gpt2_small_FR_unigram_49152_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.3251
- Accuracy: 0.4008
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 |
|---|
| 5.7223 | 1.0 | 1587 | 4.0117 | 0.3480 |
| 3.8308 | 2.0 | 3174 | 3.7050 | 0.3674 |
| 3.6873 | 3.0 | 4761 | 3.5693 | 0.3785 |
| 3.5277 | 4.0 | 6348 | 3.4856 | 0.3847 |
| 3.4708 | 5.0 | 7935 | 3.4293 | 0.3897 |
| 3.3811 | 6.0 | 9522 | 3.3881 | 0.3937 |
| 3.3248 | 7.0 | 11109 | 3.3613 | 0.3968 |
| 3.2906 | 8.0 | 12696 | 3.3423 | 0.3981 |
| 3.2547 | 9.0 | 14283 | 3.3302 | 0.3993 |
| 3.2382 | 10.0 | 15870 | 3.3251 | 0.4008 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
