library_name, license, base_model, tags, model-index

library_name	license	base_model	tags	model-index
transformers	other	Qwen/Qwen3-4B	llama-factory full generated_from_trainer	name results sft_base

sft_base

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

• Loss: 0.0087

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 16
• total_train_batch_size: 32
• total_eval_batch_size: 2
• 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: 0.1
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.0056	0.1698	92	0.0098
0.0115	0.3397	184	0.0107
0.015	0.5095	276	0.0094
0.0082	0.6794	368	0.0104
0.0094	0.8492	460	0.0095
0.0038	1.0185	552	0.0086
0.0029	1.1883	644	0.0095
0.01	1.3581	736	0.0082
0.0019	1.5280	828	0.0081
0.0045	1.6978	920	0.0080
0.0091	1.8677	1012	0.0077
0.0057	2.0369	1104	0.0081
0.0006	2.2068	1196	0.0086
0.0075	2.3766	1288	0.0088
0.0065	2.5464	1380	0.0087
0.0084	2.7163	1472	0.0087
0.0027	2.8861	1564	0.0087

Framework versions

• Transformers 4.56.2
• Pytorch 2.11.0+cu128
• Datasets 3.0.0
• Tokenizers 0.22.2