gpt2-concat-mod-datatsets-rarity-all-iorder-no-cut
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.2939
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: 0.0005
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7925 | 0.32 | 500 | 5.7247 |
| 5.4528 | 0.64 | 1000 | 5.2926 |
| 5.1083 | 0.96 | 1500 | 5.0277 |
| 4.8149 | 1.28 | 2000 | 4.8716 |
| 4.6803 | 1.6 | 2500 | 4.7459 |
| 4.5724 | 1.93 | 3000 | 4.6411 |
| 4.3471 | 2.25 | 3500 | 4.5849 |
| 4.2732 | 2.57 | 4000 | 4.5154 |
| 4.2272 | 2.89 | 4500 | 4.4580 |
| 4.0295 | 3.21 | 5000 | 4.4423 |
| 3.948 | 3.53 | 5500 | 4.4057 |
| 3.9259 | 3.85 | 6000 | 4.3781 |
| 3.7712 | 4.17 | 6500 | 4.3799 |
| 3.6471 | 4.49 | 7000 | 4.3703 |
| 3.6421 | 4.81 | 7500 | 4.3553 |
| 3.5545 | 5.13 | 8000 | 4.3638 |
| 3.447 | 5.46 | 8500 | 4.3651 |
| 3.4478 | 5.78 | 9000 | 4.3656 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
- Downloads last month
- 3