<!-- 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. -->
aochildes-guten-log-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: 4.3677
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 |
---|---|---|---|
6.7164 | 0.29 | 500 | 5.6323 |
5.3447 | 0.59 | 1000 | 5.2052 |
5.0011 | 0.88 | 1500 | 4.9552 |
4.7272 | 1.17 | 2000 | 4.8144 |
4.5727 | 1.47 | 2500 | 4.6937 |
4.4591 | 1.76 | 3000 | 4.5928 |
4.3272 | 2.05 | 3500 | 4.5232 |
4.1423 | 2.35 | 4000 | 4.4760 |
4.1152 | 2.64 | 4500 | 4.4205 |
4.0725 | 2.93 | 5000 | 4.3703 |
3.8638 | 3.23 | 5500 | 4.3718 |
3.8167 | 3.52 | 6000 | 4.3411 |
3.7993 | 3.81 | 6500 | 4.3167 |
3.6795 | 4.11 | 7000 | 4.3235 |
3.5285 | 4.4 | 7500 | 4.3099 |
3.5218 | 4.69 | 8000 | 4.3012 |
3.5096 | 4.99 | 8500 | 4.2923 |
3.3413 | 5.28 | 9000 | 4.3116 |
3.3298 | 5.57 | 9500 | 4.3113 |
3.3314 | 5.87 | 10000 | 4.3111 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3