Qin Liu
commited on
Commit
•
8f6eefa
1
Parent(s):
06b72ad
Model save
Browse files- .gitattributes +1 -0
- README.md +69 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- all_results.json +8 -0
- runs/Apr26_10-12-14_COE-CS-sv003/events.out.tfevents.1714126416.COE-CS-sv003.480309.0 +3 -0
- special_tokens_map.json +28 -0
- tokenizer.json +3 -0
- tokenizer_config.json +70 -0
- train_results.json +8 -0
- trainer_state.json +2791 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: google/gemma-7b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: zephyr-7b-gemma-sft-5p-2048
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# zephyr-7b-gemma-sft-5p-2048
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.1822
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 32
|
| 48 |
+
- total_eval_batch_size: 16
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 3
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 58 |
+
| 0.9062 | 1.0 | 651 | 1.2442 |
|
| 59 |
+
| 0.907 | 2.0 | 1303 | 1.1708 |
|
| 60 |
+
| 0.8209 | 3.0 | 1953 | 1.1822 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.7.1
|
| 66 |
+
- Transformers 4.39.0.dev0
|
| 67 |
+
- Pytorch 2.1.2
|
| 68 |
+
- Datasets 2.14.6
|
| 69 |
+
- Tokenizers 0.15.2
|
adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "google/gemma-7b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 8,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 6,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"o_proj",
|
| 23 |
+
"q_proj",
|
| 24 |
+
"gate_proj",
|
| 25 |
+
"up_proj",
|
| 26 |
+
"v_proj",
|
| 27 |
+
"down_proj",
|
| 28 |
+
"k_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM"
|
| 31 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2b089dac39d843b1ff4c1788a5cf2494adbc0a1c5f554b886906cd9d48aa612
|
| 3 |
+
size 37555048
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.1336081770219621,
|
| 4 |
+
"train_runtime": 21652.8936,
|
| 5 |
+
"train_samples": 20842,
|
| 6 |
+
"train_samples_per_second": 2.888,
|
| 7 |
+
"train_steps_per_second": 0.09
|
| 8 |
+
}
|
runs/Apr26_10-12-14_COE-CS-sv003/events.out.tfevents.1714126416.COE-CS-sv003.480309.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39c4df3285df824e39695aeb563c785b14a88c9a4f0eaad634c78e2a75accd96
|
| 3 |
+
size 88370
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|im_start|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|im_end|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
}
|
| 17 |
+
],
|
| 18 |
+
"bos_token": "<|im_start|>",
|
| 19 |
+
"eos_token": "<|im_end|>",
|
| 20 |
+
"pad_token": "<|im_end|>",
|
| 21 |
+
"unk_token": {
|
| 22 |
+
"content": "<unk>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
}
|
| 28 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:299f8e59a5c4a3b3941dbe1159a7079d69c7a8d5ca34322ace1be9140ae76cc0
|
| 3 |
+
size 17477572
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<pad>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<eos>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<bos>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"106": {
|
| 38 |
+
"content": "<|im_start|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"107": {
|
| 46 |
+
"content": "<|im_end|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
}
|
| 53 |
+
},
|
| 54 |
+
"additional_special_tokens": [
|
| 55 |
+
"<|im_start|>",
|
| 56 |
+
"<|im_end|>"
|
| 57 |
+
],
|
| 58 |
+
"bos_token": "<|im_start|>",
|
| 59 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 60 |
+
"clean_up_tokenization_spaces": false,
|
| 61 |
+
"eos_token": "<|im_end|>",
|
| 62 |
+
"legacy": null,
|
| 63 |
+
"model_max_length": 2048,
|
| 64 |
+
"pad_token": "<|im_end|>",
|
| 65 |
+
"sp_model_kwargs": {},
|
| 66 |
+
"spaces_between_special_tokens": false,
|
| 67 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 68 |
+
"unk_token": "<unk>",
|
| 69 |
+
"use_default_system_prompt": false
|
| 70 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.1336081770219621,
|
| 4 |
+
"train_runtime": 21652.8936,
|
| 5 |
+
"train_samples": 20842,
|
| 6 |
+
"train_samples_per_second": 2.888,
|
| 7 |
+
"train_steps_per_second": 0.09
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2791 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.997697620874904,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1953,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"grad_norm": 72.47212998792584,
|
| 14 |
+
"learning_rate": 1.020408163265306e-06,
|
| 15 |
+
"loss": 17.0139,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01,
|
| 20 |
+
"grad_norm": 72.19709441394318,
|
| 21 |
+
"learning_rate": 5.102040816326531e-06,
|
| 22 |
+
"loss": 16.8099,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02,
|
| 27 |
+
"grad_norm": 74.11072818499085,
|
| 28 |
+
"learning_rate": 1.0204081632653061e-05,
|
| 29 |
+
"loss": 16.7768,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02,
|
| 34 |
+
"grad_norm": 67.88747859998395,
|
| 35 |
+
"learning_rate": 1.5306122448979594e-05,
|
| 36 |
+
"loss": 15.8745,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"grad_norm": 57.15585576506464,
|
| 42 |
+
"learning_rate": 2.0408163265306123e-05,
|
| 43 |
+
"loss": 13.4197,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"grad_norm": 52.16705016348157,
|
| 49 |
+
"learning_rate": 2.5510204081632654e-05,
|
| 50 |
+
"loss": 10.8782,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"grad_norm": 34.872891212782,
|
| 56 |
+
"learning_rate": 3.061224489795919e-05,
|
| 57 |
+
"loss": 8.0236,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.05,
|
| 62 |
+
"grad_norm": 20.96608544121856,
|
| 63 |
+
"learning_rate": 3.571428571428572e-05,
|
| 64 |
+
"loss": 5.7602,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"grad_norm": 14.39165011334111,
|
| 70 |
+
"learning_rate": 4.0816326530612245e-05,
|
| 71 |
+
"loss": 4.4129,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07,
|
| 76 |
+
"grad_norm": 9.537119159153542,
|
| 77 |
+
"learning_rate": 4.591836734693878e-05,
|
| 78 |
+
"loss": 3.4241,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"grad_norm": 10.051565364855378,
|
| 84 |
+
"learning_rate": 5.102040816326531e-05,
|
| 85 |
+
"loss": 2.7711,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.08,
|
| 90 |
+
"grad_norm": 5.902644991920862,
|
| 91 |
+
"learning_rate": 5.6122448979591836e-05,
|
| 92 |
+
"loss": 2.1132,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.09,
|
| 97 |
+
"grad_norm": 5.282533867200368,
|
| 98 |
+
"learning_rate": 6.122448979591838e-05,
|
| 99 |
+
"loss": 1.7932,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1,
|
| 104 |
+
"grad_norm": 2.849440118853736,
|
| 105 |
+
"learning_rate": 6.63265306122449e-05,
|
| 106 |
+
"loss": 1.5357,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"grad_norm": 6.395981446361856,
|
| 112 |
+
"learning_rate": 7.142857142857143e-05,
|
| 113 |
+
"loss": 1.4488,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.12,
|
| 118 |
+
"grad_norm": 1.9290320550828686,
|
| 119 |
+
"learning_rate": 7.653061224489796e-05,
|
| 120 |
+
"loss": 1.3654,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.12,
|
| 125 |
+
"grad_norm": 1.9509560155111505,
|
| 126 |
+
"learning_rate": 8.163265306122449e-05,
|
| 127 |
+
"loss": 1.2287,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.13,
|
| 132 |
+
"grad_norm": 1.9800497930938712,
|
| 133 |
+
"learning_rate": 8.673469387755102e-05,
|
| 134 |
+
"loss": 1.2863,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"grad_norm": 1.7120992753385207,
|
| 140 |
+
"learning_rate": 9.183673469387756e-05,
|
| 141 |
+
"loss": 1.1712,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.15,
|
| 146 |
+
"grad_norm": 2.2653861656665453,
|
| 147 |
+
"learning_rate": 9.693877551020408e-05,
|
| 148 |
+
"loss": 1.2174,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.15,
|
| 153 |
+
"grad_norm": 3.453167617907916,
|
| 154 |
+
"learning_rate": 0.00010204081632653062,
|
| 155 |
+
"loss": 1.1714,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.16,
|
| 160 |
+
"grad_norm": 1.547620666252296,
|
| 161 |
+
"learning_rate": 0.00010714285714285715,
|
| 162 |
+
"loss": 1.1168,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.17,
|
| 167 |
+
"grad_norm": 1.6713272237258214,
|
| 168 |
+
"learning_rate": 0.00011224489795918367,
|
| 169 |
+
"loss": 1.0868,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.18,
|
| 174 |
+
"grad_norm": 1.5402309108002492,
|
| 175 |
+
"learning_rate": 0.00011734693877551022,
|
| 176 |
+
"loss": 1.1351,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.18,
|
| 181 |
+
"grad_norm": 1.5970274628335959,
|
| 182 |
+
"learning_rate": 0.00012244897959183676,
|
| 183 |
+
"loss": 1.0946,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.19,
|
| 188 |
+
"grad_norm": 1.6342350847388514,
|
| 189 |
+
"learning_rate": 0.00012755102040816328,
|
| 190 |
+
"loss": 1.1211,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2,
|
| 195 |
+
"grad_norm": 1.6632080501970268,
|
| 196 |
+
"learning_rate": 0.0001326530612244898,
|
| 197 |
+
"loss": 1.0501,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.21,
|
| 202 |
+
"grad_norm": 1.4060120288881937,
|
| 203 |
+
"learning_rate": 0.00013775510204081635,
|
| 204 |
+
"loss": 1.0956,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.21,
|
| 209 |
+
"grad_norm": 1.801531788644497,
|
| 210 |
+
"learning_rate": 0.00014285714285714287,
|
| 211 |
+
"loss": 1.072,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.22,
|
| 216 |
+
"grad_norm": 1.520734386032582,
|
| 217 |
+
"learning_rate": 0.0001479591836734694,
|
| 218 |
+
"loss": 1.1119,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.23,
|
| 223 |
+
"grad_norm": 1.840843283748014,
|
| 224 |
+
"learning_rate": 0.0001530612244897959,
|
| 225 |
+
"loss": 0.9969,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.24,
|
| 230 |
+
"grad_norm": 2.07792070441396,
|
| 231 |
+
"learning_rate": 0.00015816326530612246,
|
| 232 |
+
"loss": 1.0208,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.25,
|
| 237 |
+
"grad_norm": 1.9723613912374163,
|
| 238 |
+
"learning_rate": 0.00016326530612244898,
|
| 239 |
+
"loss": 0.9975,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.25,
|
| 244 |
+
"grad_norm": 2.8108507027898844,
|
| 245 |
+
"learning_rate": 0.00016836734693877553,
|
| 246 |
+
"loss": 1.0388,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.26,
|
| 251 |
+
"grad_norm": 2.049204774598262,
|
| 252 |
+
"learning_rate": 0.00017346938775510205,
|
| 253 |
+
"loss": 0.9837,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.27,
|
| 258 |
+
