [15:35:37<13:25:23, 23.32s/it]
 54%|█████▎    | 2405/4476 [15:36:00<13:20:45, 23.20s/it]
 54%|█████▍    | 2406/4476 [15:36:24<13:22:16, 23.25s/it]
 54%|█████▍    | 2407/4476 [15:36:47<13:26:15, 23.38s/it]
 54%|█████▍    | 2408/4476 [15:37:11<13:25:00, 23.36s/it]
 54%|█████▍    | 2409/4476 [15:37:34<13:25:43, 23.39s/it]
 54%|█████▍    | 2410/4476 [15:37:58<13:29:00, 23.49s/it]
                                                         
{'loss': 0.3871, 'learning_rate': 2.2006682755823367e-05, 'epoch': 1.62}

 54%|█████▍    | 2410/4476 [15:37:58<13:29:00, 23.49s/it]
 54%|█████▍    | 2411/4476 [15:38:21<13:28:26, 23.49s/it]
 54%|█████▍    | 2412/4476 [15:38:44<13:25:12, 23.41s/it]
 54%|█████▍    | 2413/4476 [15:39:08<13:26:03, 23.44s/it]
 54%|█████▍    | 2414/4476 [15:39:31<13:23:11, 23.37s/it]
 54%|█████▍    | 2415/4476 [15:39:55<13:27:19, 23.50s/it]
 54%|█████▍    | 2416/4476 [15:40:18<13:22:33, 23.38s/it]
 54%|█████▍    | 2417/4476 [15:40:41<13:20:20, 23.32s/it]
 54%|█████▍    | 2418/4476 [15:41:05<13:24:37, 23.46s/it]
 54%|█████▍    | 2419/4476 [15:41:28<13:19:50, 23.33s/it]
 54%|█████▍    | 2420/4476 [15:41:51<13:18:30, 23.30s/it]
                                                         
{'loss': 0.3867, 'learning_rate': 2.1832551450078594e-05, 'epoch': 1.62}

 54%|█████▍    | 2420/4476 [15:41:51<13:18:30, 23.30s/it]
 54%|█████▍    | 2421/4476 [15:42:14<13:16:35, 23.26s/it]
 54%|█████▍    | 2422/4476 [15:42:38<13:19:07, 23.34s/it]
 54%|█████▍    | 2423/4476 [15:43:02<13:23:14, 23.48s/it]
 54%|█████▍    | 2424/4476 [15:43:25<13:19:10, 23.37s/it]
 54%|█████▍    | 2425/4476 [15:43:48<13:17:43, 23.34s/it]
 54%|█████▍    | 2426/4476 [15:44:11<13:13:32, 23.23s/it]
 54%|█████▍    | 2427/4476 [15:44:34<13:13:24, 23.23s/it]
 54%|█████▍    | 2428/4476 [15:44:58<13:15:03, 23.29s/it]
 54%|█████▍    | 2429/4476 [15:45:21<13:18:26, 23.40s/it]
 54%|█████▍    | 2430/4476 [15:45:45<13:14:30, 23.30s/it]
                                                         
{'loss': 0.3878, 'learning_rate': 2.165857618125625e-05, 'epoch': 1.63}

 54%|█████▍    | 2430/4476 [15:45:45<13:14:30, 23.30s/it]
 54%|█████▍    | 2431/4476 [15:46:08<13:11:51, 23.23s/it]
 54%|█████▍    | 2432/4476 [15:46:31<13:10:37, 23.21s/it]
 54%|█████▍    | 2433/4476 [15:46:54<13:12:55, 23.29s/it]
 54%|█████▍    | 2434/4476 [15:47:18<13:14:49, 23.35s/it]
 54%|█████▍    | 2435/4476 [15:47:41<13:14:47, 23.36s/it]
 54%|█████▍    | 2436/4476 [15:48:05<13:15:12, 23.39s/it]
 54%|█████▍    | 2437/4476 [15:48:28<13:11:16, 23.28s/it]
 54%|█████▍    | 2438/4476 [15:48:51<13:13:32, 23.36s/it]
 54%|█████▍    | 2439/4476 [15:49:15<13:16:58, 23.47s/it]
 55%|█████▍    | 2440/4476 [15:49:38<13:17:02, 23.49s/it]
                                                         
{'loss': 0.386, 'learning_rate': 2.1484765519839843e-05, 'epoch': 1.64}

 55%|█████▍    | 2440/4476 [15:49:38<13:17:02, 23.49s/it]
 55%|█████▍    | 2441/4476 [15:50:02<13:14:49, 23.43s/it]
 55%|█████▍    | 2442/4476 [15:50:25<13:09:52, 23.30s/it]
 55%|█████▍    | 2443/4476 [15:50:48<13:09:15, 23.29s/it]
 55%|█████▍    | 2444/4476 [15:51:12<13:14:35, 23.46s/it]
 55%|█████▍    | 2445/4476 [15:51:35<13:15:02, 23.49s/it]
 55%|█████▍    | 2446/4476 [15:51:59<13:17:10, 23.56s/it]
 55%|█████▍    | 2447/4476 [15:52:22<13:13:31, 23.47s/it]
 55%|█████▍    | 2448/4476 [15:52:46<13:11:32, 23.42s/it]
 55%|█████▍    | 2449/4476 [15:53:09<13:11:20, 23.42s/it]
 55%|█████▍    | 2450/4476 [15:53:33<13:11:01, 23.43s/it]
                                                         
{'loss': 0.3836, 'learning_rate': 2.1311128028203863e-05, 'epoch': 1.64}

 55%|█████▍    | 2450/4476 [15:53:33<13:11:01, 23.43s/it]
 55%|█████▍    | 2451/4476 [15:53:56<13:11:59, 23.47s/it]
 55%|█████▍    | 2452/4476 [15:54:20<13:14:27, 23.55s/it]
 55%|█████▍    | 2453/4476 [15:54:43<13:07:49, 23.37s/it]
 55%|█████▍    | 2454/4476 [15:55:06<13:05:51, 23.32s/it]
 55%|█████▍    | 2455/4476 [15:55:30<13:07:37, 23.38s/it]
 55%|█████▍    | 2456/4476 [15:55:53<13:10:52, 23.49s/it]
 55%|█████▍    | 2457/4476 [15:56:17<13:08:11, 23.42s/it]
 55%|█████▍    | 2458/4476 [15:56:40<13:06:10, 23.37s/it]
 55%|█████▍    | 2459/4476 [15:57:03<13:08:32, 23.46s/it]
 55%|█████▍    | 2460/4476 [15:57:27<13:06:49, 23.42s/it]
                                                         
{'loss': 0.3849, 'learning_rate': 2.1137672260192004e-05, 'epoch': 1.65}

 55%|█████▍    | 2460/4476 [15:57:27<13:06:49, 23.42s/it]
 55%|█████▍    | 2461/4476 [15:57:50<13:07:40, 23.45s/it]
 55%|█████▌    | 2462/4476 [15:58:14<13:09:30, 23.52s/it]
 55%|█████▌    | 2463/4476 [15:58:37<13:04:49, 23.39s/it]
 55%|█████▌    | 2464/4476 [15:59:00<13:03:16, 23.36s/it]
 55%|█████▌    | 2465/4476 [15:59:24<13:03:56, 23.39s/it]
 55%|█████▌    | 2466/4476 [15:59:47<13:05:26, 23.45s/it]
 55%|█████▌    | 2467/4476 [16:00:11<13:05:59, 23.47s/it]
 55%|█████▌    | 2468/4476 [16:00:35<13:08:21, 23.56s/it]
 55%|█████▌    | 2469/4476 [16:00:58<13:07:48, 23.55s/it]
 55%|█████▌    | 2470/4476 [16:01:22<13:10:03, 23.63s/it]
                                                         
{'loss': 0.3841, 'learning_rate': 2.09644067606958e-05, 'epoch': 1.66}

 55%|█████▌    | 2470/4476 [16:01:22<13:10:03, 23.63s/it]
 55%|█████▌    | 2471/4476 [16:01:45<13:05:56, 23.52s/it]
 55%|█████▌    | 2472/4476 [16:02:08<13:00:55, 23.38s/it]
 55%|█████▌    | 2473/4476 [16:02:32<13:03:30, 23.47s/it]
 55%|█████▌    | 2474/4476 [16:02:56<13:05:39, 23.55s/it]
 55%|█████▌    | 2475/4476 [16:03:19<13:04:25, 23.52s/it]
 55%|█████▌    | 2476/4476 [16:03:43<13:02:30, 23.48s/it]
 55%|█████▌    | 2477/4476 [16:04:06<13:00:36, 23.43s/it]
 55%|█████▌    | 2478/4476 [16:04:29<13:00:12, 23.43s/it]
 55%|█████▌    | 2479/4476 [16:04:53<12:59:21, 23.42s/it]
 55%|█████▌    | 2480/4476 [16:05:16<13:00:11, 23.45s/it]
                                                         
{'loss': 0.3848, 'learning_rate': 2.079134006523359e-05, 'epoch': 1.66}

 55%|█████▌    | 2480/4476 [16:05:16<13:00:11, 23.45s/it]
 55%|█████▌    | 2481/4476 [16:05:40<13:02:13, 23.53s/it]
 55%|█████▌    | 2482/4476 [16:06:03<12:57:11, 23.39s/it]
 55%|█████▌    | 2483/4476 [16:06:26<12:57:12, 23.40s/it]
 55%|█████▌    | 2484/4476 [16:06:50<12:57:50, 23.43s/it]
 56%|█████▌    | 2485/4476 [16:07:13<12:55:40, 23.38s/it]
 56%|█████▌    | 2486/4476 [16:07:37<12:56:25, 23.41s/it]
 56%|█████▌    | 2487/4476 [16:08:00<12:53:19, 23.33s/it]
 56%|█████▌    | 2488/4476 [16:08:23<12:50:03, 23.24s/it]
 56%|█████▌    | 2489/4476 [16:08:47<12:54:29, 23.39s/it]
 56%|█████▌    | 2490/4476 [16:09:10<12:52:13, 23.33s/it]
                                                         
{'loss': 0.3817, 'learning_rate': 2.061848069953017e-05, 'epoch': 1.67}

 56%|█████▌    | 2490/4476 [16:09:10<12:52:13, 23.33s/it]
 56%|█████▌    | 2491/4476 [16:09:34<12:56:20, 23.47s/it]
 56%|█████▌    | 2492/4476 [16:09:57<12:52:28, 23.36s/it]
 56%|█████▌    | 2493/4476 [16:10:20<12:48:26, 23.25s/it]
 56%|█████▌    | 2494/4476 [16:10:43<12:50:04, 23.31s/it]
 56%|█████▌    | 2495/4476 [16:11:06<12:49:58, 23.32s/it]
 56%|█████▌    | 2496/4476 [16:11:30<12:50:51, 23.36s/it]
 56%|█████▌    | 2497/4476 [16:11:53<12:50:36, 23.36s/it]
 56%|█████▌    | 2498/4476 [16:12:17<12:52:51, 23.44s/it]
 56%|█████▌    | 2499/4476 [16:12:40<12:49:20, 23.35s/it]
 56%|█████▌    | 2500/4476 [16:13:04<12:51:09, 23.42s/it]
                                                         
{'loss': 0.3839, 'learning_rate': 2.044583717909667e-05, 'epoch': 1.68}

 56%|█████▌    | 2500/4476 [16:13:04<12:51:09, 23.42s/it]
 56%|█████▌    | 2501/4476 [16:13:27<12:48:56, 23.36s/it]
 56%|█████▌    | 2502/4476 [16:13:50<12:45:44, 23.27s/it]
 56%|█████▌    | 2503/4476 [16:14:13<12:45:20, 23.27s/it]
 56%|█████▌    | 2504/4476 [16:14:36<12:41:17, 23.16s/it]
 56%|█████▌    | 2505/4476 [16:14:59<12:42:38, 23.22s/it]
 56%|█████▌    | 2506/4476 [16:15:23<12:41:42, 23.20s/it]
 56%|█████▌    | 2507/4476 [16:15:46<12:44:20, 23.29s/it]
 56%|█████▌    | 2508/4476 [16:16:09<12:44:00, 23.29s/it]
 56%|█████▌    | 2509/4476 [16:16:33<12:41:52, 23.24s/it]
 56%|█████▌    | 2510/4476 [16:16:56<12:43:28, 23.30s/it]
                                                         
{'loss': 0.3749, 'learning_rate': 2.0273418008811125e-05, 'epoch': 1.68}

 56%|█████▌    | 2510/4476 [16:16:56<12:43:28, 23.30s/it]
 56%|█████▌    | 2511/4476 [16:17:19<12:42:45, 23.29s/it]
 56%|█████▌    | 2512/4476 [16:17:43<12:42:34, 23.30s/it]
 56%|█████▌    | 2513/4476 [16:18:06<12:41:41, 23.28s/it]
 56%|█████▌    | 2514/4476 [16:18:29<12:39:52, 23.24s/it]
 56%|█████▌    | 2515/4476 [16:18:52<12:37:34, 23.18s/it]
 56%|█████▌    | 2516/4476 [16:19:15<12:38:53, 23.23s/it]
 56%|█████▌    | 2517/4476 [16:19:39<12:39:07, 23.25s/it]
 56%|█████▋    | 2518/4476 [16:20:02<12:38:48, 23.25s/it]
 56%|█████▋    | 2519/4476 [16:20:26<12:43:43, 23.42s/it]
 56%|█████▋    | 2520/4476 [16:20:49<12:42:08, 23.38s/it]
                                                         
{'loss': 0.3824, 'learning_rate': 2.0101231682499506e-05, 'epoch': 1.69}

 56%|█████▋    | 2520/4476 [16:20:49<12:42:08, 23.38s/it]
 56%|█████▋    | 2521/4476 [16:21:12<12:42:28, 23.40s/it]
 56%|█████▋    | 2522/4476 [16:21:36<12:42:08, 23.40s/it]
 56%|█████▋    | 2523/4476 [16:21:59<12:40:35, 23.37s/it]
 56%|█████▋    | 2524/4476 [16:22:22<12:39:17, 23.34s/it]
 56%|█████▋    | 2525/4476 [16:22:46<12:37:53, 23.31s/it]
 56%|█████▋    | 2526/4476 [16:23:09<12:41:50, 23.44s/it]
 56%|█████▋    | 2527/4476 [16:23:33<12:39:32, 23.38s/it]
 56%|█████▋    | 2528/4476 [16:23:56<12:36:36, 23.30s/it]
 57%|█████▋    | 2529/4476 [16:24:19<12:36:27, 23.31s/it]
 57%|█████▋    | 2530/4476 [16:24:42<12:36:26, 23.32s/it]
                                                         
{'loss': 0.3854, 'learning_rate': 1.9929286682517235e-05, 'epoch': 1.7}

 57%|█████▋    | 2530/4476 [16:24:42<12:36:26, 23.32s/it]
 57%|█████▋    | 2531/4476 [16:25:06<12:36:29, 23.34s/it]
 57%|█████▋    | 2532/4476 [16:25:29<12:38:15, 23.40s/it]
 57%|█████▋    | 2533/4476 [16:25:53<12:35:38, 23.33s/it]
 57%|█████▋    | 2534/4476 [16:26:16<12:36:24, 23.37s/it]
 57%|█████▋    | 2535/4476 [16:26:39<12:36:38, 23.39s/it]
 57%|█████▋    | 2536/4476 [16:27:03<12:37:25, 23.43s/it]
 57%|█████▋    | 2537/4476 [16:27:26<12:37:24, 23.44s/it]
 57%|█████▋    | 2538/4476 [16:27:50<12:35:40, 23.40s/it]
 57%|█████▋    | 2539/4476 [16:28:13<12:36:10, 23.42s/it]
 57%|█████▋    | 2540/4476 [16:28:37<12:36:05, 23.43s/it]
                                                         
{'loss': 0.3894, 'learning_rate': 1.9757591479331382e-05, 'epoch': 1.7}

 57%|█████▋    | 2540/4476 [16:28:37<12:36:05, 23.43s/it]
 57%|█████▋    | 2541/4476 [16:29:00<12:34:59, 23.41s/it]
 57%|█████▋    | 2542/4476 [16:29:24<12:36:02, 23.46s/it]
 57%|█████▋    | 2543/4476 [16:29:47<12:33:53, 23.40s/it]
 57%|█████▋    | 2544/4476 [16:30:10<12:33:15, 23.39s/it]
 57%|█████▋    | 2545/4476 [16:30:33<12:30:37, 23.32s/it]
 57%|█████▋    | 2546/4476 [16:30:57<12:29:31, 23.30s/it]
 57%|█████▋    | 2547/4476 [16:31:20<12:29:40, 23.32s/it]
 57%|█████▋    | 2548/4476 [16:31:44<12:33:18, 23.44s/it]
 57%|█████▋    | 2549/4476 [16:32:07<12:33:49, 23.47s/it]
 57%|█████▋    | 2550/4476 [16:32:31<12:31:46, 23.42s/it]
                                                         
{'loss': 0.3871, 'learning_rate': 1.9586154531103373e-05, 'epoch': 1.71}

 57%|█████▋    | 2550/4476 [16:32:31<12:31:46, 23.42s/it]
 57%|█████▋    | 2551/4476 [16:32:54<12:28:53, 23.34s/it]
 57%|█████▋    | 2552/4476 [16:33:17<12:26:47, 23.29s/it]
 57%|█████▋    | 2553/4476 [16:33:40<12:25:15, 23.25s/it]
 57%|█████▋    | 2554/4476 [16:34:03<12:25:20, 23.27s/it]
 57%|█████▋    | 2555/4476 [16:34:27<12:26:16, 23.31s/it]
 57%|█████▋    | 2556/4476 [16:34:50<12:27:47, 23.37s/it]
 57%|█████▋    | 2557/4476 [16:35:14<12:29:11, 23.42s/it]
 57%|█████▋    | 2558/4476 [16:35:37<12:23:41, 23.26s/it]
 57%|█████▋    | 2559/4476 [16:36:00<12:24:19, 23.30s/it]
 57%|█████▋    | 2560/4476 [16:36:23<12:22:37, 23.26s/it]
                                                         
{'loss': 0.382, 'learning_rate': 1.9414984283272286e-05, 'epoch': 1.72}

 57%|█████▋    | 2560/4476 [16:36:23<12:22:37, 23.26s/it]
 57%|█████▋    | 2561/4476 [16:36:46<12:20:31, 23.20s/it]
 57%|█████▋    | 2562/4476 [16:37:10<12:20:24, 23.21s/it]
 57%|█████▋    | 2563/4476 [16:37:33<12:20:54, 23.24s/it]
 57%|█████▋    | 2564/4476 [16:37:56<12:17:12, 23.13s/it]
 57%|█████▋    | 2565/4476 [16:38:19<12:16:52, 23.14s/it]
 57%|█████▋    | 2566/4476 [16:38:42<12:20:25, 23.26s/it]
 57%|█████▋    | 2567/4476 [16:39:06<12:19:46, 23.25s/it]
 57%|█████▋    | 2568/4476 [16:39:29<12:19:49, 23.26s/it]
 57%|█████▋    | 2569/4476 [16:39:52<12:19:23, 23.26s/it]
 57%|█████▋    | 2570/4476 [16:40:16<12:22:24, 23.37s/it]
                                                         
{'loss': 0.3858, 'learning_rate': 1.9244089168138836e-05, 'epoch': 1.72}

 57%|█████▋    | 2570/4476 [16:40:16<12:22:24, 23.37s/it]
 57%|█████▋    | 2571/4476 [16:40:39<12:18:34, 23.26s/it]
 57%|█████▋    | 2572/4476 [16:41:02<12:18:41, 23.28s/it]
 57%|█████▋    | 2573/4476 [16:41:25<12:18:27, 23.28s/it]
 58%|█████▊    | 2574/4476 [16:41:49<12:18:10, 23.29s/it]
 58%|█████▊    | 2575/4476 [16:42:12<12:19:49, 23.35s/it]
 58%|█████▊    | 2576/4476 [16:42:35<12:16:38, 23.26s/it]
 58%|█████▊    | 2577/4476 [16:42:59<12:17:44, 23.31s/it]
 58%|█████▊    | 2578/4476 [16:43:22<12:18:58, 23.36s/it]
 58%|█████▊    | 2579/4476 [16:43:46<12:20:30, 23.42s/it]
 58%|█████▊    | 2580/4476 [16:44:09<12:16:57, 23.32s/it]
                                                         
{'loss': 0.3819, 'learning_rate': 1.9073477604449985e-05, 'epoch': 1.73}

 58%|█████▊    | 2580/4476 [16:44:09<12:16:57, 23.32s/it]
 58%|█████▊    | 2581/4476 [16:44:32<12:17:55, 23.36s/it]
 58%|█████▊    | 2582/4476 [16:44:56<12:20:27, 23.46s/it]
 58%|█████▊    | 2583/4476 [16:45:19<12:20:20, 23.47s/it]
 58%|█████▊    | 2584/4476 [16:45:43<12:18:26, 23.42s/it]
 58%|█████▊    | 2585/4476 [16:46:06<12:16:42, 23.37s/it]
 58%|█████▊    | 2586/4476 [16:46:29<12:15:47, 23.36s/it]
 58%|█████▊    | 2587/4476 [16:46:52<12:13:19, 23.29s/it]
 58%|█████▊    | 2588/4476 [16:47:16<12:16:22, 23.40s/it]
 58%|█████▊    | 2589/4476 [16:47:40<12:19:35, 23.52s/it]
 58%|█████▊    | 2590/4476 [16:48:03<12:16:57, 23.45s/it]
                                                         
{'loss': 0.3795, 'learning_rate': 1.8903157996984174e-05, 'epoch': 1.74}

 58%|█████▊    | 2590/4476 [16:48:03<12:16:57, 23.45s/it]
 58%|█████▊    | 2591/4476 [16:48:26<12:14:34, 23.38s/it]
 58%|█████▊    | 2592/4476 [16:48:50<12:13:19, 23.35s/it]
 58%|█████▊    | 2593/4476 [16:49:13<12:11:32, 23.31s/it]
 58%|█████▊    | 2594/4476 [16:49:36<12:10:31, 23.29s/it]
 58%|█████▊    | 2595/4476 [16:50:00<12:11:05, 23.32s/it]
 58%|█████▊    | 2596/4476 [16:50:23<12:10:38, 23.32s/it]
 58%|█████▊    | 2597/4476 [16:50:46<12:12:33, 23.39s/it]
 58%|█████▊    | 2598/4476 [16:51:10<12:15:47, 23.51s/it]
 58%|█████▊    | 2599/4476 [16:51:34<12:17:16, 23.57s/it]
 58%|█████▊    | 2600/4476 [16:51:57<12:16:16, 23.55s/it]
                                                         
{'loss': 0.3827, 'learning_rate': 1.873313873613733e-05, 'epoch': 1.74}

 58%|█████▊    | 2600/4476 [16:51:57<12:16:16, 23.55s/it][INFO|trainer.py:2939] 2023-11-12 20:15:45,179 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2600
[INFO|tokenization_utils_base.py:2437] 2023-11-12 20:15:45,210 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2600/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 20:15:45,210 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2600/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 20:15:45,211 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2600/added_tokens.json

 58%|█████▊    | 2601/4476 [16:52:21<12:14:11, 23.49s/it]
 58%|█████▊    | 2602/4476 [16:52:44<12:09:45, 23.36s/it]
 58%|█████▊    | 2603/4476 [16:53:07<12:08:39, 23.34s/it]
 58%|█████▊    | 2604/4476 [16:53:31<12:09:10, 23.37s/it]
 58%|█████▊    | 2605/4476 [16:53:54<12:08:12, 23.35s/it]
 58%|█████▊    | 2606/4476 [16:54:17<12:04:58, 23.26s/it]
 58%|█████▊    | 2607/4476 [16:54:40<12:05:39, 23.30s/it]
 58%|█████▊    | 2608/4476 [16:55:04<12:07:24, 23.36s/it]
 58%|█████▊    | 2609/4476 [16:55:27<12:08:20, 23.41s/it]
 58%|█████▊    | 2610/4476 [16:55:51<12:09:13, 23.45s/it]
                                                         
{'loss': 0.385, 'learning_rate': 1.8563428197509502e-05, 'epoch': 1.75}

 58%|█████▊    | 2610/4476 [16:55:51<12:09:13, 23.45s/it]
 58%|█████▊    | 2611/4476 [16:56:14<12:07:40, 23.41s/it]
 58%|█████▊    | 2612/4476 [16:56:37<12:05:56, 23.37s/it]
 58%|█████▊    | 2613/4476 [16:57:01<12:06:02, 23.38s/it]
 58%|█████▊    | 2614/4476 [16:57:24<12:03:48, 23.32s/it]
 58%|█████▊    | 2615/4476 [16:57:47<12:03:04, 23.31s/it]
 58%|█████▊    | 2616/4476 [16:58:11<12:01:35, 23.28s/it]
 58%|█████▊    | 2617/4476 [16:58:34<12:00:34, 23.26s/it]
 58%|█████▊    | 2618/4476 [16:58:57<12:02:52, 23.34s/it]
 59%|█████▊    | 2619/4476 [16:59:20<12:00:11, 23.27s/it]
 59%|█████▊    | 2620/4476 [16:59:44<12:02:01, 23.34s/it]
                                                         
{'loss': 0.3846, 'learning_rate': 1.839403474149225e-05, 'epoch': 1.76}

 59%|█████▊    | 2620/4476 [16:59:44<12:02:01, 23.34s/it]
 59%|█████▊    | 2621/4476 [17:00:07<12:01:01, 23.32s/it]
 59%|█████▊    | 2622/4476 [17:00:31<12:01:31, 23.35s/it]
 59%|█████▊    | 2623/4476 [17:00:54<11:59:02, 23.28s/it]
 59%|█████▊    | 2624/4476 [17:01:17<12:00:20, 23.34s/it]
 59%|█████▊    | 2625/4476 [17:01:40<11:56:51, 23.24s/it]
 59%|█████▊    | 2626/4476 [17:02:03<11:56:40, 23.24s/it]
 59%|█████▊    | 2627/4476 [17:02:27<11:56:48, 23.26s/it]
 59%|█████▊    | 2628/4476 [17:02:50<11:58:10, 23.32s/it]
 59%|█████▊    | 2629/4476 [17:03:14<11:59:59, 23.39s/it]
 59%|█████▉    | 2630/4476 [17:03:37<11:59:52, 23.40s/it]
                                                         
{'loss': 0.3828, 'learning_rate': 1.8224966712856806e-05, 'epoch': 1.76}

 59%|█████▉    | 2630/4476 [17:03:37<11:59:52, 23.40s/it]
 59%|█████▉    | 2631/4476 [17:04:00<11:56:32, 23.30s/it]
 59%|█████▉    | 2632/4476 [17:04:24<11:58:09, 23.37s/it]
 59%|█████▉    | 2633/4476 [17:04:47<11:56:35, 23.33s/it]
 59%|█████▉    | 2634/4476 [17:05:10<11:53:12, 23.23s/it]
 59%|█████▉    | 2635/4476 [17:05:33<11:51:25, 23.19s/it]
 59%|█████▉    | 2636/4476 [17:05:56<11:51:02, 23.19s/it]
 59%|█████▉    | 2637/4476 [17:06:19<11:50:17, 23.17s/it]
 59%|█████▉    | 2638/4476 [17:06:43<11:50:19, 23.19s/it]
 59%|█████▉    | 2639/4476 [17:07:06<11:51:13, 23.23s/it]
 59%|█████▉    | 2640/4476 [17:07:29<11:52:49, 23.30s/it]
                                                         
{'loss': 0.3792, 'learning_rate': 1.8056232440343013e-05, 'epoch': 1.77}

 59%|█████▉    | 2640/4476 [17:07:29<11:52:49, 23.30s/it]
 59%|█████▉    | 2641/4476 [17:07:53<11:51:19, 23.26s/it]
 59%|█████▉    | 2642/4476 [17:08:16<11:53:37, 23.35s/it]
 59%|█████▉    | 2643/4476 [17:08:39<11:51:05, 23.28s/it]
 59%|█████▉    | 2644/4476 [17:09:02<11:48:42, 23.21s/it]
 59%|█████▉    | 2645/4476 [17:09:26<11:48:18, 23.21s/it]
 59%|█████▉    | 2646/4476 [17:09:49<11:49:38, 23.27s/it]
 59%|█████▉    | 2647/4476 [17:10:13<11:52:58, 23.39s/it]
 59%|█████▉    | 2648/4476 [17:10:36<11:52:46, 23.40s/it]
 59%|█████▉    | 2649/4476 [17:10:59<11:49:03, 23.29s/it]
 59%|█████▉    | 2650/4476 [17:11:22<11:47:51, 23.26s/it]
                                                         
{'loss': 0.3814, 'learning_rate': 1.788784023624896e-05, 'epoch': 1.78}

 59%|█████▉    | 2650/4476 [17:11:22<11:47:51, 23.26s/it]
 59%|█████▉    | 2651/4476 [17:11:45<11:46:36, 23.23s/it]
 59%|█████▉    | 2652/4476 [17:12:09<11:48:01, 23.29s/it]
 59%|█████▉    | 2653/4476 [17:12:32<11:48:29, 23.32s/it]
 59%|█████▉    | 2654/4476 [17:12:55<11:47:22, 23.29s/it]
 59%|█████▉    | 2655/4476 [17:13:19<11:48:55, 23.36s/it]
 59%|█████▉    | 2656/4476 [17:13:42<11:48:02, 23.34s/it]
 59%|█████▉    | 2657/4476 [17:14:06<11:46:34, 23.31s/it]
 59%|█████▉    | 2658/4476 [17:14:29<11:45:49, 23.29s/it]
 59%|█████▉    | 2659/4476 [17:14:52<11:43:06, 23.22s/it]
 59%|█████▉    | 2660/4476 [17:15:15<11:46:31, 23.34s/it]
                                                         
