gpt2_small_DE_unigram_32768_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.5683
- Accuracy: 0.3887
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: 110
- 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 |
|---|
| 5.9132 | 1.0 | 1818 | 4.2924 | 0.3327 |
| 4.1656 | 2.0 | 3636 | 3.9683 | 0.3537 |
| 3.9125 | 3.0 | 5454 | 3.8254 | 0.3643 |
| 3.7773 | 4.0 | 7272 | 3.7388 | 0.3712 |
| 3.683 | 5.0 | 9090 | 3.6809 | 0.3766 |
| 3.6316 | 6.0 | 10908 | 3.6379 | 0.3801 |
| 3.5698 | 7.0 | 12726 | 3.6090 | 0.3834 |
| 3.5234 | 8.0 | 14544 | 3.5872 | 0.3858 |
| 3.488 | 9.0 | 16362 | 3.5743 | 0.3880 |
| 3.4621 | 10.0 | 18180 | 3.5683 | 0.3887 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
