<!-- 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-finetune-cl-mod-datasets-rarity1
This model was trained from scratch on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 2.6666
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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.8306 | 0.3 | 500 | 4.5102 |
| 3.9443 | 0.59 | 1000 | 4.5278 |
| 4.039 | 0.89 | 1500 | 4.4641 |
| 3.8859 | 1.18 | 2000 | 4.4729 |
| 3.8519 | 1.48 | 2500 | 4.4332 |
| 3.8622 | 1.78 | 3000 | 4.3895 |
| 3.765 | 2.07 | 3500 | 4.4085 |
| 3.5415 | 2.37 | 4000 | 4.4023 |
| 3.5631 | 2.66 | 4500 | 4.3729 |
| 3.559 | 2.96 | 5000 | 4.3403 |
| 3.2324 | 3.26 | 5500 | 4.4037 |
| 3.2011 | 3.55 | 6000 | 4.3997 |
| 3.1898 | 3.85 | 6500 | 4.3837 |
| 3.0472 | 4.14 | 7000 | 4.4190 |
| 2.9036 | 4.44 | 7500 | 4.4273 |
| 2.8985 | 4.74 | 8000 | 4.4269 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
