gpt2_small_AR_unigram_32768_parallel3_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.2522
- Accuracy: 0.4033
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 |
|---|
| 3.7072 | 1.0 | 29228 | 3.6391 | 0.3598 |
| 3.5449 | 2.0 | 58456 | 3.4841 | 0.3762 |
| 3.4627 | 3.0 | 87684 | 3.4129 | 0.3837 |
| 3.4124 | 4.0 | 116912 | 3.3683 | 0.3890 |
| 3.3715 | 5.0 | 146140 | 3.3372 | 0.3929 |
| 3.3423 | 6.0 | 175368 | 3.3114 | 0.3952 |
| 3.3149 | 7.0 | 204596 | 3.2910 | 0.3979 |
| 3.2897 | 8.0 | 233824 | 3.2746 | 0.4003 |
| 3.2681 | 9.0 | 263052 | 3.2606 | 0.4021 |
| 3.249 | 10.0 | 292280 | 3.2522 | 0.4033 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
