SFT-Qwen2.5-Coder-7B_v1.2s

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-7B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5014

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: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.8618	0.1985	20	0.7028
0.6786	0.3970	40	0.6255
0.5624	0.5955	60	0.5838
0.5455	0.7940	80	0.5579
0.531	0.9926	100	0.5395
0.3587	1.1886	120	0.5308
0.4215	1.3871	140	0.5241
0.4159	1.5856	160	0.5155
0.4262	1.7841	180	0.5045
0.4629	1.9826	200	0.5014
0.344	2.1787	220	0.5186
0.3075	2.3772	240	0.5144
0.3312	2.5757	260	0.5061

Framework versions

• PEFT 0.18.0
• Transformers 4.57.3
• Pytorch 2.9.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1