qwen3-4b-instruct-2507-3epochs-non-ood

This model is a fine-tuned version of Qwen/Qwen3-4B-Instruct-2507 on the deepcoder_partial_edits_non_ood_train dataset. It achieves the following results on the evaluation set:

• Loss: 0.0361

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: 2
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.0624	0.4779	100	0.0417
0.0461	0.9558	200	0.0352
0.0368	1.4301	300	0.0338
0.0429	1.9080	400	0.0323
0.0179	2.3823	500	0.0359
0.0204	2.8602	600	0.0360

Framework versions

• Transformers 4.55.0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1