gpt2_small_EN_superbpe_65536_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.1678
- Accuracy: 0.2341
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: 64
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 128
- total_eval_batch_size: 128
- 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 |
|---|
| 7.7731 | 1.0 | 1158 | 6.4637 | 0.1737 |
| 6.2521 | 2.0 | 2316 | 5.8133 | 0.2019 |
| 5.8042 | 3.0 | 3474 | 5.5626 | 0.2125 |
| 5.5808 | 4.0 | 4632 | 5.4218 | 0.2184 |
| 5.4392 | 5.0 | 5790 | 5.3327 | 0.2241 |
| 5.3354 | 6.0 | 6948 | 5.2679 | 0.2276 |
| 5.1893 | 7.0 | 8106 | 5.2243 | 0.2304 |
| 5.1358 | 8.0 | 9264 | 5.1952 | 0.2325 |
| 5.0973 | 9.0 | 10422 | 5.1757 | 0.2336 |
| 5.0683 | 10.0 | 11580 | 5.1678 | 0.2341 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