{'loss': 0.3819, 'learning_rate': 1.7719798396021558e-05, 'epoch': 1.78}

 59%|█████▉    | 2660/4476 [17:15:15<11:46:31, 23.34s/it]
 59%|█████▉    | 2661/4476 [17:15:39<11:45:43, 23.33s/it]
 59%|█████▉    | 2662/4476 [17:16:02<11:45:35, 23.34s/it]
 59%|█████▉    | 2663/4476 [17:16:26<11:47:28, 23.41s/it]
 60%|█████▉    | 2664/4476 [17:16:49<11:47:11, 23.42s/it]
 60%|█████▉    | 2665/4476 [17:17:12<11:45:53, 23.39s/it]
 60%|█████▉    | 2666/4476 [17:17:36<11:44:06, 23.34s/it]
 60%|█████▉    | 2667/4476 [17:17:59<11:42:53, 23.31s/it]
 60%|█████▉    | 2668/4476 [17:18:23<11:45:55, 23.43s/it]
 60%|█████▉    | 2669/4476 [17:18:46<11:46:40, 23.46s/it]
 60%|█████▉    | 2670/4476 [17:19:10<11:45:41, 23.45s/it]
                                                         
{'loss': 0.3798, 'learning_rate': 1.7552115197847884e-05, 'epoch': 1.79}

 60%|█████▉    | 2670/4476 [17:19:10<11:45:41, 23.45s/it]
 60%|█████▉    | 2671/4476 [17:19:33<11:45:25, 23.45s/it]
 60%|█████▉    | 2672/4476 [17:19:56<11:43:22, 23.39s/it]
 60%|█████▉    | 2673/4476 [17:20:20<11:43:01, 23.40s/it]
 60%|█████▉    | 2674/4476 [17:20:43<11:41:21, 23.35s/it]
 60%|█████▉    | 2675/4476 [17:21:06<11:41:51, 23.38s/it]
 60%|█████▉    | 2676/4476 [17:21:30<11:42:58, 23.43s/it]
 60%|█████▉    | 2677/4476 [17:21:53<11:40:23, 23.36s/it]
 60%|█████▉    | 2678/4476 [17:22:16<11:38:25, 23.31s/it]
 60%|█████▉    | 2679/4476 [17:22:40<11:38:28, 23.32s/it]
 60%|█████▉    | 2680/4476 [17:23:03<11:35:37, 23.24s/it]
                                                         
{'loss': 0.3772, 'learning_rate': 1.7384798902247316e-05, 'epoch': 1.8}

 60%|█████▉    | 2680/4476 [17:23:03<11:35:37, 23.24s/it]
 60%|█████▉    | 2681/4476 [17:23:26<11:34:05, 23.20s/it]
 60%|█████▉    | 2682/4476 [17:23:49<11:37:00, 23.31s/it]
 60%|█████▉    | 2683/4476 [17:24:13<11:36:38, 23.31s/it]
 60%|█████▉    | 2684/4476 [17:24:36<11:34:42, 23.26s/it]
 60%|█████▉    | 2685/4476 [17:25:00<11:38:18, 23.39s/it]
 60%|██████    | 2686/4476 [17:25:23<11:36:55, 23.36s/it]
 60%|██████    | 2687/4476 [17:25:46<11:34:43, 23.30s/it]
 60%|██████    | 2688/4476 [17:26:10<11:36:31, 23.37s/it]
 60%|██████    | 2689/4476 [17:26:33<11:35:25, 23.35s/it]
 60%|██████    | 2690/4476 [17:26:56<11:30:16, 23.19s/it]
                                                         
{'loss': 0.3758, 'learning_rate': 1.7217857751664663e-05, 'epoch': 1.8}

 60%|██████    | 2690/4476 [17:26:56<11:30:16, 23.19s/it]
 60%|██████    | 2691/4476 [17:27:19<11:30:40, 23.22s/it]
 60%|██████    | 2692/4476 [17:27:42<11:32:31, 23.29s/it]
 60%|██████    | 2693/4476 [17:28:06<11:34:09, 23.36s/it]
 60%|██████    | 2694/4476 [17:28:29<11:34:28, 23.38s/it]
 60%|██████    | 2695/4476 [17:28:53<11:35:49, 23.44s/it]
 60%|██████    | 2696/4476 [17:29:17<11:36:39, 23.48s/it]
 60%|██████    | 2697/4476 [17:29:40<11:38:27, 23.56s/it]
 60%|██████    | 2698/4476 [17:30:04<11:36:56, 23.52s/it]
 60%|██████    | 2699/4476 [17:30:27<11:34:09, 23.44s/it]
 60%|██████    | 2700/4476 [17:30:51<11:35:49, 23.51s/it]
                                                         
{'loss': 0.3706, 'learning_rate': 1.7051299970064098e-05, 'epoch': 1.81}

 60%|██████    | 2700/4476 [17:30:51<11:35:49, 23.51s/it]
 60%|██████    | 2701/4476 [17:31:14<11:37:22, 23.57s/it]
 60%|██████    | 2702/4476 [17:31:38<11:35:36, 23.53s/it]
 60%|██████    | 2703/4476 [17:32:01<11:35:34, 23.54s/it]
 60%|██████    | 2704/4476 [17:32:24<11:31:45, 23.42s/it]
 60%|██████    | 2705/4476 [17:32:48<11:32:31, 23.46s/it]
 60%|██████    | 2706/4476 [17:33:12<11:35:05, 23.56s/it]
 60%|██████    | 2707/4476 [17:33:35<11:32:22, 23.48s/it]
 61%|██████    | 2708/4476 [17:33:58<11:28:27, 23.36s/it]
 61%|██████    | 2709/4476 [17:34:22<11:29:44, 23.42s/it]
 61%|██████    | 2710/4476 [17:34:45<11:29:57, 23.44s/it]
                                                         
{'loss': 0.3805, 'learning_rate': 1.6885133762523985e-05, 'epoch': 1.82}

 61%|██████    | 2710/4476 [17:34:45<11:29:57, 23.44s/it]
 61%|██████    | 2711/4476 [17:35:09<11:29:50, 23.45s/it]
 61%|██████    | 2712/4476 [17:35:32<11:29:58, 23.47s/it]
 61%|██████    | 2713/4476 [17:35:56<11:30:45, 23.51s/it]
 61%|██████    | 2714/4476 [17:36:19<11:28:27, 23.44s/it]
 61%|██████    | 2715/4476 [17:36:42<11:27:23, 23.42s/it]
 61%|██████    | 2716/4476 [17:37:06<11:29:57, 23.52s/it]
 61%|██████    | 2717/4476 [17:37:30<11:28:27, 23.48s/it]
 61%|██████    | 2718/4476 [17:37:53<11:26:34, 23.43s/it]
 61%|██████    | 2719/4476 [17:38:16<11:23:48, 23.35s/it]
 61%|██████    | 2720/4476 [17:38:39<11:22:14, 23.31s/it]
                                                         
{'loss': 0.3892, 'learning_rate': 1.6719367314832756e-05, 'epoch': 1.82}

 61%|██████    | 2720/4476 [17:38:39<11:22:14, 23.31s/it]
 61%|██████    | 2721/4476 [17:39:03<11:22:18, 23.33s/it]
 61%|██████    | 2722/4476 [17:39:26<11:22:17, 23.34s/it]
 61%|██████    | 2723/4476 [17:39:49<11:21:49, 23.34s/it]
 61%|██████    | 2724/4476 [17:40:13<11:25:21, 23.47s/it]
 61%|██████    | 2725/4476 [17:40:37<11:25:24, 23.49s/it]
 61%|██████    | 2726/4476 [17:41:00<11:24:25, 23.47s/it]
 61%|██████    | 2727/4476 [17:41:24<11:24:20, 23.48s/it]
 61%|██████    | 2728/4476 [17:41:47<11:22:45, 23.44s/it]
 61%|██████    | 2729/4476 [17:42:10<11:18:45, 23.31s/it]
 61%|██████    | 2730/4476 [17:42:33<11:16:22, 23.24s/it]
                                                         
{'loss': 0.387, 'learning_rate': 1.65540087930856e-05, 'epoch': 1.83}

 61%|██████    | 2730/4476 [17:42:33<11:16:22, 23.24s/it]
 61%|██████    | 2731/4476 [17:42:56<11:14:39, 23.20s/it]
 61%|██████    | 2732/4476 [17:43:19<11:15:02, 23.22s/it]
 61%|██████    | 2733/4476 [17:43:43<11:17:02, 23.31s/it]
 61%|██████    | 2734/4476 [17:44:06<11:18:31, 23.37s/it]
 61%|██████    | 2735/4476 [17:44:30<11:17:44, 23.36s/it]
 61%|██████    | 2736/4476 [17:44:53<11:18:18, 23.39s/it]
 61%|██████    | 2737/4476 [17:45:17<11:19:08, 23.43s/it]
 61%|██████    | 2738/4476 [17:45:40<11:15:51, 23.33s/it]
 61%|██████    | 2739/4476 [17:46:03<11:15:07, 23.32s/it]
 61%|██████    | 2740/4476 [17:46:26<11:13:02, 23.26s/it]
                                                         
{'loss': 0.3773, 'learning_rate': 1.6389066343282168e-05, 'epoch': 1.84}

 61%|██████    | 2740/4476 [17:46:26<11:13:02, 23.26s/it]
 61%|██████    | 2741/4476 [17:46:50<11:14:42, 23.33s/it]
 61%|██████▏   | 2742/4476 [17:47:13<11:13:50, 23.32s/it]
 61%|██████▏   | 2743/4476 [17:47:36<11:10:40, 23.22s/it]
 61%|██████▏   | 2744/4476 [17:48:00<11:12:55, 23.31s/it]
 61%|██████▏   | 2745/4476 [17:48:23<11:12:36, 23.31s/it]
 61%|██████▏   | 2746/4476 [17:48:46<11:13:46, 23.37s/it]
 61%|██████▏   | 2747/4476 [17:49:10<11:13:52, 23.38s/it]
 61%|██████▏   | 2748/4476 [17:49:33<11:12:32, 23.35s/it]
 61%|██████▏   | 2749/4476 [17:49:56<11:09:34, 23.26s/it]
 61%|██████▏   | 2750/4476 [17:50:20<11:10:03, 23.29s/it]
                                                         
{'loss': 0.3829, 'learning_rate': 1.6224548090925323e-05, 'epoch': 1.84}

 61%|██████▏   | 2750/4476 [17:50:20<11:10:03, 23.29s/it]
 61%|██████▏   | 2751/4476 [17:50:43<11:07:22, 23.21s/it]
 61%|██████▏   | 2752/4476 [17:51:06<11:06:54, 23.21s/it]
 62%|██████▏   | 2753/4476 [17:51:29<11:09:33, 23.32s/it]
 62%|██████▏   | 2754/4476 [17:51:53<11:08:52, 23.31s/it]
 62%|██████▏   | 2755/4476 [17:52:16<11:06:31, 23.24s/it]
 62%|██████▏   | 2756/4476 [17:52:39<11:06:17, 23.24s/it]
 62%|██████▏   | 2757/4476 [17:53:02<11:05:01, 23.21s/it]
 62%|██████▏   | 2758/4476 [17:53:25<11:01:31, 23.10s/it]
 62%|██████▏   | 2759/4476 [17:53:48<10:58:51, 23.02s/it]
 62%|██████▏   | 2760/4476 [17:54:11<11:01:33, 23.13s/it]
                                                         
{'loss': 0.3697, 'learning_rate': 1.6060462140620835e-05, 'epoch': 1.85}

 62%|██████▏   | 2760/4476 [17:54:11<11:01:33, 23.13s/it]
 62%|██████▏   | 2761/4476 [17:54:34<11:02:42, 23.19s/it]
 62%|██████▏   | 2762/4476 [17:54:57<11:00:58, 23.14s/it]
 62%|██████▏   | 2763/4476 [17:55:21<11:02:05, 23.19s/it]
 62%|██████▏   | 2764/4476 [17:55:44<11:04:38, 23.29s/it]
 62%|██████▏   | 2765/4476 [17:56:08<11:05:21, 23.33s/it]
 62%|██████▏   | 2766/4476 [17:56:31<11:04:31, 23.32s/it]
 62%|██████▏   | 2767/4476 [17:56:54<11:02:38, 23.26s/it]
 62%|██████▏   | 2768/4476 [17:57:18<11:04:04, 23.33s/it]
 62%|██████▏   | 2769/4476 [17:57:41<11:05:14, 23.38s/it]
 62%|██████▏   | 2770/4476 [17:58:04<11:03:42, 23.34s/it]
                                                         
{'loss': 0.3817, 'learning_rate': 1.589681657567811e-05, 'epoch': 1.86}

 62%|██████▏   | 2770/4476 [17:58:04<11:03:42, 23.34s/it]
 62%|██████▏   | 2771/4476 [17:58:28<11:06:59, 23.47s/it]
 62%|██████▏   | 2772/4476 [17:58:51<11:03:35, 23.37s/it]
 62%|██████▏   | 2773/4476 [17:59:15<11:04:28, 23.41s/it]
 62%|██████▏   | 2774/4476 [17:59:38<11:02:38, 23.36s/it]
 62%|██████▏   | 2775/4476 [18:00:01<11:00:49, 23.31s/it]
 62%|██████▏   | 2776/4476 [18:00:25<11:01:14, 23.34s/it]
 62%|██████▏   | 2777/4476 [18:00:48<10:59:24, 23.29s/it]
 62%|██████▏   | 2778/4476 [18:01:11<10:59:16, 23.30s/it]
 62%|██████▏   | 2779/4476 [18:01:34<10:58:24, 23.28s/it]
 62%|██████▏   | 2780/4476 [18:01:58<10:59:39, 23.34s/it]
                                                         
{'loss': 0.3819, 'learning_rate': 1.5733619457712037e-05, 'epoch': 1.86}

 62%|██████▏   | 2780/4476 [18:01:58<10:59:39, 23.34s/it]
 62%|██████▏   | 2781/4476 [18:02:21<10:56:33, 23.24s/it]
 62%|██████▏   | 2782/4476 [18:02:44<10:58:02, 23.31s/it]
 62%|██████▏   | 2783/4476 [18:03:08<11:01:14, 23.43s/it]
 62%|██████▏   | 2784/4476 [18:03:31<10:58:30, 23.35s/it]
 62%|██████▏   | 2785/4476 [18:03:55<10:58:16, 23.36s/it]
 62%|██████▏   | 2786/4476 [18:04:18<10:55:33, 23.27s/it]
 62%|██████▏   | 2787/4476 [18:04:41<10:59:25, 23.43s/it]
 62%|██████▏   | 2788/4476 [18:05:05<10:57:58, 23.39s/it]
 62%|██████▏   | 2789/4476 [18:05:28<10:57:57, 23.40s/it]
 62%|██████▏   | 2790/4476 [18:05:52<10:57:12, 23.39s/it]
                                                         
{'loss': 0.3756, 'learning_rate': 1.5570878826245773e-05, 'epoch': 1.87}

 62%|██████▏   | 2790/4476 [18:05:52<10:57:12, 23.39s/it]
 62%|██████▏   | 2791/4476 [18:06:14<10:50:38, 23.17s/it]
 62%|██████▏   | 2792/4476 [18:06:37<10:50:58, 23.19s/it]
 62%|██████▏   | 2793/4476 [18:07:01<10:50:14, 23.18s/it]
 62%|██████▏   | 2794/4476 [18:07:24<10:52:29, 23.28s/it]
 62%|██████▏   | 2795/4476 [18:07:47<10:51:31, 23.26s/it]
 62%|██████▏   | 2796/4476 [18:08:10<10:49:37, 23.20s/it]
 62%|██████▏   | 2797/4476 [18:08:34<10:50:51, 23.26s/it]
 63%|██████▎   | 2798/4476 [18:08:57<10:52:42, 23.34s/it]
 63%|██████▎   | 2799/4476 [18:09:21<10:51:10, 23.30s/it]
 63%|██████▎   | 2800/4476 [18:09:44<10:50:19, 23.28s/it]
                                                         
{'loss': 0.3791, 'learning_rate': 1.5408602698314777e-05, 'epoch': 1.88}

 63%|██████▎   | 2800/4476 [18:09:44<10:50:19, 23.28s/it][INFO|trainer.py:2939] 2023-11-12 21:33:31,487 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2800
[INFO|tokenization_utils_base.py:2437] 2023-11-12 21:33:31,519 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2800/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 21:33:31,519 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2800/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 21:33:31,519 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-2800/added_tokens.json

 63%|██████▎   | 2801/4476 [18:10:07<10:48:08, 23.22s/it]
 63%|██████▎   | 2802/4476 [18:10:30<10:51:18, 23.34s/it]
 63%|██████▎   | 2803/4476 [18:10:54<10:50:13, 23.32s/it]
 63%|██████▎   | 2804/4476 [18:11:17<10:50:06, 23.33s/it]
 63%|██████▎   | 2805/4476 [18:11:40<10:48:07, 23.27s/it]
 63%|██████▎   | 2806/4476 [18:12:04<10:47:55, 23.28s/it]
 63%|██████▎   | 2807/4476 [18:12:27<10:48:10, 23.30s/it]
 63%|██████▎   | 2808/4476 [18:12:50<10:48:00, 23.31s/it]
 63%|██████▎   | 2809/4476 [18:13:13<10:45:05, 23.22s/it]
 63%|██████▎   | 2810/4476 [18:13:37<10:45:38, 23.25s/it]
                                                         
{'loss': 0.3765, 'learning_rate': 1.5246799068071818e-05, 'epoch': 1.88}

 63%|██████▎   | 2810/4476 [18:13:37<10:45:38, 23.25s/it]
 63%|██████▎   | 2811/4476 [18:14:00<10:45:41, 23.27s/it]
 63%|██████▎   | 2812/4476 [18:14:23<10:46:59, 23.33s/it]
 63%|██████▎   | 2813/4476 [18:14:47<10:48:31, 23.40s/it]
 63%|██████▎   | 2814/4476 [18:15:10<10:48:41, 23.42s/it]
 63%|██████▎   | 2815/4476 [18:15:33<10:45:19, 23.31s/it]
 63%|██████▎   | 2816/4476 [18:15:57<10:44:44, 23.30s/it]
 63%|██████▎   | 2817/4476 [18:16:20<10:46:07, 23.37s/it]
 63%|██████▎   | 2818/4476 [18:16:44<10:46:07, 23.38s/it]
 63%|██████▎   | 2819/4476 [18:17:07<10:44:38, 23.34s/it]
 63%|██████▎   | 2820/4476 [18:17:31<10:47:17, 23.45s/it]
                                                         
{'loss': 0.3834, 'learning_rate': 1.5085475906393153e-05, 'epoch': 1.89}

 63%|██████▎   | 2820/4476 [18:17:31<10:47:17, 23.45s/it]
 63%|██████▎   | 2821/4476 [18:17:54<10:47:48, 23.49s/it]
 63%|██████▎   | 2822/4476 [18:18:18<10:48:55, 23.54s/it]
 63%|██████▎   | 2823/4476 [18:18:41<10:45:16, 23.42s/it]
 63%|██████▎   | 2824/4476 [18:19:04<10:44:44, 23.42s/it]
 63%|██████▎   | 2825/4476 [18:19:28<10:45:19, 23.45s/it]
 63%|██████▎   | 2826/4476 [18:19:51<10:40:18, 23.28s/it]
 63%|██████▎   | 2827/4476 [18:20:14<10:40:16, 23.30s/it]
 63%|██████▎   | 2828/4476 [18:20:37<10:40:34, 23.32s/it]
 63%|██████▎   | 2829/4476 [18:21:01<10:39:24, 23.29s/it]
 63%|██████▎   | 2830/4476 [18:21:24<10:37:34, 23.24s/it]
                                                         
{'loss': 0.381, 'learning_rate': 1.4924641160485923e-05, 'epoch': 1.9}

 63%|██████▎   | 2830/4476 [18:21:24<10:37:34, 23.24s/it]
 63%|██████▎   | 2831/4476 [18:21:47<10:38:21, 23.28s/it]
 63%|██████▎   | 2832/4476 [18:22:11<10:41:40, 23.42s/it]
 63%|██████▎   | 2833/4476 [18:22:34<10:41:30, 23.43s/it]
 63%|██████▎   | 2834/4476 [18:22:58<10:42:41, 23.48s/it]
 63%|██████▎   | 2835/4476 [18:23:21<10:39:07, 23.37s/it]
 63%|██████▎   | 2836/4476 [18:23:44<10:38:35, 23.36s/it]
 63%|██████▎   | 2837/4476 [18:24:07<10:33:17, 23.18s/it]
 63%|██████▎   | 2838/4476 [18:24:31<10:33:58, 23.22s/it]
 63%|██████▎   | 2839/4476 [18:24:54<10:35:14, 23.28s/it]
 63%|██████▎   | 2840/4476 [18:25:17<10:35:59, 23.33s/it]
                                                         
{'loss': 0.3766, 'learning_rate': 1.4764302753496584e-05, 'epoch': 1.9}

 63%|██████▎   | 2840/4476 [18:25:17<10:35:59, 23.33s/it]
 63%|██████▎   | 2841/4476 [18:25:40<10:32:07, 23.20s/it]
 63%|██████▎   | 2842/4476 [18:26:04<10:32:07, 23.21s/it]
 64%|██████▎   | 2843/4476 [18:26:27<10:33:38, 23.28s/it]
 64%|██████▎   | 2844/4476 [18:26:50<10:30:24, 23.18s/it]
 64%|██████▎   | 2845/4476 [18:27:13<10:30:30, 23.19s/it]
 64%|██████▎   | 2846/4476 [18:27:36<10:30:26, 23.21s/it]
 64%|██████▎   | 2847/4476 [18:28:00<10:29:47, 23.20s/it]
 64%|██████▎   | 2848/4476 [18:28:23<10:28:20, 23.16s/it]
 64%|██████▎   | 2849/4476 [18:28:46<10:27:23, 23.14s/it]
 64%|██████▎   | 2850/4476 [18:29:09<10:27:19, 23.15s/it]
                                                         
{'loss': 0.3815, 'learning_rate': 1.4604468584120607e-05, 'epoch': 1.91}

 64%|██████▎   | 2850/4476 [18:29:09<10:27:19, 23.15s/it]
 64%|██████▎   | 2851/4476 [18:29:32<10:28:30, 23.21s/it]
 64%|██████▎   | 2852/4476 [18:29:55<10:27:24, 23.18s/it]
 64%|██████▎   | 2853/4476 [18:30:19<10:26:54, 23.18s/it]
 64%|██████▍   | 2854/4476 [18:30:42<10:27:51, 23.23s/it]
 64%|██████▍   | 2855/4476 [18:31:05<10:28:28, 23.26s/it]
 64%|██████▍   | 2856/4476 [18:31:28<10:26:30, 23.20s/it]
 64%|██████▍   | 2857/4476 [18:31:51<10:24:15, 23.13s/it]
 64%|██████▍   | 2858/4476 [18:32:15<10:26:01, 23.21s/it]
 64%|██████▍   | 2859/4476 [18:32:38<10:23:32, 23.14s/it]
 64%|██████▍   | 2860/4476 [18:33:01<10:25:30, 23.22s/it]
                                                         
{'loss': 0.3774, 'learning_rate': 1.4445146526213415e-05, 'epoch': 1.92}

 64%|██████▍   | 2860/4476 [18:33:01<10:25:30, 23.22s/it]
 64%|██████▍   | 2861/4476 [18:33:24<10:27:00, 23.29s/it]
 64%|██████▍   | 2862/4476 [18:33:48<10:26:16, 23.28s/it]
 64%|██████▍   | 2863/4476 [18:34:11<10:24:10, 23.22s/it]
 64%|██████▍   | 2864/4476 [18:34:34<10:25:03, 23.27s/it]
 64%|██████▍   | 2865/4476 [18:34:57<10:25:02, 23.28s/it]
 64%|██████▍   | 2866/4476 [18:35:21<10:25:08, 23.30s/it]
 64%|██████▍   | 2867/4476 [18:35:44<10:24:22, 23.28s/it]
 64%|██████▍   | 2868/4476 [18:36:07<10:23:39, 23.27s/it]
 64%|██████▍   | 2869/4476 [18:36:30<10:21:20, 23.20s/it]
 64%|██████▍   | 2870/4476 [18:36:54<10:22:51, 23.27s/it]
                                                         
{'loss': 0.3879, 'learning_rate': 1.4286344428402454e-05, 'epoch': 1.92}

 64%|██████▍   | 2870/4476 [18:36:54<10:22:51, 23.27s/it]
 64%|██████▍   | 2871/4476 [18:37:17<10:23:29, 23.31s/it]
 64%|██████▍   | 2872/4476 [18:37:41<10:23:11, 23.31s/it]
 64%|██████▍   | 2873/4476 [18:38:04<10:24:40, 23.38s/it]
 64%|██████▍   | 2874/4476 [18:38:27<10:23:48, 23.36s/it]
 64%|██████▍   | 2875/4476 [18:38:51<10:23:11, 23.36s/it]
 64%|██████▍   | 2876/4476 [18:39:14<10:23:59, 23.40s/it]
 64%|██████▍   | 2877/4476 [18:39:38<10:23:38, 23.40s/it]
 64%|██████▍   | 2878/4476 [18:40:01<10:23:52, 23.42s/it]
 64%|██████▍   | 2879/4476 [18:40:24<10:20:44, 23.32s/it]
 64%|██████▍   | 2880/4476 [18:40:47<10:19:18, 23.28s/it]
                                                         
{'loss': 0.3777, 'learning_rate': 1.412807011370052e-05, 'epoch': 1.93}

 64%|██████▍   | 2880/4476 [18:40:47<10:19:18, 23.28s/it]
 64%|██████▍   | 2881/4476 [18:41:10<10:15:59, 23.17s/it]
 64%|██████▍   | 2882/4476 [18:41:34<10:16:21, 23.20s/it]
 64%|██████▍   | 2883/4476 [18:41:57<10:16:26, 23.22s/it]
 64%|██████▍   | 2884/4476 [18:42:20<10:18:15, 23.30s/it]
 64%|██████▍   | 2885/4476 [18:42:44<10:18:28, 23.32s/it]
 64%|██████▍   | 2886/4476 [18:43:07<10:18:03, 23.32s/it]
 64%|██████▍   | 2887/4476 [18:43:30<10:18:59, 23.37s/it]
 65%|██████▍   | 2888/4476 [18:43:54<10:16:48, 23.31s/it]
 65%|██████▍   | 2889/4476 [18:44:17<10:17:15, 23.34s/it]
 65%|██████▍   | 2890/4476 [18:44:41<10:19:01, 23.42s/it]
                                                         
{'loss': 0.3806, 'learning_rate': 1.3970331379120455e-05, 'epoch': 1.94}

 65%|██████▍   | 2890/4476 [18:44:41<10:19:01, 23.42s/it]
 65%|██████▍   | 2891/4476 [18:45:04<10:16:49, 23.35s/it]
 65%|██████▍   | 2892/4476 [18:45:27<10:17:54, 23.41s/it]
 65%|██████▍   | 2893/4476 [18:45:51<10:16:21, 23.36s/it]
 65%|██████▍   | 2894/4476 [18:46:14<10:17:02, 23.40s/it]
 65%|██████▍   | 2895/4476 [18:46:37<10:15:17, 23.35s/it]
 65%|██████▍   | 2896/4476 [18:47:01<10:14:51, 23.35s/it]
 65%|██████▍   | 2897/4476 [18:47:24<10:13:36, 23.32s/it]
 65%|██████▍   | 2898/4476 [18:47:47<10:13:10, 23.31s/it]
 65%|██████▍   | 2899/4476 [18:48:11<10:13:44, 23.35s/it]
 65%|██████▍   | 2900/4476 [18:48:34<10:15:49, 23.44s/it]
                                                         
{'loss': 0.3848, 'learning_rate': 1.3813135995290988e-05, 'epoch': 1.94}

 65%|██████▍   | 2900/4476 [18:48:34<10:15:49, 23.44s/it]
 65%|██████▍   | 2901/4476 [18:48:58<10:15:15, 23.44s/it]
 65%|██████▍   | 2902/4476 [18:49:22<10:17:30, 23.54s/it]
 65%|██████▍   | 2903/4476 [18:49:45<10:14:04, 23.42s/it]
 65%|██████▍   | 2904/4476 [18:50:08<10:09:12, 23.25s/it]
 65%|██████▍   | 2905/4476 [18:50:31<10:08:42, 23.25s/it]
 65%|██████▍   | 2906/4476 [18:50:54<10:09:48, 23.30s/it]
 65%|██████▍   | 2907/4476 [18:51:17<10:05:43, 23.16s/it]
 65%|██████▍   | 2908/4476 [18:51:40<10:05:44, 23.18s/it]
 65%|██████▍   | 2909/4476 [18:52:04<10:08:32, 23.30s/it]
 65%|██████▌   | 2910/4476 [18:52:27<10:07:55, 23.29s/it]
                                                         
{'loss': 0.3745, 'learning_rate': 1.3656491706073935e-05, 'epoch': 1.95}

 65%|██████▌   | 2910/4476 [18:52:27<10:07:55, 23.29s/it]
 65%|██████▌   | 2911/4476 [18:52:51<10:09:05, 23.35s/it]
 65%|██████▌   | 2912/4476 [18:53:14<10:07:33, 23.31s/it]
 65%|██████▌   | 2913/4476 [18:53:37<10:06:31, 23.28s/it]
 65%|██████▌   | 2914/4476 [18:54:01<10:09:07, 23.40s/it]
 65%|██████▌   | 2915/4476 [18:54:24<10:07:38, 23.36s/it]
 65%|██████▌   | 2916/4476 [18:54:47<10:05:21, 23.28s/it]
 65%|██████▌   | 2917/4476 [18:55:11<10:08:14, 23.41s/it]
 65%|██████▌   | 2918/4476 [18:55:34<10:08:21, 23.43s/it]
 65%|██████▌   | 2919/4476 [18:55:58<10:06:34, 23.37s/it]
 65%|██████▌   | 2920/4476 [18:56:21<10:04:59, 23.33s/it]
                                                         
{'loss': 0.377, 'learning_rate': 1.350040622818275e-05, 'epoch': 1.96}

 65%|██████▌   | 2920/4476 [18:56:21<10:04:59, 23.33s/it]
 65%|██████▌   | 2921/4476 [18:56:44<10:02:20, 23.24s/it]
 65%|██████▌   | 2922/4476 [18:57:07<10:00:57, 23.20s/it]
 65%|██████▌   | 2923/4476 [18:57:30<10:02:01, 23.26s/it]
 65%|██████▌   | 2924/4476 [18:57:54<10:03:42, 23.34s/it]
 65%|██████▌   | 2925/4476 [18:58:17<10:03:24, 23.34s/it]
 65%|██████▌   | 2926/4476 [18:58:41<10:06:05, 23.46s/it]
 65%|██████▌   | 2927/4476 [18:59:04<10:02:19, 23.33s/it]
 65%|██████▌   | 2928/4476 [18:59:27<10:03:21, 23.39s/it]
 65%|██████▌   | 2929/4476 [18:59:51<10:01:01, 23.31s/it]
 65%|██████▌   | 2930/4476 [19:00:14<10:03:36, 23.43s/it]
                                                         
