SFT-Qwen2.5-Coder-3B

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-3B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9728

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.0721	0.2985	20	1.1757
0.8989	0.5970	40	1.1059
0.8293	0.8955	60	1.0656
0.787	1.1940	80	1.0364
0.7025	1.4925	100	1.0206
0.7386	1.7910	120	0.9961
0.7471	2.0896	140	0.9916
0.624	2.3881	160	0.9843
0.6839	2.6866	180	0.9728
0.6561	2.9851	200	0.9737
0.6027	3.2836	220	0.9785
0.5221	3.5821	240	0.9843

Framework versions

• PEFT 0.18.0
• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1