learning_rate: 0.0002
train_batch_size: 4
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.05
num_epochs: 1.0

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6957	0.1001	283	0.6942	0.5765
0.7	0.2001	566	0.6455	0.6316
0.6356	0.3002	849	0.6433	0.6380
0.6399	0.4003	1132	0.6376	0.6445
0.5947	0.5004	1415	0.6328	0.6482
0.6299	0.6004	1698	0.6258	0.6551
0.6279	0.7005	1981	0.6252	0.6586
0.6263	0.8006	2264	0.6205	0.6603
0.593	0.9006	2547	0.6188	0.6615

licezhang/Qwen3-8B_14367368

Qwen3-8B_14367368