gpt2_small_TR_superbpe_5826_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.7522
- Accuracy: 0.3193
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 |
|---|
| 6.3375 | 1.0 | 1537 | 4.7789 | 0.2090 |
| 4.4518 | 2.0 | 3074 | 4.2507 | 0.2618 |
| 4.2319 | 3.0 | 4611 | 4.0627 | 0.2809 |
| 4.0213 | 4.0 | 6148 | 3.9573 | 0.2933 |
| 3.9438 | 5.0 | 7685 | 3.8874 | 0.3011 |
| 3.8508 | 6.0 | 9222 | 3.8404 | 0.3073 |
| 3.805 | 7.0 | 10759 | 3.8033 | 0.3123 |
| 3.7476 | 8.0 | 12296 | 3.7773 | 0.3157 |
| 3.7215 | 9.0 | 13833 | 3.7602 | 0.3178 |
| 3.6839 | 10.0 | 15370 | 3.7522 | 0.3193 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
