gpt2_small_EN_superbpe_65536_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.8332
- Accuracy: 0.3241
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 |
|---|
| 4.4342 | 1.0 | 21906 | 4.3265 | 0.2719 |
| 4.2047 | 2.0 | 43812 | 4.1242 | 0.2911 |
| 4.0992 | 3.0 | 65718 | 4.0296 | 0.3007 |
| 4.0291 | 4.0 | 87624 | 3.9748 | 0.3068 |
| 3.9785 | 5.0 | 109530 | 3.9323 | 0.3118 |
| 3.935 | 6.0 | 131436 | 3.9015 | 0.3156 |
| 3.9056 | 7.0 | 153342 | 3.8782 | 0.3185 |
| 3.8717 | 8.0 | 175248 | 3.8578 | 0.3211 |
| 3.8477 | 9.0 | 197154 | 3.8427 | 0.3229 |
| 3.8222 | 10.0 | 219060 | 3.8332 | 0.3241 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
