gpt2_small_PL_unigram_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.1565
- Accuracy: 0.4210
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 |
|---|
| 5.4814 | 1.0 | 1746 | 4.0345 | 0.3238 |
| 3.8743 | 2.0 | 3492 | 3.6047 | 0.3670 |
| 3.5587 | 3.0 | 5238 | 3.4357 | 0.3859 |
| 3.462 | 4.0 | 6984 | 3.3393 | 0.3977 |
| 3.3398 | 5.0 | 8730 | 3.2761 | 0.4047 |
| 3.2612 | 6.0 | 10476 | 3.2322 | 0.4101 |
| 3.2049 | 7.0 | 12222 | 3.2012 | 0.4145 |
| 3.1726 | 8.0 | 13968 | 3.1782 | 0.4175 |
| 3.1298 | 9.0 | 15714 | 3.1644 | 0.4199 |
| 3.1042 | 10.0 | 17460 | 3.1565 | 0.4210 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
