gpt2_small_PL_bpe_65536_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.8618
- Accuracy: 0.4126
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 |
|---|
| 6.1529 | 1.0 | 1422 | 4.8460 | 0.3505 |
| 4.8818 | 2.0 | 2844 | 4.4452 | 0.3697 |
| 4.3739 | 3.0 | 4266 | 4.2371 | 0.3828 |
| 4.2316 | 4.0 | 5688 | 4.1053 | 0.3917 |
| 4.0528 | 5.0 | 7110 | 4.0172 | 0.3989 |
| 3.9769 | 6.0 | 8532 | 3.9563 | 0.4033 |
| 3.9256 | 7.0 | 9954 | 3.9140 | 0.4069 |
| 3.8395 | 8.0 | 11376 | 3.8865 | 0.4104 |
| 3.813 | 9.0 | 12798 | 3.8681 | 0.4117 |
| 3.7709 | 10.0 | 14220 | 3.8618 | 0.4126 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
