gpt2_small_AR_unigram_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6507
- Accuracy: 0.3573
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 |
|---|
| 5.8109 | 1.0 | 1756 | 4.4633 | 0.2834 |
| 4.3067 | 2.0 | 3512 | 4.0812 | 0.3140 |
| 4.024 | 3.0 | 5268 | 3.9249 | 0.3280 |
| 3.8805 | 4.0 | 7024 | 3.8313 | 0.3377 |
| 3.8125 | 5.0 | 8780 | 3.7695 | 0.3444 |
| 3.7364 | 6.0 | 10536 | 3.7247 | 0.3486 |
| 3.68 | 7.0 | 12292 | 3.6951 | 0.3514 |
| 3.6358 | 8.0 | 14048 | 3.6725 | 0.3541 |
| 3.6053 | 9.0 | 15804 | 3.6577 | 0.3561 |
| 3.5785 | 10.0 | 17560 | 3.6507 | 0.3573 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
