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