cds_sh1_30

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

• Loss: 3.1615

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: 30
• 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
4.8958	1.0	495	3.9315
3.6499	2.0	990	3.5397
3.3793	3.0	1485	3.3794
3.2264	4.0	1980	3.2874
3.1179	5.0	2475	3.2294
3.0343	6.0	2970	3.1836
2.9631	7.0	3465	3.1519
2.901	8.0	3960	3.1372
2.8443	9.0	4455	3.1182
2.7925	10.0	4950	3.1129
2.7436	11.0	5445	3.1106
2.6967	12.0	5940	3.1145
2.6527	13.0	6435	3.1148
2.6122	14.0	6930	3.1206
2.575	15.0	7425	3.1303
2.542	16.0	7920	3.1401
2.512	17.0	8415	3.1443
2.4866	18.0	8910	3.1503
2.4648	19.0	9405	3.1571
2.4479	20.0	9900	3.1615

Framework versions

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