gpt2_small_AR_superbpe_81920_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 5.7604
- Accuracy: 0.2301
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 |
|---|
| 8.3786 | 1.0 | 1159 | 7.2922 | 0.1727 |
| 7.0899 | 2.0 | 2318 | 6.6309 | 0.1937 |
| 6.6048 | 3.0 | 3477 | 6.3080 | 0.2063 |
| 6.3279 | 4.0 | 4636 | 6.1144 | 0.2143 |
| 6.1412 | 5.0 | 5795 | 5.9851 | 0.2191 |
| 6.005 | 6.0 | 6954 | 5.8962 | 0.2223 |
| 5.8159 | 7.0 | 8113 | 5.8372 | 0.2260 |
| 5.7514 | 8.0 | 9272 | 5.7944 | 0.2280 |
| 5.704 | 9.0 | 10431 | 5.7708 | 0.2293 |
| 5.6696 | 10.0 | 11590 | 5.7604 | 0.2301 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