"grad_norm": 2.013771271987093,
|
| 259 |
+
"learning_rate": 0.0001785714285714286,
|
| 260 |
+
"loss": 0.956,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.28,
|
| 265 |
+
"grad_norm": 1.6734633410801107,
|
| 266 |
+
"learning_rate": 0.00018367346938775512,
|
| 267 |
+
"loss": 1.0388,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.28,
|
| 272 |
+
"grad_norm": 2.3629530195645043,
|
| 273 |
+
"learning_rate": 0.00018877551020408164,
|
| 274 |
+
"loss": 1.0407,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.29,
|
| 279 |
+
"grad_norm": 2.0061060540694267,
|
| 280 |
+
"learning_rate": 0.00019387755102040816,
|
| 281 |
+
"loss": 1.0528,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.3,
|
| 286 |
+
"grad_norm": 1.8957278214013087,
|
| 287 |
+
"learning_rate": 0.0001989795918367347,
|
| 288 |
+
"loss": 1.0637,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.31,
|
| 293 |
+
"grad_norm": 1.659431161796715,
|
| 294 |
+
"learning_rate": 0.00019999744233089168,
|
| 295 |
+
"loss": 1.0197,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.31,
|
| 300 |
+
"grad_norm": 1.9623902009306882,
|
| 301 |
+
"learning_rate": 0.00019998705202436978,
|
| 302 |
+
"loss": 1.0171,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.32,
|
| 307 |
+
"grad_norm": 1.666768664331948,
|
| 308 |
+
"learning_rate": 0.0001999686700559419,
|
| 309 |
+
"loss": 0.9411,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.33,
|
| 314 |
+
"grad_norm": 1.6263956387135465,
|
| 315 |
+
"learning_rate": 0.00019994229789482308,
|
| 316 |
+
"loss": 1.0452,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.34,
|
| 321 |
+
"grad_norm": 2.104534739444239,
|
| 322 |
+
"learning_rate": 0.00019990793764886012,
|
| 323 |
+
"loss": 1.0318,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.35,
|
| 328 |
+
"grad_norm": 1.5223676518998526,
|
| 329 |
+
"learning_rate": 0.0001998655920643634,
|
| 330 |
+
"loss": 0.9393,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.35,
|
| 335 |
+
"grad_norm": 1.399019073099987,
|
| 336 |
+
"learning_rate": 0.000199815264525887,
|
| 337 |
+
"loss": 0.97,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.36,
|
| 342 |
+
"grad_norm": 1.8651896389488705,
|
| 343 |
+
"learning_rate": 0.00019975695905595855,
|
| 344 |
+
"loss": 1.0187,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.37,
|
| 349 |
+
"grad_norm": 1.536205961219932,
|
| 350 |
+
"learning_rate": 0.00019969068031475744,
|
| 351 |
+
"loss": 0.9716,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.38,
|
| 356 |
+
"grad_norm": 1.7078218228495357,
|
| 357 |
+
"learning_rate": 0.0001996164335997425,
|
| 358 |
+
"loss": 0.9959,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.38,
|
| 363 |
+
"grad_norm": 1.3467510953394168,
|
| 364 |
+
"learning_rate": 0.0001995342248452285,
|
| 365 |
+
"loss": 0.9602,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.39,
|
| 370 |
+
"grad_norm": 1.9084544882836239,
|
| 371 |
+
"learning_rate": 0.00019944406062191204,
|
| 372 |
+
"loss": 0.9775,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.4,
|
| 377 |
+
"grad_norm": 1.5574610066347279,
|
| 378 |
+
"learning_rate": 0.000199345948136346,
|
| 379 |
+
"loss": 1.0177,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.41,
|
| 384 |
+
"grad_norm": 2.109450536237562,
|
| 385 |
+
"learning_rate": 0.00019923989523036394,
|
| 386 |
+
"loss": 0.9819,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.41,
|
| 391 |
+
"grad_norm": 2.7828828000504573,
|
| 392 |
+
"learning_rate": 0.00019912591038045307,
|
| 393 |
+
"loss": 0.9707,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.42,
|
| 398 |
+
"grad_norm": 1.3138848135299093,
|
| 399 |
+
"learning_rate": 0.0001990040026970768,
|
| 400 |
+
"loss": 0.9909,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.43,
|
| 405 |
+
"grad_norm": 1.5708872027230352,
|
| 406 |
+
"learning_rate": 0.0001988741819239467,
|
| 407 |
+
"loss": 0.9744,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.44,
|
| 412 |
+
"grad_norm": 1.7915877912980807,
|
| 413 |
+
"learning_rate": 0.0001987364584372435,
|
| 414 |
+
"loss": 0.9629,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.45,
|
| 419 |
+
"grad_norm": 1.7688040681479107,
|
| 420 |
+
"learning_rate": 0.00019859084324478791,
|
| 421 |
+
"loss": 1.0066,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.45,
|
| 426 |
+
"grad_norm": 1.2087916028360217,
|
| 427 |
+
"learning_rate": 0.00019843734798516077,
|
| 428 |
+
"loss": 0.9273,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.46,
|
| 433 |
+
"grad_norm": 1.7217374396550518,
|
| 434 |
+
"learning_rate": 0.00019827598492677283,
|
| 435 |
+
"loss": 0.9599,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.47,
|
| 440 |
+
"grad_norm": 1.1827063296813014,
|
| 441 |
+
"learning_rate": 0.000198106766966884,
|
| 442 |
+
"loss": 0.9526,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.48,
|
| 447 |
+
"grad_norm": 1.275288932081838,
|
| 448 |
+
"learning_rate": 0.0001979297076305728,
|
| 449 |
+
"loss": 0.9351,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.48,
|
| 454 |
+
"grad_norm": 1.4572732094413907,
|
| 455 |
+
"learning_rate": 0.00019774482106965513,
|
| 456 |
+
"loss": 0.9916,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.49,
|
| 461 |
+
"grad_norm": 1.3548000574556105,
|
| 462 |
+
"learning_rate": 0.00019755212206155318,
|
| 463 |
+
"loss": 0.9894,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5,
|
| 468 |
+
"grad_norm": 1.4221295720770253,
|
| 469 |
+
"learning_rate": 0.00019735162600811447,
|
| 470 |
+
"loss": 0.9147,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.51,
|
| 475 |
+
"grad_norm": 1.1147927135270754,
|
| 476 |
+
"learning_rate": 0.00019714334893438062,
|
| 477 |
+
"loss": 0.9173,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.51,
|
| 482 |
+
"grad_norm": 1.5740508551311372,
|
| 483 |
+
"learning_rate": 0.00019692730748730662,
|
| 484 |
+
"loss": 1.0049,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.52,
|
| 489 |
+
"grad_norm": 1.4361894716708776,
|
| 490 |
+
"learning_rate": 0.0001967035189344303,
|
| 491 |
+
"loss": 0.9772,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.53,
|
| 496 |
+
"grad_norm": 1.5979099481664976,
|
| 497 |
+
"learning_rate": 0.00019647200116249214,
|
| 498 |
+
"loss": 0.9734,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.54,
|
| 503 |
+
"grad_norm": 1.1416579629890438,
|
| 504 |
+
"learning_rate": 0.00019623277267600574,
|
| 505 |
+
"loss": 0.9695,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.54,
|
| 510 |
+
"grad_norm": 1.2859502643647958,
|
| 511 |
+
"learning_rate": 0.0001959858525957786,
|
| 512 |
+
"loss": 0.9726,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.55,
|
| 517 |
+
"grad_norm": 1.654528585364595,
|
| 518 |
+
"learning_rate": 0.00019573126065738415,
|
| 519 |
+
"loss": 0.9099,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.56,
|
| 524 |
+
"grad_norm": 1.6432235938460993,
|
| 525 |
+
"learning_rate": 0.00019546901720958405,
|
| 526 |
+
"loss": 0.993,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.57,
|
| 531 |
+
"grad_norm": 1.184408195362666,
|
| 532 |
+
"learning_rate": 0.00019519914321270196,
|
| 533 |
+
"loss": 0.983,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.58,
|
| 538 |
+
"grad_norm": 1.364894780646226,
|
| 539 |
+
"learning_rate": 0.00019492166023694823,
|
| 540 |
+
"loss": 0.9385,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.58,
|
| 545 |
+
"grad_norm": 1.2140596195120288,
|
| 546 |
+
"learning_rate": 0.0001946365904606957,
|
| 547 |
+
"loss": 0.928,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.59,
|
| 552 |
+
"grad_norm": 1.2023189029574775,
|
| 553 |
+
"learning_rate": 0.00019434395666870734,
|
| 554 |
+
"loss": 0.9497,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.6,
|
| 559 |
+
"grad_norm": 1.5293781545066814,
|
| 560 |
+
"learning_rate": 0.00019404378225031482,
|
| 561 |
+
"loss": 0.9845,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.61,
|
| 566 |
+
"grad_norm": 1.2719601216605032,
|
| 567 |
+
"learning_rate": 0.00019373609119754926,
|
| 568 |
+
"loss": 0.9535,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.61,
|
| 573 |
+
"grad_norm": 1.1767287730617413,
|
| 574 |
+
"learning_rate": 0.00019342090810322361,
|
| 575 |
+
"loss": 0.9669,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.62,
|
| 580 |
+
"grad_norm": 1.5092663681250142,
|
| 581 |
+
"learning_rate": 0.00019309825815896697,
|
| 582 |
+
"loss": 0.9097,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.63,
|
| 587 |
+
"grad_norm": 1.065331569108817,
|
| 588 |
+
"learning_rate": 0.00019276816715321107,
|
| 589 |
+
"loss": 0.9257,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.64,
|
| 594 |
+
"grad_norm": 1.0551712975999572,
|
| 595 |
+
"learning_rate": 0.00019243066146912914,
|
| 596 |
+
"loss": 0.9179,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.64,
|
| 601 |
+
"grad_norm": 1.280851911643696,
|
| 602 |
+
"learning_rate": 0.00019208576808252726,
|
| 603 |
+
"loss": 0.9322,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.65,
|
| 608 |
+
"grad_norm": 1.1957795064970969,
|
| 609 |
+
"learning_rate": 0.00019173351455968805,
|
| 610 |
+
"loss": 0.9472,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.66,
|
| 615 |
+
"grad_norm": 1.6691262118179164,
|
| 616 |
+
"learning_rate": 0.00019137392905516757,
|
| 617 |
+
"loss": 0.9833,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.67,
|
| 622 |
+
"grad_norm": 1.2087456453830114,
|
| 623 |
+
"learning_rate": 0.0001910070403095449,
|
| 624 |
+
"loss": 0.9554,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.68,
|
| 629 |
+
"grad_norm": 1.2175939573659607,
|
| 630 |
+
"learning_rate": 0.00019063287764712513,
|
| 631 |
+
"loss": 0.9844,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.68,
|
| 636 |
+
"grad_norm": 1.4498433195544413,
|
| 637 |
+
"learning_rate": 0.00019025147097359528,
|
| 638 |
+
"loss": 0.9467,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.69,
|
| 643 |
+
"grad_norm": 1.429864892305334,
|
| 644 |
+
"learning_rate": 0.00018986285077363446,
|
| 645 |
+
"loss": 0.9309,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.7,
|
| 650 |
+
"grad_norm": 1.170651014792718,
|
| 651 |
+
"learning_rate": 0.00018946704810847689,
|
| 652 |
+
"loss": 0.9234,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.71,
|
| 657 |
+
"grad_norm": 1.3282215747364223,
|
| 658 |
+
"learning_rate": 0.00018906409461342952,
|
| 659 |
+
"loss": 0.9536,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.71,
|
| 664 |
+
"grad_norm": 1.1845632567557303,
|
| 665 |
+
