See axolotl config

axolotl version: 0.6.0

base_model: Qwen/Qwen2.5-7B-Instruct
trust_remote_code: true
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: 
load_in_4bit:
strict: false

datasets:
  - path: medalpaca/medical_meadow_medqa
    type: alpaca
dataset_prepared_path:
val_set_size: 0.1
output_dir: ./sft-qwen25

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


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

gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.000005

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

gradient_checkpointing: true
  
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps:
eval_steps: 
save_steps:

evals_per_epoch: 
saves_per_epoch: 

debug:
deepspeed: deepspeed_configs/zero2.json
weight_decay:
fsdp:
fsdp_config:
special_tokens:

hub_model_id: neginashz/sft-qwen-25-7b-instruct-2
hub_strategy: 
early_stopping_patience:

resume_from_checkpoint:
auto_resume_from_checkpoints: true

sft-qwen-25-7b-instruct-2

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: 0.1054

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: 5e-06
• 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: 4
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
0.1381	0.1235	10	0.1342
0.1495	0.2469	20	0.1229
0.1215	0.3704	30	0.1246
0.1354	0.4938	40	0.1175
0.1223	0.6173	50	0.1115
0.1068	0.7407	60	0.1101
0.1061	0.8642	70	0.1056
0.118	0.9877	80	0.1055
0.0644	1.1111	90	0.1054
0.0554	1.2346	100	0.1054
0.0564	1.3580	110	0.1054
0.0601	1.4815	120	0.1054
0.0482	2.0	162	0.1054

Framework versions

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