<!-- 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-all-indv-rarity-all-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: 4.8764
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 |
|---|---|---|---|
| 6.7423 | 0.31 | 500 | 5.7056 |
| 5.4151 | 0.62 | 1000 | 5.3509 |
| 5.0793 | 0.93 | 1500 | 5.1405 |
| 4.8005 | 1.25 | 2000 | 5.0365 |
| 4.6822 | 1.56 | 2500 | 4.9729 |
| 4.5889 | 1.87 | 3000 | 4.8875 |
| 4.4127 | 2.18 | 3500 | 4.8860 |
| 4.321 | 2.49 | 4000 | 4.8474 |
| 4.291 | 2.8 | 4500 | 4.8098 |
| 4.1671 | 3.12 | 5000 | 4.8175 |
| 4.03 | 3.43 | 5500 | 4.7896 |
| 4.0205 | 3.74 | 6000 | 4.7708 |
| 3.9611 | 4.05 | 6500 | 4.7766 |
| 3.7474 | 4.36 | 7000 | 4.7885 |
| 3.7534 | 4.67 | 7500 | 4.7768 |
| 3.7348 | 4.98 | 8000 | 4.7623 |
| 3.505 | 5.3 | 8500 | 4.8038 |
| 3.4914 | 5.61 | 9000 | 4.8014 |
| 3.4892 | 5.92 | 9500 | 4.8054 |
| 3.3732 | 6.23 | 10000 | 4.8209 |
| 3.3408 | 6.54 | 10500 | 4.8276 |
| 3.3367 | 6.85 | 11000 | 4.8293 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