"learning_rate": 0.00018865402249534347,
|
| 666 |
+
"loss": 0.9772,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.72,
|
| 671 |
+
"grad_norm": 1.2127271540342577,
|
| 672 |
+
"learning_rate": 0.00018823686453003973,
|
| 673 |
+
"loss": 0.9523,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.73,
|
| 678 |
+
"grad_norm": 1.2360135960506637,
|
| 679 |
+
"learning_rate": 0.00018781265405968972,
|
| 680 |
+
"loss": 0.9135,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.74,
|
| 685 |
+
"grad_norm": 1.5041667778892749,
|
| 686 |
+
"learning_rate": 0.0001873814249901501,
|
| 687 |
+
"loss": 0.9625,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.74,
|
| 692 |
+
"grad_norm": 1.1504429942546481,
|
| 693 |
+
"learning_rate": 0.00018694321178825286,
|
| 694 |
+
"loss": 0.9363,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.75,
|
| 699 |
+
"grad_norm": 1.0319491329942774,
|
| 700 |
+
"learning_rate": 0.00018649804947905055,
|
| 701 |
+
"loss": 0.9054,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.76,
|
| 706 |
+
"grad_norm": 1.2204274089754297,
|
| 707 |
+
"learning_rate": 0.0001860459736430169,
|
| 708 |
+
"loss": 0.9635,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.77,
|
| 713 |
+
"grad_norm": 1.0611368485238843,
|
| 714 |
+
"learning_rate": 0.00018558702041320273,
|
| 715 |
+
"loss": 0.9445,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.78,
|
| 720 |
+
"grad_norm": 1.1018739389426906,
|
| 721 |
+
"learning_rate": 0.00018512122647234812,
|
| 722 |
+
"loss": 0.9289,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.78,
|
| 727 |
+
"grad_norm": 0.9882389573927103,
|
| 728 |
+
"learning_rate": 0.0001846486290499505,
|
| 729 |
+
"loss": 0.9911,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.79,
|
| 734 |
+
"grad_norm": 1.2994396811323112,
|
| 735 |
+
"learning_rate": 0.0001841692659192889,
|
| 736 |
+
"loss": 0.9264,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.8,
|
| 741 |
+
"grad_norm": 1.5236972647383147,
|
| 742 |
+
"learning_rate": 0.00018368317539440492,
|
| 743 |
+
"loss": 0.9563,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.81,
|
| 748 |
+
"grad_norm": 1.1383444262695794,
|
| 749 |
+
"learning_rate": 0.0001831903963270404,
|
| 750 |
+
"loss": 0.977,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.81,
|
| 755 |
+
"grad_norm": 1.110789605109835,
|
| 756 |
+
"learning_rate": 0.00018269096810353205,
|
| 757 |
+
"loss": 0.9388,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.82,
|
| 762 |
+
"grad_norm": 1.6690032269651018,
|
| 763 |
+
"learning_rate": 0.00018218493064166353,
|
| 764 |
+
"loss": 0.923,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.83,
|
| 769 |
+
"grad_norm": 1.2741603156202803,
|
| 770 |
+
"learning_rate": 0.00018167232438747485,
|
| 771 |
+
"loss": 0.959,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.84,
|
| 776 |
+
"grad_norm": 1.1359315426783947,
|
| 777 |
+
"learning_rate": 0.00018115319031202965,
|
| 778 |
+
"loss": 0.958,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.84,
|
| 783 |
+
"grad_norm": 1.1462965424102658,
|
| 784 |
+
"learning_rate": 0.00018062756990814058,
|
| 785 |
+
"loss": 0.9206,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.85,
|
| 790 |
+
"grad_norm": 0.9795818536509971,
|
| 791 |
+
"learning_rate": 0.00018009550518705285,
|
| 792 |
+
"loss": 0.9027,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.86,
|
| 797 |
+
"grad_norm": 0.9867688236460216,
|
| 798 |
+
"learning_rate": 0.00017955703867508633,
|
| 799 |
+
"loss": 0.9283,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.87,
|
| 804 |
+
"grad_norm": 1.0077012911846719,
|
| 805 |
+
"learning_rate": 0.00017901221341023673,
|
| 806 |
+
"loss": 0.9516,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.87,
|
| 811 |
+
"grad_norm": 1.3421028406940887,
|
| 812 |
+
"learning_rate": 0.00017846107293873555,
|
| 813 |
+
"loss": 0.9121,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.88,
|
| 818 |
+
"grad_norm": 1.4547843706488663,
|
| 819 |
+
"learning_rate": 0.0001779036613115696,
|
| 820 |
+
"loss": 0.8875,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.89,
|
| 825 |
+
"grad_norm": 1.1060535265201301,
|
| 826 |
+
"learning_rate": 0.00017734002308096014,
|
| 827 |
+
"loss": 0.9554,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.9,
|
| 832 |
+
"grad_norm": 1.5055847688809278,
|
| 833 |
+
"learning_rate": 0.00017677020329680203,
|
| 834 |
+
"loss": 0.9173,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.91,
|
| 839 |
+
"grad_norm": 1.1611583032421355,
|
| 840 |
+
"learning_rate": 0.00017619424750306287,
|
| 841 |
+
"loss": 0.9086,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.91,
|
| 846 |
+
"grad_norm": 1.1353445717142476,
|
| 847 |
+
"learning_rate": 0.00017561220173414297,
|
| 848 |
+
"loss": 0.967,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.92,
|
| 853 |
+
"grad_norm": 0.9563792857101335,
|
| 854 |
+
"learning_rate": 0.00017502411251119586,
|
| 855 |
+
"loss": 0.9155,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.93,
|
| 860 |
+
"grad_norm": 1.1337314999879342,
|
| 861 |
+
"learning_rate": 0.00017443002683841002,
|
| 862 |
+
"loss": 0.8905,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.94,
|
| 867 |
+
"grad_norm": 1.0287944149688226,
|
| 868 |
+
"learning_rate": 0.00017382999219925203,
|
| 869 |
+
"loss": 0.9092,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.94,
|
| 874 |
+
"grad_norm": 1.2022884801513027,
|
| 875 |
+
"learning_rate": 0.00017322405655267122,
|
| 876 |
+
"loss": 0.8703,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.95,
|
| 881 |
+
"grad_norm": 1.2018380484271611,
|
| 882 |
+
"learning_rate": 0.0001726122683292667,
|
| 883 |
+
"loss": 0.9769,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.96,
|
| 888 |
+
"grad_norm": 1.2278521149338097,
|
| 889 |
+
"learning_rate": 0.0001719946764274162,
|
| 890 |
+
"loss": 0.9632,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.97,
|
| 895 |
+
"grad_norm": 1.132615620517034,
|
| 896 |
+
"learning_rate": 0.00017137133020936782,
|
| 897 |
+
"loss": 0.943,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.97,
|
| 902 |
+
"grad_norm": 1.0891584621905475,
|
| 903 |
+
"learning_rate": 0.00017074227949729481,
|
| 904 |
+
"loss": 0.9249,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.98,
|
| 909 |
+
"grad_norm": 1.1417202547948113,
|
| 910 |
+
"learning_rate": 0.00017010757456931334,
|
| 911 |
+
"loss": 0.9055,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.99,
|
| 916 |
+
"grad_norm": 0.9346881604970835,
|
| 917 |
+
"learning_rate": 0.0001694672661554638,
|
| 918 |
+
"loss": 0.9336,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.0,
|
| 923 |
+
"grad_norm": 1.1301994877879067,
|
| 924 |
+
"learning_rate": 0.0001688214054336563,
|
| 925 |
+
"loss": 0.9062,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.0,
|
| 930 |
+
"eval_loss": 1.2441989183425903,
|
| 931 |
+
"eval_runtime": 252.9523,
|
| 932 |
+
"eval_samples_per_second": 9.132,
|
| 933 |
+
"eval_steps_per_second": 0.573,
|
| 934 |
+
"step": 651
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.01,
|
| 938 |
+
"grad_norm": 1.179898601273253,
|
| 939 |
+
"learning_rate": 0.00016817004402558012,
|
| 940 |
+
"loss": 0.9027,
|
| 941 |
+
"step": 655
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.01,
|
| 945 |
+
"grad_norm": 0.9417315541112381,
|
| 946 |
+
"learning_rate": 0.0001675132339925776,
|
| 947 |
+
"loss": 0.9119,
|
| 948 |
+
"step": 660
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.02,
|
| 952 |
+
"grad_norm": 1.0837703721555252,
|
| 953 |
+
"learning_rate": 0.0001668510278314833,
|
| 954 |
+
"loss": 0.86,
|
| 955 |
+
"step": 665
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.03,
|
| 959 |
+
"grad_norm": 1.1272174437541118,
|
| 960 |
+
"learning_rate": 0.00016618347847042778,
|
| 961 |
+
"loss": 0.8711,
|
| 962 |
+
"step": 670
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.04,
|
| 966 |
+
"grad_norm": 1.1359544360875102,
|
| 967 |
+
"learning_rate": 0.00016551063926460748,
|
| 968 |
+
"loss": 0.8776,
|
| 969 |
+
"step": 675
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.04,
|
| 973 |
+
"grad_norm": 1.1917117624970714,
|
| 974 |
+
"learning_rate": 0.00016483256399202006,
|
| 975 |
+
"loss": 0.9209,
|
| 976 |
+
"step": 680
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.05,
|
| 980 |
+
"grad_norm": 1.0249074428213159,
|
| 981 |
+
"learning_rate": 0.00016414930684916613,
|
| 982 |
+
"loss": 0.8196,
|
| 983 |
+
"step": 685
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.06,
|
| 987 |
+
"grad_norm": 1.1816776012865573,
|
| 988 |
+
"learning_rate": 0.00016346092244671746,
|
| 989 |
+
"loss": 0.8279,
|
| 990 |
+
"step": 690
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.07,
|
| 994 |
+
"grad_norm": 1.077779310150086,
|
| 995 |
+
"learning_rate": 0.00016276746580515218,
|
| 996 |
+
"loss": 0.8997,
|
| 997 |
+
"step": 695
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.07,
|
| 1001 |
+
"grad_norm": 1.2009550281356633,
|
| 1002 |
+
"learning_rate": 0.00016206899235035702,
|
| 1003 |
+
"loss": 0.8938,
|
| 1004 |
+
"step": 700
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.08,
|
| 1008 |
+
"grad_norm": 1.0925262506944138,
|
| 1009 |
+
"learning_rate": 0.00016136555790919748,
|
| 1010 |
+
"loss": 0.8856,
|
| 1011 |
+
"step": 705
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.09,
|
| 1015 |
+
"grad_norm": 1.0198591870033458,
|
| 1016 |
+
"learning_rate": 0.0001606572187050556,
|
| 1017 |
+
"loss": 0.8978,
|
| 1018 |
+
"step": 710
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.1,
|
| 1022 |
+
"grad_norm": 1.1119975974231693,
|
| 1023 |
+
"learning_rate": 0.0001599440313533363,
|
| 1024 |
+
"loss": 0.8876,
|
| 1025 |
+
"step": 715
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.11,
|
| 1029 |
+
"grad_norm": 1.190419344020418,
|
| 1030 |
+
"learning_rate": 0.00015922605285694215,
|
| 1031 |
+
"loss": 0.904,
|
| 1032 |
+
"step": 720
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.11,
|
| 1036 |
+
"grad_norm": 2.1639792062638574,
|
| 1037 |
+
"learning_rate": 0.0001585033406017175,
|
| 1038 |
+
"loss": 0.9021,
|
| 1039 |
+
"step": 725
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.12,
|
| 1043 |
+
"grad_norm": 1.2040942925245026,
|
| 1044 |
+
"learning_rate": 0.0001577759523518616,
|
| 1045 |
+
"loss": 0.9092,
|
| 1046 |
+
"step": 730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.13,
|
| 1050 |
+
"grad_norm": 1.0704322799863495,
|
| 1051 |
+
"learning_rate": 0.00015704394624531184,
|
| 1052 |
+
"loss": 0.8274,
|
