gpt2_medium_EN_bpe_32768_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.5927
- Accuracy: 0.4958
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 |
|---|
| 2.9632 | 1.0 | 30922 | 2.9081 | 0.4489 |
| 2.8217 | 2.0 | 61844 | 2.7863 | 0.4640 |
| 2.7441 | 3.0 | 92766 | 2.7279 | 0.4725 |
| 2.6889 | 4.0 | 123688 | 2.6886 | 0.4787 |
| 2.6446 | 5.0 | 154610 | 2.6615 | 0.4831 |
| 2.6087 | 6.0 | 185532 | 2.6406 | 0.4862 |
| 2.5691 | 7.0 | 216454 | 2.6218 | 0.4894 |
| 2.536 | 8.0 | 247376 | 2.6106 | 0.4920 |
| 2.5001 | 9.0 | 278298 | 2.5986 | 0.4941 |
| 2.4664 | 10.0 | 309220 | 2.5927 | 0.4958 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
