gpt2_small_FR_bpe_81920_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 2.9927
- Accuracy: 0.4421
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 |
|---|
| 5.3352 | 1.0 | 1741 | 3.5464 | 0.3952 |
| 3.435 | 2.0 | 3482 | 3.3018 | 0.4126 |
| 3.2414 | 3.0 | 5223 | 3.1924 | 0.4215 |
| 3.1737 | 4.0 | 6964 | 3.1232 | 0.4277 |
| 3.08 | 5.0 | 8705 | 3.0777 | 0.4321 |
| 3.0195 | 6.0 | 10446 | 3.0449 | 0.4353 |
| 2.9711 | 7.0 | 12187 | 3.0219 | 0.4380 |
| 2.942 | 8.0 | 13928 | 3.0064 | 0.4403 |
| 2.9075 | 9.0 | 15669 | 2.9965 | 0.4415 |
| 2.886 | 10.0 | 17410 | 2.9927 | 0.4421 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
