gpt2_small_FI_superbpe_65536_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.5005
- Accuracy: 0.2587
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 |
|---|
| 8.1372 | 1.0 | 1045 | 7.2798 | 0.1877 |
| 6.9472 | 2.0 | 2090 | 6.5549 | 0.2123 |
| 6.4273 | 3.0 | 3135 | 6.1838 | 0.2238 |
| 6.1142 | 4.0 | 4180 | 5.9413 | 0.2345 |
| 5.8941 | 5.0 | 5225 | 5.7791 | 0.2425 |
| 5.7353 | 6.0 | 6270 | 5.6718 | 0.2485 |
| 5.6185 | 7.0 | 7315 | 5.5928 | 0.2530 |
| 5.5304 | 8.0 | 8360 | 5.5436 | 0.2559 |
| 5.4677 | 9.0 | 9405 | 5.5121 | 0.2580 |
| 5.4223 | 10.0 | 10450 | 5.5005 | 0.2587 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
