Hugging Face's logo

neginashz
/
qlora-qwen-25-7b-instruct-3

PEFT
Safetensors
qwen2
axolotl
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Community
qlora-qwen-25-7b-instruct-3 / README.md
neginashz's picture
neginashz
Model save
38fcbe9 verified
preview code
|
raw
history blame contribute delete
4.19 kB
metadata
library_name: peft
license: apache-2.0
base_model: Qwen/Qwen2.5-7B-Instruct
tags:
  - axolotl
  - generated_from_trainer
datasets:
  - medalpaca/medical_meadow_medqa
model-index:
  - name: qlora-qwen-25-7b-instruct-3
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.6.0

base_model: Qwen/Qwen2.5-7B-Instruct
trust_remote_code: true

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: medalpaca/medical_meadow_medqa
    type: alpaca
dataset_prepared_path:
val_set_size: 0.2
output_dir: ./qlora-qwen25

sequence_len: 8192
sample_packing: true
eval_sample_packing: true
pad_to_sequence_len: true

adapter: qlora
lora_model_dir:
lora_r: 256
lora_alpha: 128
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.00002

train_on_inputs: false
group_by_length: false
bf16: true
fp16:
tf32: 

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps:
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdp_limit_all_gathers: true
  fsdp_sync_module_states: true
  fsdp_offload_params: true
  fsdp_use_orig_params: false
  fsdp_cpu_ram_efficient_loading: true
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
  fsdp_transformer_layer_cls_to_wrap: Qwen2DecoderLayer
  fsdp_state_dict_type: FULL_STATE_DICT
  fsdp_sharding_strategy: FULL_SHARD
special_tokens:

wandb_project: qwen-25-7b-instruct
wandb_entity: 
wandb_watch:
wandb_name: 
wandb_log_model: 

hub_model_id: neginashz/qlora-qwen-25-7b-instruct-3
hub_strategy: 
early_stopping_patience:

resume_from_checkpoint:
auto_resume_from_checkpoints: true

qlora-qwen-25-7b-instruct-3

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct on the medalpaca/medical_meadow_medqa dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1238

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 4
  • total_train_batch_size: 4
  • total_eval_batch_size: 4
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 6
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.1473 0.25 18 0.1576
0.1456 0.5 36 0.1333
0.121 0.75 54 0.1312
0.1328 1.0 72 0.1303
0.1336 1.25 90 0.1276
0.1228 1.5 108 0.1263
0.1199 1.75 126 0.1260
0.1393 2.0 144 0.1257
0.1146 2.25 162 0.1244
0.1161 2.5 180 0.1238
0.139 2.75 198 0.1238
0.0927 3.0 216 0.1238

Framework versions

  • PEFT 0.14.0
  • Transformers 4.47.0
  • Pytorch 2.5.1+cu124
  • Datasets 3.1.0
  • Tokenizers 0.21.0