yil384/Qwen3-0.6B-full

sft

This model is a fine-tuned version of Qwen/Qwen3-0.6B on the codev_r1_sft_python_passed_sharegpt_skeleton_balanced dataset. It achieves the following results on the evaluation set:

• Loss: 0.5792

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 16
• total_train_batch_size: 64
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.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_ratio: 0.03
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.6889	0.3541	100	0.6773
0.6452	0.7083	200	0.6334
0.6014	1.0602	300	0.6110
0.5915	1.4143	400	0.5974
0.5889	1.7685	500	0.5879
0.5644	2.1204	600	0.5828
0.5587	2.4745	700	0.5802
0.5635	2.8287	800	0.5792

Framework versions

• Transformers 4.57.1
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.22.1