{'loss': 0.3783, 'learning_rate': 1.3344887250802345e-05, 'epoch': 1.96}

 65%|██████▌   | 2930/4476 [19:00:14<10:03:36, 23.43s/it]
 65%|██████▌   | 2931/4476 [19:00:38<10:05:41, 23.52s/it]
 66%|██████▌   | 2932/4476 [19:01:01<10:02:14, 23.40s/it]
 66%|██████▌   | 2933/4476 [19:01:24<10:00:28, 23.35s/it]
 66%|██████▌   | 2934/4476 [19:01:48<9:59:30, 23.33s/it] 
 66%|██████▌   | 2935/4476 [19:02:11<10:00:03, 23.36s/it]
 66%|██████▌   | 2936/4476 [19:02:34<9:58:40, 23.33s/it] 
 66%|██████▌   | 2937/4476 [19:02:58<9:59:31, 23.37s/it]
 66%|██████▌   | 2938/4476 [19:03:21<10:00:00, 23.41s/it]
 66%|██████▌   | 2939/4476 [19:03:45<9:58:18, 23.36s/it] 
 66%|██████▌   | 2940/4476 [19:04:08<10:00:59, 23.48s/it]
                                                         
{'loss': 0.3768, 'learning_rate': 1.3189942435210301e-05, 'epoch': 1.97}

 66%|██████▌   | 2940/4476 [19:04:08<10:00:59, 23.48s/it]
 66%|██████▌   | 2941/4476 [19:04:32<9:59:00, 23.41s/it] 
 66%|██████▌   | 2942/4476 [19:04:55<9:57:44, 23.38s/it]
 66%|██████▌   | 2943/4476 [19:05:18<9:54:45, 23.28s/it]
 66%|██████▌   | 2944/4476 [19:05:41<9:56:44, 23.37s/it]
 66%|██████▌   | 2945/4476 [19:06:05<9:57:55, 23.43s/it]
 66%|██████▌   | 2946/4476 [19:06:28<9:56:22, 23.39s/it]
 66%|██████▌   | 2947/4476 [19:06:52<9:54:55, 23.35s/it]
 66%|██████▌   | 2948/4476 [19:07:15<9:56:16, 23.41s/it]
 66%|██████▌   | 2949/4476 [19:07:39<9:57:20, 23.47s/it]
 66%|██████▌   | 2950/4476 [19:08:02<9:55:34, 23.42s/it]
                                                        
{'loss': 0.3744, 'learning_rate': 1.303557941439949e-05, 'epoch': 1.98}

 66%|██████▌   | 2950/4476 [19:08:02<9:55:34, 23.42s/it]
 66%|██████▌   | 2951/4476 [19:08:25<9:54:24, 23.39s/it]
 66%|██████▌   | 2952/4476 [19:08:49<9:53:29, 23.37s/it]
 66%|██████▌   | 2953/4476 [19:09:12<9:51:57, 23.32s/it]
 66%|██████▌   | 2954/4476 [19:09:35<9:49:58, 23.26s/it]
 66%|██████▌   | 2955/4476 [19:09:59<9:53:04, 23.40s/it]
 66%|██████▌   | 2956/4476 [19:10:22<9:52:31, 23.39s/it]
 66%|██████▌   | 2957/4476 [19:10:46<9:53:20, 23.44s/it]
 66%|██████▌   | 2958/4476 [19:11:09<9:51:46, 23.39s/it]
 66%|██████▌   | 2959/4476 [19:11:32<9:49:41, 23.32s/it]
 66%|██████▌   | 2960/4476 [19:11:56<9:52:44, 23.46s/it]
                                                        
{'loss': 0.3788, 'learning_rate': 1.2881805792702031e-05, 'epoch': 1.98}

 66%|██████▌   | 2960/4476 [19:11:56<9:52:44, 23.46s/it]
 66%|██████▌   | 2961/4476 [19:12:19<9:51:07, 23.41s/it]
 66%|██████▌   | 2962/4476 [19:12:42<9:49:50, 23.38s/it]
 66%|██████▌   | 2963/4476 [19:13:06<9:47:47, 23.31s/it]
 66%|██████▌   | 2964/4476 [19:13:29<9:47:59, 23.33s/it]
 66%|██████▌   | 2965/4476 [19:13:52<9:48:31, 23.37s/it]
 66%|██████▋   | 2966/4476 [19:14:16<9:48:32, 23.39s/it]
 66%|██████▋   | 2967/4476 [19:14:39<9:49:12, 23.43s/it]
 66%|██████▋   | 2968/4476 [19:15:03<9:50:52, 23.51s/it]
 66%|██████▋   | 2969/4476 [19:15:27<9:51:04, 23.53s/it]
 66%|██████▋   | 2970/4476 [19:15:50<9:49:52, 23.50s/it]
                                                        
{'loss': 0.3735, 'learning_rate': 1.2728629145414645e-05, 'epoch': 1.99}

 66%|██████▋   | 2970/4476 [19:15:50<9:49:52, 23.50s/it]
 66%|██████▋   | 2971/4476 [19:16:13<9:46:45, 23.39s/it]
 66%|██████▋   | 2972/4476 [19:16:37<9:45:18, 23.35s/it]
 66%|██████▋   | 2973/4476 [19:17:00<9:47:41, 23.46s/it]
 66%|██████▋   | 2974/4476 [19:17:24<9:47:27, 23.47s/it]
 66%|██████▋   | 2975/4476 [19:17:47<9:45:22, 23.40s/it]
 66%|██████▋   | 2976/4476 [19:18:10<9:42:29, 23.30s/it]
 67%|██████▋   | 2977/4476 [19:18:34<9:44:28, 23.39s/it]
 67%|██████▋   | 2978/4476 [19:18:56<9:39:31, 23.21s/it]
 67%|██████▋   | 2979/4476 [19:19:20<9:39:57, 23.24s/it]
 67%|██████▋   | 2980/4476 [19:19:43<9:37:14, 23.15s/it]
                                                        
{'loss': 0.3819, 'learning_rate': 1.257605701842554e-05, 'epoch': 2.0}

 67%|██████▋   | 2980/4476 [19:19:43<9:37:14, 23.15s/it]
 67%|██████▋   | 2981/4476 [19:20:06<9:38:41, 23.23s/it]
 67%|██████▋   | 2982/4476 [19:20:29<9:37:40, 23.20s/it]
 67%|██████▋   | 2983/4476 [19:20:53<9:38:09, 23.23s/it]
 67%|██████▋   | 2984/4476 [19:21:16<9:38:40, 23.27s/it]
 67%|██████▋   | 2985/4476 [19:21:39<9:38:44, 23.29s/it]
 67%|██████▋   | 2986/4476 [19:22:02<9:36:34, 23.22s/it]
 67%|██████▋   | 2987/4476 [19:22:26<9:36:58, 23.25s/it]
 67%|██████▋   | 2988/4476 [19:22:49<9:37:03, 23.27s/it]
 67%|██████▋   | 2989/4476 [19:23:12<9:36:41, 23.27s/it]
 67%|██████▋   | 2990/4476 [19:23:36<9:36:34, 23.28s/it]
                                                        
{'loss': 0.3812, 'learning_rate': 1.242409692784265e-05, 'epoch': 2.0}

 67%|██████▋   | 2990/4476 [19:23:36<9:36:34, 23.28s/it]
 67%|██████▋   | 2991/4476 [19:23:59<9:36:34, 23.30s/it]
 67%|██████▋   | 2992/4476 [19:24:22<9:36:33, 23.31s/it]
 67%|██████▋   | 2993/4476 [19:24:46<9:36:56, 23.34s/it]
 67%|██████▋   | 2994/4476 [19:25:09<9:36:08, 23.33s/it]
 67%|██████▋   | 2995/4476 [19:25:32<9:35:01, 23.30s/it]
 67%|██████▋   | 2996/4476 [19:25:55<9:34:35, 23.29s/it]
 67%|██████▋   | 2997/4476 [19:26:19<9:33:52, 23.28s/it]
 67%|██████▋   | 2998/4476 [19:26:42<9:34:40, 23.33s/it]
 67%|██████▋   | 2999/4476 [19:27:05<9:34:25, 23.33s/it]
 67%|██████▋   | 3000/4476 [19:27:29<9:35:34, 23.40s/it]
                                                        
{'loss': 0.3769, 'learning_rate': 1.2272756359623342e-05, 'epoch': 2.01}

 67%|██████▋   | 3000/4476 [19:27:29<9:35:34, 23.40s/it][INFO|trainer.py:2939] 2023-11-12 22:51:16,724 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3000
[INFO|tokenization_utils_base.py:2437] 2023-11-12 22:51:16,763 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3000/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-12 22:51:16,763 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3000/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-12 22:51:16,763 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3000/added_tokens.json

 67%|██████▋   | 3001/4476 [19:27:52<9:35:49, 23.42s/it]
 67%|██████▋   | 3002/4476 [19:28:16<9:37:38, 23.51s/it]
 67%|██████▋   | 3003/4476 [19:28:40<9:36:29, 23.48s/it]
 67%|██████▋   | 3004/4476 [19:29:03<9:34:26, 23.41s/it]
 67%|██████▋   | 3005/4476 [19:29:26<9:31:54, 23.33s/it]
 67%|██████▋   | 3006/4476 [19:29:49<9:32:41, 23.38s/it]
 67%|██████▋   | 3007/4476 [19:30:12<9:29:14, 23.25s/it]
 67%|██████▋   | 3008/4476 [19:30:36<9:28:27, 23.23s/it]
 67%|██████▋   | 3009/4476 [19:30:59<9:26:42, 23.18s/it]
 67%|██████▋   | 3010/4476 [19:31:22<9:25:59, 23.16s/it]
                                                        
{'loss': 0.3779, 'learning_rate': 1.2122042769205702e-05, 'epoch': 2.02}

 67%|██████▋   | 3010/4476 [19:31:22<9:25:59, 23.16s/it]
 67%|██████▋   | 3011/4476 [19:31:45<9:28:21, 23.28s/it]
 67%|██████▋   | 3012/4476 [19:32:09<9:28:12, 23.29s/it]
 67%|██████▋   | 3013/4476 [19:32:32<9:29:33, 23.36s/it]
 67%|██████▋   | 3014/4476 [19:32:55<9:28:43, 23.34s/it]
 67%|██████▋   | 3015/4476 [19:33:19<9:26:22, 23.26s/it]
 67%|██████▋   | 3016/4476 [19:33:42<9:27:31, 23.32s/it]
 67%|██████▋   | 3017/4476 [19:34:05<9:27:30, 23.34s/it]
 67%|██████▋   | 3018/4476 [19:34:29<9:29:29, 23.44s/it]
 67%|██████▋   | 3019/4476 [19:34:52<9:27:19, 23.36s/it]
 67%|██████▋   | 3020/4476 [19:35:16<9:27:26, 23.38s/it]
                                                        
{'loss': 0.3817, 'learning_rate': 1.1971963581141196e-05, 'epoch': 2.02}

 67%|██████▋   | 3020/4476 [19:35:16<9:27:26, 23.38s/it]
 67%|██████▋   | 3021/4476 [19:35:39<9:25:40, 23.33s/it]
 68%|██████▊   | 3022/4476 [19:36:03<9:27:32, 23.42s/it]
 68%|██████▊   | 3023/4476 [19:36:26<9:28:40, 23.48s/it]
 68%|██████▊   | 3024/4476 [19:36:50<9:28:34, 23.50s/it]
 68%|██████▊   | 3025/4476 [19:37:13<9:26:59, 23.45s/it]
 68%|██████▊   | 3026/4476 [19:37:37<9:27:02, 23.46s/it]
 68%|██████▊   | 3027/4476 [19:38:00<9:27:00, 23.48s/it]
 68%|██████▊   | 3028/4476 [19:38:23<9:24:01, 23.37s/it]
 68%|██████▊   | 3029/4476 [19:38:46<9:22:22, 23.32s/it]
 68%|██████▊   | 3030/4476 [19:39:09<9:18:36, 23.18s/it]
                                                        
{'loss': 0.3788, 'learning_rate': 1.1822526188728966e-05, 'epoch': 2.03}

 68%|██████▊   | 3030/4476 [19:39:09<9:18:36, 23.18s/it]
 68%|██████▊   | 3031/4476 [19:39:32<9:17:11, 23.14s/it]
 68%|██████▊   | 3032/4476 [19:39:56<9:19:04, 23.23s/it]
 68%|██████▊   | 3033/4476 [19:40:19<9:21:33, 23.35s/it]
 68%|██████▊   | 3034/4476 [19:40:43<9:20:44, 23.33s/it]
 68%|██████▊   | 3035/4476 [19:41:06<9:21:30, 23.38s/it]
 68%|██████▊   | 3036/4476 [19:41:29<9:21:10, 23.38s/it]
 68%|██████▊   | 3037/4476 [19:41:53<9:22:43, 23.46s/it]
 68%|██████▊   | 3038/4476 [19:42:17<9:21:53, 23.44s/it]
 68%|██████▊   | 3039/4476 [19:42:40<9:20:36, 23.41s/it]
 68%|██████▊   | 3040/4476 [19:43:03<9:19:27, 23.38s/it]
                                                        
{'loss': 0.3776, 'learning_rate': 1.1673737953651601e-05, 'epoch': 2.04}

 68%|██████▊   | 3040/4476 [19:43:03<9:19:27, 23.38s/it]
 68%|██████▊   | 3041/4476 [19:43:27<9:19:13, 23.38s/it]
 68%|██████▊   | 3042/4476 [19:43:50<9:17:16, 23.32s/it]
 68%|██████▊   | 3043/4476 [19:44:13<9:18:39, 23.39s/it]
 68%|██████▊   | 3044/4476 [19:44:36<9:16:39, 23.32s/it]
 68%|██████▊   | 3045/4476 [19:45:00<9:17:57, 23.39s/it]
 68%|██████▊   | 3046/4476 [19:45:24<9:18:27, 23.43s/it]
 68%|██████▊   | 3047/4476 [19:45:47<9:19:06, 23.48s/it]
 68%|██████▊   | 3048/4476 [19:46:11<9:18:36, 23.47s/it]
 68%|██████▊   | 3049/4476 [19:46:34<9:20:06, 23.55s/it]
 68%|██████▊   | 3050/4476 [19:46:58<9:19:47, 23.55s/it]
                                                        
{'loss': 0.3785, 'learning_rate': 1.1525606205612447e-05, 'epoch': 2.04}

 68%|██████▊   | 3050/4476 [19:46:58<9:19:47, 23.55s/it]
 68%|██████▊   | 3051/4476 [19:47:22<9:20:46, 23.61s/it]
 68%|██████▊   | 3052/4476 [19:47:45<9:20:44, 23.63s/it]
 68%|██████▊   | 3053/4476 [19:48:08<9:16:38, 23.47s/it]
 68%|██████▊   | 3054/4476 [19:48:32<9:16:23, 23.48s/it]
 68%|██████▊   | 3055/4476 [19:48:56<9:17:17, 23.53s/it]
 68%|██████▊   | 3056/4476 [19:49:19<9:14:59, 23.45s/it]
 68%|██████▊   | 3057/4476 [19:49:42<9:13:53, 23.42s/it]
 68%|██████▊   | 3058/4476 [19:50:05<9:10:39, 23.30s/it]
 68%|██████▊   | 3059/4476 [19:50:29<9:11:47, 23.36s/it]
 68%|██████▊   | 3060/4476 [19:50:52<9:11:52, 23.38s/it]
                                                        
{'loss': 0.3858, 'learning_rate': 1.1378138241974595e-05, 'epoch': 2.05}

 68%|██████▊   | 3060/4476 [19:50:52<9:11:52, 23.38s/it]
 68%|██████▊   | 3061/4476 [19:51:15<9:10:45, 23.35s/it]
 68%|██████▊   | 3062/4476 [19:51:38<9:08:32, 23.28s/it]
 68%|██████▊   | 3063/4476 [19:52:02<9:10:57, 23.40s/it]
 68%|██████▊   | 3064/4476 [19:52:26<9:12:03, 23.46s/it]
 68%|██████▊   | 3065/4476 [19:52:49<9:10:48, 23.42s/it]
 68%|██████▊   | 3066/4476 [19:53:13<9:11:10, 23.45s/it]
 69%|██████▊   | 3067/4476 [19:53:36<9:09:36, 23.40s/it]
 69%|██████▊   | 3068/4476 [19:53:59<9:06:01, 23.27s/it]
 69%|██████▊   | 3069/4476 [19:54:22<9:05:24, 23.26s/it]
 69%|██████▊   | 3070/4476 [19:54:46<9:06:04, 23.30s/it]
                                                        
{'loss': 0.3766, 'learning_rate': 1.1231341327401323e-05, 'epoch': 2.06}

 69%|██████▊   | 3070/4476 [19:54:46<9:06:04, 23.30s/it]
 69%|██████▊   | 3071/4476 [19:55:09<9:05:17, 23.29s/it]
 69%|██████▊   | 3072/4476 [19:55:32<9:03:04, 23.21s/it]
 69%|██████▊   | 3073/4476 [19:55:55<9:04:09, 23.27s/it]
 69%|██████▊   | 3074/4476 [19:56:19<9:04:38, 23.31s/it]
 69%|██████▊   | 3075/4476 [19:56:42<9:04:10, 23.31s/it]
 69%|██████▊   | 3076/4476 [19:57:05<9:03:42, 23.30s/it]
 69%|██████▊   | 3077/4476 [19:57:28<9:00:48, 23.19s/it]
 69%|██████▉   | 3078/4476 [19:57:52<9:02:05, 23.27s/it]
 69%|██████▉   | 3079/4476 [19:58:15<9:04:40, 23.39s/it]
 69%|██████▉   | 3080/4476 [19:58:39<9:05:06, 23.43s/it]
                                                        
{'loss': 0.3766, 'learning_rate': 1.1085222693498256e-05, 'epoch': 2.06}

 69%|██████▉   | 3080/4476 [19:58:39<9:05:06, 23.43s/it]
 69%|██████▉   | 3081/4476 [19:59:02<9:04:58, 23.44s/it]
 69%|██████▉   | 3082/4476 [19:59:26<9:04:55, 23.45s/it]
 69%|██████▉   | 3083/4476 [19:59:49<9:04:55, 23.47s/it]
 69%|██████▉   | 3084/4476 [20:00:13<9:03:38, 23.43s/it]
 69%|██████▉   | 3085/4476 [20:00:36<9:02:27, 23.40s/it]
 69%|██████▉   | 3086/4476 [20:00:59<9:02:31, 23.42s/it]
 69%|██████▉   | 3087/4476 [20:01:23<9:00:39, 23.35s/it]
 69%|██████▉   | 3088/4476 [20:01:46<8:58:18, 23.27s/it]
 69%|██████▉   | 3089/4476 [20:02:09<8:58:36, 23.30s/it]
 69%|██████▉   | 3090/4476 [20:02:32<8:58:09, 23.30s/it]
                                                        
{'loss': 0.3795, 'learning_rate': 1.093978953845713e-05, 'epoch': 2.07}

 69%|██████▉   | 3090/4476 [20:02:32<8:58:09, 23.30s/it]
 69%|██████▉   | 3091/4476 [20:02:56<8:57:22, 23.28s/it]
 69%|██████▉   | 3092/4476 [20:03:19<8:56:12, 23.25s/it]
 69%|██████▉   | 3093/4476 [20:03:42<8:55:59, 23.25s/it]
 69%|██████▉   | 3094/4476 [20:04:06<8:57:35, 23.34s/it]
 69%|██████▉   | 3095/4476 [20:04:29<8:56:35, 23.31s/it]
 69%|██████▉   | 3096/4476 [20:04:52<8:54:34, 23.24s/it]
 69%|██████▉   | 3097/4476 [20:05:15<8:54:09, 23.24s/it]
 69%|██████▉   | 3098/4476 [20:05:39<8:55:08, 23.30s/it]
 69%|██████▉   | 3099/4476 [20:06:02<8:54:40, 23.30s/it]
 69%|██████▉   | 3100/4476 [20:06:25<8:54:07, 23.29s/it]
                                                        
{'loss': 0.3837, 'learning_rate': 1.079504902670117e-05, 'epoch': 2.08}

 69%|██████▉   | 3100/4476 [20:06:25<8:54:07, 23.29s/it]
 69%|██████▉   | 3101/4476 [20:06:49<8:55:00, 23.35s/it]
 69%|██████▉   | 3102/4476 [20:07:12<8:55:56, 23.40s/it]
 69%|██████▉   | 3103/4476 [20:07:35<8:54:09, 23.34s/it]
 69%|██████▉   | 3104/4476 [20:07:58<8:51:09, 23.23s/it]
 69%|██████▉   | 3105/4476 [20:08:22<8:51:02, 23.24s/it]
 69%|██████▉   | 3106/4476 [20:08:45<8:49:54, 23.21s/it]
 69%|██████▉   | 3107/4476 [20:09:08<8:52:41, 23.35s/it]
 69%|██████▉   | 3108/4476 [20:09:32<8:52:52, 23.37s/it]
 69%|██████▉   | 3109/4476 [20:09:55<8:51:56, 23.35s/it]
 69%|██████▉   | 3110/4476 [20:10:18<8:51:30, 23.35s/it]
                                                        
{'loss': 0.377, 'learning_rate': 1.065100828853213e-05, 'epoch': 2.08}

 69%|██████▉   | 3110/4476 [20:10:18<8:51:30, 23.35s/it]
 70%|██████▉   | 3111/4476 [20:10:42<8:50:26, 23.32s/it]
 70%|██████▉   | 3112/4476 [20:11:05<8:49:05, 23.27s/it]
 70%|██████▉   | 3113/4476 [20:11:28<8:48:50, 23.28s/it]
 70%|██████▉   | 3114/4476 [20:11:51<8:48:57, 23.30s/it]
 70%|██████▉   | 3115/4476 [20:12:15<8:51:07, 23.41s/it]
 70%|██████▉   | 3116/4476 [20:12:38<8:50:19, 23.40s/it]
 70%|██████▉   | 3117/4476 [20:13:02<8:49:53, 23.39s/it]
 70%|██████▉   | 3118/4476 [20:13:25<8:49:38, 23.40s/it]
 70%|██████▉   | 3119/4476 [20:13:48<8:46:45, 23.29s/it]
 70%|██████▉   | 3120/4476 [20:14:12<8:46:07, 23.28s/it]
                                                        
{'loss': 0.3759, 'learning_rate': 1.0507674419779085e-05, 'epoch': 2.09}

 70%|██████▉   | 3120/4476 [20:14:12<8:46:07, 23.28s/it]
 70%|██████▉   | 3121/4476 [20:14:35<8:45:22, 23.26s/it]
 70%|██████▉   | 3122/4476 [20:14:58<8:44:57, 23.26s/it]
 70%|██████▉   | 3123/4476 [20:15:21<8:44:05, 23.24s/it]
 70%|██████▉   | 3124/4476 [20:15:45<8:46:01, 23.34s/it]
 70%|██████▉   | 3125/4476 [20:16:08<8:45:33, 23.34s/it]
 70%|██████▉   | 3126/4476 [20:16:31<8:44:44, 23.32s/it]
 70%|██████▉   | 3127/4476 [20:16:55<8:43:46, 23.30s/it]
 70%|██████▉   | 3128/4476 [20:17:18<8:44:30, 23.35s/it]
 70%|██████▉   | 3129/4476 [20:17:41<8:41:01, 23.21s/it]
 70%|██████▉   | 3130/4476 [20:18:05<8:42:44, 23.30s/it]
                                                        
{'loss': 0.3704, 'learning_rate': 1.0365054481448849e-05, 'epoch': 2.1}

 70%|██████▉   | 3130/4476 [20:18:05<8:42:44, 23.30s/it]
 70%|██████▉   | 3131/4476 [20:18:28<8:41:47, 23.28s/it]
 70%|██████▉   | 3132/4476 [20:18:51<8:42:33, 23.33s/it]
 70%|██████▉   | 3133/4476 [20:19:15<8:44:31, 23.43s/it]
 70%|███████   | 3134/4476 [20:19:39<8:45:52, 23.51s/it]
 70%|███████   | 3135/4476 [20:20:02<8:43:56, 23.44s/it]
 70%|███████   | 3136/4476 [20:20:25<8:42:51, 23.41s/it]
 70%|███████   | 3137/4476 [20:20:49<8:43:57, 23.48s/it]
 70%|███████   | 3138/4476 [20:21:12<8:44:13, 23.51s/it]
 70%|███████   | 3139/4476 [20:21:36<8:42:29, 23.45s/it]
 70%|███████   | 3140/4476 [20:21:59<8:41:02, 23.40s/it]
                                                        
{'loss': 0.3751, 'learning_rate': 1.02231554993781e-05, 'epoch': 2.1}

 70%|███████   | 3140/4476 [20:21:59<8:41:02, 23.40s/it]
 70%|███████   | 3141/4476 [20:22:23<8:42:05, 23.46s/it]
 70%|███████   | 3142/4476 [20:22:46<8:38:43, 23.33s/it]
 70%|███████   | 3143/4476 [20:23:09<8:37:19, 23.29s/it]
 70%|███████   | 3144/4476 [20:23:32<8:38:58, 23.38s/it]
 70%|███████   | 3145/4476 [20:23:56<8:38:13, 23.36s/it]
 70%|███████   | 3146/4476 [20:24:19<8:37:35, 23.35s/it]
 70%|███████   | 3147/4476 [20:24:43<8:39:03, 23.43s/it]
 70%|███████   | 3148/4476 [20:25:06<8:37:42, 23.39s/it]
 70%|███████   | 3149/4476 [20:25:29<8:37:09, 23.38s/it]
 70%|███████   | 3150/4476 [20:25:53<8:37:35, 23.42s/it]
                                                        
{'loss': 0.396, 'learning_rate': 1.0081984463887325e-05, 'epoch': 2.11}

 70%|███████   | 3150/4476 [20:25:53<8:37:35, 23.42s/it]
 70%|███████   | 3151/4476 [20:26:16<8:36:45, 23.40s/it]
 70%|███████   | 3152/4476 [20:26:40<8:38:15, 23.49s/it]
 70%|███████   | 3153/4476 [20:27:03<8:34:36, 23.34s/it]
 70%|███████   | 3154/4476 [20:27:26<8:33:47, 23.32s/it]
 70%|███████   | 3155/4476 [20:27:49<8:32:19, 23.27s/it]
 71%|███████   | 3156/4476 [20:28:13<8:35:05, 23.41s/it]
 71%|███████   | 3157/4476 [20:28:36<8:34:21, 23.40s/it]
 71%|███████   | 3158/4476 [20:29:00<8:33:13, 23.36s/it]
 71%|███████   | 3159/4476 [20:29:23<8:29:09, 23.20s/it]
 71%|███████   | 3160/4476 [20:29:46<8:29:29, 23.23s/it]
                                                        
{'loss': 0.3788, 'learning_rate': 9.941548329436425e-06, 'epoch': 2.12}

 71%|███████   | 3160/4476 [20:29:46<8:29:29, 23.23s/it]
 71%|███████   | 3161/4476 [20:30:09<8:30:17, 23.28s/it]
 71%|███████   | 3162/4476 [20:30:33<8:32:30, 23.40s/it]
 71%|███████   | 3163/4476 [20:30:56<8:31:33, 23.38s/it]
 71%|███████   | 3164/4476 [20:31:20<8:31:43, 23.40s/it]
 71%|███████   | 3165/4476 [20:31:43<8:29:37, 23.32s/it]
 71%|███████   | 3166/4476 [20:32:06<8:30:17, 23.37s/it]
 71%|███████   | 3167/4476 [20:32:30<8:30:07, 23.38s/it]
 71%|███████   | 3168/4476 [20:32:53<8:29:21, 23.37s/it]
 71%|███████   | 3169/4476 [20:33:16<8:25:54, 23.22s/it]
 71%|███████   | 3170/4476 [20:33:39<8:27:21, 23.31s/it]
                                                        
{'loss': 0.3767, 'learning_rate': 9.801854014282108e-06, 'epoch': 2.12}

 71%|███████   | 3170/4476 [20:33:39<8:27:21, 23.31s/it]
 71%|███████   | 3171/4476 [20:34:03<8:26:53, 23.31s/it]
 71%|███████   | 3172/4476 [20:34:26<8:27:45, 23.36s/it]
 71%|███████   | 3173/4476 [20:34:49<8:26:31, 23.32s/it]
 71%|███████   | 3174/4476 [20:35:13<8:24:46, 23.26s/it]
 71%|███████   | 3175/4476 [20:35:36<8:24:45, 23.28s/it]
 71%|███████   | 3176/4476 [20:35:59<8:24:29, 23.28s/it]
 71%|███████   | 3177/4476 [20:36:23<8:24:03, 23.28s/it]
 71%|███████   | 3178/4476 [20:36:46<8:23:45, 23.29s/it]
 71%|███████   | 3179/4476 [20:37:09<8:22:50, 23.26s/it]
 71%|███████   | 3180/4476 [20:37:32<8:21:39, 23.22s/it]
                                                        
{'loss': 0.3783, 'learning_rate': 9.662908400137125e-06, 'epoch': 2.13}

 71%|███████   | 3180/4476 [20:37:32<8:21:39, 23.22s/it]
 71%|███████   | 3181/4476 [20:37:55<8:21:53, 23.25s/it]
 71%|███████   | 3182/4476 [20:38:19<8:22:10, 23.29s/it]
 71%|███████   | 3183/4476 [20:38:42<8:21:43, 23.28s/it]
 71%|███████   | 3184/4476 [20:39:05<8:21:25, 23.29s/it]
 71%|███████   | 3185/4476 [20:39:29<8:21:08, 23.29s/it]
 71%|███████   | 3186/4476 [20:39:52<8:21:29, 23.33s/it]
 71%|███████   | 3187/4476 [20:40:16<8:22:20, 23.38s/it]
 71%|███████   | 3188/4476 [20:40:39<8:21:31, 23.36s/it]
 71%|███████   | 3189/4476 [20:41:02<8:19:15, 23.28s/it]
 71%|███████▏  | 3190/4476 [20:41:25<8:18:15, 23.25s/it]
                                                        
