gpt2_small_DE_bpe_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.4685
- Accuracy: 0.4047
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 |
|---|
| 5.7244 | 1.0 | 1605 | 4.1809 | 0.3488 |
| 4.0185 | 2.0 | 3210 | 3.8692 | 0.3698 |
| 3.8701 | 3.0 | 4815 | 3.7307 | 0.3803 |
| 3.7007 | 4.0 | 6420 | 3.6446 | 0.3879 |
| 3.5963 | 5.0 | 8025 | 3.5851 | 0.3925 |
| 3.5393 | 6.0 | 9630 | 3.5400 | 0.3974 |
| 3.4771 | 7.0 | 11235 | 3.5095 | 0.3996 |
| 3.4435 | 8.0 | 12840 | 3.4886 | 0.4024 |
| 3.4007 | 9.0 | 14445 | 3.4742 | 0.4041 |
| 3.3755 | 10.0 | 16050 | 3.4685 | 0.4047 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
