<!-- 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. -->
rarity-all-guten-2p5k-cbt-p5k-mixed
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3206
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 |
|---|---|---|---|
| 6.6916 | 0.29 | 500 | 5.6242 |
| 5.3287 | 0.59 | 1000 | 5.1956 |
| 4.9961 | 0.88 | 1500 | 4.9421 |
| 4.7198 | 1.17 | 2000 | 4.8015 |
| 4.5643 | 1.47 | 2500 | 4.6835 |
| 4.4523 | 1.76 | 3000 | 4.5745 |
| 4.3273 | 2.06 | 3500 | 4.4993 |
| 4.1372 | 2.35 | 4000 | 4.4498 |
| 4.1052 | 2.64 | 4500 | 4.3880 |
| 4.0721 | 2.94 | 5000 | 4.3409 |
| 3.8586 | 3.23 | 5500 | 4.3325 |
| 3.8079 | 3.52 | 6000 | 4.3061 |
| 3.7897 | 3.82 | 6500 | 4.2690 |
| 3.678 | 4.11 | 7000 | 4.2702 |
| 3.5266 | 4.4 | 7500 | 4.2641 |
| 3.5165 | 4.7 | 8000 | 4.2488 |
| 3.5069 | 4.99 | 8500 | 4.2361 |
| 3.3367 | 5.28 | 9000 | 4.2512 |
| 3.3295 | 5.58 | 9500 | 4.2494 |
| 3.3275 | 5.87 | 10000 | 4.2480 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