{'loss': 0.3775, 'learning_rate': 9.524718331831186e-06, 'epoch': 2.14}

 71%|███████▏  | 3190/4476 [20:41:25<8:18:15, 23.25s/it]
 71%|███████▏  | 3191/4476 [20:41:49<8:19:42, 23.33s/it]
 71%|███████▏  | 3192/4476 [20:42:12<8:19:09, 23.33s/it]
 71%|███████▏  | 3193/4476 [20:42:35<8:17:20, 23.26s/it]
 71%|███████▏  | 3194/4476 [20:42:59<8:18:20, 23.32s/it]
 71%|███████▏  | 3195/4476 [20:43:22<8:18:21, 23.34s/it]
 71%|███████▏  | 3196/4476 [20:43:45<8:18:51, 23.38s/it]
 71%|███████▏  | 3197/4476 [20:44:09<8:20:23, 23.47s/it]
 71%|███████▏  | 3198/4476 [20:44:32<8:18:39, 23.41s/it]
 71%|███████▏  | 3199/4476 [20:44:56<8:18:12, 23.41s/it]
 71%|███████▏  | 3200/4476 [20:45:19<8:17:10, 23.38s/it]
                                                        
{'loss': 0.3789, 'learning_rate': 9.387290616973859e-06, 'epoch': 2.14}

 71%|███████▏  | 3200/4476 [20:45:19<8:17:10, 23.38s/it][INFO|trainer.py:2939] 2023-11-13 00:09:06,861 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3200
[INFO|tokenization_utils_base.py:2437] 2023-11-13 00:09:06,893 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3200/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 00:09:06,893 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3200/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 00:09:06,893 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3200/added_tokens.json

 72%|███████▏  | 3201/4476 [20:45:42<8:15:08, 23.30s/it]
 72%|███████▏  | 3202/4476 [20:46:06<8:17:58, 23.45s/it]
 72%|███████▏  | 3203/4476 [20:46:29<8:15:11, 23.34s/it]
 72%|███████▏  | 3204/4476 [20:46:53<8:15:59, 23.40s/it]
 72%|███████▏  | 3205/4476 [20:47:16<8:16:37, 23.44s/it]
 72%|███████▏  | 3206/4476 [20:47:40<8:17:49, 23.52s/it]
 72%|███████▏  | 3207/4476 [20:48:03<8:16:57, 23.50s/it]
 72%|███████▏  | 3208/4476 [20:48:27<8:15:23, 23.44s/it]
 72%|███████▏  | 3209/4476 [20:48:50<8:14:13, 23.40s/it]
 72%|███████▏  | 3210/4476 [20:49:13<8:13:01, 23.37s/it]
                                                        
{'loss': 0.3776, 'learning_rate': 9.250632025619104e-06, 'epoch': 2.15}

 72%|███████▏  | 3210/4476 [20:49:13<8:13:01, 23.37s/it]
 72%|███████▏  | 3211/4476 [20:49:36<8:11:27, 23.31s/it]
 72%|███████▏  | 3212/4476 [20:50:00<8:10:54, 23.30s/it]
 72%|███████▏  | 3213/4476 [20:50:23<8:11:21, 23.34s/it]
 72%|███████▏  | 3214/4476 [20:50:46<8:08:50, 23.24s/it]
 72%|███████▏  | 3215/4476 [20:51:10<8:09:49, 23.31s/it]
 72%|███████▏  | 3216/4476 [20:51:33<8:07:58, 23.24s/it]
 72%|███████▏  | 3217/4476 [20:51:56<8:10:45, 23.39s/it]
 72%|███████▏  | 3218/4476 [20:52:19<8:07:56, 23.27s/it]
 72%|███████▏  | 3219/4476 [20:52:43<8:09:25, 23.36s/it]
 72%|███████▏  | 3220/4476 [20:53:06<8:08:30, 23.34s/it]
                                                        
{'loss': 0.368, 'learning_rate': 9.11474928993187e-06, 'epoch': 2.16}

 72%|███████▏  | 3220/4476 [20:53:06<8:08:30, 23.34s/it]
 72%|███████▏  | 3221/4476 [20:53:30<8:07:41, 23.32s/it]
 72%|███████▏  | 3222/4476 [20:53:53<8:07:45, 23.34s/it]
 72%|███████▏  | 3223/4476 [20:54:16<8:06:41, 23.31s/it]
 72%|███████▏  | 3224/4476 [20:54:39<8:04:16, 23.21s/it]
 72%|███████▏  | 3225/4476 [20:55:02<8:04:25, 23.23s/it]
 72%|███████▏  | 3226/4476 [20:55:26<8:05:11, 23.29s/it]
 72%|███████▏  | 3227/4476 [20:55:49<8:03:02, 23.20s/it]
 72%|███████▏  | 3228/4476 [20:56:12<8:03:11, 23.23s/it]
 72%|███████▏  | 3229/4476 [20:56:36<8:03:32, 23.27s/it]
 72%|███████▏  | 3230/4476 [20:56:59<8:04:35, 23.34s/it]
                                                        
{'loss': 0.378, 'learning_rate': 8.979649103856345e-06, 'epoch': 2.16}

 72%|███████▏  | 3230/4476 [20:56:59<8:04:35, 23.34s/it]
 72%|███████▏  | 3231/4476 [20:57:22<8:04:22, 23.34s/it]
 72%|███████▏  | 3232/4476 [20:57:46<8:05:50, 23.43s/it]
 72%|███████▏  | 3233/4476 [20:58:09<8:03:34, 23.34s/it]
 72%|███████▏  | 3234/4476 [20:58:32<8:02:57, 23.33s/it]
 72%|███████▏  | 3235/4476 [20:58:56<8:02:55, 23.35s/it]
 72%|███████▏  | 3236/4476 [20:59:19<8:02:42, 23.36s/it]
 72%|███████▏  | 3237/4476 [20:59:43<8:03:37, 23.42s/it]
 72%|███████▏  | 3238/4476 [21:00:06<8:03:02, 23.41s/it]
 72%|███████▏  | 3239/4476 [21:00:30<8:02:10, 23.39s/it]
 72%|███████▏  | 3240/4476 [21:00:53<8:02:16, 23.41s/it]
                                                        
{'loss': 0.3776, 'learning_rate': 8.84533812278629e-06, 'epoch': 2.17}

 72%|███████▏  | 3240/4476 [21:00:53<8:02:16, 23.41s/it]
 72%|███████▏  | 3241/4476 [21:01:17<8:02:47, 23.46s/it]
 72%|███████▏  | 3242/4476 [21:01:40<8:01:20, 23.40s/it]
 72%|███████▏  | 3243/4476 [21:02:03<8:01:38, 23.44s/it]
 72%|███████▏  | 3244/4476 [21:02:26<7:58:40, 23.31s/it]
 72%|███████▏  | 3245/4476 [21:02:50<7:59:03, 23.35s/it]
 73%|███████▎  | 3246/4476 [21:03:13<7:58:35, 23.35s/it]
 73%|███████▎  | 3247/4476 [21:03:37<7:58:44, 23.37s/it]
 73%|███████▎  | 3248/4476 [21:04:00<7:59:57, 23.45s/it]
 73%|███████▎  | 3249/4476 [21:04:24<7:59:46, 23.46s/it]
 73%|███████▎  | 3250/4476 [21:04:47<7:57:12, 23.35s/it]
                                                        
{'loss': 0.3731, 'learning_rate': 8.711822963237093e-06, 'epoch': 2.18}

 73%|███████▎  | 3250/4476 [21:04:47<7:57:12, 23.35s/it]
 73%|███████▎  | 3251/4476 [21:05:11<7:59:09, 23.47s/it]
 73%|███████▎  | 3252/4476 [21:05:34<7:56:41, 23.37s/it]
 73%|███████▎  | 3253/4476 [21:05:57<7:54:03, 23.26s/it]
 73%|███████▎  | 3254/4476 [21:06:20<7:54:36, 23.30s/it]
 73%|███████▎  | 3255/4476 [21:06:44<7:55:52, 23.38s/it]
 73%|███████▎  | 3256/4476 [21:07:07<7:54:41, 23.35s/it]
 73%|███████▎  | 3257/4476 [21:07:30<7:54:42, 23.37s/it]
 73%|███████▎  | 3258/4476 [21:07:53<7:53:13, 23.31s/it]
 73%|███████▎  | 3259/4476 [21:08:17<7:53:53, 23.36s/it]
 73%|███████▎  | 3260/4476 [21:08:40<7:53:27, 23.36s/it]
                                                        
{'loss': 0.3827, 'learning_rate': 8.579110202519894e-06, 'epoch': 2.18}

 73%|███████▎  | 3260/4476 [21:08:40<7:53:27, 23.36s/it]
 73%|███████▎  | 3261/4476 [21:09:04<7:54:27, 23.43s/it]
 73%|███████▎  | 3262/4476 [21:09:27<7:52:56, 23.37s/it]
 73%|███████▎  | 3263/4476 [21:09:50<7:51:24, 23.32s/it]
 73%|███████▎  | 3264/4476 [21:10:14<7:54:00, 23.47s/it]
 73%|███████▎  | 3265/4476 [21:10:37<7:52:26, 23.41s/it]
 73%|███████▎  | 3266/4476 [21:11:00<7:47:40, 23.19s/it]
 73%|███████▎  | 3267/4476 [21:11:24<7:50:44, 23.36s/it]
 73%|███████▎  | 3268/4476 [21:11:47<7:51:02, 23.40s/it]
 73%|███████▎  | 3269/4476 [21:12:10<7:47:17, 23.23s/it]
 73%|███████▎  | 3270/4476 [21:12:34<7:48:15, 23.30s/it]
                                                        
{'loss': 0.3725, 'learning_rate': 8.447206378417533e-06, 'epoch': 2.19}

 73%|███████▎  | 3270/4476 [21:12:34<7:48:15, 23.30s/it]
 73%|███████▎  | 3271/4476 [21:12:57<7:49:07, 23.36s/it]
 73%|███████▎  | 3272/4476 [21:13:20<7:47:40, 23.31s/it]
 73%|███████▎  | 3273/4476 [21:13:43<7:45:37, 23.22s/it]
 73%|███████▎  | 3274/4476 [21:14:07<7:45:02, 23.21s/it]
 73%|███████▎  | 3275/4476 [21:14:30<7:45:03, 23.23s/it]
 73%|███████▎  | 3276/4476 [21:14:53<7:45:40, 23.28s/it]
 73%|███████▎  | 3277/4476 [21:15:17<7:46:07, 23.33s/it]
 73%|███████▎  | 3278/4476 [21:15:40<7:46:52, 23.38s/it]
 73%|███████▎  | 3279/4476 [21:16:04<7:47:22, 23.43s/it]
 73%|███████▎  | 3280/4476 [21:16:27<7:46:24, 23.40s/it]
                                                        
{'loss': 0.372, 'learning_rate': 8.31611798886246e-06, 'epoch': 2.2}

 73%|███████▎  | 3280/4476 [21:16:27<7:46:24, 23.40s/it]
 73%|███████▎  | 3281/4476 [21:16:50<7:45:19, 23.36s/it]
 73%|███████▎  | 3282/4476 [21:17:14<7:44:24, 23.34s/it]
 73%|███████▎  | 3283/4476 [21:17:37<7:43:24, 23.31s/it]
 73%|███████▎  | 3284/4476 [21:18:00<7:42:28, 23.28s/it]
 73%|███████▎  | 3285/4476 [21:18:24<7:44:54, 23.42s/it]
 73%|███████▎  | 3286/4476 [21:18:48<7:46:43, 23.53s/it]
 73%|███████▎  | 3287/4476 [21:19:11<7:44:32, 23.44s/it]
 73%|███████▎  | 3288/4476 [21:19:34<7:43:49, 23.43s/it]
 73%|███████▎  | 3289/4476 [21:19:58<7:43:00, 23.40s/it]
 74%|███████▎  | 3290/4476 [21:20:21<7:42:24, 23.39s/it]
                                                        
{'loss': 0.3753, 'learning_rate': 8.185851491616677e-06, 'epoch': 2.2}

 74%|███████▎  | 3290/4476 [21:20:21<7:42:24, 23.39s/it]
 74%|███████▎  | 3291/4476 [21:20:45<7:43:15, 23.46s/it]
 74%|███████▎  | 3292/4476 [21:21:08<7:40:39, 23.34s/it]
 74%|███████▎  | 3293/4476 [21:21:31<7:41:36, 23.41s/it]
 74%|███████▎  | 3294/4476 [21:21:55<7:43:20, 23.52s/it]
 74%|███████▎  | 3295/4476 [21:22:19<7:43:48, 23.56s/it]
 74%|███████▎  | 3296/4476 [21:22:42<7:43:12, 23.55s/it]
 74%|███████▎  | 3297/4476 [21:23:06<7:42:38, 23.54s/it]
 74%|███████▎  | 3298/4476 [21:23:29<7:40:52, 23.47s/it]
 74%|███████▎  | 3299/4476 [21:23:52<7:39:25, 23.42s/it]
 74%|███████▎  | 3300/4476 [21:24:16<7:38:56, 23.42s/it]
                                                        
{'loss': 0.3732, 'learning_rate': 8.0564133039536e-06, 'epoch': 2.21}

 74%|███████▎  | 3300/4476 [21:24:16<7:38:56, 23.42s/it]
 74%|███████▎  | 3301/4476 [21:24:39<7:38:54, 23.43s/it]
 74%|███████▍  | 3302/4476 [21:25:02<7:33:10, 23.16s/it]
 74%|███████▍  | 3303/4476 [21:25:25<7:34:34, 23.25s/it]
 74%|███████▍  | 3304/4476 [21:25:49<7:36:11, 23.35s/it]
 74%|███████▍  | 3305/4476 [21:26:12<7:36:51, 23.41s/it]
 74%|███████▍  | 3306/4476 [21:26:36<7:35:38, 23.37s/it]
 74%|███████▍  | 3307/4476 [21:26:59<7:35:56, 23.40s/it]
 74%|███████▍  | 3308/4476 [21:27:23<7:36:35, 23.46s/it]
 74%|███████▍  | 3309/4476 [21:27:46<7:34:38, 23.38s/it]
 74%|███████▍  | 3310/4476 [21:28:09<7:34:29, 23.39s/it]
                                                        
{'loss': 0.37, 'learning_rate': 7.927809802341876e-06, 'epoch': 2.22}

 74%|███████▍  | 3310/4476 [21:28:09<7:34:29, 23.39s/it]
 74%|███████▍  | 3311/4476 [21:28:33<7:34:39, 23.42s/it]
 74%|███████▍  | 3312/4476 [21:28:56<7:33:51, 23.39s/it]
 74%|███████▍  | 3313/4476 [21:29:20<7:33:58, 23.42s/it]
 74%|███████▍  | 3314/4476 [21:29:43<7:34:07, 23.45s/it]
 74%|███████▍  | 3315/4476 [21:30:06<7:32:44, 23.40s/it]
 74%|███████▍  | 3316/4476 [21:30:30<7:32:20, 23.40s/it]
 74%|███████▍  | 3317/4476 [21:30:53<7:32:00, 23.40s/it]
 74%|███████▍  | 3318/4476 [21:31:16<7:30:54, 23.36s/it]
 74%|███████▍  | 3319/4476 [21:31:40<7:30:17, 23.35s/it]
 74%|███████▍  | 3320/4476 [21:32:03<7:30:23, 23.38s/it]
                                                        
{'loss': 0.372, 'learning_rate': 7.800047322131346e-06, 'epoch': 2.22}

 74%|███████▍  | 3320/4476 [21:32:03<7:30:23, 23.38s/it]
 74%|███████▍  | 3321/4476 [21:32:27<7:29:54, 23.37s/it]
 74%|███████▍  | 3322/4476 [21:32:49<7:26:35, 23.22s/it]
 74%|███████▍  | 3323/4476 [21:33:12<7:24:41, 23.14s/it]
 74%|███████▍  | 3324/4476 [21:33:36<7:28:21, 23.35s/it]
 74%|███████▍  | 3325/4476 [21:33:59<7:26:25, 23.27s/it]
 74%|███████▍  | 3326/4476 [21:34:22<7:24:38, 23.20s/it]
 74%|███████▍  | 3327/4476 [21:34:46<7:24:18, 23.20s/it]
 74%|███████▍  | 3328/4476 [21:35:09<7:25:30, 23.28s/it]
 74%|███████▍  | 3329/4476 [21:35:32<7:25:14, 23.29s/it]
 74%|███████▍  | 3330/4476 [21:35:56<7:24:40, 23.28s/it]
                                                        
{'loss': 0.3734, 'learning_rate': 7.673132157240877e-06, 'epoch': 2.23}

 74%|███████▍  | 3330/4476 [21:35:56<7:24:40, 23.28s/it]
 74%|███████▍  | 3331/4476 [21:36:19<7:26:34, 23.40s/it]
 74%|███████▍  | 3332/4476 [21:36:43<7:26:44, 23.43s/it]
 74%|███████▍  | 3333/4476 [21:37:06<7:26:46, 23.45s/it]
 74%|███████▍  | 3334/4476 [21:37:30<7:26:41, 23.47s/it]
 75%|███████▍  | 3335/4476 [21:37:53<7:26:48, 23.50s/it]
 75%|███████▍  | 3336/4476 [21:38:17<7:25:58, 23.47s/it]
 75%|███████▍  | 3337/4476 [21:38:40<7:24:09, 23.40s/it]
 75%|███████▍  | 3338/4476 [21:39:03<7:22:43, 23.34s/it]
 75%|███████▍  | 3339/4476 [21:39:27<7:24:50, 23.47s/it]
 75%|███████▍  | 3340/4476 [21:39:51<7:25:58, 23.56s/it]
                                                        
{'loss': 0.3734, 'learning_rate': 7.5470705598483405e-06, 'epoch': 2.24}

 75%|███████▍  | 3340/4476 [21:39:51<7:25:58, 23.56s/it]
 75%|███████▍  | 3341/4476 [21:40:14<7:24:32, 23.50s/it]
 75%|███████▍  | 3342/4476 [21:40:37<7:21:29, 23.36s/it]
 75%|███████▍  | 3343/4476 [21:41:01<7:22:28, 23.43s/it]
 75%|███████▍  | 3344/4476 [21:41:24<7:22:08, 23.44s/it]
 75%|███████▍  | 3345/4476 [21:41:48<7:21:47, 23.44s/it]
 75%|███████▍  | 3346/4476 [21:42:11<7:21:30, 23.44s/it]
 75%|███████▍  | 3347/4476 [21:42:34<7:20:17, 23.40s/it]
 75%|███████▍  | 3348/4476 [21:42:58<7:19:04, 23.35s/it]
 75%|███████▍  | 3349/4476 [21:43:21<7:21:11, 23.49s/it]
 75%|███████▍  | 3350/4476 [21:43:45<7:21:08, 23.51s/it]
                                                        
{'loss': 0.3784, 'learning_rate': 7.4218687400826075e-06, 'epoch': 2.24}

 75%|███████▍  | 3350/4476 [21:43:45<7:21:08, 23.51s/it]
 75%|███████▍  | 3351/4476 [21:44:09<7:22:24, 23.60s/it]
 75%|███████▍  | 3352/4476 [21:44:32<7:21:11, 23.55s/it]
 75%|███████▍  | 3353/4476 [21:44:56<7:22:03, 23.62s/it]
 75%|███████▍  | 3354/4476 [21:45:19<7:20:23, 23.55s/it]
 75%|███████▍  | 3355/4476 [21:45:43<7:19:34, 23.53s/it]
 75%|███████▍  | 3356/4476 [21:46:06<7:17:34, 23.44s/it]
 75%|███████▌  | 3357/4476 [21:46:30<7:19:22, 23.56s/it]
 75%|███████▌  | 3358/4476 [21:46:53<7:16:42, 23.44s/it]
 75%|███████▌  | 3359/4476 [21:47:16<7:15:36, 23.40s/it]
 75%|███████▌  | 3360/4476 [21:47:40<7:13:55, 23.33s/it]
                                                        
{'loss': 0.3715, 'learning_rate': 7.297532865717638e-06, 'epoch': 2.25}

 75%|███████▌  | 3360/4476 [21:47:40<7:13:55, 23.33s/it]
 75%|███████▌  | 3361/4476 [21:48:03<7:13:15, 23.31s/it]
 75%|███████▌  | 3362/4476 [21:48:27<7:15:16, 23.44s/it]
 75%|███████▌  | 3363/4476 [21:48:50<7:13:30, 23.37s/it]
 75%|███████▌  | 3364/4476 [21:49:13<7:12:36, 23.34s/it]
 75%|███████▌  | 3365/4476 [21:49:36<7:11:38, 23.31s/it]
 75%|███████▌  | 3366/4476 [21:50:00<7:11:48, 23.34s/it]
 75%|███████▌  | 3367/4476 [21:50:23<7:12:19, 23.39s/it]
 75%|███████▌  | 3368/4476 [21:50:46<7:11:06, 23.35s/it]
 75%|███████▌  | 3369/4476 [21:51:10<7:11:18, 23.38s/it]
 75%|███████▌  | 3370/4476 [21:51:33<7:09:56, 23.32s/it]
                                                        
{'loss': 0.3836, 'learning_rate': 7.174069061868591e-06, 'epoch': 2.26}

 75%|███████▌  | 3370/4476 [21:51:33<7:09:56, 23.32s/it]
 75%|███████▌  | 3371/4476 [21:51:57<7:10:38, 23.38s/it]
 75%|███████▌  | 3372/4476 [21:52:20<7:11:18, 23.44s/it]
 75%|███████▌  | 3373/4476 [21:52:44<7:12:34, 23.53s/it]
 75%|███████▌  | 3374/4476 [21:53:07<7:09:03, 23.36s/it]
 75%|███████▌  | 3375/4476 [21:53:30<7:08:12, 23.34s/it]
 75%|███████▌  | 3376/4476 [21:53:53<7:06:07, 23.24s/it]
 75%|███████▌  | 3377/4476 [21:54:16<7:05:12, 23.21s/it]
 75%|███████▌  | 3378/4476 [21:54:40<7:06:44, 23.32s/it]
 75%|███████▌  | 3379/4476 [21:55:03<7:06:15, 23.31s/it]
 76%|███████▌  | 3380/4476 [21:55:26<7:05:13, 23.28s/it]
                                                        
{'loss': 0.3784, 'learning_rate': 7.05148341069014e-06, 'epoch': 2.27}

 76%|███████▌  | 3380/4476 [21:55:26<7:05:13, 23.28s/it]
 76%|███████▌  | 3381/4476 [21:55:49<7:03:55, 23.23s/it]
 76%|███████▌  | 3382/4476 [21:56:13<7:04:56, 23.31s/it]
 76%|███████▌  | 3383/4476 [21:56:36<7:04:37, 23.31s/it]
 76%|███████▌  | 3384/4476 [21:56:59<7:03:19, 23.26s/it]
 76%|███████▌  | 3385/4476 [21:57:23<7:02:06, 23.21s/it]
 76%|███████▌  | 3386/4476 [21:57:46<7:04:44, 23.38s/it]
 76%|███████▌  | 3387/4476 [21:58:09<7:02:57, 23.30s/it]
 76%|███████▌  | 3388/4476 [21:58:33<7:02:58, 23.33s/it]
 76%|███████▌  | 3389/4476 [21:58:56<7:02:26, 23.32s/it]
 76%|███████▌  | 3390/4476 [21:59:19<7:01:14, 23.27s/it]
                                                        
{'loss': 0.3737, 'learning_rate': 6.929781951076836e-06, 'epoch': 2.27}

 76%|███████▌  | 3390/4476 [21:59:19<7:01:14, 23.27s/it]
 76%|███████▌  | 3391/4476 [21:59:43<7:01:36, 23.31s/it]
 76%|███████▌  | 3392/4476 [22:00:06<7:03:19, 23.43s/it]
 76%|███████▌  | 3393/4476 [22:00:29<7:00:57, 23.32s/it]
 76%|███████▌  | 3394/4476 [22:00:53<7:00:52, 23.34s/it]
 76%|███████▌  | 3395/4476 [22:01:16<7:02:10, 23.43s/it]
 76%|███████▌  | 3396/4476 [22:01:39<6:58:48, 23.27s/it]
 76%|███████▌  | 3397/4476 [22:02:03<6:59:04, 23.30s/it]
 76%|███████▌  | 3398/4476 [22:02:26<6:58:46, 23.31s/it]
 76%|███████▌  | 3399/4476 [22:02:49<6:58:43, 23.33s/it]
 76%|███████▌  | 3400/4476 [22:03:13<6:57:35, 23.29s/it]
                                                        
{'loss': 0.3827, 'learning_rate': 6.80897067836557e-06, 'epoch': 2.28}

 76%|███████▌  | 3400/4476 [22:03:13<6:57:35, 23.29s/it][INFO|trainer.py:2939] 2023-11-13 01:27:00,375 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3400
[INFO|tokenization_utils_base.py:2437] 2023-11-13 01:27:00,406 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3400/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 01:27:00,406 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3400/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 01:27:00,406 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3400/added_tokens.json

 76%|███████▌  | 3401/4476 [22:03:36<6:59:48, 23.43s/it]
 76%|███████▌  | 3402/4476 [22:04:00<6:59:31, 23.44s/it]
 76%|███████▌  | 3403/4476 [22:04:23<6:58:24, 23.40s/it]
 76%|███████▌  | 3404/4476 [22:04:47<6:58:44, 23.44s/it]
 76%|███████▌  | 3405/4476 [22:05:10<6:58:53, 23.47s/it]
 76%|███████▌  | 3406/4476 [22:05:34<6:59:25, 23.52s/it]
 76%|███████▌  | 3407/4476 [22:05:57<6:58:43, 23.50s/it]
 76%|███████▌  | 3408/4476 [22:06:21<6:58:13, 23.50s/it]
 76%|███████▌  | 3409/4476 [22:06:44<6:56:08, 23.40s/it]
 76%|███████▌  | 3410/4476 [22:07:08<6:57:37, 23.51s/it]
                                                        
{'loss': 0.3808, 'learning_rate': 6.6890555440403015e-06, 'epoch': 2.29}

 76%|███████▌  | 3410/4476 [22:07:08<6:57:37, 23.51s/it]
 76%|███████▌  | 3411/4476 [22:07:31<6:57:52, 23.54s/it]
 76%|███████▌  | 3412/4476 [22:07:55<6:56:29, 23.49s/it]
 76%|███████▋  | 3413/4476 [22:08:18<6:56:34, 23.51s/it]
 76%|███████▋  | 3414/4476 [22:08:42<6:55:03, 23.45s/it]
 76%|███████▋  | 3415/4476 [22:09:05<6:54:02, 23.41s/it]
 76%|███████▋  | 3416/4476 [22:09:28<6:53:02, 23.38s/it]
 76%|███████▋  | 3417/4476 [22:09:52<6:53:04, 23.40s/it]
 76%|███████▋  | 3418/4476 [22:10:15<6:52:19, 23.38s/it]
 76%|███████▋  | 3419/4476 [22:10:39<6:53:40, 23.48s/it]
 76%|███████▋  | 3420/4476 [22:11:02<6:51:58, 23.41s/it]
                                                        
{'loss': 0.3797, 'learning_rate': 6.570042455438822e-06, 'epoch': 2.29}

 76%|███████▋  | 3420/4476 [22:11:02<6:51:58, 23.41s/it]
 76%|███████▋  | 3421/4476 [22:11:26<6:53:28, 23.52s/it]
 76%|███████▋  | 3422/4476 [22:11:49<6:53:09, 23.52s/it]
 76%|███████▋  | 3423/4476 [22:12:12<6:50:18, 23.38s/it]
 76%|███████▋  | 3424/4476 [22:12:36<6:49:04, 23.33s/it]
 77%|███████▋  | 3425/4476 [22:12:59<6:49:17, 23.37s/it]
 77%|███████▋  | 3426/4476 [22:13:22<6:49:24, 23.40s/it]
 77%|███████▋  | 3427/4476 [22:13:46<6:48:12, 23.35s/it]
 77%|███████▋  | 3428/4476 [22:14:09<6:47:59, 23.36s/it]
 77%|███████▋  | 3429/4476 [22:14:32<6:46:43, 23.31s/it]
 77%|███████▋  | 3430/4476 [22:14:56<6:47:29, 23.37s/it]
                                                        
{'loss': 0.3739, 'learning_rate': 6.451937275461736e-06, 'epoch': 2.3}

 77%|███████▋  | 3430/4476 [22:14:56<6:47:29, 23.37s/it]
 77%|███████▋  | 3431/4476 [22:15:19<6:47:32, 23.40s/it]
 77%|███████▋  | 3432/4476 [22:15:43<6:47:02, 23.39s/it]
 77%|███████▋  | 3433/4476 [22:16:06<6:46:20, 23.38s/it]
 77%|███████▋  | 3434/4476 [22:16:29<6:45:57, 23.38s/it]
 77%|███████▋  | 3435/4476 [22:16:53<6:44:43, 23.33s/it]
 77%|███████▋  | 3436/4476 [22:17:16<6:45:40, 23.40s/it]
 77%|███████▋  | 3437/4476 [22:17:39<6:44:56, 23.38s/it]
 77%|███████▋  | 3438/4476 [22:18:03<6:44:49, 23.40s/it]
 77%|███████▋  | 3439/4476 [22:18:26<6:43:34, 23.35s/it]
 77%|███████▋  | 3440/4476 [22:18:50<6:43:37, 23.38s/it]
                                                        
