<!-- 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-bnc-rarity-all-15k-1k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1958
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: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7306 | 0.29 | 500 | 5.6425 |
| 5.383 | 0.59 | 1000 | 5.2067 |
| 5.0354 | 0.88 | 1500 | 4.9561 |
| 4.7644 | 1.18 | 2000 | 4.8074 |
| 4.5957 | 1.47 | 2500 | 4.6852 |
| 4.4924 | 1.77 | 3000 | 4.5774 |
| 4.3417 | 2.06 | 3500 | 4.5014 |
| 4.1694 | 2.36 | 4000 | 4.4512 |
| 4.1359 | 2.65 | 4500 | 4.3948 |
| 4.0992 | 2.94 | 5000 | 4.3417 |
| 3.8748 | 3.24 | 5500 | 4.3363 |
| 3.8362 | 3.53 | 6000 | 4.3051 |
| 3.8169 | 3.83 | 6500 | 4.2728 |
| 3.6885 | 4.12 | 7000 | 4.2740 |
| 3.5458 | 4.42 | 7500 | 4.2658 |
| 3.5385 | 4.71 | 8000 | 4.2522 |
| 3.524 | 5.01 | 8500 | 4.2427 |
| 3.347 | 5.3 | 9000 | 4.2535 |
| 3.3478 | 5.59 | 9500 | 4.2521 |
| 3.3503 | 5.89 | 10000 | 4.2510 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
