<!-- 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-concat-cl-rarity-11-135k-mod-datasets-rarity1-root3
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.7842
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 |
|---|---|---|---|
| 6.6744 | 0.05 | 500 | 5.8476 |
| 5.4179 | 0.11 | 1000 | 5.4438 |
| 5.0636 | 0.16 | 1500 | 5.2289 |
| 4.8263 | 0.21 | 2000 | 5.0980 |
| 4.6688 | 0.27 | 2500 | 5.0253 |
| 4.5377 | 0.32 | 3000 | 4.9573 |
| 4.427 | 0.37 | 3500 | 4.9024 |
| 4.3275 | 0.43 | 4000 | 4.8730 |
| 4.2279 | 0.48 | 4500 | 4.8377 |
| 4.1432 | 0.53 | 5000 | 4.8105 |
| 4.0517 | 0.59 | 5500 | 4.7917 |
| 3.9751 | 0.64 | 6000 | 4.7647 |
| 3.893 | 0.69 | 6500 | 4.7603 |
| 3.8238 | 0.74 | 7000 | 4.7474 |
| 3.771 | 0.8 | 7500 | 4.7374 |
| 3.7292 | 0.85 | 8000 | 4.7341 |
| 3.6984 | 0.9 | 8500 | 4.7283 |
| 3.6812 | 0.96 | 9000 | 4.7263 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
