<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
gpt2-cl-concat-rarity-138k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.7621
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: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.599 | 0.06 | 500 | 5.8546 |
| 5.3587 | 0.12 | 1000 | 5.4652 |
| 5.0281 | 0.18 | 1500 | 5.2287 |
| 4.8017 | 0.24 | 2000 | 5.0937 |
| 4.6386 | 0.3 | 2500 | 5.0005 |
| 4.5143 | 0.36 | 3000 | 4.9333 |
| 4.4073 | 0.42 | 3500 | 4.8969 |
| 4.3024 | 0.48 | 4000 | 4.8546 |
| 4.2073 | 0.54 | 4500 | 4.8167 |
| 4.1186 | 0.6 | 5000 | 4.7919 |
| 4.0392 | 0.66 | 5500 | 4.7687 |
| 3.9579 | 0.72 | 6000 | 4.7472 |
| 3.8922 | 0.78 | 6500 | 4.7355 |
| 3.8472 | 0.84 | 7000 | 4.7231 |
| 3.8129 | 0.9 | 7500 | 4.7162 |
| 3.7885 | 0.96 | 8000 | 4.7134 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
