cds_sh1_51

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

• Loss: 3.1650

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: 256
• eval_batch_size: 256
• seed: 51
• optimizer: Use 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: 500
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.8772	1.0	495	3.9269
3.6539	2.0	990	3.5510
3.3838	3.0	1485	3.3823
3.2305	4.0	1980	3.2957
3.1233	5.0	2475	3.2330
3.0397	6.0	2970	3.1851
2.9689	7.0	3465	3.1579
2.9059	8.0	3960	3.1408
2.8508	9.0	4455	3.1233
2.7979	10.0	4950	3.1176
2.7495	11.0	5445	3.1153
2.7035	12.0	5940	3.1164
2.6583	13.0	6435	3.1205
2.6178	14.0	6930	3.1245
2.5807	15.0	7425	3.1350
2.5479	16.0	7920	3.1412
2.5183	17.0	8415	3.1502
2.4933	18.0	8910	3.1565
2.4713	19.0	9405	3.1624
2.4539	20.0	9900	3.1650

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.0