<!-- 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-dp-mod-datasets-rarity2
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 2.9689
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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.6964 | 0.28 | 500 | 5.6571 |
| 5.3695 | 0.56 | 1000 | 5.2302 |
| 5.0252 | 0.83 | 1500 | 4.9783 |
| 4.7727 | 1.11 | 2000 | 4.8337 |
| 4.6037 | 1.39 | 2500 | 4.7203 |
| 4.4995 | 1.67 | 3000 | 4.6237 |
| 4.4109 | 1.94 | 3500 | 4.5399 |
| 4.1994 | 2.22 | 4000 | 4.5071 |
| 4.1606 | 2.5 | 4500 | 4.4425 |
| 4.1134 | 2.78 | 5000 | 4.3980 |
| 4.0337 | 3.05 | 5500 | 4.3731 |
| 3.8408 | 3.33 | 6000 | 4.3581 |
| 3.8431 | 3.61 | 6500 | 4.3268 |
| 3.8253 | 3.89 | 7000 | 4.2934 |
| 3.6561 | 4.16 | 7500 | 4.3160 |
| 3.5535 | 4.44 | 8000 | 4.3077 |
| 3.5564 | 4.72 | 8500 | 4.2849 |
| 3.5441 | 5.0 | 9000 | 4.2669 |
| 3.296 | 5.27 | 9500 | 4.3047 |
| 3.2948 | 5.55 | 10000 | 4.2986 |
| 3.2913 | 5.83 | 10500 | 4.2950 |
| 3.2305 | 6.11 | 11000 | 4.3041 |
| 3.1394 | 6.39 | 11500 | 4.3095 |
| 3.1341 | 6.66 | 12000 | 4.3099 |
| 3.1359 | 6.94 | 12500 | 4.3096 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