{'loss': 0.3748, 'learning_rate': 6.334745822283699e-06, 'epoch': 2.31}

 77%|███████▋  | 3440/4476 [22:18:50<6:43:37, 23.38s/it]
 77%|███████▋  | 3441/4476 [22:19:13<6:41:46, 23.29s/it]
 77%|███████▋  | 3442/4476 [22:19:36<6:41:33, 23.30s/it]
 77%|███████▋  | 3443/4476 [22:19:59<6:40:38, 23.27s/it]
 77%|███████▋  | 3444/4476 [22:20:23<6:40:33, 23.29s/it]
 77%|███████▋  | 3445/4476 [22:20:46<6:39:01, 23.22s/it]
 77%|███████▋  | 3446/4476 [22:21:09<6:39:34, 23.28s/it]
 77%|███████▋  | 3447/4476 [22:21:33<6:41:35, 23.42s/it]
 77%|███████▋  | 3448/4476 [22:21:56<6:41:20, 23.42s/it]
 77%|███████▋  | 3449/4476 [22:22:20<6:41:02, 23.43s/it]
 77%|███████▋  | 3450/4476 [22:22:43<6:39:20, 23.35s/it]
                                                        
{'loss': 0.3769, 'learning_rate': 6.2184738690667214e-06, 'epoch': 2.31}

 77%|███████▋  | 3450/4476 [22:22:43<6:39:20, 23.35s/it]
 77%|███████▋  | 3451/4476 [22:23:06<6:38:00, 23.30s/it]
 77%|███████▋  | 3452/4476 [22:23:29<6:37:56, 23.32s/it]
 77%|███████▋  | 3453/4476 [22:23:53<6:38:30, 23.37s/it]
 77%|███████▋  | 3454/4476 [22:24:16<6:37:25, 23.33s/it]
 77%|███████▋  | 3455/4476 [22:24:39<6:37:07, 23.34s/it]
 77%|███████▋  | 3456/4476 [22:25:03<6:36:29, 23.32s/it]
 77%|███████▋  | 3457/4476 [22:25:26<6:38:21, 23.46s/it]
 77%|███████▋  | 3458/4476 [22:25:50<6:39:12, 23.53s/it]
 77%|███████▋  | 3459/4476 [22:26:14<6:39:43, 23.58s/it]
 77%|███████▋  | 3460/4476 [22:26:37<6:38:23, 23.53s/it]
                                                        
{'loss': 0.3756, 'learning_rate': 6.103127143675832e-06, 'epoch': 2.32}

 77%|███████▋  | 3460/4476 [22:26:37<6:38:23, 23.53s/it]
 77%|███████▋  | 3461/4476 [22:27:00<6:35:59, 23.41s/it]
 77%|███████▋  | 3462/4476 [22:27:24<6:33:56, 23.31s/it]
 77%|███████▋  | 3463/4476 [22:27:47<6:34:49, 23.39s/it]
 77%|███████▋  | 3464/4476 [22:28:11<6:34:53, 23.41s/it]
 77%|███████▋  | 3465/4476 [22:28:34<6:33:06, 23.33s/it]
 77%|███████▋  | 3466/4476 [22:28:57<6:31:43, 23.27s/it]
 77%|███████▋  | 3467/4476 [22:29:20<6:32:44, 23.35s/it]
 77%|███████▋  | 3468/4476 [22:29:44<6:34:24, 23.48s/it]
 78%|███████▊  | 3469/4476 [22:30:07<6:31:59, 23.36s/it]
 78%|███████▊  | 3470/4476 [22:30:30<6:30:15, 23.28s/it]
                                                        
{'loss': 0.3738, 'learning_rate': 5.988711328396859e-06, 'epoch': 2.33}

 78%|███████▊  | 3470/4476 [22:30:30<6:30:15, 23.28s/it]
 78%|███████▊  | 3471/4476 [22:30:54<6:30:54, 23.34s/it]
 78%|███████▊  | 3472/4476 [22:31:17<6:29:36, 23.28s/it]
 78%|███████▊  | 3473/4476 [22:31:40<6:29:41, 23.31s/it]
 78%|███████▊  | 3474/4476 [22:32:03<6:28:44, 23.28s/it]
 78%|███████▊  | 3475/4476 [22:32:27<6:28:41, 23.30s/it]
 78%|███████▊  | 3476/4476 [22:32:50<6:27:06, 23.23s/it]
 78%|███████▊  | 3477/4476 [22:33:13<6:26:16, 23.20s/it]
 78%|███████▊  | 3478/4476 [22:33:36<6:25:46, 23.19s/it]
 78%|███████▊  | 3479/4476 [22:33:59<6:25:03, 23.17s/it]
 78%|███████▊  | 3480/4476 [22:34:23<6:26:09, 23.26s/it]
                                                        
{'loss': 0.3676, 'learning_rate': 5.875232059656552e-06, 'epoch': 2.33}

 78%|███████▊  | 3480/4476 [22:34:23<6:26:09, 23.26s/it]
 78%|███████▊  | 3481/4476 [22:34:46<6:26:03, 23.28s/it]
 78%|███████▊  | 3482/4476 [22:35:09<6:25:00, 23.24s/it]
 78%|███████▊  | 3483/4476 [22:35:33<6:26:04, 23.33s/it]
 78%|███████▊  | 3484/4476 [22:35:56<6:24:11, 23.24s/it]
 78%|███████▊  | 3485/4476 [22:36:20<6:26:03, 23.37s/it]
 78%|███████▊  | 3486/4476 [22:36:43<6:25:55, 23.39s/it]
 78%|███████▊  | 3487/4476 [22:37:06<6:25:25, 23.38s/it]
 78%|███████▊  | 3488/4476 [22:37:30<6:24:16, 23.34s/it]
 78%|███████▊  | 3489/4476 [22:37:53<6:22:49, 23.27s/it]
 78%|███████▊  | 3490/4476 [22:38:16<6:22:07, 23.25s/it]
                                                        
{'loss': 0.3737, 'learning_rate': 5.762694927744866e-06, 'epoch': 2.34}

 78%|███████▊  | 3490/4476 [22:38:16<6:22:07, 23.25s/it]
 78%|███████▊  | 3491/4476 [22:38:39<6:22:39, 23.31s/it]
 78%|███████▊  | 3492/4476 [22:39:03<6:22:15, 23.31s/it]
 78%|███████▊  | 3493/4476 [22:39:26<6:22:36, 23.35s/it]
 78%|███████▊  | 3494/4476 [22:39:50<6:22:41, 23.38s/it]
 78%|███████▊  | 3495/4476 [22:40:13<6:20:39, 23.28s/it]
 78%|███████▊  | 3496/4476 [22:40:36<6:20:38, 23.30s/it]
 78%|███████▊  | 3497/4476 [22:40:59<6:19:46, 23.28s/it]
 78%|███████▊  | 3498/4476 [22:41:22<6:18:15, 23.21s/it]
 78%|███████▊  | 3499/4476 [22:41:46<6:18:47, 23.26s/it]
 78%|███████▊  | 3500/4476 [22:42:09<6:18:51, 23.29s/it]
                                                        
{'loss': 0.369, 'learning_rate': 5.651105476539623e-06, 'epoch': 2.35}

 78%|███████▊  | 3500/4476 [22:42:09<6:18:51, 23.29s/it]
 78%|███████▊  | 3501/4476 [22:42:32<6:18:29, 23.29s/it]
 78%|███████▊  | 3502/4476 [22:42:56<6:18:41, 23.33s/it]
 78%|███████▊  | 3503/4476 [22:43:19<6:18:18, 23.33s/it]
 78%|███████▊  | 3504/4476 [22:43:43<6:19:54, 23.45s/it]
 78%|███████▊  | 3505/4476 [22:44:06<6:19:38, 23.46s/it]
 78%|███████▊  | 3506/4476 [22:44:29<6:17:22, 23.34s/it]
 78%|███████▊  | 3507/4476 [22:44:53<6:16:26, 23.31s/it]
 78%|███████▊  | 3508/4476 [22:45:16<6:16:52, 23.36s/it]
 78%|███████▊  | 3509/4476 [22:45:39<6:16:11, 23.34s/it]
 78%|███████▊  | 3510/4476 [22:46:03<6:15:50, 23.34s/it]
                                                        
{'loss': 0.3723, 'learning_rate': 5.540469203233347e-06, 'epoch': 2.35}

 78%|███████▊  | 3510/4476 [22:46:03<6:15:50, 23.34s/it]
 78%|███████▊  | 3511/4476 [22:46:26<6:15:56, 23.37s/it]
 78%|███████▊  | 3512/4476 [22:46:49<6:15:34, 23.38s/it]
 78%|███████▊  | 3513/4476 [22:47:13<6:14:16, 23.32s/it]
 79%|███████▊  | 3514/4476 [22:47:36<6:14:44, 23.37s/it]
 79%|███████▊  | 3515/4476 [22:47:59<6:13:17, 23.31s/it]
 79%|███████▊  | 3516/4476 [22:48:23<6:12:56, 23.31s/it]
 79%|███████▊  | 3517/4476 [22:48:46<6:13:31, 23.37s/it]
 79%|███████▊  | 3518/4476 [22:49:09<6:13:08, 23.37s/it]
 79%|███████▊  | 3519/4476 [22:49:33<6:14:21, 23.47s/it]
 79%|███████▊  | 3520/4476 [22:49:57<6:13:21, 23.43s/it]
                                                        
{'loss': 0.3791, 'learning_rate': 5.430791558062518e-06, 'epoch': 2.36}

 79%|███████▊  | 3520/4476 [22:49:57<6:13:21, 23.43s/it]
 79%|███████▊  | 3521/4476 [22:50:20<6:12:57, 23.43s/it]
 79%|███████▊  | 3522/4476 [22:50:43<6:11:42, 23.38s/it]
 79%|███████▊  | 3523/4476 [22:51:06<6:10:47, 23.35s/it]
 79%|███████▊  | 3524/4476 [22:51:30<6:09:19, 23.28s/it]
 79%|███████▉  | 3525/4476 [22:51:53<6:09:43, 23.33s/it]
 79%|███████▉  | 3526/4476 [22:52:16<6:09:23, 23.33s/it]
 79%|███████▉  | 3527/4476 [22:52:40<6:08:18, 23.29s/it]
 79%|███████▉  | 3528/4476 [22:53:03<6:08:15, 23.31s/it]
 79%|███████▉  | 3529/4476 [22:53:26<6:07:28, 23.28s/it]
 79%|███████▉  | 3530/4476 [22:53:50<6:08:51, 23.39s/it]
                                                        
{'loss': 0.3753, 'learning_rate': 5.322077944039039e-06, 'epoch': 2.37}

 79%|███████▉  | 3530/4476 [22:53:50<6:08:51, 23.39s/it]
 79%|███████▉  | 3531/4476 [22:54:13<6:08:01, 23.37s/it]
 79%|███████▉  | 3532/4476 [22:54:36<6:06:22, 23.29s/it]
 79%|███████▉  | 3533/4476 [22:55:00<6:07:54, 23.41s/it]
 79%|███████▉  | 3534/4476 [22:55:23<6:06:39, 23.35s/it]
 79%|███████▉  | 3535/4476 [22:55:46<6:05:40, 23.32s/it]
 79%|███████▉  | 3536/4476 [22:56:10<6:05:23, 23.32s/it]
 79%|███████▉  | 3537/4476 [22:56:33<6:04:32, 23.29s/it]
 79%|███████▉  | 3538/4476 [22:56:56<6:03:56, 23.28s/it]
 79%|███████▉  | 3539/4476 [22:57:20<6:05:20, 23.39s/it]
 79%|███████▉  | 3540/4476 [22:57:43<6:04:03, 23.34s/it]
                                                        
{'loss': 0.3703, 'learning_rate': 5.21433371668407e-06, 'epoch': 2.37}

 79%|███████▉  | 3540/4476 [22:57:43<6:04:03, 23.34s/it]
 79%|███████▉  | 3541/4476 [22:58:06<6:04:14, 23.37s/it]
 79%|███████▉  | 3542/4476 [22:58:30<6:03:44, 23.37s/it]
 79%|███████▉  | 3543/4476 [22:58:53<6:03:01, 23.35s/it]
 79%|███████▉  | 3544/4476 [22:59:17<6:03:35, 23.41s/it]
 79%|███████▉  | 3545/4476 [22:59:40<6:04:27, 23.49s/it]
 79%|███████▉  | 3546/4476 [23:00:04<6:04:45, 23.53s/it]
 79%|███████▉  | 3547/4476 [23:00:28<6:04:35, 23.55s/it]
 79%|███████▉  | 3548/4476 [23:00:51<6:02:43, 23.45s/it]
 79%|███████▉  | 3549/4476 [23:01:14<6:02:01, 23.43s/it]
 79%|███████▉  | 3550/4476 [23:01:37<5:59:50, 23.32s/it]
                                                        
{'loss': 0.3781, 'learning_rate': 5.107564183764219e-06, 'epoch': 2.38}

 79%|███████▉  | 3550/4476 [23:01:37<5:59:50, 23.32s/it]
 79%|███████▉  | 3551/4476 [23:02:01<5:59:50, 23.34s/it]
 79%|███████▉  | 3552/4476 [23:02:24<6:01:06, 23.45s/it]
 79%|███████▉  | 3553/4476 [23:02:47<5:58:58, 23.34s/it]
 79%|███████▉  | 3554/4476 [23:03:11<5:58:06, 23.30s/it]
 79%|███████▉  | 3555/4476 [23:03:34<5:57:10, 23.27s/it]
 79%|███████▉  | 3556/4476 [23:03:57<5:56:33, 23.25s/it]
 79%|███████▉  | 3557/4476 [23:04:21<5:57:36, 23.35s/it]
 79%|███████▉  | 3558/4476 [23:04:44<5:57:53, 23.39s/it]
 80%|███████▉  | 3559/4476 [23:05:07<5:55:17, 23.25s/it]
 80%|███████▉  | 3560/4476 [23:05:30<5:54:00, 23.19s/it]
                                                        
{'loss': 0.3766, 'learning_rate': 5.001774605030074e-06, 'epoch': 2.39}

 80%|███████▉  | 3560/4476 [23:05:30<5:54:00, 23.19s/it]
 80%|███████▉  | 3561/4476 [23:05:54<5:55:02, 23.28s/it]
 80%|███████▉  | 3562/4476 [23:06:17<5:55:26, 23.33s/it]
 80%|███████▉  | 3563/4476 [23:06:40<5:55:20, 23.35s/it]
 80%|███████▉  | 3564/4476 [23:07:04<5:53:43, 23.27s/it]
 80%|███████▉  | 3565/4476 [23:07:27<5:53:33, 23.29s/it]
 80%|███████▉  | 3566/4476 [23:07:50<5:52:37, 23.25s/it]
 80%|███████▉  | 3567/4476 [23:08:13<5:53:04, 23.31s/it]
 80%|███████▉  | 3568/4476 [23:08:37<5:52:30, 23.29s/it]
 80%|███████▉  | 3569/4476 [23:09:00<5:52:29, 23.32s/it]
 80%|███████▉  | 3570/4476 [23:09:23<5:50:53, 23.24s/it]
                                                        
{'loss': 0.38, 'learning_rate': 4.8969701919570454e-06, 'epoch': 2.39}

 80%|███████▉  | 3570/4476 [23:09:23<5:50:53, 23.24s/it]
 80%|███████▉  | 3571/4476 [23:09:47<5:51:57, 23.33s/it]
 80%|███████▉  | 3572/4476 [23:10:10<5:51:33, 23.33s/it]
 80%|███████▉  | 3573/4476 [23:10:33<5:51:21, 23.35s/it]
 80%|███████▉  | 3574/4476 [23:10:57<5:50:38, 23.32s/it]
 80%|███████▉  | 3575/4476 [23:11:20<5:49:30, 23.27s/it]
 80%|███████▉  | 3576/4476 [23:11:43<5:48:19, 23.22s/it]
 80%|███████▉  | 3577/4476 [23:12:06<5:47:17, 23.18s/it]
 80%|███████▉  | 3578/4476 [23:12:29<5:48:09, 23.26s/it]
 80%|███████▉  | 3579/4476 [23:12:53<5:47:26, 23.24s/it]
 80%|███████▉  | 3580/4476 [23:13:16<5:47:45, 23.29s/it]
                                                        
{'loss': 0.3681, 'learning_rate': 4.7931561074887e-06, 'epoch': 2.4}

 80%|███████▉  | 3580/4476 [23:13:16<5:47:45, 23.29s/it]
 80%|████████  | 3581/4476 [23:13:39<5:47:44, 23.31s/it]
 80%|████████  | 3582/4476 [23:14:03<5:47:23, 23.32s/it]
 80%|████████  | 3583/4476 [23:14:26<5:45:52, 23.24s/it]
 80%|████████  | 3584/4476 [23:14:49<5:46:36, 23.31s/it]
 80%|████████  | 3585/4476 [23:15:12<5:44:47, 23.22s/it]
 80%|████████  | 3586/4476 [23:15:36<5:44:29, 23.22s/it]
 80%|████████  | 3587/4476 [23:15:59<5:44:15, 23.23s/it]
 80%|████████  | 3588/4476 [23:16:22<5:45:57, 23.38s/it]
 80%|████████  | 3589/4476 [23:16:46<5:44:57, 23.33s/it]
 80%|████████  | 3590/4476 [23:17:08<5:41:34, 23.13s/it]
                                                        
{'loss': 0.3752, 'learning_rate': 4.690337465782366e-06, 'epoch': 2.41}

 80%|████████  | 3590/4476 [23:17:08<5:41:34, 23.13s/it]
 80%|████████  | 3591/4476 [23:17:32<5:41:33, 23.16s/it]
 80%|████████  | 3592/4476 [23:17:55<5:40:44, 23.13s/it]
 80%|████████  | 3593/4476 [23:18:18<5:41:48, 23.23s/it]
 80%|████████  | 3594/4476 [23:18:42<5:43:49, 23.39s/it]
 80%|████████  | 3595/4476 [23:19:05<5:44:11, 23.44s/it]
 80%|████████  | 3596/4476 [23:19:29<5:42:56, 23.38s/it]
 80%|████████  | 3597/4476 [23:19:52<5:42:14, 23.36s/it]
 80%|████████  | 3598/4476 [23:20:15<5:40:56, 23.30s/it]
 80%|████████  | 3599/4476 [23:20:39<5:41:00, 23.33s/it]
 80%|████████  | 3600/4476 [23:21:02<5:42:07, 23.43s/it]
                                                        
{'loss': 0.3775, 'learning_rate': 4.588519331957241e-06, 'epoch': 2.41}

 80%|████████  | 3600/4476 [23:21:02<5:42:07, 23.43s/it][INFO|trainer.py:2939] 2023-11-13 02:44:49,960 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3600
[INFO|tokenization_utils_base.py:2437] 2023-11-13 02:44:49,998 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3600/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 02:44:49,998 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3600/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 02:44:49,999 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3600/added_tokens.json

 80%|████████  | 3601/4476 [23:21:26<5:41:52, 23.44s/it]
 80%|████████  | 3602/4476 [23:21:49<5:41:53, 23.47s/it]
 80%|████████  | 3603/4476 [23:22:12<5:40:26, 23.40s/it]
 81%|████████  | 3604/4476 [23:22:36<5:39:25, 23.36s/it]
 81%|████████  | 3605/4476 [23:22:59<5:39:45, 23.41s/it]
 81%|████████  | 3606/4476 [23:23:23<5:39:30, 23.41s/it]
 81%|████████  | 3607/4476 [23:23:46<5:38:45, 23.39s/it]
 81%|████████  | 3608/4476 [23:24:09<5:38:20, 23.39s/it]
 81%|████████  | 3609/4476 [23:24:32<5:36:41, 23.30s/it]
 81%|████████  | 3610/4476 [23:24:56<5:36:03, 23.28s/it]
                                                        
{'loss': 0.3677, 'learning_rate': 4.4877067218448285e-06, 'epoch': 2.42}

 81%|████████  | 3610/4476 [23:24:56<5:36:03, 23.28s/it]
 81%|████████  | 3611/4476 [23:25:19<5:34:57, 23.23s/it]
 81%|████████  | 3612/4476 [23:25:42<5:34:13, 23.21s/it]
 81%|████████  | 3613/4476 [23:26:05<5:34:10, 23.23s/it]
 81%|████████  | 3614/4476 [23:26:29<5:33:58, 23.25s/it]
 81%|████████  | 3615/4476 [23:26:52<5:34:52, 23.34s/it]
 81%|████████  | 3616/4476 [23:27:15<5:32:57, 23.23s/it]
 81%|████████  | 3617/4476 [23:27:38<5:31:24, 23.15s/it]
 81%|████████  | 3618/4476 [23:28:01<5:29:48, 23.06s/it]
 81%|████████  | 3619/4476 [23:28:24<5:29:48, 23.09s/it]
 81%|████████  | 3620/4476 [23:28:47<5:29:12, 23.08s/it]
                                                        
{'loss': 0.3718, 'learning_rate': 4.38790460174188e-06, 'epoch': 2.43}

 81%|████████  | 3620/4476 [23:28:47<5:29:12, 23.08s/it]
 81%|████████  | 3621/4476 [23:29:11<5:30:37, 23.20s/it]
 81%|████████  | 3622/4476 [23:29:34<5:31:30, 23.29s/it]
 81%|████████  | 3623/4476 [23:29:58<5:32:55, 23.42s/it]
 81%|████████  | 3624/4476 [23:30:21<5:31:25, 23.34s/it]
 81%|████████  | 3625/4476 [23:30:44<5:31:00, 23.34s/it]
 81%|████████  | 3626/4476 [23:31:08<5:30:28, 23.33s/it]
 81%|████████  | 3627/4476 [23:31:31<5:28:37, 23.22s/it]
 81%|████████  | 3628/4476 [23:31:54<5:30:06, 23.36s/it]
 81%|████████  | 3629/4476 [23:32:18<5:29:15, 23.32s/it]
 81%|████████  | 3630/4476 [23:32:41<5:29:30, 23.37s/it]
                                                        
{'loss': 0.3671, 'learning_rate': 4.289117888165708e-06, 'epoch': 2.43}

 81%|████████  | 3630/4476 [23:32:41<5:29:30, 23.37s/it]
 81%|████████  | 3631/4476 [23:33:04<5:28:47, 23.35s/it]
 81%|████████  | 3632/4476 [23:33:27<5:27:32, 23.28s/it]
 81%|████████  | 3633/4476 [23:33:51<5:28:47, 23.40s/it]
 81%|████████  | 3634/4476 [23:34:14<5:28:09, 23.38s/it]
 81%|████████  | 3635/4476 [23:34:37<5:25:13, 23.20s/it]
 81%|████████  | 3636/4476 [23:35:01<5:25:24, 23.24s/it]
 81%|████████▏ | 3637/4476 [23:35:24<5:25:23, 23.27s/it]
 81%|████████▏ | 3638/4476 [23:35:47<5:24:42, 23.25s/it]
 81%|████████▏ | 3639/4476 [23:36:10<5:24:06, 23.23s/it]
 81%|████████▏ | 3640/4476 [23:36:33<5:23:24, 23.21s/it]
                                                        
{'loss': 0.3728, 'learning_rate': 4.191351447612032e-06, 'epoch': 2.44}

 81%|████████▏ | 3640/4476 [23:36:33<5:23:24, 23.21s/it]
 81%|████████▏ | 3641/4476 [23:36:56<5:22:23, 23.17s/it]
 81%|████████▏ | 3642/4476 [23:37:20<5:24:39, 23.36s/it]
 81%|████████▏ | 3643/4476 [23:37:44<5:24:10, 23.35s/it]
 81%|████████▏ | 3644/4476 [23:38:07<5:23:36, 23.34s/it]
 81%|████████▏ | 3645/4476 [23:38:31<5:24:35, 23.44s/it]
 81%|████████▏ | 3646/4476 [23:38:54<5:23:05, 23.36s/it]
 81%|████████▏ | 3647/4476 [23:39:17<5:22:35, 23.35s/it]
 82%|████████▏ | 3648/4476 [23:39:40<5:22:01, 23.34s/it]
 82%|████████▏ | 3649/4476 [23:40:04<5:21:09, 23.30s/it]
 82%|████████▏ | 3650/4476 [23:40:27<5:19:43, 23.22s/it]
                                                        
{'loss': 0.3769, 'learning_rate': 4.094610096315199e-06, 'epoch': 2.45}

 82%|████████▏ | 3650/4476 [23:40:27<5:19:43, 23.22s/it]
 82%|████████▏ | 3651/4476 [23:40:50<5:19:53, 23.27s/it]
 82%|████████▏ | 3652/4476 [23:41:14<5:21:40, 23.42s/it]
 82%|████████▏ | 3653/4476 [23:41:37<5:21:11, 23.42s/it]
 82%|████████▏ | 3654/4476 [23:42:01<5:21:09, 23.44s/it]
 82%|████████▏ | 3655/4476 [23:42:24<5:19:44, 23.37s/it]
 82%|████████▏ | 3656/4476 [23:42:47<5:18:39, 23.32s/it]
 82%|████████▏ | 3657/4476 [23:43:10<5:18:17, 23.32s/it]
 82%|████████▏ | 3658/4476 [23:43:34<5:17:31, 23.29s/it]
 82%|████████▏ | 3659/4476 [23:43:57<5:16:02, 23.21s/it]
 82%|████████▏ | 3660/4476 [23:44:20<5:15:30, 23.20s/it]
                                                        
{'loss': 0.3777, 'learning_rate': 3.998898600010928e-06, 'epoch': 2.45}

 82%|████████▏ | 3660/4476 [23:44:20<5:15:30, 23.20s/it]
 82%|████████▏ | 3661/4476 [23:44:43<5:16:21, 23.29s/it]
 82%|████████▏ | 3662/4476 [23:45:07<5:15:45, 23.27s/it]
 82%|████████▏ | 3663/4476 [23:45:30<5:15:46, 23.30s/it]
 82%|████████▏ | 3664/4476 [23:45:53<5:15:58, 23.35s/it]
 82%|████████▏ | 3665/4476 [23:46:17<5:16:19, 23.40s/it]
 82%|████████▏ | 3666/4476 [23:46:40<5:15:29, 23.37s/it]
 82%|████████▏ | 3667/4476 [23:47:04<5:15:26, 23.40s/it]
 82%|████████▏ | 3668/4476 [23:47:27<5:14:34, 23.36s/it]
 82%|████████▏ | 3669/4476 [23:47:51<5:15:05, 23.43s/it]
 82%|████████▏ | 3670/4476 [23:48:14<5:15:50, 23.51s/it]
                                                        
{'loss': 0.3817, 'learning_rate': 3.904221673701566e-06, 'epoch': 2.46}

 82%|████████▏ | 3670/4476 [23:48:14<5:15:50, 23.51s/it]
 82%|████████▏ | 3671/4476 [23:48:38<5:16:16, 23.57s/it]
 82%|████████▏ | 3672/4476 [23:49:01<5:14:38, 23.48s/it]
 82%|████████▏ | 3673/4476 [23:49:25<5:13:25, 23.42s/it]
 82%|████████▏ | 3674/4476 [23:49:48<5:13:27, 23.45s/it]
 82%|████████▏ | 3675/4476 [23:50:11<5:09:43, 23.20s/it]
 82%|████████▏ | 3676/4476 [23:50:34<5:10:37, 23.30s/it]
 82%|████████▏ | 3677/4476 [23:50:57<5:08:28, 23.16s/it]
 82%|████████▏ | 3678/4476 [23:51:21<5:10:09, 23.32s/it]
 82%|████████▏ | 3679/4476 [23:51:44<5:09:43, 23.32s/it]
 82%|████████▏ | 3680/4476 [23:52:07<5:08:36, 23.26s/it]
                                                        
{'loss': 0.383, 'learning_rate': 3.810583981423796e-06, 'epoch': 2.47}

 82%|████████▏ | 3680/4476 [23:52:07<5:08:36, 23.26s/it]
 82%|████████▏ | 3681/4476 [23:52:31<5:09:52, 23.39s/it]
 82%|████████▏ | 3682/4476 [23:52:54<5:09:23, 23.38s/it]
 82%|████████▏ | 3683/4476 [23:53:17<5:07:45, 23.29s/it]
 82%|████████▏ | 3684/4476 [23:53:41<5:07:03, 23.26s/it]
 82%|████████▏ | 3685/4476 [23:54:04<5:06:43, 23.27s/it]
 82%|████████▏ | 3686/4476 [23:54:27<5:07:13, 23.33s/it]
 82%|████████▏ | 3687/4476 [23:54:51<5:06:32, 23.31s/it]
 82%|████████▏ | 3688/4476 [23:55:14<5:07:36, 23.42s/it]
 82%|████████▏ | 3689/4476 [23:55:38<5:08:16, 23.50s/it]
 82%|████████▏ | 3690/4476 [23:56:01<5:07:28, 23.47s/it]
                                                        
{'loss': 0.3719, 'learning_rate': 3.7179901360188533e-06, 'epoch': 2.47}

 82%|████████▏ | 3690/4476 [23:56:01<5:07:28, 23.47s/it]
 82%|████████▏ | 3691/4476 [23:56:24<5:05:29, 23.35s/it]
 82%|████████▏ | 3692/4476 [23:56:48<5:05:40, 23.39s/it]
 83%|████████▎ | 3693/4476 [23:57:11<5:04:52, 23.36s/it]
 83%|████████▎ | 3694/4476 [23:57:34<5:03:58, 23.32s/it]
 83%|████████▎ | 3695/4476 [23:57:58<5:04:28, 23.39s/it]
 83%|████████▎ | 3696/4476 [23:58:21<5:03:28, 23.34s/it]
 83%|████████▎ | 3697/4476 [23:58:45<5:03:43, 23.39s/it]
 83%|████████▎ | 3698/4476 [23:59:08<5:03:41, 23.42s/it]
 83%|████████▎ | 3699/4476 [23:59:31<5:02:50, 23.39s/it]
 83%|████████▎ | 3700/4476 [23:59:55<5:02:37, 23.40s/it]
                                                        
{'loss': 0.3716, 'learning_rate': 3.626444698905329e-06, 'epoch': 2.48}

 83%|████████▎ | 3700/4476 [23:59:55<5:02:37, 23.40s/it]
 83%|████████▎ | 3701/4476 [24:00:18<5:02:50, 23.45s/it]
 83%|████████▎ | 3702/4476 [24:00:42<5:01:51, 23.40s/it]
 83%|████████▎ | 3703/4476 [24:01:05<4:59:51, 23.28s/it]
 83%|████████▎ | 3704/4476 [24:01:28<4:59:36, 23.29s/it]
 83%|████████▎ | 3705/4476 [24:01:51<4:59:15, 23.29s/it]
 83%|████████▎ | 3706/4476 [24:02:14<4:58:20, 23.25s/it]
 83%|████████▎ | 3707/4476 [24:02:38<4:58:23, 23.28s/it]
 83%|████████▎ | 3708/4476 [24:03:01<4:57:57, 23.28s/it]
 83%|████████▎ | 3709/4476 [24:03:24<4:57:56, 23.31s/it]
 83%|████████▎ | 3710/4476 [24:03:48<4:56:38, 23.24s/it]
                                                        
