qwen3-2b-ablation-C-multi-answer

This model is a fine-tuned version of Qwen/Qwen3-2B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7672

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: 4
• eval_batch_size: 8
• seed: 46
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• 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: linear
• lr_scheduler_warmup_steps: 0.03
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.0877	0.2257	20	1.0368
0.9775	0.4513	40	0.9480
0.9055	0.6770	60	0.8906
0.8374	0.9027	80	0.8423
0.6946	1.1241	100	0.8146
0.6537	1.3498	120	0.7890
0.6290	1.5755	140	0.7754
0.6405	1.8011	160	0.7682
0.6170	2.0	178	0.7672

Framework versions

• Transformers 5.4.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.4
• Tokenizers 0.22.2