gpt2_small_EN_unigram_32768_parallel3-100_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.0879
- Accuracy: 0.4268
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 |
|---|
| 3.7506 | 1.0 | 7711 | 3.6186 | 0.3720 |
| 3.4588 | 2.0 | 15422 | 3.3861 | 0.3931 |
| 3.3296 | 3.0 | 23133 | 3.2820 | 0.4038 |
| 3.2528 | 4.0 | 30844 | 3.2183 | 0.4106 |
| 3.1978 | 5.0 | 38555 | 3.1763 | 0.4150 |
| 3.1499 | 6.0 | 46266 | 3.1465 | 0.4185 |
| 3.1117 | 7.0 | 53977 | 3.1254 | 0.4213 |
| 3.0799 | 8.0 | 61688 | 3.1074 | 0.4238 |
| 3.0548 | 9.0 | 69399 | 3.0943 | 0.4258 |
| 3.033 | 10.0 | 77110 | 3.0879 | 0.4268 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