{'loss': 0.3736, 'learning_rate': 3.5359521798544347e-06, 'epoch': 2.49}

 83%|████████▎ | 3710/4476 [24:03:48<4:56:38, 23.24s/it]
 83%|████████▎ | 3711/4476 [24:04:11<4:57:12, 23.31s/it]
 83%|████████▎ | 3712/4476 [24:04:34<4:56:25, 23.28s/it]
 83%|████████▎ | 3713/4476 [24:04:57<4:55:40, 23.25s/it]
 83%|████████▎ | 3714/4476 [24:05:21<4:56:28, 23.34s/it]
 83%|████████▎ | 3715/4476 [24:05:44<4:56:40, 23.39s/it]
 83%|████████▎ | 3716/4476 [24:06:08<4:55:42, 23.35s/it]
 83%|████████▎ | 3717/4476 [24:06:31<4:54:18, 23.27s/it]
 83%|████████▎ | 3718/4476 [24:06:54<4:54:17, 23.30s/it]
 83%|████████▎ | 3719/4476 [24:07:17<4:53:36, 23.27s/it]
 83%|████████▎ | 3720/4476 [24:07:41<4:53:01, 23.26s/it]
                                                        
{'loss': 0.3741, 'learning_rate': 3.4465170367678294e-06, 'epoch': 2.49}

 83%|████████▎ | 3720/4476 [24:07:41<4:53:01, 23.26s/it]
 83%|████████▎ | 3721/4476 [24:08:04<4:52:32, 23.25s/it]
 83%|████████▎ | 3722/4476 [24:08:27<4:53:22, 23.35s/it]
 83%|████████▎ | 3723/4476 [24:08:51<4:52:18, 23.29s/it]
 83%|████████▎ | 3724/4476 [24:09:14<4:53:37, 23.43s/it]
 83%|████████▎ | 3725/4476 [24:09:38<4:53:36, 23.46s/it]
 83%|████████▎ | 3726/4476 [24:10:01<4:52:46, 23.42s/it]
 83%|████████▎ | 3727/4476 [24:10:24<4:51:52, 23.38s/it]
 83%|████████▎ | 3728/4476 [24:10:48<4:51:59, 23.42s/it]
 83%|████████▎ | 3729/4476 [24:11:11<4:51:31, 23.42s/it]
 83%|████████▎ | 3730/4476 [24:11:35<4:50:30, 23.37s/it]
                                                        
{'loss': 0.3756, 'learning_rate': 3.3581436754580363e-06, 'epoch': 2.5}

 83%|████████▎ | 3730/4476 [24:11:35<4:50:30, 23.37s/it]
 83%|████████▎ | 3731/4476 [24:11:58<4:50:21, 23.38s/it]
 83%|████████▎ | 3732/4476 [24:12:21<4:49:01, 23.31s/it]
 83%|████████▎ | 3733/4476 [24:12:44<4:47:15, 23.20s/it]
 83%|████████▎ | 3734/4476 [24:13:07<4:47:19, 23.23s/it]
 83%|████████▎ | 3735/4476 [24:13:31<4:46:33, 23.20s/it]
 83%|████████▎ | 3736/4476 [24:13:54<4:47:04, 23.28s/it]
 83%|████████▎ | 3737/4476 [24:14:17<4:46:43, 23.28s/it]
 84%|████████▎ | 3738/4476 [24:14:41<4:46:41, 23.31s/it]
 84%|████████▎ | 3739/4476 [24:15:04<4:47:04, 23.37s/it]
 84%|████████▎ | 3740/4476 [24:15:28<4:46:38, 23.37s/it]
                                                        
{'loss': 0.3777, 'learning_rate': 3.270836449431397e-06, 'epoch': 2.51}

 84%|████████▎ | 3740/4476 [24:15:28<4:46:38, 23.37s/it]
 84%|████████▎ | 3741/4476 [24:15:51<4:47:41, 23.48s/it]
 84%|████████▎ | 3742/4476 [24:16:15<4:47:41, 23.52s/it]
 84%|████████▎ | 3743/4476 [24:16:38<4:46:40, 23.47s/it]
 84%|████████▎ | 3744/4476 [24:17:02<4:45:55, 23.44s/it]
 84%|████████▎ | 3745/4476 [24:17:25<4:44:30, 23.35s/it]
 84%|████████▎ | 3746/4476 [24:17:48<4:43:47, 23.33s/it]
 84%|████████▎ | 3747/4476 [24:18:11<4:43:36, 23.34s/it]
 84%|████████▎ | 3748/4476 [24:18:35<4:43:58, 23.40s/it]
 84%|████████▍ | 3749/4476 [24:18:59<4:44:37, 23.49s/it]
 84%|████████▍ | 3750/4476 [24:19:22<4:44:23, 23.50s/it]
                                                        
{'loss': 0.3774, 'learning_rate': 3.184599659673579e-06, 'epoch': 2.51}

 84%|████████▍ | 3750/4476 [24:19:22<4:44:23, 23.50s/it]
 84%|████████▍ | 3751/4476 [24:19:46<4:44:05, 23.51s/it]
 84%|████████▍ | 3752/4476 [24:20:10<4:44:46, 23.60s/it]
 84%|████████▍ | 3753/4476 [24:20:33<4:43:55, 23.56s/it]
 84%|████████▍ | 3754/4476 [24:20:57<4:43:21, 23.55s/it]
 84%|████████▍ | 3755/4476 [24:21:20<4:43:36, 23.60s/it]
 84%|████████▍ | 3756/4476 [24:21:44<4:41:58, 23.50s/it]
 84%|████████▍ | 3757/4476 [24:22:07<4:40:54, 23.44s/it]
 84%|████████▍ | 3758/4476 [24:22:30<4:40:07, 23.41s/it]
 84%|████████▍ | 3759/4476 [24:22:54<4:39:36, 23.40s/it]
 84%|████████▍ | 3760/4476 [24:23:17<4:39:22, 23.41s/it]
                                                        
{'loss': 0.3785, 'learning_rate': 3.0994375544377424e-06, 'epoch': 2.52}

 84%|████████▍ | 3760/4476 [24:23:17<4:39:22, 23.41s/it]
 84%|████████▍ | 3761/4476 [24:23:40<4:39:15, 23.43s/it]
 84%|████████▍ | 3762/4476 [24:24:04<4:37:48, 23.35s/it]
 84%|████████▍ | 3763/4476 [24:24:27<4:38:55, 23.47s/it]
 84%|████████▍ | 3764/4476 [24:24:51<4:39:22, 23.54s/it]
 84%|████████▍ | 3765/4476 [24:25:14<4:37:46, 23.44s/it]
 84%|████████▍ | 3766/4476 [24:25:38<4:36:46, 23.39s/it]
 84%|████████▍ | 3767/4476 [24:26:01<4:36:12, 23.37s/it]
 84%|████████▍ | 3768/4476 [24:26:24<4:36:03, 23.40s/it]
 84%|████████▍ | 3769/4476 [24:26:48<4:35:27, 23.38s/it]
 84%|████████▍ | 3770/4476 [24:27:11<4:34:57, 23.37s/it]
                                                        
{'loss': 0.3768, 'learning_rate': 3.0153543290352164e-06, 'epoch': 2.53}

 84%|████████▍ | 3770/4476 [24:27:11<4:34:57, 23.37s/it]
 84%|████████▍ | 3771/4476 [24:27:34<4:34:12, 23.34s/it]
 84%|████████▍ | 3772/4476 [24:27:58<4:34:36, 23.40s/it]
 84%|████████▍ | 3773/4476 [24:28:21<4:33:54, 23.38s/it]
 84%|████████▍ | 3774/4476 [24:28:45<4:33:49, 23.40s/it]
 84%|████████▍ | 3775/4476 [24:29:08<4:34:45, 23.52s/it]
 84%|████████▍ | 3776/4476 [24:29:32<4:33:40, 23.46s/it]
 84%|████████▍ | 3777/4476 [24:29:55<4:32:36, 23.40s/it]
 84%|████████▍ | 3778/4476 [24:30:18<4:31:42, 23.36s/it]
 84%|████████▍ | 3779/4476 [24:30:42<4:31:54, 23.41s/it]
 84%|████████▍ | 3780/4476 [24:31:05<4:30:34, 23.33s/it]
                                                        
{'loss': 0.377, 'learning_rate': 2.932354125628853e-06, 'epoch': 2.53}

 84%|████████▍ | 3780/4476 [24:31:05<4:30:34, 23.33s/it]
 84%|████████▍ | 3781/4476 [24:31:28<4:29:38, 23.28s/it]
 84%|████████▍ | 3782/4476 [24:31:51<4:29:08, 23.27s/it]
 85%|████████▍ | 3783/4476 [24:32:15<4:28:44, 23.27s/it]
 85%|████████▍ | 3784/4476 [24:32:38<4:28:16, 23.26s/it]
 85%|████████▍ | 3785/4476 [24:33:01<4:28:13, 23.29s/it]
 85%|████████▍ | 3786/4476 [24:33:24<4:27:53, 23.29s/it]
 85%|████████▍ | 3787/4476 [24:33:48<4:27:00, 23.25s/it]
 85%|████████▍ | 3788/4476 [24:34:11<4:27:17, 23.31s/it]
 85%|████████▍ | 3789/4476 [24:34:35<4:27:49, 23.39s/it]
 85%|████████▍ | 3790/4476 [24:34:58<4:27:07, 23.36s/it]
                                                        
{'loss': 0.3803, 'learning_rate': 2.8504410330289778e-06, 'epoch': 2.54}

 85%|████████▍ | 3790/4476 [24:34:58<4:27:07, 23.36s/it]
 85%|████████▍ | 3791/4476 [24:35:22<4:28:02, 23.48s/it]
 85%|████████▍ | 3792/4476 [24:35:45<4:26:46, 23.40s/it]
 85%|████████▍ | 3793/4476 [24:36:08<4:25:15, 23.30s/it]
 85%|████████▍ | 3794/4476 [24:36:31<4:23:39, 23.20s/it]
 85%|████████▍ | 3795/4476 [24:36:54<4:23:48, 23.24s/it]
 85%|████████▍ | 3796/4476 [24:37:18<4:23:21, 23.24s/it]
 85%|████████▍ | 3797/4476 [24:37:41<4:23:06, 23.25s/it]
 85%|████████▍ | 3798/4476 [24:38:04<4:23:03, 23.28s/it]
 85%|████████▍ | 3799/4476 [24:38:27<4:22:40, 23.28s/it]
 85%|████████▍ | 3800/4476 [24:38:51<4:22:21, 23.29s/it]
                                                        
{'loss': 0.3706, 'learning_rate': 2.769619086491923e-06, 'epoch': 2.55}

 85%|████████▍ | 3800/4476 [24:38:51<4:22:21, 23.29s/it][INFO|trainer.py:2939] 2023-11-13 04:02:38,481 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3800
[INFO|tokenization_utils_base.py:2437] 2023-11-13 04:02:38,512 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3800/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 04:02:38,512 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3800/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 04:02:38,513 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-3800/added_tokens.json

 85%|████████▍ | 3801/4476 [24:39:14<4:22:12, 23.31s/it]
 85%|████████▍ | 3802/4476 [24:39:38<4:22:30, 23.37s/it]
 85%|████████▍ | 3803/4476 [24:40:01<4:21:32, 23.32s/it]
 85%|████████▍ | 3804/4476 [24:40:24<4:21:07, 23.32s/it]
 85%|████████▌ | 3805/4476 [24:40:47<4:20:18, 23.28s/it]
 85%|████████▌ | 3806/4476 [24:41:11<4:20:12, 23.30s/it]
 85%|████████▌ | 3807/4476 [24:41:34<4:19:54, 23.31s/it]
 85%|████████▌ | 3808/4476 [24:41:57<4:20:02, 23.36s/it]
 85%|████████▌ | 3809/4476 [24:42:21<4:19:35, 23.35s/it]
 85%|████████▌ | 3810/4476 [24:42:45<4:20:21, 23.46s/it]
                                                        
{'loss': 0.3712, 'learning_rate': 2.6898922675213016e-06, 'epoch': 2.55}

 85%|████████▌ | 3810/4476 [24:42:45<4:20:21, 23.46s/it]
 85%|████████▌ | 3811/4476 [24:43:08<4:19:28, 23.41s/it]
 85%|████████▌ | 3812/4476 [24:43:31<4:17:57, 23.31s/it]
 85%|████████▌ | 3813/4476 [24:43:54<4:17:17, 23.28s/it]
 85%|████████▌ | 3814/4476 [24:44:17<4:17:14, 23.32s/it]
 85%|████████▌ | 3815/4476 [24:44:41<4:18:00, 23.42s/it]
 85%|████████▌ | 3816/4476 [24:45:04<4:17:04, 23.37s/it]
 85%|████████▌ | 3817/4476 [24:45:28<4:17:05, 23.41s/it]
 85%|████████▌ | 3818/4476 [24:45:51<4:16:35, 23.40s/it]
 85%|████████▌ | 3819/4476 [24:46:15<4:15:50, 23.36s/it]
 85%|████████▌ | 3820/4476 [24:46:38<4:16:01, 23.42s/it]
                                                        
{'loss': 0.369, 'learning_rate': 2.611264503671823e-06, 'epoch': 2.56}

 85%|████████▌ | 3820/4476 [24:46:38<4:16:01, 23.42s/it]
 85%|████████▌ | 3821/4476 [24:47:01<4:15:18, 23.39s/it]
 85%|████████▌ | 3822/4476 [24:47:25<4:15:33, 23.45s/it]
 85%|████████▌ | 3823/4476 [24:47:48<4:14:13, 23.36s/it]
 85%|████████▌ | 3824/4476 [24:48:11<4:13:33, 23.33s/it]
 85%|████████▌ | 3825/4476 [24:48:35<4:13:10, 23.33s/it]
 85%|████████▌ | 3826/4476 [24:48:58<4:11:27, 23.21s/it]
 86%|████████▌ | 3827/4476 [24:49:21<4:11:35, 23.26s/it]
 86%|████████▌ | 3828/4476 [24:49:45<4:11:51, 23.32s/it]
 86%|████████▌ | 3829/4476 [24:50:08<4:11:24, 23.31s/it]
 86%|████████▌ | 3830/4476 [24:50:31<4:10:57, 23.31s/it]
                                                        
{'loss': 0.3726, 'learning_rate': 2.533739668355814e-06, 'epoch': 2.57}

 86%|████████▌ | 3830/4476 [24:50:31<4:10:57, 23.31s/it]
 86%|████████▌ | 3831/4476 [24:50:54<4:10:36, 23.31s/it]
 86%|████████▌ | 3832/4476 [24:51:18<4:10:13, 23.31s/it]
 86%|████████▌ | 3833/4476 [24:51:41<4:10:06, 23.34s/it]
 86%|████████▌ | 3834/4476 [24:52:05<4:10:58, 23.46s/it]
 86%|████████▌ | 3835/4476 [24:52:28<4:11:01, 23.50s/it]
 86%|████████▌ | 3836/4476 [24:52:52<4:10:45, 23.51s/it]
 86%|████████▌ | 3837/4476 [24:53:15<4:09:35, 23.44s/it]
 86%|████████▌ | 3838/4476 [24:53:38<4:07:57, 23.32s/it]
 86%|████████▌ | 3839/4476 [24:54:01<4:06:45, 23.24s/it]
 86%|████████▌ | 3840/4476 [24:54:25<4:07:16, 23.33s/it]
                                                        
{'loss': 0.3746, 'learning_rate': 2.45732158065243e-06, 'epoch': 2.57}

 86%|████████▌ | 3840/4476 [24:54:25<4:07:16, 23.33s/it]
 86%|████████▌ | 3841/4476 [24:54:48<4:06:40, 23.31s/it]
 86%|████████▌ | 3842/4476 [24:55:11<4:05:22, 23.22s/it]
 86%|████████▌ | 3843/4476 [24:55:35<4:05:39, 23.29s/it]
 86%|████████▌ | 3844/4476 [24:55:58<4:04:52, 23.25s/it]
 86%|████████▌ | 3845/4476 [24:56:21<4:05:01, 23.30s/it]
 86%|████████▌ | 3846/4476 [24:56:44<4:03:49, 23.22s/it]
 86%|████████▌ | 3847/4476 [24:57:08<4:04:00, 23.28s/it]
 86%|████████▌ | 3848/4476 [24:57:31<4:04:08, 23.33s/it]
 86%|████████▌ | 3849/4476 [24:57:55<4:04:06, 23.36s/it]
 86%|████████▌ | 3850/4476 [24:58:18<4:03:17, 23.32s/it]
                                                        
{'loss': 0.3701, 'learning_rate': 2.382014005119501e-06, 'epoch': 2.58}

 86%|████████▌ | 3850/4476 [24:58:18<4:03:17, 23.32s/it]
 86%|████████▌ | 3851/4476 [24:58:41<4:02:17, 23.26s/it]
 86%|████████▌ | 3852/4476 [24:59:04<4:02:35, 23.33s/it]
 86%|████████▌ | 3853/4476 [24:59:28<4:02:05, 23.32s/it]
 86%|████████▌ | 3854/4476 [24:59:51<4:00:45, 23.22s/it]
 86%|████████▌ | 3855/4476 [25:00:14<4:01:28, 23.33s/it]
 86%|████████▌ | 3856/4476 [25:00:38<4:01:44, 23.39s/it]
 86%|████████▌ | 3857/4476 [25:01:01<4:00:46, 23.34s/it]
 86%|████████▌ | 3858/4476 [25:01:25<4:01:42, 23.47s/it]
 86%|████████▌ | 3859/4476 [25:01:48<4:00:26, 23.38s/it]
 86%|████████▌ | 3860/4476 [25:02:11<3:58:34, 23.24s/it]
                                                        
{'loss': 0.38, 'learning_rate': 2.3078206516080695e-06, 'epoch': 2.59}

 86%|████████▌ | 3860/4476 [25:02:11<3:58:34, 23.24s/it]
 86%|████████▋ | 3861/4476 [25:02:34<3:58:39, 23.28s/it]
 86%|████████▋ | 3862/4476 [25:02:58<3:58:12, 23.28s/it]
 86%|████████▋ | 3863/4476 [25:03:21<3:58:26, 23.34s/it]
 86%|████████▋ | 3864/4476 [25:03:45<3:58:50, 23.42s/it]
 86%|████████▋ | 3865/4476 [25:04:08<3:58:02, 23.38s/it]
 86%|████████▋ | 3866/4476 [25:04:31<3:58:09, 23.42s/it]
 86%|████████▋ | 3867/4476 [25:04:55<3:58:24, 23.49s/it]
 86%|████████▋ | 3868/4476 [25:05:18<3:57:24, 23.43s/it]
 86%|████████▋ | 3869/4476 [25:05:42<3:57:08, 23.44s/it]
 86%|████████▋ | 3870/4476 [25:06:05<3:57:10, 23.48s/it]
                                                        
{'loss': 0.3725, 'learning_rate': 2.2347451750796474e-06, 'epoch': 2.59}

 86%|████████▋ | 3870/4476 [25:06:05<3:57:10, 23.48s/it]
 86%|████████▋ | 3871/4476 [25:06:28<3:55:11, 23.32s/it]
 87%|████████▋ | 3872/4476 [25:06:52<3:55:26, 23.39s/it]
 87%|████████▋ | 3873/4476 [25:07:15<3:55:08, 23.40s/it]
 87%|████████▋ | 3874/4476 [25:07:38<3:53:12, 23.24s/it]
 87%|████████▋ | 3875/4476 [25:08:01<3:52:33, 23.22s/it]
 87%|████████▋ | 3876/4476 [25:08:24<3:51:55, 23.19s/it]
 87%|████████▋ | 3877/4476 [25:08:48<3:51:46, 23.22s/it]
 87%|████████▋ | 3878/4476 [25:09:11<3:52:14, 23.30s/it]
 87%|████████▋ | 3879/4476 [25:09:35<3:51:42, 23.29s/it]
 87%|████████▋ | 3880/4476 [25:09:58<3:51:04, 23.26s/it]
                                                        
{'loss': 0.3825, 'learning_rate': 2.1627911754261653e-06, 'epoch': 2.6}

 87%|████████▋ | 3880/4476 [25:09:58<3:51:04, 23.26s/it]
 87%|████████▋ | 3881/4476 [25:10:21<3:50:39, 23.26s/it]
 87%|████████▋ | 3882/4476 [25:10:44<3:50:32, 23.29s/it]
 87%|████████▋ | 3883/4476 [25:11:08<3:50:10, 23.29s/it]
 87%|████████▋ | 3884/4476 [25:11:31<3:50:08, 23.32s/it]
 87%|████████▋ | 3885/4476 [25:11:54<3:49:23, 23.29s/it]
 87%|████████▋ | 3886/4476 [25:12:18<3:49:19, 23.32s/it]
 87%|████████▋ | 3887/4476 [25:12:41<3:48:31, 23.28s/it]
 87%|████████▋ | 3888/4476 [25:13:04<3:48:38, 23.33s/it]
 87%|████████▋ | 3889/4476 [25:13:27<3:47:20, 23.24s/it]
 87%|████████▋ | 3890/4476 [25:13:51<3:47:43, 23.32s/it]
                                                        
{'loss': 0.3791, 'learning_rate': 2.0919621972926156e-06, 'epoch': 2.61}

 87%|████████▋ | 3890/4476 [25:13:51<3:47:43, 23.32s/it]
 87%|████████▋ | 3891/4476 [25:14:14<3:47:53, 23.37s/it]
 87%|████████▋ | 3892/4476 [25:14:38<3:47:09, 23.34s/it]
 87%|████████▋ | 3893/4476 [25:15:01<3:46:18, 23.29s/it]
 87%|████████▋ | 3894/4476 [25:15:24<3:46:11, 23.32s/it]
 87%|████████▋ | 3895/4476 [25:15:48<3:46:15, 23.37s/it]
 87%|████████▋ | 3896/4476 [25:16:11<3:44:55, 23.27s/it]
 87%|████████▋ | 3897/4476 [25:16:34<3:43:52, 23.20s/it]
 87%|████████▋ | 3898/4476 [25:16:57<3:43:58, 23.25s/it]
 87%|████████▋ | 3899/4476 [25:17:20<3:43:53, 23.28s/it]
 87%|████████▋ | 3900/4476 [25:17:44<3:43:53, 23.32s/it]
                                                        
{'loss': 0.3778, 'learning_rate': 2.022261729902458e-06, 'epoch': 2.61}

 87%|████████▋ | 3900/4476 [25:17:44<3:43:53, 23.32s/it]
 87%|████████▋ | 3901/4476 [25:18:07<3:43:18, 23.30s/it]
 87%|████████▋ | 3902/4476 [25:18:31<3:43:38, 23.38s/it]
 87%|████████▋ | 3903/4476 [25:18:54<3:43:36, 23.41s/it]
 87%|████████▋ | 3904/4476 [25:19:18<3:43:24, 23.43s/it]
 87%|████████▋ | 3905/4476 [25:19:41<3:42:24, 23.37s/it]
 87%|████████▋ | 3906/4476 [25:20:04<3:42:53, 23.46s/it]
 87%|████████▋ | 3907/4476 [25:20:28<3:42:03, 23.42s/it]
 87%|████████▋ | 3908/4476 [25:20:51<3:40:36, 23.30s/it]
 87%|████████▋ | 3909/4476 [25:21:14<3:39:54, 23.27s/it]
 87%|████████▋ | 3910/4476 [25:21:38<3:40:46, 23.40s/it]
                                                        
{'loss': 0.3735, 'learning_rate': 1.953693206885715e-06, 'epoch': 2.62}

 87%|████████▋ | 3910/4476 [25:21:38<3:40:46, 23.40s/it]
 87%|████████▋ | 3911/4476 [25:22:01<3:39:26, 23.30s/it]
 87%|████████▋ | 3912/4476 [25:22:24<3:38:47, 23.28s/it]
 87%|████████▋ | 3913/4476 [25:22:47<3:37:55, 23.22s/it]
 87%|████████▋ | 3914/4476 [25:23:11<3:38:10, 23.29s/it]
 87%|████████▋ | 3915/4476 [25:23:34<3:39:05, 23.43s/it]
 87%|████████▋ | 3916/4476 [25:23:58<3:38:45, 23.44s/it]
 88%|████████▊ | 3917/4476 [25:24:21<3:37:51, 23.38s/it]
 88%|████████▊ | 3918/4476 [25:24:44<3:36:17, 23.26s/it]
 88%|████████▊ | 3919/4476 [25:25:07<3:35:53, 23.26s/it]
 88%|████████▊ | 3920/4476 [25:25:31<3:35:35, 23.26s/it]
                                                        
{'loss': 0.3816, 'learning_rate': 1.8862600061098106e-06, 'epoch': 2.63}

 88%|████████▊ | 3920/4476 [25:25:31<3:35:35, 23.26s/it]
 88%|████████▊ | 3921/4476 [25:25:54<3:36:15, 23.38s/it]
 88%|████████▊ | 3922/4476 [25:26:18<3:36:25, 23.44s/it]
 88%|████████▊ | 3923/4476 [25:26:41<3:34:42, 23.29s/it]
 88%|████████▊ | 3924/4476 [25:27:04<3:33:40, 23.23s/it]
 88%|████████▊ | 3925/4476 [25:27:27<3:33:22, 23.24s/it]
 88%|████████▊ | 3926/4476 [25:27:51<3:33:37, 23.30s/it]
 88%|████████▊ | 3927/4476 [25:28:14<3:34:32, 23.45s/it]
 88%|████████▊ | 3928/4476 [25:28:38<3:33:30, 23.38s/it]
 88%|████████▊ | 3929/4476 [25:29:00<3:31:48, 23.23s/it]
 88%|████████▊ | 3930/4476 [25:29:24<3:31:42, 23.26s/it]
                                                        
{'loss': 0.3752, 'learning_rate': 1.8199654495131974e-06, 'epoch': 2.63}

 88%|████████▊ | 3930/4476 [25:29:24<3:31:42, 23.26s/it]
 88%|████████▊ | 3931/4476 [25:29:48<3:32:41, 23.42s/it]
 88%|████████▊ | 3932/4476 [25:30:11<3:32:12, 23.40s/it]
 88%|████████▊ | 3933/4476 [25:30:34<3:31:40, 23.39s/it]
 88%|████████▊ | 3934/4476 [25:30:57<3:30:49, 23.34s/it]
 88%|████████▊ | 3935/4476 [25:31:21<3:29:47, 23.27s/it]
 88%|████████▊ | 3936/4476 [25:31:44<3:30:55, 23.44s/it]
 88%|████████▊ | 3937/4476 [25:32:08<3:30:48, 23.47s/it]
 88%|████████▊ | 3938/4476 [25:32:31<3:30:00, 23.42s/it]
 88%|████████▊ | 3939/4476 [25:32:55<3:29:55, 23.46s/it]
 88%|████████▊ | 3940/4476 [25:33:18<3:28:24, 23.33s/it]
                                                        
{'loss': 0.3739, 'learning_rate': 1.754812802941691e-06, 'epoch': 2.64}

 88%|████████▊ | 3940/4476 [25:33:18<3:28:24, 23.33s/it]
 88%|████████▊ | 3941/4476 [25:33:41<3:28:41, 23.41s/it]
 88%|████████▊ | 3942/4476 [25:34:05<3:27:58, 23.37s/it]
 88%|████████▊ | 3943/4476 [25:34:28<3:27:49, 23.39s/it]
 88%|████████▊ | 3944/4476 [25:34:52<3:28:21, 23.50s/it]
 88%|████████▊ | 3945/4476 [25:35:15<3:27:30, 23.45s/it]
 88%|████████▊ | 3946/4476 [25:35:38<3:26:19, 23.36s/it]
 88%|████████▊ | 3947/4476 [25:36:02<3:26:42, 23.45s/it]
 88%|████████▊ | 3948/4476 [25:36:25<3:26:22, 23.45s/it]
 88%|████████▊ | 3949/4476 [25:36:49<3:25:25, 23.39s/it]
 88%|████████▊ | 3950/4476 [25:37:12<3:24:49, 23.36s/it]
                                                        
{'loss': 0.3745, 'learning_rate': 1.6908052759875836e-06, 'epoch': 2.65}

 88%|████████▊ | 3950/4476 [25:37:12<3:24:49, 23.36s/it]
 88%|████████▊ | 3951/4476 [25:37:36<3:25:23, 23.47s/it]
 88%|████████▊ | 3952/4476 [25:37:59<3:24:30, 23.42s/it]
 88%|████████▊ | 3953/4476 [25:38:22<3:23:53, 23.39s/it]
 88%|████████▊ | 3954/4476 [25:38:46<3:23:18, 23.37s/it]
 88%|████████▊ | 3955/4476 [25:39:09<3:22:44, 23.35s/it]
 88%|████████▊ | 3956/4476 [25:39:32<3:22:26, 23.36s/it]
 88%|████████▊ | 3957/4476 [25:39:56<3:21:57, 23.35s/it]
 88%|████████▊ | 3958/4476 [25:40:19<3:21:55, 23.39s/it]
 88%|████████▊ | 3959/4476 [25:40:43<3:21:58, 23.44s/it]
 88%|████████▊ | 3960/4476 [25:41:06<3:21:54, 23.48s/it]
                                                        
{'loss': 0.3753, 'learning_rate': 1.6279460218315361e-06, 'epoch': 2.65}

 88%|████████▊ | 3960/4476 [25:41:06<3:21:54, 23.48s/it]
 88%|████████▊ | 3961/4476 [25:41:30<3:21:44, 23.50s/it]
 89%|████████▊ | 3962/4476 [25:41:54<3:22:14, 23.61s/it]
 89%|████████▊ | 3963/4476 [25:42:17<3:21:17, 23.54s/it]
 89%|████████▊ | 3964/4476 [25:42:40<3:19:55, 23.43s/it]
 89%|████████▊ | 3965/4476 [25:43:04<3:19:33, 23.43s/it]
 89%|████████▊ | 3966/4476 [25:43:27<3:20:05, 23.54s/it]
 89%|████████▊ | 3967/4476 [25:43:51<3:19:29, 23.52s/it]
 89%|████████▊ | 3968/4476 [25:44:14<3:19:06, 23.52s/it]
 89%|████████▊ | 3969/4476 [25:44:38<3:17:50, 23.41s/it]
 89%|████████▊ | 3970/4476 [25:45:01<3:17:10, 23.38s/it]
                                                        
