learning_rate: 1e-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
num_epochs: 1

Training Loss	Epoch	Step	Validation Loss
0.1836	0.0676	100	0.2156
0.2063	0.1351	200	0.2099
0.1713	0.2027	300	0.2057
0.2054	0.2703	400	0.2089
0.143	0.3378	500	0.1921
0.1687	0.4054	600	0.1971
0.2407	0.4730	700	0.1927
0.1904	0.5405	800	0.1867
0.1565	0.6081	900	0.1772
0.1581	0.6757	1000	0.1763
0.1486	0.7432	1100	0.1727
0.1386	0.8108	1200	0.1690
0.1285	0.8784	1300	0.1682
0.1637	0.9459	1400	0.1683

cutelemonlili/Qwen-7B-instruct-20percent-6k

20percent_6k