poem-gen-gpt2-small-spanish
This model is a fine-tuned version of datificate/gpt2-small-spanish on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.9229
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: 2e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.2121 | 1.0 | 2569 | 3.9954 |
| 4.0612 | 2.0 | 5138 | 3.9375 |
| 3.9988 | 3.0 | 7707 | 3.9229 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.10.0+cu111
- Datasets 2.0.0
- Tokenizers 0.11.6