{'loss': 0.3736, 'learning_rate': 1.5662381370872532e-06, 'epoch': 2.66}

 89%|████████▊ | 3970/4476 [25:45:01<3:17:10, 23.38s/it]
 89%|████████▊ | 3971/4476 [25:45:24<3:16:55, 23.40s/it]
 89%|████████▊ | 3972/4476 [25:45:48<3:16:47, 23.43s/it]
 89%|████████▉ | 3973/4476 [25:46:12<3:17:04, 23.51s/it]
 89%|████████▉ | 3974/4476 [25:46:35<3:16:37, 23.50s/it]
 89%|████████▉ | 3975/4476 [25:46:58<3:15:49, 23.45s/it]
 89%|████████▉ | 3976/4476 [25:47:22<3:16:15, 23.55s/it]
 89%|████████▉ | 3977/4476 [25:47:46<3:15:20, 23.49s/it]
 89%|████████▉ | 3978/4476 [25:48:09<3:14:59, 23.49s/it]
 89%|████████▉ | 3979/4476 [25:48:32<3:13:56, 23.41s/it]
 89%|████████▉ | 3980/4476 [25:48:56<3:13:09, 23.37s/it]
                                                        
{'loss': 0.3755, 'learning_rate': 1.5056846616489124e-06, 'epoch': 2.67}

 89%|████████▉ | 3980/4476 [25:48:56<3:13:09, 23.37s/it]
 89%|████████▉ | 3981/4476 [25:49:19<3:13:43, 23.48s/it]
 89%|████████▉ | 3982/4476 [25:49:43<3:12:45, 23.41s/it]
 89%|████████▉ | 3983/4476 [25:50:06<3:13:13, 23.52s/it]
 89%|████████▉ | 3984/4476 [25:50:30<3:12:27, 23.47s/it]
 89%|████████▉ | 3985/4476 [25:50:53<3:10:46, 23.31s/it]
 89%|████████▉ | 3986/4476 [25:51:16<3:11:18, 23.42s/it]
 89%|████████▉ | 3987/4476 [25:51:39<3:09:55, 23.30s/it]
 89%|████████▉ | 3988/4476 [25:52:03<3:09:55, 23.35s/it]
 89%|████████▉ | 3989/4476 [25:52:26<3:09:39, 23.37s/it]
 89%|████████▉ | 3990/4476 [25:52:50<3:09:52, 23.44s/it]
                                                        
{'loss': 0.3741, 'learning_rate': 1.4462885785414327e-06, 'epoch': 2.67}

 89%|████████▉ | 3990/4476 [25:52:50<3:09:52, 23.44s/it]
 89%|████████▉ | 3991/4476 [25:53:13<3:08:37, 23.34s/it]
 89%|████████▉ | 3992/4476 [25:53:36<3:08:02, 23.31s/it]
 89%|████████▉ | 3993/4476 [25:53:59<3:07:28, 23.29s/it]
 89%|████████▉ | 3994/4476 [25:54:23<3:06:58, 23.27s/it]
 89%|████████▉ | 3995/4476 [25:54:46<3:06:35, 23.28s/it]
 89%|████████▉ | 3996/4476 [25:55:09<3:05:26, 23.18s/it]
 89%|████████▉ | 3997/4476 [25:55:32<3:05:27, 23.23s/it]
 89%|████████▉ | 3998/4476 [25:55:55<3:05:10, 23.24s/it]
 89%|████████▉ | 3999/4476 [25:56:19<3:05:25, 23.32s/it]
 89%|████████▉ | 4000/4476 [25:56:42<3:05:08, 23.34s/it]
                                                        
{'loss': 0.3708, 'learning_rate': 1.3880528137735132e-06, 'epoch': 2.68}

 89%|████████▉ | 4000/4476 [25:56:42<3:05:08, 23.34s/it][INFO|trainer.py:2939] 2023-11-13 05:20:30,061 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4000
[INFO|tokenization_utils_base.py:2437] 2023-11-13 05:20:30,092 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4000/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 05:20:30,092 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4000/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 05:20:30,092 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4000/added_tokens.json

 89%|████████▉ | 4001/4476 [25:57:06<3:04:24, 23.29s/it]
 89%|████████▉ | 4002/4476 [25:57:29<3:05:00, 23.42s/it]
 89%|████████▉ | 4003/4476 [25:57:52<3:03:47, 23.31s/it]
 89%|████████▉ | 4004/4476 [25:58:16<3:03:22, 23.31s/it]
 89%|████████▉ | 4005/4476 [25:58:39<3:02:52, 23.30s/it]
 89%|████████▉ | 4006/4476 [25:59:02<3:02:04, 23.24s/it]
 90%|████████▉ | 4007/4476 [25:59:26<3:02:50, 23.39s/it]
 90%|████████▉ | 4008/4476 [25:59:49<3:01:59, 23.33s/it]
 90%|████████▉ | 4009/4476 [26:00:12<3:01:07, 23.27s/it]
 90%|████████▉ | 4010/4476 [26:00:35<3:00:39, 23.26s/it]
                                                        
{'loss': 0.3703, 'learning_rate': 1.3309802361934936e-06, 'epoch': 2.69}

 90%|████████▉ | 4010/4476 [26:00:35<3:00:39, 23.26s/it]
 90%|████████▉ | 4011/4476 [26:00:59<3:01:28, 23.42s/it]
 90%|████████▉ | 4012/4476 [26:01:22<3:00:57, 23.40s/it]
 90%|████████▉ | 4013/4476 [26:01:46<3:00:46, 23.43s/it]
 90%|████████▉ | 4014/4476 [26:02:09<3:00:00, 23.38s/it]
 90%|████████▉ | 4015/4476 [26:02:32<2:59:19, 23.34s/it]
 90%|████████▉ | 4016/4476 [26:02:56<2:58:51, 23.33s/it]
 90%|████████▉ | 4017/4476 [26:03:19<2:58:19, 23.31s/it]
 90%|████████▉ | 4018/4476 [26:03:43<2:58:58, 23.45s/it]
 90%|████████▉ | 4019/4476 [26:04:06<2:58:14, 23.40s/it]
 90%|████████▉ | 4020/4476 [26:04:29<2:57:14, 23.32s/it]
                                                        
{'loss': 0.3784, 'learning_rate': 1.2750736573480248e-06, 'epoch': 2.69}

 90%|████████▉ | 4020/4476 [26:04:29<2:57:14, 23.32s/it]
 90%|████████▉ | 4021/4476 [26:04:52<2:56:44, 23.31s/it]
 90%|████████▉ | 4022/4476 [26:05:16<2:55:51, 23.24s/it]
 90%|████████▉ | 4023/4476 [26:05:39<2:56:01, 23.31s/it]
 90%|████████▉ | 4024/4476 [26:06:02<2:55:27, 23.29s/it]
 90%|████████▉ | 4025/4476 [26:06:26<2:55:28, 23.34s/it]
 90%|████████▉ | 4026/4476 [26:06:49<2:55:13, 23.36s/it]
 90%|████████▉ | 4027/4476 [26:07:12<2:54:44, 23.35s/it]
 90%|████████▉ | 4028/4476 [26:07:36<2:54:44, 23.40s/it]
 90%|█████████ | 4029/4476 [26:07:59<2:54:32, 23.43s/it]
 90%|█████████ | 4030/4476 [26:08:23<2:53:54, 23.40s/it]
                                                        
{'loss': 0.3785, 'learning_rate': 1.2203358313435609e-06, 'epoch': 2.7}

 90%|█████████ | 4030/4476 [26:08:23<2:53:54, 23.40s/it]
 90%|█████████ | 4031/4476 [26:08:46<2:53:31, 23.40s/it]
 90%|█████████ | 4032/4476 [26:09:09<2:52:39, 23.33s/it]
 90%|█████████ | 4033/4476 [26:09:33<2:52:19, 23.34s/it]
 90%|█████████ | 4034/4476 [26:09:56<2:52:13, 23.38s/it]
 90%|█████████ | 4035/4476 [26:10:20<2:52:14, 23.43s/it]
 90%|█████████ | 4036/4476 [26:10:43<2:51:04, 23.33s/it]
 90%|█████████ | 4037/4476 [26:11:06<2:50:23, 23.29s/it]
 90%|█████████ | 4038/4476 [26:11:29<2:49:41, 23.25s/it]
 90%|█████████ | 4039/4476 [26:11:53<2:50:15, 23.38s/it]
 90%|█████████ | 4040/4476 [26:12:16<2:50:03, 23.40s/it]
                                                        
{'loss': 0.3832, 'learning_rate': 1.1667694547106978e-06, 'epoch': 2.71}

 90%|█████████ | 4040/4476 [26:12:16<2:50:03, 23.40s/it]
 90%|█████████ | 4041/4476 [26:12:40<2:50:03, 23.46s/it]
 90%|█████████ | 4042/4476 [26:13:03<2:49:20, 23.41s/it]
 90%|█████████ | 4043/4476 [26:13:27<2:49:15, 23.45s/it]
 90%|█████████ | 4044/4476 [26:13:50<2:47:53, 23.32s/it]
 90%|█████████ | 4045/4476 [26:14:13<2:47:44, 23.35s/it]
 90%|█████████ | 4046/4476 [26:14:37<2:47:14, 23.34s/it]
 90%|█████████ | 4047/4476 [26:15:00<2:47:40, 23.45s/it]
 90%|█████████ | 4048/4476 [26:15:23<2:46:41, 23.37s/it]
 90%|█████████ | 4049/4476 [26:15:47<2:46:34, 23.41s/it]
 90%|█████████ | 4050/4476 [26:16:10<2:45:31, 23.31s/it]
                                                        
{'loss': 0.3708, 'learning_rate': 1.1143771662713214e-06, 'epoch': 2.71}

 90%|█████████ | 4050/4476 [26:16:10<2:45:31, 23.31s/it]
 91%|█████████ | 4051/4476 [26:16:33<2:44:38, 23.24s/it]
 91%|█████████ | 4052/4476 [26:16:56<2:44:16, 23.25s/it]
 91%|█████████ | 4053/4476 [26:17:19<2:43:25, 23.18s/it]
 91%|█████████ | 4054/4476 [26:17:42<2:42:54, 23.16s/it]
 91%|█████████ | 4055/4476 [26:18:06<2:42:51, 23.21s/it]
 91%|█████████ | 4056/4476 [26:18:29<2:42:43, 23.25s/it]
 91%|█████████ | 4057/4476 [26:18:52<2:41:52, 23.18s/it]
 91%|█████████ | 4058/4476 [26:19:16<2:41:56, 23.24s/it]
 91%|█████████ | 4059/4476 [26:19:39<2:41:52, 23.29s/it]
 91%|█████████ | 4060/4476 [26:20:02<2:41:31, 23.30s/it]
                                                        
{'loss': 0.3777, 'learning_rate': 1.063161547008612e-06, 'epoch': 2.72}

 91%|█████████ | 4060/4476 [26:20:02<2:41:31, 23.30s/it]
 91%|█████████ | 4061/4476 [26:20:25<2:40:47, 23.25s/it]
 91%|█████████ | 4062/4476 [26:20:49<2:40:57, 23.33s/it]
 91%|█████████ | 4063/4476 [26:21:12<2:40:11, 23.27s/it]
 91%|█████████ | 4064/4476 [26:21:35<2:39:57, 23.30s/it]
 91%|█████████ | 4065/4476 [26:21:59<2:39:33, 23.29s/it]
 91%|█████████ | 4066/4476 [26:22:22<2:39:37, 23.36s/it]
 91%|█████████ | 4067/4476 [26:22:46<2:39:35, 23.41s/it]
 91%|█████████ | 4068/4476 [26:23:09<2:39:03, 23.39s/it]
 91%|█████████ | 4069/4476 [26:23:32<2:38:30, 23.37s/it]
 91%|█████████ | 4070/4476 [26:23:56<2:38:46, 23.46s/it]
                                                        
{'loss': 0.375, 'learning_rate': 1.0131251199399089e-06, 'epoch': 2.73}

 91%|█████████ | 4070/4476 [26:23:56<2:38:46, 23.46s/it]
 91%|█████████ | 4071/4476 [26:24:19<2:38:16, 23.45s/it]
 91%|█████████ | 4072/4476 [26:24:43<2:37:05, 23.33s/it]
 91%|█████████ | 4073/4476 [26:25:06<2:37:33, 23.46s/it]
 91%|█████████ | 4074/4476 [26:25:30<2:36:40, 23.38s/it]
 91%|█████████ | 4075/4476 [26:25:53<2:36:46, 23.46s/it]
 91%|█████████ | 4076/4476 [26:26:16<2:36:03, 23.41s/it]
 91%|█████████ | 4077/4476 [26:26:40<2:35:34, 23.39s/it]
 91%|█████████ | 4078/4476 [26:27:03<2:34:38, 23.31s/it]
 91%|█████████ | 4079/4476 [26:27:26<2:34:22, 23.33s/it]
 91%|█████████ | 4080/4476 [26:27:50<2:33:59, 23.33s/it]
                                                        
{'loss': 0.3719, 'learning_rate': 9.642703499924216e-07, 'epoch': 2.73}

 91%|█████████ | 4080/4476 [26:27:50<2:33:59, 23.33s/it]
 91%|█████████ | 4081/4476 [26:28:13<2:34:24, 23.45s/it]
 91%|█████████ | 4082/4476 [26:28:37<2:33:51, 23.43s/it]
 91%|█████████ | 4083/4476 [26:29:00<2:33:08, 23.38s/it]
 91%|█████████ | 4084/4476 [26:29:23<2:32:37, 23.36s/it]
 91%|█████████▏| 4085/4476 [26:29:47<2:31:57, 23.32s/it]
 91%|█████████▏| 4086/4476 [26:30:09<2:30:49, 23.20s/it]
 91%|█████████▏| 4087/4476 [26:30:33<2:31:00, 23.29s/it]
 91%|█████████▏| 4088/4476 [26:30:56<2:30:34, 23.28s/it]
 91%|█████████▏| 4089/4476 [26:31:20<2:30:43, 23.37s/it]
 91%|█████████▏| 4090/4476 [26:31:44<2:31:05, 23.49s/it]
                                                        
{'loss': 0.3776, 'learning_rate': 9.16599643881777e-07, 'epoch': 2.74}

 91%|█████████▏| 4090/4476 [26:31:44<2:31:05, 23.49s/it]
 91%|█████████▏| 4091/4476 [26:32:07<2:31:08, 23.55s/it]
 91%|█████████▏| 4092/4476 [26:32:31<2:30:40, 23.54s/it]
 91%|█████████▏| 4093/4476 [26:32:54<2:29:04, 23.35s/it]
 91%|█████████▏| 4094/4476 [26:33:17<2:28:24, 23.31s/it]
 91%|█████████▏| 4095/4476 [26:33:40<2:27:59, 23.30s/it]
 92%|█████████▏| 4096/4476 [26:34:04<2:27:59, 23.37s/it]
 92%|█████████▏| 4097/4476 [26:34:27<2:27:30, 23.35s/it]
 92%|█████████▏| 4098/4476 [26:34:50<2:26:43, 23.29s/it]
 92%|█████████▏| 4099/4476 [26:35:13<2:26:07, 23.26s/it]
 92%|█████████▏| 4100/4476 [26:35:37<2:26:16, 23.34s/it]
                                                        
{'loss': 0.377, 'learning_rate': 8.701153499934833e-07, 'epoch': 2.75}

 92%|█████████▏| 4100/4476 [26:35:37<2:26:16, 23.34s/it]
 92%|█████████▏| 4101/4476 [26:36:00<2:26:14, 23.40s/it]
 92%|█████████▏| 4102/4476 [26:36:24<2:25:39, 23.37s/it]
 92%|█████████▏| 4103/4476 [26:36:47<2:25:11, 23.35s/it]
 92%|█████████▏| 4104/4476 [26:37:10<2:24:42, 23.34s/it]
 92%|█████████▏| 4105/4476 [26:37:34<2:24:05, 23.30s/it]
 92%|█████████▏| 4106/4476 [26:37:57<2:23:36, 23.29s/it]
 92%|█████████▏| 4107/4476 [26:38:20<2:23:13, 23.29s/it]
 92%|█████████▏| 4108/4476 [26:38:44<2:23:06, 23.33s/it]
 92%|█████████▏| 4109/4476 [26:39:07<2:22:31, 23.30s/it]
 92%|█████████▏| 4110/4476 [26:39:30<2:22:40, 23.39s/it]
                                                        
{'loss': 0.3759, 'learning_rate': 8.248197582672395e-07, 'epoch': 2.75}

 92%|█████████▏| 4110/4476 [26:39:30<2:22:40, 23.39s/it]
 92%|█████████▏| 4111/4476 [26:39:54<2:23:02, 23.51s/it]
 92%|█████████▏| 4112/4476 [26:40:17<2:22:00, 23.41s/it]
 92%|█████████▏| 4113/4476 [26:40:41<2:21:19, 23.36s/it]
 92%|█████████▏| 4114/4476 [26:41:04<2:20:49, 23.34s/it]
 92%|█████████▏| 4115/4476 [26:41:27<2:20:34, 23.36s/it]
 92%|█████████▏| 4116/4476 [26:41:50<2:19:45, 23.29s/it]
 92%|█████████▏| 4117/4476 [26:42:14<2:18:57, 23.22s/it]
 92%|█████████▏| 4118/4476 [26:42:37<2:18:38, 23.24s/it]
 92%|█████████▏| 4119/4476 [26:43:00<2:18:19, 23.25s/it]
 92%|█████████▏| 4120/4476 [26:43:24<2:18:21, 23.32s/it]
                                                        
{'loss': 0.3727, 'learning_rate': 7.807151000841118e-07, 'epoch': 2.76}

 92%|█████████▏| 4120/4476 [26:43:24<2:18:21, 23.32s/it]
 92%|█████████▏| 4121/4476 [26:43:47<2:18:23, 23.39s/it]
 92%|█████████▏| 4122/4476 [26:44:11<2:18:08, 23.41s/it]
 92%|█████████▏| 4123/4476 [26:44:34<2:17:42, 23.41s/it]
 92%|█████████▏| 4124/4476 [26:44:57<2:17:23, 23.42s/it]
 92%|█████████▏| 4125/4476 [26:45:21<2:16:42, 23.37s/it]
 92%|█████████▏| 4126/4476 [26:45:44<2:16:15, 23.36s/it]
 92%|█████████▏| 4127/4476 [26:46:07<2:15:22, 23.27s/it]
 92%|█████████▏| 4128/4476 [26:46:31<2:15:18, 23.33s/it]
 92%|█████████▏| 4129/4476 [26:46:54<2:14:46, 23.30s/it]
 92%|█████████▏| 4130/4476 [26:47:17<2:13:56, 23.23s/it]
                                                        
{'loss': 0.374, 'learning_rate': 7.378035481566181e-07, 'epoch': 2.77}

 92%|█████████▏| 4130/4476 [26:47:17<2:13:56, 23.23s/it]
 92%|█████████▏| 4131/4476 [26:47:40<2:14:00, 23.31s/it]
 92%|█████████▏| 4132/4476 [26:48:04<2:14:22, 23.44s/it]
 92%|█████████▏| 4133/4476 [26:48:27<2:13:31, 23.36s/it]
 92%|█████████▏| 4134/4476 [26:48:51<2:13:41, 23.45s/it]
 92%|█████████▏| 4135/4476 [26:49:14<2:12:40, 23.34s/it]
 92%|█████████▏| 4136/4476 [26:49:37<2:11:59, 23.29s/it]
 92%|█████████▏| 4137/4476 [26:50:00<2:11:30, 23.28s/it]
 92%|█████████▏| 4138/4476 [26:50:23<2:10:41, 23.20s/it]
 92%|█████████▏| 4139/4476 [26:50:47<2:11:06, 23.34s/it]
 92%|█████████▏| 4140/4476 [26:51:11<2:11:20, 23.45s/it]
                                                        
{'loss': 0.3792, 'learning_rate': 6.960872164217064e-07, 'epoch': 2.77}

 92%|█████████▏| 4140/4476 [26:51:11<2:11:20, 23.45s/it]
 93%|█████████▎| 4141/4476 [26:51:34<2:10:34, 23.39s/it]
 93%|█████████▎| 4142/4476 [26:51:57<2:09:48, 23.32s/it]
 93%|█████████▎| 4143/4476 [26:52:20<2:08:58, 23.24s/it]
 93%|█████████▎| 4144/4476 [26:52:44<2:08:55, 23.30s/it]
 93%|█████████▎| 4145/4476 [26:53:07<2:08:39, 23.32s/it]
 93%|█████████▎| 4146/4476 [26:53:30<2:08:14, 23.32s/it]
 93%|█████████▎| 4147/4476 [26:53:54<2:08:10, 23.37s/it]
 93%|█████████▎| 4148/4476 [26:54:17<2:07:41, 23.36s/it]
 93%|█████████▎| 4149/4476 [26:54:40<2:07:13, 23.34s/it]
 93%|█████████▎| 4150/4476 [26:55:04<2:06:35, 23.30s/it]
                                                        
{'loss': 0.3692, 'learning_rate': 6.555681599365926e-07, 'epoch': 2.78}

 93%|█████████▎| 4150/4476 [26:55:04<2:06:35, 23.30s/it]
 93%|█████████▎| 4151/4476 [26:55:27<2:06:16, 23.31s/it]
 93%|█████████▎| 4152/4476 [26:55:50<2:05:57, 23.33s/it]
 93%|█████████▎| 4153/4476 [26:56:14<2:05:54, 23.39s/it]
 93%|█████████▎| 4154/4476 [26:56:37<2:05:27, 23.38s/it]
 93%|█████████▎| 4155/4476 [26:57:01<2:04:50, 23.34s/it]
 93%|█████████▎| 4156/4476 [26:57:24<2:04:26, 23.33s/it]
 93%|█████████▎| 4157/4476 [26:57:47<2:04:04, 23.34s/it]
 93%|█████████▎| 4158/4476 [26:58:10<2:03:24, 23.29s/it]
 93%|█████████▎| 4159/4476 [26:58:34<2:02:57, 23.27s/it]
 93%|█████████▎| 4160/4476 [26:58:57<2:03:23, 23.43s/it]
                                                        
{'loss': 0.3736, 'learning_rate': 6.16248374777545e-07, 'epoch': 2.79}

 93%|█████████▎| 4160/4476 [26:58:57<2:03:23, 23.43s/it]
 93%|█████████▎| 4161/4476 [26:59:21<2:03:07, 23.45s/it]
 93%|█████████▎| 4162/4476 [26:59:44<2:02:45, 23.46s/it]
 93%|█████████▎| 4163/4476 [27:00:08<2:02:02, 23.40s/it]
 93%|█████████▎| 4164/4476 [27:00:31<2:01:31, 23.37s/it]
 93%|█████████▎| 4165/4476 [27:00:54<2:01:04, 23.36s/it]
 93%|█████████▎| 4166/4476 [27:01:18<2:00:44, 23.37s/it]
 93%|█████████▎| 4167/4476 [27:01:41<2:00:01, 23.31s/it]
 93%|█████████▎| 4168/4476 [27:02:04<2:00:05, 23.40s/it]
 93%|█████████▎| 4169/4476 [27:02:28<1:59:19, 23.32s/it]
 93%|█████████▎| 4170/4476 [27:02:51<1:58:40, 23.27s/it]
                                                        
{'loss': 0.3695, 'learning_rate': 5.781297979415456e-07, 'epoch': 2.79}

 93%|█████████▎| 4170/4476 [27:02:51<1:58:40, 23.27s/it]
 93%|█████████▎| 4171/4476 [27:03:14<1:58:24, 23.29s/it]
 93%|█████████▎| 4172/4476 [27:03:37<1:57:23, 23.17s/it]
 93%|█████████▎| 4173/4476 [27:04:00<1:56:46, 23.13s/it]
 93%|█████████▎| 4174/4476 [27:04:23<1:56:55, 23.23s/it]
 93%|█████████▎| 4175/4476 [27:04:47<1:56:40, 23.26s/it]
 93%|█████████▎| 4176/4476 [27:05:10<1:56:04, 23.22s/it]
 93%|█████████▎| 4177/4476 [27:05:34<1:56:25, 23.36s/it]
 93%|█████████▎| 4178/4476 [27:05:57<1:55:44, 23.31s/it]
 93%|█████████▎| 4179/4476 [27:06:20<1:55:25, 23.32s/it]
 93%|█████████▎| 4180/4476 [27:06:44<1:55:22, 23.39s/it]
                                                        
{'loss': 0.3716, 'learning_rate': 5.412143072508563e-07, 'epoch': 2.8}

 93%|█████████▎| 4180/4476 [27:06:44<1:55:22, 23.39s/it]
 93%|█████████▎| 4181/4476 [27:07:07<1:55:36, 23.52s/it]
 93%|█████████▎| 4182/4476 [27:07:31<1:54:31, 23.37s/it]
 93%|█████████▎| 4183/4476 [27:07:54<1:53:54, 23.32s/it]
 93%|█████████▎| 4184/4476 [27:08:17<1:53:45, 23.38s/it]
 93%|█████████▎| 4185/4476 [27:08:41<1:53:29, 23.40s/it]
 94%|█████████▎| 4186/4476 [27:09:04<1:53:15, 23.43s/it]
 94%|█████████▎| 4187/4476 [27:09:27<1:52:42, 23.40s/it]
 94%|█████████▎| 4188/4476 [27:09:51<1:52:48, 23.50s/it]
 94%|█████████▎| 4189/4476 [27:10:15<1:52:26, 23.51s/it]
 94%|█████████▎| 4190/4476 [27:10:38<1:52:00, 23.50s/it]
                                                        
{'loss': 0.364, 'learning_rate': 5.055037212605279e-07, 'epoch': 2.81}

 94%|█████████▎| 4190/4476 [27:10:38<1:52:00, 23.50s/it]
 94%|█████████▎| 4191/4476 [27:11:02<1:51:19, 23.44s/it]
 94%|█████████▎| 4192/4476 [27:11:25<1:51:17, 23.51s/it]
 94%|█████████▎| 4193/4476 [27:11:49<1:51:13, 23.58s/it]
 94%|█████████▎| 4194/4476 [27:12:13<1:51:03, 23.63s/it]
 94%|█████████▎| 4195/4476 [27:12:36<1:50:25, 23.58s/it]
 94%|█████████▎| 4196/4476 [27:13:00<1:49:46, 23.52s/it]
 94%|█████████▍| 4197/4476 [27:13:23<1:48:52, 23.42s/it]
 94%|█████████▍| 4198/4476 [27:13:46<1:48:45, 23.47s/it]
 94%|█████████▍| 4199/4476 [27:14:10<1:48:36, 23.52s/it]
 94%|█████████▍| 4200/4476 [27:14:33<1:47:40, 23.41s/it]
                                                        
{'loss': 0.3707, 'learning_rate': 4.709997991688114e-07, 'epoch': 2.81}

 94%|█████████▍| 4200/4476 [27:14:33<1:47:40, 23.41s/it][INFO|trainer.py:2939] 2023-11-13 06:38:20,840 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4200
[INFO|tokenization_utils_base.py:2437] 2023-11-13 06:38:20,872 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4200/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 06:38:20,872 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4200/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 06:38:20,872 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4200/added_tokens.json

 94%|█████████▍| 4201/4476 [27:14:57<1:47:18, 23.41s/it]
 94%|█████████▍| 4202/4476 [27:15:20<1:46:58, 23.42s/it]
 94%|█████████▍| 4203/4476 [27:15:43<1:46:40, 23.45s/it]
 94%|█████████▍| 4204/4476 [27:16:07<1:46:23, 23.47s/it]
 94%|█████████▍| 4205/4476 [27:16:31<1:46:10, 23.51s/it]
 94%|█████████▍| 4206/4476 [27:16:54<1:45:29, 23.44s/it]
 94%|█████████▍| 4207/4476 [27:17:17<1:44:55, 23.40s/it]
 94%|█████████▍| 4208/4476 [27:17:40<1:44:16, 23.35s/it]
 94%|█████████▍| 4209/4476 [27:18:04<1:43:58, 23.37s/it]
 94%|█████████▍| 4210/4476 [27:18:27<1:43:27, 23.34s/it]
                                                        
{'loss': 0.3833, 'learning_rate': 4.377042407304827e-07, 'epoch': 2.82}

 94%|█████████▍| 4210/4476 [27:18:27<1:43:27, 23.34s/it]
 94%|█████████▍| 4211/4476 [27:18:50<1:43:03, 23.33s/it]
 94%|█████████▍| 4212/4476 [27:19:13<1:42:19, 23.26s/it]
 94%|█████████▍| 4213/4476 [27:19:37<1:42:22, 23.35s/it]
 94%|█████████▍| 4214/4476 [27:20:00<1:42:00, 23.36s/it]
 94%|█████████▍| 4215/4476 [27:20:24<1:41:36, 23.36s/it]
 94%|█████████▍| 4216/4476 [27:20:48<1:41:53, 23.51s/it]
 94%|█████████▍| 4217/4476 [27:21:11<1:41:35, 23.53s/it]
 94%|█████████▍| 4218/4476 [27:21:35<1:40:59, 23.49s/it]
 94%|█████████▍| 4219/4476 [27:21:58<1:40:24, 23.44s/it]
 94%|█████████▍| 4220/4476 [27:22:21<1:39:44, 23.38s/it]
                                                        
