gpt2-french-rap

<!-- 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. -->

gpt2-french-rap

This model is a fine-tuned version of dbddv01/gpt2-french-small on the None dataset.

Model description

It's trained on a database of 37 000 texts of French rap lyrics to generate lyrics from a word or a sentence. It's limited but functionable.

Intended uses & limitations

More information needed

Training and evaluation data

A corpus of 37 000 lyrics of French rap music.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Framework versions

• Transformers 4.54.0
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.2