| 1053 |
+
"step": 735
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.14,
|
| 1057 |
+
"grad_norm": 1.0627822260269795,
|
| 1058 |
+
"learning_rate": 0.00015630738078909685,
|
| 1059 |
+
"loss": 0.8776,
|
| 1060 |
+
"step": 740
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.14,
|
| 1064 |
+
"grad_norm": 0.9905597794182393,
|
| 1065 |
+
"learning_rate": 0.00015556631485466027,
|
| 1066 |
+
"loss": 0.9246,
|
| 1067 |
+
"step": 745
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.15,
|
| 1071 |
+
"grad_norm": 1.173456919024251,
|
| 1072 |
+
"learning_rate": 0.00015482080767315528,
|
| 1073 |
+
"loss": 0.9656,
|
| 1074 |
+
"step": 750
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.16,
|
| 1078 |
+
"grad_norm": 1.0873012836166025,
|
| 1079 |
+
"learning_rate": 0.00015407091883071054,
|
| 1080 |
+
"loss": 0.9464,
|
| 1081 |
+
"step": 755
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.17,
|
| 1085 |
+
"grad_norm": 1.1473043693064626,
|
| 1086 |
+
"learning_rate": 0.00015331670826366754,
|
| 1087 |
+
"loss": 0.8496,
|
| 1088 |
+
"step": 760
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.17,
|
| 1092 |
+
"grad_norm": 0.985606297895901,
|
| 1093 |
+
"learning_rate": 0.00015255823625379017,
|
| 1094 |
+
"loss": 0.853,
|
| 1095 |
+
"step": 765
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.18,
|
| 1099 |
+
"grad_norm": 1.231267953892478,
|
| 1100 |
+
"learning_rate": 0.00015179556342344644,
|
| 1101 |
+
"loss": 0.8652,
|
| 1102 |
+
"step": 770
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.19,
|
| 1106 |
+
"grad_norm": 1.0503811328789359,
|
| 1107 |
+
"learning_rate": 0.00015102875073076324,
|
| 1108 |
+
"loss": 0.9447,
|
| 1109 |
+
"step": 775
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.2,
|
| 1113 |
+
"grad_norm": 1.0561678099502803,
|
| 1114 |
+
"learning_rate": 0.00015025785946475408,
|
| 1115 |
+
"loss": 0.879,
|
| 1116 |
+
"step": 780
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2,
|
| 1120 |
+
"grad_norm": 0.9127623837677151,
|
| 1121 |
+
"learning_rate": 0.00014948295124042057,
|
| 1122 |
+
"loss": 0.9144,
|
| 1123 |
+
"step": 785
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.21,
|
| 1127 |
+
"grad_norm": 1.147282495230838,
|
| 1128 |
+
"learning_rate": 0.00014870408799382752,
|
| 1129 |
+
"loss": 0.9404,
|
| 1130 |
+
"step": 790
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.22,
|
| 1134 |
+
"grad_norm": 1.0608780257484842,
|
| 1135 |
+
"learning_rate": 0.00014792133197715266,
|
| 1136 |
+
"loss": 0.9021,
|
| 1137 |
+
"step": 795
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.23,
|
| 1141 |
+
"grad_norm": 1.2163076965365305,
|
| 1142 |
+
"learning_rate": 0.0001471347457537111,
|
| 1143 |
+
"loss": 0.915,
|
| 1144 |
+
"step": 800
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.24,
|
| 1148 |
+
"grad_norm": 1.1177869922249701,
|
| 1149 |
+
"learning_rate": 0.00014634439219295478,
|
| 1150 |
+
"loss": 0.8648,
|
| 1151 |
+
"step": 805
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.24,
|
| 1155 |
+
"grad_norm": 1.194327396103283,
|
| 1156 |
+
"learning_rate": 0.0001455503344654474,
|
| 1157 |
+
"loss": 0.9526,
|
| 1158 |
+
"step": 810
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.25,
|
| 1162 |
+
"grad_norm": 1.0905268007576805,
|
| 1163 |
+
"learning_rate": 0.00014475263603781554,
|
| 1164 |
+
"loss": 0.8757,
|
| 1165 |
+
"step": 815
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.26,
|
| 1169 |
+
"grad_norm": 0.9152140993127001,
|
| 1170 |
+
"learning_rate": 0.0001439513606676759,
|
| 1171 |
+
"loss": 0.8722,
|
| 1172 |
+
"step": 820
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.27,
|
| 1176 |
+
"grad_norm": 0.9900853991789728,
|
| 1177 |
+
"learning_rate": 0.00014314657239853927,
|
| 1178 |
+
"loss": 0.8669,
|
| 1179 |
+
"step": 825
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.27,
|
| 1183 |
+
"grad_norm": 1.0767926512086485,
|
| 1184 |
+
"learning_rate": 0.000142338335554692,
|
| 1185 |
+
"loss": 0.8841,
|
| 1186 |
+
"step": 830
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.28,
|
| 1190 |
+
"grad_norm": 0.8724281805811482,
|
| 1191 |
+
"learning_rate": 0.00014152671473605428,
|
| 1192 |
+
"loss": 0.826,
|
| 1193 |
+
"step": 835
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.29,
|
| 1197 |
+
"grad_norm": 0.9929237657179188,
|
| 1198 |
+
"learning_rate": 0.0001407117748130174,
|
| 1199 |
+
"loss": 0.8765,
|
| 1200 |
+
"step": 840
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.3,
|
| 1204 |
+
"grad_norm": 1.0329119732814056,
|
| 1205 |
+
"learning_rate": 0.00013989358092125843,
|
| 1206 |
+
"loss": 0.879,
|
| 1207 |
+
"step": 845
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.3,
|
| 1211 |
+
"grad_norm": 0.8699115976888,
|
| 1212 |
+
"learning_rate": 0.00013907219845653442,
|
| 1213 |
+
"loss": 0.8871,
|
| 1214 |
+
"step": 850
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.31,
|
| 1218 |
+
"grad_norm": 2.0311070826655615,
|
| 1219 |
+
"learning_rate": 0.00013824769306945532,
|
| 1220 |
+
"loss": 0.9038,
|
| 1221 |
+
"step": 855
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.32,
|
| 1225 |
+
"grad_norm": 1.2265447591069667,
|
| 1226 |
+
"learning_rate": 0.00013742013066023678,
|
| 1227 |
+
"loss": 0.8918,
|
| 1228 |
+
"step": 860
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.33,
|
| 1232 |
+
"grad_norm": 1.2487739691504016,
|
| 1233 |
+
"learning_rate": 0.00013658957737343298,
|
| 1234 |
+
"loss": 0.8986,
|
| 1235 |
+
"step": 865
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.34,
|
| 1239 |
+
"grad_norm": 0.9999833105305884,
|
| 1240 |
+
"learning_rate": 0.00013575609959264994,
|
| 1241 |
+
"loss": 0.9054,
|
| 1242 |
+
"step": 870
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.34,
|
| 1246 |
+
"grad_norm": 1.0265246633147944,
|
| 1247 |
+
"learning_rate": 0.0001349197639352395,
|
| 1248 |
+
"loss": 0.8781,
|
| 1249 |
+
"step": 875
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.35,
|
| 1253 |
+
"grad_norm": 1.2752835605974824,
|
| 1254 |
+
"learning_rate": 0.00013408063724697499,
|
| 1255 |
+
"loss": 0.9096,
|
| 1256 |
+
"step": 880
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.36,
|
| 1260 |
+
"grad_norm": 1.0841900048671298,
|
| 1261 |
+
"learning_rate": 0.00013323878659670836,
|
| 1262 |
+
"loss": 0.8954,
|
| 1263 |
+
"step": 885
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.37,
|
| 1267 |
+
"grad_norm": 0.9822221386317229,
|
| 1268 |
+
"learning_rate": 0.00013239427927100964,
|
| 1269 |
+
"loss": 0.9197,
|
| 1270 |
+
"step": 890
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.37,
|
| 1274 |
+
"grad_norm": 1.0664898243823369,
|
| 1275 |
+
"learning_rate": 0.00013154718276878872,
|
| 1276 |
+
"loss": 0.8101,
|
| 1277 |
+
"step": 895
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.38,
|
| 1281 |
+
"grad_norm": 0.9728647259678721,
|
| 1282 |
+
"learning_rate": 0.00013069756479590065,
|
| 1283 |
+
"loss": 0.8808,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.39,
|
| 1288 |
+
"grad_norm": 1.233908623068181,
|
| 1289 |
+
"learning_rate": 0.00012984549325973394,
|
| 1290 |
+
"loss": 0.8942,
|
| 1291 |
+
"step": 905
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.4,
|
| 1295 |
+
"grad_norm": 1.0507024710570716,
|
| 1296 |
+
"learning_rate": 0.000128991036263783,
|
| 1297 |
+
"loss": 0.8719,
|
| 1298 |
+
"step": 910
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.4,
|
| 1302 |
+
"grad_norm": 0.9728551705551435,
|
| 1303 |
+
"learning_rate": 0.0001281342621022048,
|
| 1304 |
+
"loss": 0.8734,
|
| 1305 |
+
"step": 915
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.41,
|
| 1309 |
+
"grad_norm": 0.9127829023580948,
|
| 1310 |
+
"learning_rate": 0.00012727523925436026,
|
| 1311 |
+
"loss": 0.8641,
|
| 1312 |
+
"step": 920
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.42,
|
| 1316 |
+
"grad_norm": 1.114858935845165,
|
| 1317 |
+
"learning_rate": 0.00012641403637934112,
|
| 1318 |
+
"loss": 0.8989,
|
| 1319 |
+
"step": 925
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.43,
|
| 1323 |
+
"grad_norm": 1.0021048323600934,
|
| 1324 |
+
"learning_rate": 0.00012555072231048192,
|
| 1325 |
+
"loss": 0.8757,
|
| 1326 |
+
"step": 930
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.44,
|
| 1330 |
+
"grad_norm": 0.9850639243326992,
|
| 1331 |
+
"learning_rate": 0.00012468536604985867,
|
| 1332 |
+
"loss": 0.8595,
|
| 1333 |
+
"step": 935
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.44,
|
| 1337 |
+
"grad_norm": 0.8816979962904702,
|
| 1338 |
+
"learning_rate": 0.00012381803676277345,
|
| 1339 |
+
"loss": 0.8854,
|
| 1340 |
+
"step": 940
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.45,
|
| 1344 |
+
"grad_norm": 1.052381113604825,
|
| 1345 |
+
"learning_rate": 0.00012294880377222649,
|
| 1346 |
+
"loss": 0.8966,
|
| 1347 |
+
"step": 945
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.46,
|
| 1351 |
+
"grad_norm": 0.9471916607767124,
|
| 1352 |
+
"learning_rate": 0.0001220777365533751,
|
| 1353 |
+
"loss": 0.8977,
|
| 1354 |
+
"step": 950
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.47,
|
| 1358 |
+
"grad_norm": 0.9648558401941475,
|
| 1359 |
+
"learning_rate": 0.00012120490472798112,
|
| 1360 |
+
"loss": 0.8837,
|
| 1361 |
+
"step": 955
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.47,
|
| 1365 |
+
"grad_norm": 0.9430401800136761,
|
| 1366 |
+
"learning_rate": 0.0001203303780588458,
|
| 1367 |
+
"loss": 0.9009,
|
| 1368 |
+
"step": 960
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.48,
|
| 1372 |
+
"grad_norm": 0.9624197502185033,
|
| 1373 |
+
"learning_rate": 0.00011945422644423425,
|
| 1374 |
+
"loss": 0.8645,
|
| 1375 |
+
"step": 965
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.49,
|
| 1379 |
+
"grad_norm": 0.9260810941977449,
|
| 1380 |
+
"learning_rate": 0.00011857651991228855,
|
| 1381 |
+
"loss": 0.8243,
|
| 1382 |
+
"step": 970
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.5,
|
| 1386 |
+
"grad_norm": 0.9362374564085868,
|
| 1387 |
+
"learning_rate": 0.00011769732861543057,
|
| 1388 |
+
"loss": 0.886,
|
| 1389 |
+
"step": 975
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.5,
|
| 1393 |
+
"grad_norm": 0.9293220698915177,
|
| 1394 |
+
"learning_rate": 0.00011681672282475495,
|
| 1395 |
+
"loss": 0.9028,
|
| 1396 |
+
"step": 980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.51,
|
| 1400 |
+
"grad_norm": 0.9026227086147156,
|
| 1401 |
+
"learning_rate": 0.00011593477292441251,
|
| 1402 |
+
"loss": 0.8253,
|
| 1403 |
+
"step": 985
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.52,