{'loss': 0.3791, 'learning_rate': 4.0561868617312316e-07, 'epoch': 2.83}

 94%|█████████▍| 4220/4476 [27:22:21<1:39:44, 23.38s/it]
 94%|█████████▍| 4221/4476 [27:22:44<1:39:13, 23.35s/it]
 94%|█████████▍| 4222/4476 [27:23:08<1:38:28, 23.26s/it]
 94%|█████████▍| 4223/4476 [27:23:31<1:38:01, 23.25s/it]
 94%|█████████▍| 4224/4476 [27:23:54<1:37:46, 23.28s/it]
 94%|█████████▍| 4225/4476 [27:24:18<1:37:35, 23.33s/it]
 94%|█████████▍| 4226/4476 [27:24:41<1:37:22, 23.37s/it]
 94%|█████████▍| 4227/4476 [27:25:04<1:37:03, 23.39s/it]
 94%|█████████▍| 4228/4476 [27:25:28<1:36:39, 23.39s/it]
 94%|█████████▍| 4229/4476 [27:25:51<1:36:05, 23.34s/it]
 95%|█████████▍| 4230/4476 [27:26:14<1:35:34, 23.31s/it]
                                                        
{'loss': 0.3743, 'learning_rate': 3.747447161163126e-07, 'epoch': 2.83}

 95%|█████████▍| 4230/4476 [27:26:14<1:35:34, 23.31s/it]
 95%|█████████▍| 4231/4476 [27:26:38<1:35:28, 23.38s/it]
 95%|█████████▍| 4232/4476 [27:27:01<1:34:57, 23.35s/it]
 95%|█████████▍| 4233/4476 [27:27:25<1:34:40, 23.38s/it]
 95%|█████████▍| 4234/4476 [27:27:48<1:34:25, 23.41s/it]
 95%|█████████▍| 4235/4476 [27:28:11<1:33:27, 23.27s/it]
 95%|█████████▍| 4236/4476 [27:28:34<1:33:08, 23.28s/it]
 95%|█████████▍| 4237/4476 [27:28:58<1:32:46, 23.29s/it]
 95%|█████████▍| 4238/4476 [27:29:21<1:32:23, 23.29s/it]
 95%|█████████▍| 4239/4476 [27:29:44<1:32:02, 23.30s/it]
 95%|█████████▍| 4240/4476 [27:30:08<1:31:51, 23.36s/it]
                                                        
{'loss': 0.3759, 'learning_rate': 3.4508385149375764e-07, 'epoch': 2.84}

 95%|█████████▍| 4240/4476 [27:30:08<1:31:51, 23.36s/it]
 95%|█████████▍| 4241/4476 [27:30:31<1:30:59, 23.23s/it]
 95%|█████████▍| 4242/4476 [27:30:54<1:31:13, 23.39s/it]
 95%|█████████▍| 4243/4476 [27:31:17<1:30:22, 23.27s/it]
 95%|█████████▍| 4244/4476 [27:31:41<1:30:00, 23.28s/it]
 95%|█████████▍| 4245/4476 [27:32:04<1:29:52, 23.34s/it]
 95%|█████████▍| 4246/4476 [27:32:27<1:29:06, 23.25s/it]
 95%|█████████▍| 4247/4476 [27:32:51<1:28:49, 23.27s/it]
 95%|█████████▍| 4248/4476 [27:33:14<1:28:12, 23.21s/it]
 95%|█████████▍| 4249/4476 [27:33:37<1:27:34, 23.15s/it]
 95%|█████████▍| 4250/4476 [27:34:00<1:27:18, 23.18s/it]
                                                        
{'loss': 0.3667, 'learning_rate': 3.166375534783717e-07, 'epoch': 2.85}

 95%|█████████▍| 4250/4476 [27:34:00<1:27:18, 23.18s/it]
 95%|█████████▍| 4251/4476 [27:34:23<1:26:44, 23.13s/it]
 95%|█████████▍| 4252/4476 [27:34:46<1:26:16, 23.11s/it]
 95%|█████████▌| 4253/4476 [27:35:09<1:26:07, 23.17s/it]
 95%|█████████▌| 4254/4476 [27:35:33<1:25:55, 23.22s/it]
 95%|█████████▌| 4255/4476 [27:35:56<1:25:32, 23.22s/it]
 95%|█████████▌| 4256/4476 [27:36:19<1:25:18, 23.27s/it]
 95%|█████████▌| 4257/4476 [27:36:43<1:24:58, 23.28s/it]
 95%|█████████▌| 4258/4476 [27:37:06<1:25:09, 23.44s/it]
 95%|█████████▌| 4259/4476 [27:37:30<1:24:39, 23.41s/it]
 95%|█████████▌| 4260/4476 [27:37:53<1:23:55, 23.31s/it]
                                                        
{'loss': 0.3725, 'learning_rate': 2.8940722341030126e-07, 'epoch': 2.85}

 95%|█████████▌| 4260/4476 [27:37:53<1:23:55, 23.31s/it]
 95%|█████████▌| 4261/4476 [27:38:16<1:23:46, 23.38s/it]
 95%|█████████▌| 4262/4476 [27:38:40<1:23:16, 23.35s/it]
 95%|█████████▌| 4263/4476 [27:39:03<1:23:19, 23.47s/it]
 95%|█████████▌| 4264/4476 [27:39:27<1:22:36, 23.38s/it]
 95%|█████████▌| 4265/4476 [27:39:50<1:22:22, 23.42s/it]
 95%|█████████▌| 4266/4476 [27:40:13<1:21:48, 23.38s/it]
 95%|█████████▌| 4267/4476 [27:40:37<1:21:16, 23.33s/it]
 95%|█████████▌| 4268/4476 [27:41:00<1:21:13, 23.43s/it]
 95%|█████████▌| 4269/4476 [27:41:24<1:20:49, 23.43s/it]
 95%|█████████▌| 4270/4476 [27:41:47<1:20:19, 23.40s/it]
                                                        
{'loss': 0.3733, 'learning_rate': 2.6339420272787074e-07, 'epoch': 2.86}

 95%|█████████▌| 4270/4476 [27:41:47<1:20:19, 23.40s/it]
 95%|█████████▌| 4271/4476 [27:42:10<1:19:43, 23.33s/it]
 95%|█████████▌| 4272/4476 [27:42:34<1:19:41, 23.44s/it]
 95%|█████████▌| 4273/4476 [27:42:57<1:19:23, 23.47s/it]
 95%|█████████▌| 4274/4476 [27:43:20<1:18:41, 23.38s/it]
 96%|█████████▌| 4275/4476 [27:43:44<1:18:40, 23.48s/it]
 96%|█████████▌| 4276/4476 [27:44:08<1:18:09, 23.45s/it]
 96%|█████████▌| 4277/4476 [27:44:31<1:17:37, 23.41s/it]
 96%|█████████▌| 4278/4476 [27:44:54<1:17:02, 23.34s/it]
 96%|█████████▌| 4279/4476 [27:45:18<1:16:56, 23.44s/it]
 96%|█████████▌| 4280/4476 [27:45:41<1:16:44, 23.49s/it]
                                                        
{'loss': 0.373, 'learning_rate': 2.3859977290152935e-07, 'epoch': 2.87}

 96%|█████████▌| 4280/4476 [27:45:41<1:16:44, 23.49s/it]
 96%|█████████▌| 4281/4476 [27:46:05<1:16:03, 23.40s/it]
 96%|█████████▌| 4282/4476 [27:46:28<1:15:38, 23.39s/it]
 96%|█████████▌| 4283/4476 [27:46:51<1:14:53, 23.28s/it]
 96%|█████████▌| 4284/4476 [27:47:14<1:14:16, 23.21s/it]
 96%|█████████▌| 4285/4476 [27:47:37<1:13:52, 23.21s/it]
 96%|█████████▌| 4286/4476 [27:48:01<1:13:43, 23.28s/it]
 96%|█████████▌| 4287/4476 [27:48:24<1:13:28, 23.33s/it]
 96%|█████████▌| 4288/4476 [27:48:47<1:12:47, 23.23s/it]
 96%|█████████▌| 4289/4476 [27:49:10<1:12:25, 23.24s/it]
 96%|█████████▌| 4290/4476 [27:49:34<1:12:18, 23.32s/it]
                                                        
{'loss': 0.3725, 'learning_rate': 2.1502515537069334e-07, 'epoch': 2.87}

 96%|█████████▌| 4290/4476 [27:49:34<1:12:18, 23.32s/it]
 96%|█████████▌| 4291/4476 [27:49:57<1:11:55, 23.33s/it]
 96%|█████████▌| 4292/4476 [27:50:20<1:11:03, 23.17s/it]
 96%|█████████▌| 4293/4476 [27:50:43<1:10:44, 23.19s/it]
 96%|█████████▌| 4294/4476 [27:51:07<1:10:52, 23.36s/it]
 96%|█████████▌| 4295/4476 [27:51:30<1:10:24, 23.34s/it]
 96%|█████████▌| 4296/4476 [27:51:54<1:09:58, 23.32s/it]
 96%|█████████▌| 4297/4476 [27:52:17<1:09:29, 23.29s/it]
 96%|█████████▌| 4298/4476 [27:52:41<1:09:28, 23.42s/it]
 96%|█████████▌| 4299/4476 [27:53:04<1:08:52, 23.35s/it]
 96%|█████████▌| 4300/4476 [27:53:27<1:08:29, 23.35s/it]
                                                        
{'loss': 0.3729, 'learning_rate': 1.926715114835914e-07, 'epoch': 2.88}

 96%|█████████▌| 4300/4476 [27:53:27<1:08:29, 23.35s/it]
 96%|█████████▌| 4301/4476 [27:53:50<1:08:07, 23.36s/it]
 96%|█████████▌| 4302/4476 [27:54:14<1:07:42, 23.35s/it]
 96%|█████████▌| 4303/4476 [27:54:37<1:07:36, 23.45s/it]
 96%|█████████▌| 4304/4476 [27:55:01<1:07:10, 23.43s/it]
 96%|█████████▌| 4305/4476 [27:55:24<1:06:48, 23.44s/it]
 96%|█████████▌| 4306/4476 [27:55:48<1:06:40, 23.53s/it]
 96%|█████████▌| 4307/4476 [27:56:11<1:06:05, 23.46s/it]
 96%|█████████▌| 4308/4476 [27:56:35<1:05:31, 23.40s/it]
 96%|█████████▋| 4309/4476 [27:56:58<1:05:16, 23.45s/it]
 96%|█████████▋| 4310/4476 [27:57:22<1:04:51, 23.44s/it]
                                                        
{'loss': 0.3742, 'learning_rate': 1.7153994244005766e-07, 'epoch': 2.89}

 96%|█████████▋| 4310/4476 [27:57:22<1:04:51, 23.44s/it]
 96%|█████████▋| 4311/4476 [27:57:45<1:04:20, 23.40s/it]
 96%|█████████▋| 4312/4476 [27:58:09<1:04:16, 23.52s/it]
 96%|█████████▋| 4313/4476 [27:58:32<1:03:49, 23.49s/it]
 96%|█████████▋| 4314/4476 [27:58:56<1:03:39, 23.58s/it]
 96%|█████████▋| 4315/4476 [27:59:19<1:02:55, 23.45s/it]
 96%|█████████▋| 4316/4476 [27:59:42<1:02:18, 23.37s/it]
 96%|█████████▋| 4317/4476 [28:00:06<1:02:02, 23.41s/it]
 96%|█████████▋| 4318/4476 [28:00:29<1:01:25, 23.33s/it]
 96%|█████████▋| 4319/4476 [28:00:52<1:00:57, 23.30s/it]
 97%|█████████▋| 4320/4476 [28:01:15<1:00:31, 23.28s/it]
                                                        
{'loss': 0.3739, 'learning_rate': 1.516314892372639e-07, 'epoch': 2.89}

 97%|█████████▋| 4320/4476 [28:01:15<1:00:31, 23.28s/it]
 97%|█████████▋| 4321/4476 [28:01:39<1:00:08, 23.28s/it]
 97%|█████████▋| 4322/4476 [28:02:02<59:45, 23.28s/it]  
 97%|█████████▋| 4323/4476 [28:02:25<59:13, 23.22s/it]
 97%|█████████▋| 4324/4476 [28:02:49<59:05, 23.33s/it]
 97%|█████████▋| 4325/4476 [28:03:12<58:48, 23.37s/it]
 97%|█████████▋| 4326/4476 [28:03:36<58:37, 23.45s/it]
 97%|█████████▋| 4327/4476 [28:03:59<58:06, 23.40s/it]
 97%|█████████▋| 4328/4476 [28:04:22<57:37, 23.36s/it]
 97%|█████████▋| 4329/4476 [28:04:46<57:13, 23.36s/it]
 97%|█████████▋| 4330/4476 [28:05:09<56:37, 23.27s/it]
                                                      
{'loss': 0.3755, 'learning_rate': 1.3294713261845503e-07, 'epoch': 2.9}

 97%|█████████▋| 4330/4476 [28:05:09<56:37, 23.27s/it]
 97%|█████████▋| 4331/4476 [28:05:32<56:18, 23.30s/it]
 97%|█████████▋| 4332/4476 [28:05:55<56:02, 23.35s/it]
 97%|█████████▋| 4333/4476 [28:06:19<55:33, 23.31s/it]
 97%|█████████▋| 4334/4476 [28:06:42<55:11, 23.32s/it]
 97%|█████████▋| 4335/4476 [28:07:05<54:47, 23.32s/it]
 97%|█████████▋| 4336/4476 [28:07:29<54:23, 23.31s/it]
 97%|█████████▋| 4337/4476 [28:07:52<53:57, 23.29s/it]
 97%|█████████▋| 4338/4476 [28:08:15<53:29, 23.26s/it]
 97%|█████████▋| 4339/4476 [28:08:39<53:17, 23.34s/it]
 97%|█████████▋| 4340/4476 [28:09:01<52:35, 23.20s/it]
                                                      
{'loss': 0.3702, 'learning_rate': 1.1548779302463231e-07, 'epoch': 2.91}

 97%|█████████▋| 4340/4476 [28:09:01<52:35, 23.20s/it]
 97%|█████████▋| 4341/4476 [28:09:24<52:03, 23.14s/it]
 97%|█████████▋| 4342/4476 [28:09:48<51:39, 23.13s/it]
 97%|█████████▋| 4343/4476 [28:10:11<51:17, 23.14s/it]
 97%|█████████▋| 4344/4476 [28:10:34<50:52, 23.12s/it]
 97%|█████████▋| 4345/4476 [28:10:57<50:34, 23.16s/it]
 97%|█████████▋| 4346/4476 [28:11:20<50:16, 23.21s/it]
 97%|█████████▋| 4347/4476 [28:11:44<49:57, 23.23s/it]
 97%|█████████▋| 4348/4476 [28:12:07<49:33, 23.23s/it]
 97%|█████████▋| 4349/4476 [28:12:30<49:10, 23.23s/it]
 97%|█████████▋| 4350/4476 [28:12:53<48:45, 23.22s/it]
                                                      
{'loss': 0.3761, 'learning_rate': 9.92543305492033e-08, 'epoch': 2.92}

 97%|█████████▋| 4350/4476 [28:12:53<48:45, 23.22s/it]
 97%|█████████▋| 4351/4476 [28:13:17<48:24, 23.24s/it]
 97%|█████████▋| 4352/4476 [28:13:40<48:10, 23.31s/it]
 97%|█████████▋| 4353/4476 [28:14:04<48:01, 23.43s/it]
 97%|█████████▋| 4354/4476 [28:14:27<47:33, 23.39s/it]
 97%|█████████▋| 4355/4476 [28:14:50<47:08, 23.38s/it]
 97%|█████████▋| 4356/4476 [28:15:14<46:35, 23.29s/it]
 97%|█████████▋| 4357/4476 [28:15:37<46:17, 23.34s/it]
 97%|█████████▋| 4358/4476 [28:16:01<46:05, 23.44s/it]
 97%|█████████▋| 4359/4476 [28:16:24<45:38, 23.41s/it]
 97%|█████████▋| 4360/4476 [28:16:48<45:24, 23.49s/it]
                                                      
{'loss': 0.38, 'learning_rate': 8.424754489561038e-08, 'epoch': 2.92}

 97%|█████████▋| 4360/4476 [28:16:48<45:24, 23.49s/it]
 97%|█████████▋| 4361/4476 [28:17:11<44:53, 23.42s/it]
 97%|█████████▋| 4362/4476 [28:17:34<44:26, 23.39s/it]
 97%|█████████▋| 4363/4476 [28:17:58<44:15, 23.50s/it]
 97%|█████████▋| 4364/4476 [28:18:21<43:43, 23.43s/it]
 98%|█████████▊| 4365/4476 [28:18:45<43:24, 23.46s/it]
 98%|█████████▊| 4366/4476 [28:19:08<42:56, 23.43s/it]
 98%|█████████▊| 4367/4476 [28:19:32<42:31, 23.40s/it]
 98%|█████████▊| 4368/4476 [28:19:55<42:19, 23.51s/it]
 98%|█████████▊| 4369/4476 [28:20:19<41:46, 23.43s/it]
 98%|█████████▊| 4370/4476 [28:20:42<41:25, 23.45s/it]
                                                      
{'loss': 0.3737, 'learning_rate': 7.046817533795102e-08, 'epoch': 2.93}

 98%|█████████▊| 4370/4476 [28:20:42<41:25, 23.45s/it]
 98%|█████████▊| 4371/4476 [28:21:05<40:56, 23.40s/it]
 98%|█████████▊| 4372/4476 [28:21:29<40:30, 23.37s/it]
 98%|█████████▊| 4373/4476 [28:21:52<40:08, 23.39s/it]
 98%|█████████▊| 4374/4476 [28:22:15<39:35, 23.29s/it]
 98%|█████████▊| 4375/4476 [28:22:38<39:11, 23.28s/it]
 98%|█████████▊| 4376/4476 [28:23:02<38:47, 23.27s/it]
 98%|█████████▊| 4377/4476 [28:23:25<38:27, 23.31s/it]
 98%|█████████▊| 4378/4476 [28:23:48<38:00, 23.27s/it]
 98%|█████████▊| 4379/4476 [28:24:12<37:45, 23.35s/it]
 98%|█████████▊| 4380/4476 [28:24:35<37:12, 23.26s/it]
                                                      
{'loss': 0.3722, 'learning_rate': 5.7916900684540366e-08, 'epoch': 2.94}

 98%|█████████▊| 4380/4476 [28:24:35<37:12, 23.26s/it]
 98%|█████████▊| 4381/4476 [28:24:58<36:44, 23.20s/it]
 98%|█████████▊| 4382/4476 [28:25:21<36:33, 23.33s/it]
 98%|█████████▊| 4383/4476 [28:25:45<36:08, 23.31s/it]
 98%|█████████▊| 4384/4476 [28:26:08<35:50, 23.37s/it]
 98%|█████████▊| 4385/4476 [28:26:32<35:34, 23.46s/it]
 98%|█████████▊| 4386/4476 [28:26:55<35:06, 23.41s/it]
 98%|█████████▊| 4387/4476 [28:27:18<34:39, 23.37s/it]
 98%|█████████▊| 4388/4476 [28:27:42<34:16, 23.37s/it]
 98%|█████████▊| 4389/4476 [28:28:05<33:53, 23.37s/it]
 98%|█████████▊| 4390/4476 [28:28:28<33:27, 23.35s/it]
                                                      
{'loss': 0.3805, 'learning_rate': 4.6594339244479536e-08, 'epoch': 2.94}

 98%|█████████▊| 4390/4476 [28:28:29<33:27, 23.35s/it]
 98%|█████████▊| 4391/4476 [28:28:52<33:07, 23.38s/it]
 98%|█████████▊| 4392/4476 [28:29:16<32:50, 23.46s/it]
 98%|█████████▊| 4393/4476 [28:29:39<32:28, 23.48s/it]
 98%|█████████▊| 4394/4476 [28:30:03<32:10, 23.54s/it]
 98%|█████████▊| 4395/4476 [28:30:26<31:33, 23.38s/it]
 98%|█████████▊| 4396/4476 [28:30:49<31:03, 23.29s/it]
 98%|█████████▊| 4397/4476 [28:31:12<30:38, 23.27s/it]
 98%|█████████▊| 4398/4476 [28:31:36<30:20, 23.33s/it]
 98%|█████████▊| 4399/4476 [28:31:59<29:51, 23.27s/it]
 98%|█████████▊| 4400/4476 [28:32:22<29:28, 23.27s/it]
                                                      
{'loss': 0.3736, 'learning_rate': 3.650104879719951e-08, 'epoch': 2.95}

 98%|█████████▊| 4400/4476 [28:32:22<29:28, 23.27s/it][INFO|trainer.py:2939] 2023-11-13 07:56:09,732 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4400
[INFO|tokenization_utils_base.py:2437] 2023-11-13 07:56:09,764 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4400/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 07:56:09,764 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4400/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 07:56:09,764 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/checkpoint-4400/added_tokens.json

 98%|█████████▊| 4401/4476 [28:32:46<29:16, 23.42s/it]
 98%|█████████▊| 4402/4476 [28:33:09<28:55, 23.45s/it]
 98%|█████████▊| 4403/4476 [28:33:33<28:27, 23.39s/it]
 98%|█████████▊| 4404/4476 [28:33:56<28:00, 23.33s/it]
 98%|█████████▊| 4405/4476 [28:34:19<27:35, 23.31s/it]
 98%|█████████▊| 4406/4476 [28:34:42<27:12, 23.32s/it]
 98%|█████████▊| 4407/4476 [28:35:06<26:47, 23.30s/it]
 98%|█████████▊| 4408/4476 [28:35:29<26:20, 23.25s/it]
 99%|█████████▊| 4409/4476 [28:35:52<26:05, 23.37s/it]
 99%|█████████▊| 4410/4476 [28:36:16<25:42, 23.37s/it]
                                                      
{'loss': 0.37, 'learning_rate': 2.7637526564971982e-08, 'epoch': 2.96}

 99%|█████████▊| 4410/4476 [28:36:16<25:42, 23.37s/it]
 99%|█████████▊| 4411/4476 [28:36:39<25:16, 23.33s/it]
 99%|█████████▊| 4412/4476 [28:37:02<24:51, 23.30s/it]
 99%|█████████▊| 4413/4476 [28:37:25<24:21, 23.21s/it]
 99%|█████████▊| 4414/4476 [28:37:48<23:57, 23.18s/it]
 99%|█████████▊| 4415/4476 [28:38:12<23:39, 23.26s/it]
 99%|█████████▊| 4416/4476 [28:38:35<23:20, 23.33s/it]
 99%|█████████▊| 4417/4476 [28:38:59<22:57, 23.35s/it]
 99%|█████████▊| 4418/4476 [28:39:22<22:41, 23.48s/it]
 99%|█████████▊| 4419/4476 [28:39:46<22:14, 23.41s/it]
 99%|█████████▊| 4420/4476 [28:40:09<21:49, 23.38s/it]
                                                      
{'loss': 0.374, 'learning_rate': 2.0004209188428937e-08, 'epoch': 2.96}

 99%|█████████▊| 4420/4476 [28:40:09<21:49, 23.38s/it]
 99%|█████████▉| 4421/4476 [28:40:32<21:24, 23.36s/it]
 99%|█████████▉| 4422/4476 [28:40:56<21:07, 23.47s/it]
 99%|█████████▉| 4423/4476 [28:41:20<20:48, 23.56s/it]
 99%|█████████▉| 4424/4476 [28:41:43<20:22, 23.52s/it]
 99%|█████████▉| 4425/4476 [28:42:06<19:54, 23.43s/it]
 99%|█████████▉| 4426/4476 [28:42:30<19:30, 23.41s/it]
 99%|█████████▉| 4427/4476 [28:42:53<19:06, 23.40s/it]
 99%|█████████▉| 4428/4476 [28:43:16<18:40, 23.35s/it]
 99%|█████████▉| 4429/4476 [28:43:40<18:19, 23.40s/it]
 99%|█████████▉| 4430/4476 [28:44:03<17:58, 23.44s/it]
                                                      
{'loss': 0.3713, 'learning_rate': 1.3601472705046525e-08, 'epoch': 2.97}

 99%|█████████▉| 4430/4476 [28:44:03<17:58, 23.44s/it]
 99%|█████████▉| 4431/4476 [28:44:27<17:32, 23.39s/it]
 99%|█████████▉| 4432/4476 [28:44:51<17:14, 23.50s/it]
 99%|█████████▉| 4433/4476 [28:45:14<16:47, 23.42s/it]
 99%|█████████▉| 4434/4476 [28:45:37<16:22, 23.39s/it]
 99%|█████████▉| 4435/4476 [28:46:00<15:54, 23.29s/it]
 99%|█████████▉| 4436/4476 [28:46:24<15:36, 23.41s/it]
 99%|█████████▉| 4437/4476 [28:46:47<15:14, 23.45s/it]
 99%|█████████▉| 4438/4476 [28:47:11<14:54, 23.55s/it]
 99%|█████████▉| 4439/4476 [28:47:34<14:27, 23.44s/it]
 99%|█████████▉| 4440/4476 [28:47:58<14:06, 23.51s/it]
                                                      
{'loss': 0.3686, 'learning_rate': 8.429632530618236e-09, 'epoch': 2.98}

 99%|█████████▉| 4440/4476 [28:47:58<14:06, 23.51s/it]
 99%|█████████▉| 4441/4476 [28:48:22<13:44, 23.56s/it]
 99%|█████████▉| 4442/4476 [28:48:45<13:19, 23.51s/it]
 99%|█████████▉| 4443/4476 [28:49:08<12:53, 23.45s/it]
 99%|█████████▉| 4444/4476 [28:49:32<12:28, 23.38s/it]
 99%|█████████▉| 4445/4476 [28:49:55<12:01, 23.29s/it]
 99%|█████████▉| 4446/4476 [28:50:18<11:40, 23.34s/it]
 99%|█████████▉| 4447/4476 [28:50:42<11:18, 23.40s/it]
 99%|█████████▉| 4448/4476 [28:51:05<10:54, 23.37s/it]
 99%|█████████▉| 4449/4476 [28:51:29<10:32, 23.43s/it]
 99%|█████████▉| 4450/4476 [28:51:52<10:08, 23.40s/it]
                                                      
{'loss': 0.3721, 'learning_rate': 4.488943443711757e-09, 'epoch': 2.98}

 99%|█████████▉| 4450/4476 [28:51:52<10:08, 23.40s/it]
 99%|█████████▉| 4451/4476 [28:52:15<09:43, 23.34s/it]
 99%|█████████▉| 4452/4476 [28:52:39<09:23, 23.48s/it]
 99%|█████████▉| 4453/4476 [28:53:02<08:59, 23.44s/it]
100%|█████████▉| 4454/4476 [28:53:26<08:36, 23.47s/it]
100%|█████████▉| 4455/4476 [28:53:49<08:11, 23.41s/it]
100%|█████████▉| 4456/4476 [28:54:12<07:46, 23.30s/it]
100%|█████████▉| 4457/4476 [28:54:35<07:23, 23.33s/it]
100%|█████████▉| 4458/4476 [28:54:59<07:02, 23.47s/it]
100%|█████████▉| 4459/4476 [28:55:23<06:37, 23.41s/it]
100%|█████████▉| 4460/4476 [28:55:46<06:16, 23.52s/it]
                                                      
{'loss': 0.3766, 'learning_rate': 1.779599573137336e-09, 'epoch': 2.99}

100%|█████████▉| 4460/4476 [28:55:46<06:16, 23.52s/it]
100%|█████████▉| 4461/4476 [28:56:10<05:52, 23.53s/it]
100%|█████████▉| 4462/4476 [28:56:33<05:29, 23.53s/it]
100%|█████████▉| 4463/4476 [28:56:57<05:04, 23.44s/it]
100%|█████████▉| 4464/4476 [28:57:20<04:41, 23.43s/it]
100%|█████████▉| 4465/4476 [28:57:43<04:17, 23.40s/it]
100%|█████████▉| 4466/4476 [28:58:07<03:54, 23.43s/it]
100%|█████████▉| 4467/4476 [28:58:30<03:30, 23.37s/it]
100%|█████████▉| 4468/4476 [28:58:53<03:06, 23.31s/it]
100%|█████████▉| 4469/4476 [28:59:16<02:42, 23.25s/it]
100%|█████████▉| 4470/4476 [28:59:39<02:18, 23.12s/it]
                                                      
{'loss': 0.376, 'learning_rate': 3.017343883637835e-10, 'epoch': 3.0}

100%|█████████▉| 4470/4476 [28:59:39<02:18, 23.12s/it]
100%|█████████▉| 4471/4476 [29:00:03<01:55, 23.19s/it]
100%|█████████▉| 4472/4476 [29:00:26<01:32, 23.18s/it]
100%|█████████▉| 4473/4476 [29:00:49<01:09, 23.28s/it]
100%|█████████▉| 4474/4476 [29:01:12<00:46, 23.22s/it]
100%|█████████▉| 4475/4476 [29:01:36<00:23, 23.25s/it]
100%|██████████| 4476/4476 [29:01:59<00:00, 23.29s/it][INFO|trainer.py:2017] 2023-11-13 08:25:46,723 >> 

Training completed. Do not forget to share your model on huggingface.co/models =)



                                                      
{'train_runtime': 104521.3326, 'train_samples_per_second': 10.964, 'train_steps_per_second': 0.043, 'train_loss': 0.40742741023141216, 'epoch': 3.0}

100%|██████████| 4476/4476 [29:01:59<00:00, 23.29s/it]
100%|██████████| 4476/4476 [29:01:59<00:00, 23.35s/it]
[INFO|trainer.py:2939] 2023-11-13 08:25:46,728 >> Saving model checkpoint to /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned
[INFO|tokenization_utils_base.py:2437] 2023-11-13 08:25:46,761 >> tokenizer config file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/tokenizer_config.json
[INFO|tokenization_utils_base.py:2446] 2023-11-13 08:25:46,761 >> Special tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/special_tokens_map.json
[INFO|tokenization_utils_base.py:2496] 2023-11-13 08:25:46,761 >> added tokens file saved in /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/added_tokens.json
***** train metrics *****
  epoch                    =               3.0
  train_loss               =            0.4074
  train_runtime            = 1 day, 5:02:01.33
  train_samples_per_second =            10.964
  train_steps_per_second   =             0.043
Figure saved: /home/hz/projects/Project/chatglm3-6b-32k-wenshu-finetuned/training_loss.png
11/13/2023 08:25:46 - WARNING - llmtuner.extras.ploting - No metric eval_loss to plot.
[INFO|modelcard.py:452] 2023-11-13 08:25:46,887 >> Dropping the following result as it does not have all the necessary fields:
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}