gpt2_small_EN_superbpe_65536_parallel3-100_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.1713
- Accuracy: 0.2893
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: 5e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|
| 5.2549 | 1.0 | 5715 | 4.9966 | 0.2206 |
| 4.7303 | 2.0 | 11430 | 4.6117 | 0.2490 |
| 4.5204 | 3.0 | 17145 | 4.4498 | 0.2626 |
| 4.407 | 4.0 | 22860 | 4.3565 | 0.2707 |
| 4.3159 | 5.0 | 28575 | 4.2921 | 0.2764 |
| 4.2498 | 6.0 | 34290 | 4.2496 | 0.2814 |
| 4.1921 | 7.0 | 40005 | 4.2193 | 0.2845 |
| 4.1524 | 8.0 | 45720 | 4.1942 | 0.2870 |
| 4.1149 | 9.0 | 51435 | 4.1784 | 0.2889 |
| 4.0861 | 10.0 | 57150 | 4.1713 | 0.2893 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