|
| 1407 |
+
"grad_norm": 0.9630673700134131,
|
| 1408 |
+
"learning_rate": 0.00011505154940598468,
|
| 1409 |
+
"loss": 0.8686,
|
| 1410 |
+
"step": 990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.53,
|
| 1414 |
+
"grad_norm": 0.8694549872905225,
|
| 1415 |
+
"learning_rate": 0.00011416712286284943,
|
| 1416 |
+
"loss": 0.8782,
|
| 1417 |
+
"step": 995
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.53,
|
| 1421 |
+
"grad_norm": 1.007303227993832,
|
| 1422 |
+
"learning_rate": 0.00011328156398453864,
|
| 1423 |
+
"loss": 0.8633,
|
| 1424 |
+
"step": 1000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.54,
|
| 1428 |
+
"grad_norm": 0.9083011875988664,
|
| 1429 |
+
"learning_rate": 0.00011239494355108848,
|
| 1430 |
+
"loss": 0.9039,
|
| 1431 |
+
"step": 1005
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.55,
|
| 1435 |
+
"grad_norm": 1.0177463749412752,
|
| 1436 |
+
"learning_rate": 0.00011150733242738198,
|
| 1437 |
+
"loss": 0.9029,
|
| 1438 |
+
"step": 1010
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.56,
|
| 1442 |
+
"grad_norm": 0.8631790225704151,
|
| 1443 |
+
"learning_rate": 0.00011061880155748497,
|
| 1444 |
+
"loss": 0.8385,
|
| 1445 |
+
"step": 1015
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.57,
|
| 1449 |
+
"grad_norm": 1.5230132007370771,
|
| 1450 |
+
"learning_rate": 0.00010972942195897582,
|
| 1451 |
+
"loss": 0.9055,
|
| 1452 |
+
"step": 1020
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.57,
|
| 1456 |
+
"grad_norm": 0.8997463696925398,
|
| 1457 |
+
"learning_rate": 0.00010883926471726926,
|
| 1458 |
+
"loss": 0.8656,
|
| 1459 |
+
"step": 1025
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.58,
|
| 1463 |
+
"grad_norm": 0.9377827912335661,
|
| 1464 |
+
"learning_rate": 0.00010794840097993466,
|
| 1465 |
+
"loss": 0.9163,
|
| 1466 |
+
"step": 1030
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.59,
|
| 1470 |
+
"grad_norm": 0.9887953595089554,
|
| 1471 |
+
"learning_rate": 0.00010705690195100939,
|
| 1472 |
+
"loss": 0.8789,
|
| 1473 |
+
"step": 1035
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.6,
|
| 1477 |
+
"grad_norm": 1.0514798236504352,
|
| 1478 |
+
"learning_rate": 0.00010616483888530781,
|
| 1479 |
+
"loss": 0.9027,
|
| 1480 |
+
"step": 1040
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.6,
|
| 1484 |
+
"grad_norm": 0.9297004988976107,
|
| 1485 |
+
"learning_rate": 0.00010527228308272605,
|
| 1486 |
+
"loss": 0.9473,
|
| 1487 |
+
"step": 1045
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.61,
|
| 1491 |
+
"grad_norm": 1.079573310483164,
|
| 1492 |
+
"learning_rate": 0.0001043793058825431,
|
| 1493 |
+
"loss": 0.8308,
|
| 1494 |
+
"step": 1050
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.62,
|
| 1498 |
+
"grad_norm": 0.8437863342382683,
|
| 1499 |
+
"learning_rate": 0.00010348597865771909,
|
| 1500 |
+
"loss": 0.9183,
|
| 1501 |
+
"step": 1055
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.63,
|
| 1505 |
+
"grad_norm": 1.0314459958872142,
|
| 1506 |
+
"learning_rate": 0.00010259237280919054,
|
| 1507 |
+
"loss": 0.8965,
|
| 1508 |
+
"step": 1060
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.63,
|
| 1512 |
+
"grad_norm": 1.1300004513750828,
|
| 1513 |
+
"learning_rate": 0.00010169855976016345,
|
| 1514 |
+
"loss": 0.9058,
|
| 1515 |
+
"step": 1065
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.64,
|
| 1519 |
+
"grad_norm": 0.915751139105661,
|
| 1520 |
+
"learning_rate": 0.00010080461095040476,
|
| 1521 |
+
"loss": 0.8522,
|
| 1522 |
+
"step": 1070
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.65,
|
| 1526 |
+
"grad_norm": 0.9282214710118697,
|
| 1527 |
+
"learning_rate": 9.991059783053244e-05,
|
| 1528 |
+
"loss": 0.893,
|
| 1529 |
+
"step": 1075
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.66,
|
| 1533 |
+
"grad_norm": 0.9267500154050946,
|
| 1534 |
+
"learning_rate": 9.901659185630445e-05,
|
| 1535 |
+
"loss": 0.9187,
|
| 1536 |
+
"step": 1080
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.67,
|
| 1540 |
+
"grad_norm": 0.9851835947209288,
|
| 1541 |
+
"learning_rate": 9.812266448290767e-05,
|
| 1542 |
+
"loss": 0.8489,
|
| 1543 |
+
"step": 1085
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.67,
|
| 1547 |
+
"grad_norm": 0.9072684185548759,
|
| 1548 |
+
"learning_rate": 9.722888715924664e-05,
|
| 1549 |
+
"loss": 0.8598,
|
| 1550 |
+
"step": 1090
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.68,
|
| 1554 |
+
"grad_norm": 1.0793886242538702,
|
| 1555 |
+
"learning_rate": 9.633533132223293e-05,
|
| 1556 |
+
"loss": 0.9136,
|
| 1557 |
+
"step": 1095
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.69,
|
| 1561 |
+
"grad_norm": 0.9100627615470746,
|
| 1562 |
+
"learning_rate": 9.54420683910753e-05,
|
| 1563 |
+
"loss": 0.894,
|
| 1564 |
+
"step": 1100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.7,
|
| 1568 |
+
"grad_norm": 1.053250149548252,
|
| 1569 |
+
"learning_rate": 9.454916976157144e-05,
|
| 1570 |
+
"loss": 0.8604,
|
| 1571 |
+
"step": 1105
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.7,
|
| 1575 |
+
"grad_norm": 1.054945196161409,
|
| 1576 |
+
"learning_rate": 9.365670680040157e-05,
|
| 1577 |
+
"loss": 0.8875,
|
| 1578 |
+
"step": 1110
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.71,
|
| 1582 |
+
"grad_norm": 1.0344061549877617,
|
| 1583 |
+
"learning_rate": 9.276475083942416e-05,
|
| 1584 |
+
"loss": 0.8612,
|
| 1585 |
+
"step": 1115
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.72,
|
| 1589 |
+
"grad_norm": 1.034253018495178,
|
| 1590 |
+
"learning_rate": 9.187337316997476e-05,
|
| 1591 |
+
"loss": 0.8901,
|
| 1592 |
+
"step": 1120
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.73,
|
| 1596 |
+
"grad_norm": 0.9350006681122276,
|
| 1597 |
+
"learning_rate": 9.09826450371678e-05,
|
| 1598 |
+
"loss": 0.8913,
|
| 1599 |
+
"step": 1125
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.73,
|
| 1603 |
+
"grad_norm": 1.1854492776108583,
|
| 1604 |
+
"learning_rate": 9.009263763420228e-05,
|
| 1605 |
+
"loss": 0.9029,
|
| 1606 |
+
"step": 1130
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.74,
|
| 1610 |
+
"grad_norm": 1.0313744595376304,
|
| 1611 |
+
"learning_rate": 8.920342209667136e-05,
|
| 1612 |
+
"loss": 0.8399,
|
| 1613 |
+
"step": 1135
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.75,
|
| 1617 |
+
"grad_norm": 1.109169472722535,
|
| 1618 |
+
"learning_rate": 8.831506949687685e-05,
|
| 1619 |
+
"loss": 0.8517,
|
| 1620 |
+
"step": 1140
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.76,
|
| 1624 |
+
"grad_norm": 0.9979711332157661,
|
| 1625 |
+
"learning_rate": 8.74276508381486e-05,
|
| 1626 |
+
"loss": 0.8773,
|
| 1627 |
+
"step": 1145
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.77,
|
| 1631 |
+
"grad_norm": 0.9829724669057092,
|
| 1632 |
+
"learning_rate": 8.654123704916927e-05,
|
| 1633 |
+
"loss": 0.879,
|
| 1634 |
+
"step": 1150
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.77,
|
| 1638 |
+
"grad_norm": 0.99994654751558,
|
| 1639 |
+
"learning_rate": 8.565589897830543e-05,
|
| 1640 |
+
"loss": 0.8523,
|
| 1641 |
+
"step": 1155
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.78,
|
| 1645 |
+
"grad_norm": 0.9478833521675312,
|
| 1646 |
+
"learning_rate": 8.47717073879447e-05,
|
| 1647 |
+
"loss": 0.8541,
|
| 1648 |
+
"step": 1160
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.79,
|
| 1652 |
+
"grad_norm": 0.9034620075081865,
|
| 1653 |
+
"learning_rate": 8.388873294884e-05,
|
| 1654 |
+
"loss": 0.8918,
|
| 1655 |
+
"step": 1165
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.8,
|
| 1659 |
+
"grad_norm": 0.8723934369993729,
|
| 1660 |
+
"learning_rate": 8.300704623446111e-05,
|
| 1661 |
+
"loss": 0.9133,
|
| 1662 |
+
"step": 1170
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.8,
|
| 1666 |
+
"grad_norm": 0.9525657563232377,
|
| 1667 |
+
"learning_rate": 8.212671771535379e-05,
|
| 1668 |
+
"loss": 0.8997,
|
| 1669 |
+
"step": 1175
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.81,
|
| 1673 |
+
"grad_norm": 0.9812331418181283,
|
| 1674 |
+
"learning_rate": 8.124781775350741e-05,
|
| 1675 |
+
"loss": 0.8877,
|
| 1676 |
+
"step": 1180
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.82,
|
| 1680 |
+
"grad_norm": 1.0880452788188242,
|
| 1681 |
+
"learning_rate": 8.037041659673105e-05,
|
| 1682 |
+
"loss": 0.9202,
|
| 1683 |
+
"step": 1185
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.83,
|
| 1687 |
+
"grad_norm": 1.097063406936711,
|
| 1688 |
+
"learning_rate": 7.949458437303891e-05,
|
| 1689 |
+
"loss": 0.9068,
|
| 1690 |
+
"step": 1190
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.83,
|
| 1694 |
+
"grad_norm": 0.816681343841031,
|
| 1695 |
+
"learning_rate": 7.862039108504513e-05,
|
| 1696 |
+
"loss": 0.8774,
|
| 1697 |
+
"step": 1195
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.84,
|
| 1701 |
+
"grad_norm": 1.0107773305774448,
|
| 1702 |
+
"learning_rate": 7.774790660436858e-05,
|
| 1703 |
+
"loss": 0.8973,
|
| 1704 |
+
"step": 1200
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.85,
|
| 1708 |
+
"grad_norm": 1.151927912306129,
|
| 1709 |
+
"learning_rate": 7.687720066604844e-05,
|
| 1710 |
+
"loss": 0.8857,
|
| 1711 |
+
"step": 1205
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.86,
|
| 1715 |
+
"grad_norm": 0.7574607659244479,
|
| 1716 |
+
"learning_rate": 7.600834286297035e-05,
|
| 1717 |
+
"loss": 0.8681,
|
| 1718 |
+
"step": 1210
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.86,
|
| 1722 |
+
"grad_norm": 0.993971472764409,
|
| 1723 |
+
"learning_rate": 7.514140264030413e-05,
|
| 1724 |
+
"loss": 0.9421,
|
| 1725 |
+
"step": 1215
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.87,
|
| 1729 |
+
"grad_norm": 1.1317554618491943,
|
| 1730 |
+
"learning_rate": 7.427644928995326e-05,
|
| 1731 |
+
"loss": 0.9151,
|
| 1732 |
+
"step": 1220
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.88,
|
| 1736 |
+
"grad_norm": 0.9946380695782382,
|
| 1737 |
+
"learning_rate": 7.341355194501638e-05,
|
| 1738 |
+
"loss": 0.9331,
|
| 1739 |
+
"step": 1225
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.89,
|
| 1743 |
+
"grad_norm": 0.9307204785839222,
|
| 1744 |
+
"learning_rate": 7.2552779574262e-05,
|
| 1745 |
+
"loss": 0.9324,
|
| 1746 |
+
"step": 1230
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.9,
|
| 1750 |
+
"grad_norm": 0.8165240509268185,
|
| 1751 |
+
"learning_rate": 7.16942009766159e-05,
|
| 1752 |
+
"loss": 0.8972,
|
| 1753 |
+
"step": 1235
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.9,
|
| 1757 |
+
