gpt2_small_DE_bpe_81920_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.5090
- Accuracy: 0.4061
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 |
|---|
| 5.8385 | 1.0 | 1577 | 4.1871 | 0.3535 |
| 4.0115 | 2.0 | 3154 | 3.8827 | 0.3751 |
| 3.8623 | 3.0 | 4731 | 3.7481 | 0.3850 |
| 3.7024 | 4.0 | 6308 | 3.6647 | 0.3916 |
| 3.641 | 5.0 | 7885 | 3.6111 | 0.3959 |
| 3.5509 | 6.0 | 9462 | 3.5721 | 0.3997 |
| 3.4924 | 7.0 | 11039 | 3.5446 | 0.4023 |
| 3.455 | 8.0 | 12616 | 3.5248 | 0.4040 |
| 3.4206 | 9.0 | 14193 | 3.5134 | 0.4056 |
| 3.4012 | 10.0 | 15770 | 3.5090 | 0.4061 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
