wiki_sh1_30
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 6.0193
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 |
|---|---|---|---|
| 7.3374 | 1.0 | 399 | 6.7886 |
| 6.5631 | 2.0 | 798 | 6.5030 |
| 6.3611 | 3.0 | 1197 | 6.3739 |
| 6.2399 | 4.0 | 1596 | 6.2857 |
| 6.1437 | 5.0 | 1995 | 6.2176 |
| 6.0582 | 6.0 | 2394 | 6.1570 |
| 5.9795 | 7.0 | 2793 | 6.1109 |
| 5.9051 | 8.0 | 3192 | 6.0746 |
| 5.8345 | 9.0 | 3591 | 6.0432 |
| 5.7674 | 10.0 | 3990 | 6.0230 |
| 5.7039 | 11.0 | 4389 | 6.0070 |
| 5.6435 | 12.0 | 4788 | 5.9996 |
| 5.5868 | 13.0 | 5187 | 5.9982 |
| 5.5342 | 14.0 | 5586 | 5.9972 |
| 5.4864 | 15.0 | 5985 | 5.9970 |
| 5.4421 | 16.0 | 6384 | 6.0018 |
| 5.4034 | 17.0 | 6783 | 6.0083 |
| 5.3716 | 18.0 | 7182 | 6.0128 |
| 5.3466 | 19.0 | 7581 | 6.0176 |
| 5.3284 | 20.0 | 7980 | 6.0193 |
Framework versions
- Transformers 4.56.1
- Pytorch 2.8.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.0