"grad_norm": 1.0160286848183477,
|
| 1758 |
+
"learning_rate": 7.083788477566206e-05,
|
| 1759 |
+
"loss": 0.888,
|
| 1760 |
+
"step": 1240
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.91,
|
| 1764 |
+
"grad_norm": 0.9588077943326088,
|
| 1765 |
+
"learning_rate": 6.998389941415811e-05,
|
| 1766 |
+
"loss": 0.8776,
|
| 1767 |
+
"step": 1245
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.92,
|
| 1771 |
+
"grad_norm": 1.0027507311971422,
|
| 1772 |
+
"learning_rate": 6.913231314856467e-05,
|
| 1773 |
+
"loss": 0.892,
|
| 1774 |
+
"step": 1250
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.93,
|
| 1778 |
+
"grad_norm": 0.8157424560652131,
|
| 1779 |
+
"learning_rate": 6.828319404358998e-05,
|
| 1780 |
+
"loss": 0.8611,
|
| 1781 |
+
"step": 1255
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.93,
|
| 1785 |
+
"grad_norm": 0.9346949667365426,
|
| 1786 |
+
"learning_rate": 6.74366099667495e-05,
|
| 1787 |
+
"loss": 0.9041,
|
| 1788 |
+
"step": 1260
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.94,
|
| 1792 |
+
"grad_norm": 0.9310369545608014,
|
| 1793 |
+
"learning_rate": 6.659262858294167e-05,
|
| 1794 |
+
"loss": 0.8348,
|
| 1795 |
+
"step": 1265
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.95,
|
| 1799 |
+
"grad_norm": 0.916679588029739,
|
| 1800 |
+
"learning_rate": 6.575131734903952e-05,
|
| 1801 |
+
"loss": 0.8665,
|
| 1802 |
+
"step": 1270
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.96,
|
| 1806 |
+
"grad_norm": 0.8335818591855719,
|
| 1807 |
+
"learning_rate": 6.491274350849914e-05,
|
| 1808 |
+
"loss": 0.8892,
|
| 1809 |
+
"step": 1275
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.96,
|
| 1813 |
+
"grad_norm": 0.9831047716703826,
|
| 1814 |
+
"learning_rate": 6.407697408598497e-05,
|
| 1815 |
+
"loss": 0.8944,
|
| 1816 |
+
"step": 1280
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.97,
|
| 1820 |
+
"grad_norm": 0.9268335345851625,
|
| 1821 |
+
"learning_rate": 6.324407588201292e-05,
|
| 1822 |
+
"loss": 0.8536,
|
| 1823 |
+
"step": 1285
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.98,
|
| 1827 |
+
"grad_norm": 1.0086597114412514,
|
| 1828 |
+
"learning_rate": 6.241411546761109e-05,
|
| 1829 |
+
"loss": 0.8983,
|
| 1830 |
+
"step": 1290
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.99,
|
| 1834 |
+
"grad_norm": 0.9455374773689034,
|
| 1835 |
+
"learning_rate": 6.158715917899893e-05,
|
| 1836 |
+
"loss": 0.8638,
|
| 1837 |
+
"step": 1295
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.0,
|
| 1841 |
+
"grad_norm": 0.9581394982958279,
|
| 1842 |
+
"learning_rate": 6.076327311228522e-05,
|
| 1843 |
+
"loss": 0.907,
|
| 1844 |
+
"step": 1300
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.0,
|
| 1848 |
+
"eval_loss": 1.1707841157913208,
|
| 1849 |
+
"eval_runtime": 252.7994,
|
| 1850 |
+
"eval_samples_per_second": 9.138,
|
| 1851 |
+
"eval_steps_per_second": 0.574,
|
| 1852 |
+
"step": 1303
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 2.0,
|
| 1856 |
+
"grad_norm": 1.0117168856081882,
|
| 1857 |
+
"learning_rate": 5.99425231181853e-05,
|
| 1858 |
+
"loss": 0.8658,
|
| 1859 |
+
"step": 1305
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 2.01,
|
| 1863 |
+
"grad_norm": 0.8313758221887009,
|
| 1864 |
+
"learning_rate": 5.9124974796757614e-05,
|
| 1865 |
+
"loss": 0.803,
|
| 1866 |
+
"step": 1310
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 2.02,
|
| 1870 |
+
"grad_norm": 1.1429742233636946,
|
| 1871 |
+
"learning_rate": 5.831069349216069e-05,
|
| 1872 |
+
"loss": 0.854,
|
| 1873 |
+
"step": 1315
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 2.03,
|
| 1877 |
+
"grad_norm": 1.2005681342348582,
|
| 1878 |
+
"learning_rate": 5.7499744287430366e-05,
|
| 1879 |
+
"loss": 0.8209,
|
| 1880 |
+
"step": 1320
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.03,
|
| 1884 |
+
"grad_norm": 1.0882362071151326,
|
| 1885 |
+
"learning_rate": 5.6692191999277614e-05,
|
| 1886 |
+
"loss": 0.8182,
|
| 1887 |
+
"step": 1325
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 2.04,
|
| 1891 |
+
"grad_norm": 1.0316332870317524,
|
| 1892 |
+
"learning_rate": 5.588810117290843e-05,
|
| 1893 |
+
"loss": 0.888,
|
| 1894 |
+
"step": 1330
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 2.05,
|
| 1898 |
+
"grad_norm": 0.8415700670907011,
|
| 1899 |
+
"learning_rate": 5.508753607686452e-05,
|
| 1900 |
+
"loss": 0.8274,
|
| 1901 |
+
"step": 1335
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.06,
|
| 1905 |
+
"grad_norm": 1.19035695841182,
|
| 1906 |
+
"learning_rate": 5.429056069788663e-05,
|
| 1907 |
+
"loss": 0.8587,
|
| 1908 |
+
"step": 1340
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 2.06,
|
| 1912 |
+
"grad_norm": 0.8282530747648139,
|
| 1913 |
+
"learning_rate": 5.3497238735800456e-05,
|
| 1914 |
+
"loss": 0.8582,
|
| 1915 |
+
"step": 1345
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 2.07,
|
| 1919 |
+
"grad_norm": 0.8236642419616373,
|
| 1920 |
+
"learning_rate": 5.2707633598425023e-05,
|
| 1921 |
+
"loss": 0.8242,
|
| 1922 |
+
"step": 1350
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 2.08,
|
| 1926 |
+
"grad_norm": 0.9499844942270589,
|
| 1927 |
+
"learning_rate": 5.192180839650482e-05,
|
| 1928 |
+
"loss": 0.8419,
|
| 1929 |
+
"step": 1355
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.09,
|
| 1933 |
+
"grad_norm": 0.9665172203162608,
|
| 1934 |
+
"learning_rate": 5.1139825938665706e-05,
|
| 1935 |
+
"loss": 0.8168,
|
| 1936 |
+
"step": 1360
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.1,
|
| 1940 |
+
"grad_norm": 0.9570953639965174,
|
| 1941 |
+
"learning_rate": 5.036174872639443e-05,
|
| 1942 |
+
"loss": 0.7975,
|
| 1943 |
+
"step": 1365
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 2.1,
|
| 1947 |
+
"grad_norm": 1.1769715973472896,
|
| 1948 |
+
"learning_rate": 4.95876389490435e-05,
|
| 1949 |
+
"loss": 0.8803,
|
| 1950 |
+
"step": 1370
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 2.11,
|
| 1954 |
+
"grad_norm": 1.0737767072312128,
|
| 1955 |
+
"learning_rate": 4.8817558478860316e-05,
|
| 1956 |
+
"loss": 0.8392,
|
| 1957 |
+
"step": 1375
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 2.12,
|
| 1961 |
+
"grad_norm": 1.0356976294210105,
|
| 1962 |
+
"learning_rate": 4.805156886604192e-05,
|
| 1963 |
+
"loss": 0.8427,
|
| 1964 |
+
"step": 1380
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 2.13,
|
| 1968 |
+
"grad_norm": 0.9731310972525462,
|
| 1969 |
+
"learning_rate": 4.728973133381557e-05,
|
| 1970 |
+
"loss": 0.8422,
|
| 1971 |
+
"step": 1385
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 2.13,
|
| 1975 |
+
"grad_norm": 1.0915183222326856,
|
| 1976 |
+
"learning_rate": 4.6532106773545356e-05,
|
| 1977 |
+
"loss": 0.8002,
|
| 1978 |
+
"step": 1390
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.14,
|
| 1982 |
+
"grad_norm": 1.0419392375432739,
|
| 1983 |
+
"learning_rate": 4.5778755739865234e-05,
|
| 1984 |
+
"loss": 0.8035,
|
| 1985 |
+
"step": 1395
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.15,
|
| 1989 |
+
"grad_norm": 1.034845804276428,
|
| 1990 |
+
"learning_rate": 4.5029738445839143e-05,
|
| 1991 |
+
"loss": 0.8633,
|
| 1992 |
+
"step": 1400
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.16,
|
| 1996 |
+
"grad_norm": 0.9808221912256501,
|
| 1997 |
+
"learning_rate": 4.4285114758148385e-05,
|
| 1998 |
+
"loss": 0.8238,
|
| 1999 |
+
"step": 1405
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 2.16,
|
| 2003 |
+
"grad_norm": 0.9518518180660357,
|
| 2004 |
+
"learning_rate": 4.3544944192306536e-05,
|
| 2005 |
+
"loss": 0.821,
|
| 2006 |
+
"step": 1410
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 2.17,
|
| 2010 |
+
"grad_norm": 1.0002459852548866,
|
| 2011 |
+
"learning_rate": 4.2809285907902804e-05,
|
| 2012 |
+
"loss": 0.8153,
|
| 2013 |
+
"step": 1415
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 2.18,
|
| 2017 |
+
"grad_norm": 0.9090185258143434,
|
| 2018 |
+
"learning_rate": 4.207819870387331e-05,
|
| 2019 |
+
"loss": 0.8724,
|
| 2020 |
+
"step": 1420
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 2.19,
|
| 2024 |
+
"grad_norm": 0.8794828030312869,
|
| 2025 |
+
"learning_rate": 4.135174101380154e-05,
|
| 2026 |
+
"loss": 0.7995,
|
| 2027 |
+
"step": 1425
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.19,
|
| 2031 |
+
"grad_norm": 0.8426944935252626,
|
| 2032 |
+
"learning_rate": 4.0629970901248125e-05,
|
| 2033 |
+
"loss": 0.878,
|
| 2034 |
+
"step": 1430
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 2.2,
|
| 2038 |
+
"grad_norm": 1.0954146157200644,
|
| 2039 |
+
"learning_rate": 3.991294605510969e-05,
|
| 2040 |
+
"loss": 0.8605,
|
| 2041 |
+
"step": 1435
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 2.21,
|
| 2045 |
+
"grad_norm": 1.0887405499363092,
|
| 2046 |
+
"learning_rate": 3.920072378500814e-05,
|
| 2047 |
+
"loss": 0.853,
|
| 2048 |
+
"step": 1440
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 2.22,
|
| 2052 |
+
"grad_norm": 1.0547395334098018,
|
| 2053 |
+
"learning_rate": 3.849336101671015e-05,
|
| 2054 |
+
"loss": 0.8921,
|
| 2055 |
+
"step": 1445
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.23,
|
| 2059 |
+
"grad_norm": 1.069888150135721,
|
| 2060 |
+
"learning_rate": 3.779091428757692e-05,
|
| 2061 |
+
"loss": 0.8161,
|
| 2062 |
+
"step": 1450
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 2.23,
|
| 2066 |
+
"grad_norm": 0.9353341753814841,
|
| 2067 |
+
"learning_rate": 3.709343974204577e-05,
|
| 2068 |
+
"loss": 0.8179,
|
| 2069 |
+
"step": 1455
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.24,
|
| 2073 |
+
"grad_norm": 1.0108851490383193,
|
| 2074 |
+
"learning_rate": 3.640099312714235e-05,
|
| 2075 |
+
"loss": 0.8385,
|
| 2076 |
+
"step": 1460
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 2.25,
|
| 2080 |
+
"grad_norm": 1.064940293841458,
|
| 2081 |
+
"learning_rate": 3.5713629788025036e-05,
|
| 2082 |
+
"loss": 0.8135,
|
| 2083 |
+
"step": 1465
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 2.26,
|
| 2087 |
+
"grad_norm": 1.1487280639099855,
|
| 2088 |
+
"learning_rate": 3.503140466356151e-05,
|
| 2089 |
+
"loss": 0.8021,
|
| 2090 |
+
"step": 1470
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.26,
|
| 2094 |
+
"grad_norm": 0.9701769325400604,
|
| 2095 |
+
"learning_rate": 3.435437228193741e-05,
|
| 2096 |
+
"loss": 0.8399,
|
| 2097 |
+
"step": 1475
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 2.27,
|
| 2101 |
+
"grad_norm": 0.9942305006661968,
|
| 2102 |
+
"learning_rate": 3.3682586756298185e-05,
|
| 2103 |
+
"loss": 0.8427,
|
| 2104 |
+
