<!-- 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-cbt-rarity-all-end-p5k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3074
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 |
|---|---|---|---|
| 6.7056 | 0.29 | 500 | 5.6370 |
| 5.3385 | 0.58 | 1000 | 5.2064 |
| 4.9896 | 0.87 | 1500 | 4.9537 |
| 4.7182 | 1.17 | 2000 | 4.8020 |
| 4.5599 | 1.46 | 2500 | 4.6783 |
| 4.448 | 1.75 | 3000 | 4.5720 |
| 4.3216 | 2.04 | 3500 | 4.4947 |
| 4.1234 | 2.33 | 4000 | 4.4478 |
| 4.0942 | 2.62 | 4500 | 4.3887 |
| 4.0687 | 2.92 | 5000 | 4.3369 |
| 3.8617 | 3.21 | 5500 | 4.3314 |
| 3.7939 | 3.5 | 6000 | 4.2976 |
| 3.7824 | 3.79 | 6500 | 4.2652 |
| 3.6941 | 4.08 | 7000 | 4.2608 |
| 3.5091 | 4.37 | 7500 | 4.2583 |
| 3.5104 | 4.66 | 8000 | 4.2411 |
| 3.5011 | 4.96 | 8500 | 4.2278 |
| 3.3458 | 5.25 | 9000 | 4.2417 |
| 3.319 | 5.54 | 9500 | 4.2395 |
| 3.3151 | 5.83 | 10000 | 4.2387 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
