gpt2_small_PL_superbpe_6137_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.5677
- Accuracy: 0.3493
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 |
|---|
| 6.2972 | 1.0 | 1565 | 4.7139 | 0.2188 |
| 4.3555 | 2.0 | 3130 | 4.1169 | 0.2800 |
| 4.1043 | 3.0 | 4695 | 3.8945 | 0.3077 |
| 3.8667 | 4.0 | 6260 | 3.7821 | 0.3211 |
| 3.7855 | 5.0 | 7825 | 3.7074 | 0.3302 |
| 3.6798 | 6.0 | 9390 | 3.6564 | 0.3370 |
| 3.6374 | 7.0 | 10955 | 3.6201 | 0.3422 |
| 3.5691 | 8.0 | 12520 | 3.5946 | 0.3453 |
| 3.5289 | 9.0 | 14085 | 3.5755 | 0.3481 |
| 3.5042 | 10.0 | 15650 | 3.5677 | 0.3493 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