"step": 1480
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 2.28,
|
| 2108 |
+
"grad_norm": 1.0099321403910597,
|
| 2109 |
+
"learning_rate": 3.3016101780424146e-05,
|
| 2110 |
+
"loss": 0.8693,
|
| 2111 |
+
"step": 1485
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 2.29,
|
| 2115 |
+
"grad_norm": 1.169305983316981,
|
| 2116 |
+
"learning_rate": 3.235497062443852e-05,
|
| 2117 |
+
"loss": 0.8545,
|
| 2118 |
+
"step": 1490
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 2.29,
|
| 2122 |
+
"grad_norm": 0.9289433553993577,
|
| 2123 |
+
"learning_rate": 3.169924613055003e-05,
|
| 2124 |
+
"loss": 0.7912,
|
| 2125 |
+
"step": 1495
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 2.3,
|
| 2129 |
+
"grad_norm": 1.1184033184773026,
|
| 2130 |
+
"learning_rate": 3.10489807088294e-05,
|
| 2131 |
+
"loss": 0.8488,
|
| 2132 |
+
"step": 1500
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 2.31,
|
| 2136 |
+
"grad_norm": 0.8651320149450171,
|
| 2137 |
+
"learning_rate": 3.0404226333020114e-05,
|
| 2138 |
+
"loss": 0.8063,
|
| 2139 |
+
"step": 1505
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 2.32,
|
| 2143 |
+
"grad_norm": 1.015380797479418,
|
| 2144 |
+
"learning_rate": 2.976503453638452e-05,
|
| 2145 |
+
"loss": 0.8281,
|
| 2146 |
+
"step": 1510
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 2.33,
|
| 2150 |
+
"grad_norm": 0.9105715510091448,
|
| 2151 |
+
"learning_rate": 2.9131456407584912e-05,
|
| 2152 |
+
"loss": 0.8396,
|
| 2153 |
+
"step": 1515
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 2.33,
|
| 2157 |
+
"grad_norm": 0.8953944449056994,
|
| 2158 |
+
"learning_rate": 2.8503542586600095e-05,
|
| 2159 |
+
"loss": 0.8059,
|
| 2160 |
+
"step": 1520
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 2.34,
|
| 2164 |
+
"grad_norm": 0.9445583773531011,
|
| 2165 |
+
"learning_rate": 2.7881343260677938e-05,
|
| 2166 |
+
"loss": 0.7555,
|
| 2167 |
+
"step": 1525
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 2.35,
|
| 2171 |
+
"grad_norm": 0.9760863299166253,
|
| 2172 |
+
"learning_rate": 2.7264908160324044e-05,
|
| 2173 |
+
"loss": 0.872,
|
| 2174 |
+
"step": 1530
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 2.36,
|
| 2178 |
+
"grad_norm": 0.9217062841634305,
|
| 2179 |
+
"learning_rate": 2.66542865553269e-05,
|
| 2180 |
+
"loss": 0.8772,
|
| 2181 |
+
"step": 1535
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 2.36,
|
| 2185 |
+
"grad_norm": 0.9862000270745448,
|
| 2186 |
+
"learning_rate": 2.6049527250820048e-05,
|
| 2187 |
+
"loss": 0.8042,
|
| 2188 |
+
"step": 1540
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 2.37,
|
| 2192 |
+
"grad_norm": 0.9372789886762245,
|
| 2193 |
+
"learning_rate": 2.5450678583381037e-05,
|
| 2194 |
+
"loss": 0.8373,
|
| 2195 |
+
"step": 1545
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 2.38,
|
| 2199 |
+
"grad_norm": 0.9849583322670422,
|
| 2200 |
+
"learning_rate": 2.4857788417168082e-05,
|
| 2201 |
+
"loss": 0.8449,
|
| 2202 |
+
"step": 1550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 2.39,
|
| 2206 |
+
"grad_norm": 0.9531252163149414,
|
| 2207 |
+
"learning_rate": 2.4270904140094597e-05,
|
| 2208 |
+
"loss": 0.8204,
|
| 2209 |
+
"step": 1555
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 2.39,
|
| 2213 |
+
"grad_norm": 1.1160019622624047,
|
| 2214 |
+
"learning_rate": 2.3690072660041373e-05,
|
| 2215 |
+
"loss": 0.857,
|
| 2216 |
+
"step": 1560
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 2.4,
|
| 2220 |
+
"grad_norm": 1.0574865907973916,
|
| 2221 |
+
"learning_rate": 2.3115340401107487e-05,
|
| 2222 |
+
"loss": 0.8154,
|
| 2223 |
+
"step": 1565
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 2.41,
|
| 2227 |
+
"grad_norm": 1.0014445052570997,
|
| 2228 |
+
"learning_rate": 2.254675329989988e-05,
|
| 2229 |
+
"loss": 0.8526,
|
| 2230 |
+
"step": 1570
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 2.42,
|
| 2234 |
+
"grad_norm": 1.1337662282318417,
|
| 2235 |
+
"learning_rate": 2.1984356801861506e-05,
|
| 2236 |
+
"loss": 0.7529,
|
| 2237 |
+
"step": 1575
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 2.43,
|
| 2241 |
+
"grad_norm": 1.0240125708817747,
|
| 2242 |
+
"learning_rate": 2.1428195857639256e-05,
|
| 2243 |
+
"loss": 0.8252,
|
| 2244 |
+
"step": 1580
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 2.43,
|
| 2248 |
+
"grad_norm": 0.8777879796341344,
|
| 2249 |
+
"learning_rate": 2.0878314919491183e-05,
|
| 2250 |
+
"loss": 0.8349,
|
| 2251 |
+
"step": 1585
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 2.44,
|
| 2255 |
+
"grad_norm": 0.9749914994991105,
|
| 2256 |
+
"learning_rate": 2.0334757937733374e-05,
|
| 2257 |
+
"loss": 0.8147,
|
| 2258 |
+
"step": 1590
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 2.45,
|
| 2262 |
+
"grad_norm": 1.0047182891795483,
|
| 2263 |
+
"learning_rate": 1.9797568357227293e-05,
|
| 2264 |
+
"loss": 0.8225,
|
| 2265 |
+
"step": 1595
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 2.46,
|
| 2269 |
+
"grad_norm": 1.102505598943093,
|
| 2270 |
+
"learning_rate": 1.92667891139074e-05,
|
| 2271 |
+
"loss": 0.8356,
|
| 2272 |
+
"step": 1600
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 2.46,
|
| 2276 |
+
"grad_norm": 0.9052644924404957,
|
| 2277 |
+
"learning_rate": 1.8742462631349246e-05,
|
| 2278 |
+
"loss": 0.8509,
|
| 2279 |
+
"step": 1605
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 2.47,
|
| 2283 |
+
"grad_norm": 0.9927305294402076,
|
| 2284 |
+
"learning_rate": 1.822463081737883e-05,
|
| 2285 |
+
"loss": 0.8182,
|
| 2286 |
+
"step": 1610
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 2.48,
|
| 2290 |
+
"grad_norm": 1.0563552318088962,
|
| 2291 |
+
"learning_rate": 1.7713335060722946e-05,
|
| 2292 |
+
"loss": 0.7578,
|
| 2293 |
+
"step": 1615
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 2.49,
|
| 2297 |
+
"grad_norm": 1.0170016112678426,
|
| 2298 |
+
"learning_rate": 1.720861622770116e-05,
|
| 2299 |
+
"loss": 0.7562,
|
| 2300 |
+
"step": 1620
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 2.49,
|
| 2304 |
+
"grad_norm": 0.9404587306406071,
|
| 2305 |
+
"learning_rate": 1.671051465895953e-05,
|
| 2306 |
+
"loss": 0.786,
|
| 2307 |
+
"step": 1625
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 2.5,
|
| 2311 |
+
"grad_norm": 1.0171241452919628,
|
| 2312 |
+
"learning_rate": 1.6219070166246154e-05,
|
| 2313 |
+
"loss": 0.8616,
|
| 2314 |
+
"step": 1630
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 2.51,
|
| 2318 |
+
"grad_norm": 1.0322115593813674,
|
| 2319 |
+
"learning_rate": 1.5734322029229253e-05,
|
| 2320 |
+
"loss": 0.8592,
|
| 2321 |
+
"step": 1635
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.52,
|
| 2325 |
+
"grad_norm": 1.0192391783063095,
|
| 2326 |
+
"learning_rate": 1.5256308992357716e-05,
|
| 2327 |
+
"loss": 0.8372,
|
| 2328 |
+
"step": 1640
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 2.52,
|
| 2332 |
+
"grad_norm": 1.17212669377524,
|
| 2333 |
+
"learning_rate": 1.4785069261764184e-05,
|
| 2334 |
+
"loss": 0.7713,
|
| 2335 |
+
"step": 1645
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 2.53,
|
| 2339 |
+
"grad_norm": 1.0439043355220827,
|
| 2340 |
+
"learning_rate": 1.4320640502211536e-05,
|
| 2341 |
+
"loss": 0.8379,
|
| 2342 |
+
"step": 1650
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 2.54,
|
| 2346 |
+
"grad_norm": 0.9140306646774222,
|
| 2347 |
+
"learning_rate": 1.386305983408236e-05,
|
| 2348 |
+
"loss": 0.8041,
|
| 2349 |
+
"step": 1655
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 2.55,
|
| 2353 |
+
"grad_norm": 1.061796317725875,
|
| 2354 |
+
"learning_rate": 1.3412363830412078e-05,
|
| 2355 |
+
"loss": 0.8318,
|
| 2356 |
+
"step": 1660
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 2.56,
|
| 2360 |
+
"grad_norm": 1.0834691203563807,
|
| 2361 |
+
"learning_rate": 1.2968588513965706e-05,
|
| 2362 |
+
"loss": 0.851,
|
| 2363 |
+
"step": 1665
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 2.56,
|
| 2367 |
+
"grad_norm": 0.9887481512630522,
|
| 2368 |
+
"learning_rate": 1.2531769354358825e-05,
|
| 2369 |
+
"loss": 0.8728,
|
| 2370 |
+
"step": 1670
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 2.57,
|
| 2374 |
+
"grad_norm": 1.058272033476656,
|
| 2375 |
+
"learning_rate": 1.2101941265222373e-05,
|
| 2376 |
+
"loss": 0.8016,
|
| 2377 |
+
"step": 1675
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 2.58,
|
| 2381 |
+
"grad_norm": 1.072187651154863,
|
| 2382 |
+
"learning_rate": 1.1679138601412255e-05,
|
| 2383 |
+
"loss": 0.8505,
|
| 2384 |
+
"step": 1680
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 2.59,
|
| 2388 |
+
"grad_norm": 1.2017961960493708,
|
| 2389 |
+
"learning_rate": 1.126339515626349e-05,
|
| 2390 |
+
"loss": 0.8601,
|
| 2391 |
+
"step": 1685
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 2.59,
|
| 2395 |
+
"grad_norm": 0.9334846571871207,
|
| 2396 |
+
"learning_rate": 1.0854744158889085e-05,
|
| 2397 |
+
"loss": 0.8178,
|
| 2398 |
+
"step": 1690
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 2.6,
|
| 2402 |
+
"grad_norm": 0.9385801565830063,
|
| 2403 |
+
"learning_rate": 1.0453218271524224e-05,
|
| 2404 |
+
"loss": 0.8155,
|
| 2405 |
+
"step": 1695
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 2.61,
|
| 2409 |
+
"grad_norm": 1.0026372920490085,
|
| 2410 |
+
"learning_rate": 1.0058849586915653e-05,
|
| 2411 |
+
"loss": 0.8463,
|
| 2412 |
+
"step": 1700
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 2.62,
|
| 2416 |
+
"grad_norm": 1.0094111622202961,
|
| 2417 |
+
"learning_rate": 9.671669625756574e-06,
|
| 2418 |
+
"loss": 0.8291,
|
| 2419 |
+
"step": 1705
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 2.62,
|
| 2423 |
+
"grad_norm": 1.0498204906448678,
|
| 2424 |
+
"learning_rate": 9.291709334167397e-06,
|
| 2425 |
+
"loss": 0.8694,
|
| 2426 |
+
"step": 1710
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 2.63,
|
| 2430 |
+
"grad_norm": 0.9108848375007711,
|
| 2431 |
+
"learning_rate": 8.918999081222156e-06,
|
| 2432 |
+
"loss": 0.8154,
|
| 2433 |
+
"step": 1715
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.64,
|
| 2437 |
+
"grad_norm": 1.1045628537182532,
|
| 2438 |
+
"learning_rate": 8.553568656521293e-06,
|
| 2439 |
+
"loss": 0.8542,
|
| 2440 |
+
"step": 1720
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.65,
|
| 2444 |
+
"grad_norm": 0.9187751707089163,
|
| 2445 |
+
"learning_rate": 8.195447267810686e-06,
|
| 2446 |
+
"loss": 0.8847,
|
| 2447 |
+
"step": 1725
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.66,
|
| 2451 |
+
"grad_norm": 1.0976708579458017,
