<!-- 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. -->
all-guten-merged
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.0973
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 |
---|---|---|---|
6.343 | 0.29 | 500 | 5.3343 |
5.029 | 0.58 | 1000 | 4.9248 |
4.6949 | 0.87 | 1500 | 4.6784 |
4.4411 | 1.16 | 2000 | 4.5350 |
4.2847 | 1.46 | 2500 | 4.4263 |
4.1881 | 1.75 | 3000 | 4.3229 |
4.0768 | 2.04 | 3500 | 4.2482 |
3.8868 | 2.33 | 4000 | 4.2016 |
3.854 | 2.62 | 4500 | 4.1449 |
3.8184 | 2.91 | 5000 | 4.0992 |
3.6422 | 3.2 | 5500 | 4.0917 |
3.5736 | 3.49 | 6000 | 4.0606 |
3.5562 | 3.78 | 6500 | 4.0323 |
3.4752 | 4.07 | 7000 | 4.0253 |
3.3047 | 4.37 | 7500 | 4.0219 |
3.3036 | 4.66 | 8000 | 4.0090 |
3.291 | 4.95 | 8500 | 3.9985 |
3.1484 | 5.24 | 9000 | 4.0090 |
3.1239 | 5.53 | 9500 | 4.0082 |
3.1224 | 5.82 | 10000 | 4.0074 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3