gpt2_small_DE_superbpe_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.9537
- Accuracy: 0.2567
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 |
|---|
| 7.7031 | 1.0 | 1163 | 6.4830 | 0.1816 |
| 6.2253 | 2.0 | 2326 | 5.7138 | 0.2130 |
| 5.7017 | 3.0 | 3489 | 5.4223 | 0.2263 |
| 5.4483 | 4.0 | 4652 | 5.2586 | 0.2355 |
| 5.2893 | 5.0 | 5815 | 5.1533 | 0.2421 |
| 5.1709 | 6.0 | 6978 | 5.0760 | 0.2483 |
| 5.0008 | 7.0 | 8141 | 5.0218 | 0.2522 |
| 4.9439 | 8.0 | 9304 | 4.9859 | 0.2541 |
| 4.9 | 9.0 | 10467 | 4.9618 | 0.2563 |
| 4.8685 | 10.0 | 11630 | 4.9537 | 0.2567 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