|
| 2452 |
+
"learning_rate": 7.844663538647101e-06,
|
| 2453 |
+
"loss": 0.8168,
|
| 2454 |
+
"step": 1730
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 2.66,
|
| 2458 |
+
"grad_norm": 1.0277757622039416,
|
| 2459 |
+
"learning_rate": 7.501245506110433e-06,
|
| 2460 |
+
"loss": 0.824,
|
| 2461 |
+
"step": 1735
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 2.67,
|
| 2465 |
+
"grad_norm": 0.9800736010664038,
|
| 2466 |
+
"learning_rate": 7.165220618562751e-06,
|
| 2467 |
+
"loss": 0.8499,
|
| 2468 |
+
"step": 1740
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 2.68,
|
| 2472 |
+
"grad_norm": 1.1066605466255004,
|
| 2473 |
+
"learning_rate": 6.83661573345451e-06,
|
| 2474 |
+
"loss": 0.869,
|
| 2475 |
+
"step": 1745
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 2.69,
|
| 2479 |
+
"grad_norm": 1.0217949991902664,
|
| 2480 |
+
"learning_rate": 6.515457115177803e-06,
|
| 2481 |
+
"loss": 0.8308,
|
| 2482 |
+
"step": 1750
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 2.69,
|
| 2486 |
+
"grad_norm": 1.0285372086807927,
|
| 2487 |
+
"learning_rate": 6.20177043296728e-06,
|
| 2488 |
+
"loss": 0.8513,
|
| 2489 |
+
"step": 1755
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.7,
|
| 2493 |
+
"grad_norm": 0.9228807374530837,
|
| 2494 |
+
"learning_rate": 5.895580758848318e-06,
|
| 2495 |
+
"loss": 0.8359,
|
| 2496 |
+
"step": 1760
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 2.71,
|
| 2500 |
+
"grad_norm": 1.042562235261924,
|
| 2501 |
+
"learning_rate": 5.596912565633184e-06,
|
| 2502 |
+
"loss": 0.8144,
|
| 2503 |
+
"step": 1765
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 2.72,
|
| 2507 |
+
"grad_norm": 1.098352136508281,
|
| 2508 |
+
"learning_rate": 5.305789724965038e-06,
|
| 2509 |
+
"loss": 0.8253,
|
| 2510 |
+
"step": 1770
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.72,
|
| 2514 |
+
"grad_norm": 0.9667182261715308,
|
| 2515 |
+
"learning_rate": 5.022235505409823e-06,
|
| 2516 |
+
"loss": 0.8672,
|
| 2517 |
+
"step": 1775
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 2.73,
|
| 2521 |
+
"grad_norm": 1.019562250529469,
|
| 2522 |
+
"learning_rate": 4.746272570596555e-06,
|
| 2523 |
+
"loss": 0.8214,
|
| 2524 |
+
"step": 1780
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 2.74,
|
| 2528 |
+
"grad_norm": 1.2161564555157138,
|
| 2529 |
+
"learning_rate": 4.477922977405913e-06,
|
| 2530 |
+
"loss": 0.8113,
|
| 2531 |
+
"step": 1785
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 2.75,
|
| 2535 |
+
"grad_norm": 1.0015900069881538,
|
| 2536 |
+
"learning_rate": 4.217208174207199e-06,
|
| 2537 |
+
"loss": 0.8484,
|
| 2538 |
+
"step": 1790
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 2.76,
|
| 2542 |
+
"grad_norm": 0.9631542498199687,
|
| 2543 |
+
"learning_rate": 3.964148999144202e-06,
|
| 2544 |
+
"loss": 0.853,
|
| 2545 |
+
"step": 1795
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 2.76,
|
| 2549 |
+
"grad_norm": 0.9926634902794105,
|
| 2550 |
+
"learning_rate": 3.71876567846946e-06,
|
| 2551 |
+
"loss": 0.8488,
|
| 2552 |
+
"step": 1800
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.77,
|
| 2556 |
+
"grad_norm": 0.8438960246283821,
|
| 2557 |
+
"learning_rate": 3.481077824927792e-06,
|
| 2558 |
+
"loss": 0.829,
|
| 2559 |
+
"step": 1805
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 2.78,
|
| 2563 |
+
"grad_norm": 1.0027991142986536,
|
| 2564 |
+
"learning_rate": 3.251104436188679e-06,
|
| 2565 |
+
"loss": 0.8416,
|
| 2566 |
+
"step": 1810
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 2.79,
|
| 2570 |
+
"grad_norm": 1.0903635867609474,
|
| 2571 |
+
"learning_rate": 3.0288638933277934e-06,
|
| 2572 |
+
"loss": 0.8065,
|
| 2573 |
+
"step": 1815
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 2.79,
|
| 2577 |
+
"grad_norm": 1.0073217221247301,
|
| 2578 |
+
"learning_rate": 2.8143739593578856e-06,
|
| 2579 |
+
"loss": 0.7876,
|
| 2580 |
+
"step": 1820
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 2.8,
|
| 2584 |
+
"grad_norm": 0.8171096137144128,
|
| 2585 |
+
"learning_rate": 2.607651777809039e-06,
|
| 2586 |
+
"loss": 0.7574,
|
| 2587 |
+
"step": 1825
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 2.81,
|
| 2591 |
+
"grad_norm": 0.9374489147765585,
|
| 2592 |
+
"learning_rate": 2.4087138713584367e-06,
|
| 2593 |
+
"loss": 0.8652,
|
| 2594 |
+
"step": 1830
|
| 2595 |
+
},
|
| 2596 |
+
{
|
| 2597 |
+
"epoch": 2.82,
|
| 2598 |
+
"grad_norm": 1.054601585067846,
|
| 2599 |
+
"learning_rate": 2.2175761405097584e-06,
|
| 2600 |
+
"loss": 0.8471,
|
| 2601 |
+
"step": 1835
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 2.82,
|
| 2605 |
+
"grad_norm": 1.064632352850785,
|
| 2606 |
+
"learning_rate": 2.0342538623222997e-06,
|
| 2607 |
+
"loss": 0.8189,
|
| 2608 |
+
"step": 1840
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 2.83,
|
| 2612 |
+
"grad_norm": 0.8546672110619797,
|
| 2613 |
+
"learning_rate": 1.8587616891899363e-06,
|
| 2614 |
+
"loss": 0.7818,
|
| 2615 |
+
"step": 1845
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 2.84,
|
| 2619 |
+
"grad_norm": 0.9645255792031215,
|
| 2620 |
+
"learning_rate": 1.6911136476699508e-06,
|
| 2621 |
+
"loss": 0.7776,
|
| 2622 |
+
"step": 1850
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 2.85,
|
| 2626 |
+
"grad_norm": 1.068610540574793,
|
| 2627 |
+
"learning_rate": 1.5313231373619952e-06,
|
| 2628 |
+
"loss": 0.7857,
|
| 2629 |
+
"step": 1855
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 2.85,
|
| 2633 |
+
"grad_norm": 1.100889943747955,
|
| 2634 |
+
"learning_rate": 1.3794029298370814e-06,
|
| 2635 |
+
"loss": 0.8141,
|
| 2636 |
+
"step": 1860
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 2.86,
|
| 2640 |
+
"grad_norm": 0.9783722379990912,
|
| 2641 |
+
"learning_rate": 1.2353651676167643e-06,
|
| 2642 |
+
"loss": 0.8358,
|
| 2643 |
+
"step": 1865
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 2.87,
|
| 2647 |
+
"grad_norm": 0.8803200725105156,
|
| 2648 |
+
"learning_rate": 1.0992213632026517e-06,
|
| 2649 |
+
"loss": 0.8678,
|
| 2650 |
+
"step": 1870
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 2.88,
|
| 2654 |
+
"grad_norm": 1.1249544133659088,
|
| 2655 |
+
"learning_rate": 9.709823981562282e-07,
|
| 2656 |
+
"loss": 0.8078,
|
| 2657 |
+
"step": 1875
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 2.89,
|
| 2661 |
+
"grad_norm": 1.056395258554381,
|
| 2662 |
+
"learning_rate": 8.506585222291752e-07,
|
| 2663 |
+
"loss": 0.7749,
|
| 2664 |
+
"step": 1880
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 2.89,
|
| 2668 |
+
"grad_norm": 0.9253933119169012,
|
| 2669 |
+
"learning_rate": 7.382593525440573e-07,
|
| 2670 |
+
"loss": 0.8008,
|
| 2671 |
+
"step": 1885
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"epoch": 2.9,
|
| 2675 |
+
"grad_norm": 0.9212088582083587,
|
| 2676 |
+
"learning_rate": 6.337938728257054e-07,
|
| 2677 |
+
"loss": 0.8333,
|
| 2678 |
+
"step": 1890
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 2.91,
|
| 2682 |
+
"grad_norm": 1.100818174265163,
|
| 2683 |
+
"learning_rate": 5.372704326831901e-07,
|
| 2684 |
+
"loss": 0.8342,
|
| 2685 |
+
"step": 1895
|
| 2686 |
+
},
|
| 2687 |
+
{
|
| 2688 |
+
"epoch": 2.92,
|
| 2689 |
+
"grad_norm": 0.8164157006379008,
|
| 2690 |
+
"learning_rate": 4.486967469424008e-07,
|
| 2691 |
+
"loss": 0.8583,
|
| 2692 |
+
"step": 1900
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 2.92,
|
| 2696 |
+
"grad_norm": 1.0674250153511942,
|
| 2697 |
+
"learning_rate": 3.6807989502949394e-07,
|
| 2698 |
+
"loss": 0.8213,
|
| 2699 |
+
"step": 1905
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 2.93,
|
| 2703 |
+
"grad_norm": 0.9095567427952401,
|
| 2704 |
+
"learning_rate": 2.954263204050123e-07,
|
| 2705 |
+
"loss": 0.8252,
|
| 2706 |
+
"step": 1910
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 2.94,
|
| 2710 |
+
"grad_norm": 1.0738638532567681,
|
| 2711 |
+
"learning_rate": 2.3074183004887505e-07,
|
| 2712 |
+
"loss": 0.8448,
|
| 2713 |
+
"step": 1915
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 2.95,
|
| 2717 |
+
"grad_norm": 1.1019761677075681,
|
| 2718 |
+
"learning_rate": 1.7403159399629332e-07,
|
| 2719 |
+
"loss": 0.8088,
|
| 2720 |
+
"step": 1920
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 2.95,
|
| 2724 |
+
"grad_norm": 0.9934974979812884,
|
| 2725 |
+
"learning_rate": 1.2530014492446728e-07,
|
| 2726 |
+
"loss": 0.8029,
|
| 2727 |
+
"step": 1925
|
| 2728 |
+
},
|
| 2729 |
+
{
|
| 2730 |
+
"epoch": 2.96,
|
| 2731 |
+
"grad_norm": 1.0433381527945313,
|
| 2732 |
+
"learning_rate": 8.455137779038724e-08,
|
| 2733 |
+
"loss": 0.8209,
|
| 2734 |
+
"step": 1930
|
| 2735 |
+
},
|
| 2736 |
+
{
|
| 2737 |
+
"epoch": 2.97,
|
| 2738 |
+
"grad_norm": 1.032700859857025,
|
| 2739 |
+
"learning_rate": 5.1788549519438124e-08,
|
| 2740 |
+
"loss": 0.7365,
|
| 2741 |
+
"step": 1935
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 2.98,
|
| 2745 |
+
"grad_norm": 0.9408894102132038,
|
| 2746 |
+
"learning_rate": 2.7014278745163268e-08,
|
| 2747 |
+
"loss": 0.8451,
|
| 2748 |
+
"step": 1940
|
| 2749 |
+
},
|
| 2750 |
+
{
|
| 2751 |
+
"epoch": 2.99,
|
| 2752 |
+
"grad_norm": 1.0657773929136594,
|
| 2753 |
+
"learning_rate": 1.0230545599909658e-08,
|
| 2754 |
+
"loss": 0.8246,
|
| 2755 |
+
"step": 1945
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 2.99,
|
| 2759 |
+
"grad_norm": 1.0107360164317165,
|
| 2760 |
+
"learning_rate": 1.438691556565619e-09,
|
| 2761 |
+
"loss": 0.8209,
|
| 2762 |
+
"step": 1950
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 3.0,
|
| 2766 |
+
"eval_loss": 1.1822351217269897,
|
| 2767 |
+
"eval_runtime": 252.7889,
|
| 2768 |
+
"eval_samples_per_second": 9.138,
|
| 2769 |
+
"eval_steps_per_second": 0.574,
|
| 2770 |
+
"step": 1953
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 3.0,
|
| 2774 |
+
"step": 1953,
|
| 2775 |
+
"total_flos": 4109935435055104.0,
|
| 2776 |
+
"train_loss": 1.1336081770219621,
|
| 2777 |
+
"train_runtime": 21652.8936,
|
| 2778 |
+
"train_samples_per_second": 2.888,
|
| 2779 |
+
"train_steps_per_second": 0.09
|
| 2780 |
+
}
|
| 2781 |
+
],
|
| 2782 |
+
"logging_steps": 5,
|
| 2783 |
+
"max_steps": 1953,
|
| 2784 |
+
"num_input_tokens_seen": 0,
|
| 2785 |
+
"num_train_epochs": 3,
|
| 2786 |
+
"save_steps": 500,
|
| 2787 |
+
"total_flos": 4109935435055104.0,
|
| 2788 |
+
"train_batch_size": 4,
|
| 2789 |
+
"trial_name": null,
|
| 2790 |
+
"trial_params": null
|
| 2791 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fb6c59092bfc2a53cbab03013e22bef025c879fec236faee27f03a7847365e0
|
| 3 |
+
size 6200
|