gpt2_small_AR_bpe_32768_parallel3-100_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 3.3159
- Accuracy: 0.4091
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 |
|---|
| 3.7832 | 1.0 | 28760 | 3.7149 | 0.3649 |
| 3.613 | 2.0 | 57520 | 3.5566 | 0.3805 |
| 3.526 | 3.0 | 86280 | 3.4824 | 0.3885 |
| 3.4717 | 4.0 | 115040 | 3.4356 | 0.3943 |
| 3.4305 | 5.0 | 143800 | 3.4031 | 0.3976 |
| 3.3999 | 6.0 | 172560 | 3.3766 | 0.4009 |
| 3.37 | 7.0 | 201320 | 3.3563 | 0.4032 |
| 3.3463 | 8.0 | 230080 | 3.3386 | 0.4058 |
| 3.3245 | 9.0 | 258840 | 3.3257 | 0.4079 |
| 3.3067 | 10.0 | 287600 | 3.3159 | 0.4091 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.19.1
