gpt2_small_FI_unigram_8192_parallel10_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.3385
- Accuracy: 0.4131
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 |
|---|
| 5.6446 | 1.0 | 1644 | 4.2595 | 0.3212 |
| 4.0436 | 2.0 | 3288 | 3.8290 | 0.3578 |
| 3.844 | 3.0 | 4932 | 3.6395 | 0.3774 |
| 3.6228 | 4.0 | 6576 | 3.5326 | 0.3891 |
| 3.5024 | 5.0 | 8220 | 3.4664 | 0.3969 |
| 3.4474 | 6.0 | 9864 | 3.4185 | 0.4024 |
| 3.3756 | 7.0 | 11508 | 3.3853 | 0.4063 |
| 3.3289 | 8.0 | 13152 | 3.3607 | 0.4099 |
| 3.2997 | 9.0 | 14796 | 3.3459 | 0.4119 |
| 3.2682 | 10.0 | 16440 | 3.3385 | 0.4131 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
