fine_tuned_hswag_callback10

This model is a fine-tuned version of Qwen/Qwen2-1.5B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1861
• Accuracy: 0.9602

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: 8
• eval_batch_size: 8
• seed: 42
• 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: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7551	0.0322	100	0.4489	0.9012
0.3977	0.0644	200	0.5959	0.8943
0.3608	0.0966	300	0.2267	0.9258
0.3092	0.1287	400	0.1801	0.9374
0.1932	0.1609	500	0.1921	0.9562
0.1405	0.1931	600	0.2487	0.9573
0.3093	0.2253	700	0.1245	0.9573
0.1804	0.2575	800	0.1496	0.9602
0.1717	0.2897	900	0.1923	0.9573
0.1986	0.3219	1000	0.4235	0.9167
0.1786	0.3540	1100	0.1436	0.9591
0.1563	0.3862	1200	0.2635	0.9468
0.188	0.4184	1300	0.1891	0.9540
0.137	0.4506	1400	0.2017	0.9348
0.1438	0.4828	1500	0.1510	0.9660
0.1241	0.5150	1600	0.2152	0.9551
0.1793	0.5472	1700	0.1861	0.9602

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu126
• Datasets 3.3.2
• Tokenizers 0.21.0