mbart-neutralization

This model is a fine-tuned version of facebook/mbart-large-50 on Spanish Gender Neutralization dataset. It achieves the following results on the evaluation set:

• Loss: 0.0104
• Bleu: 98.9269
• Gen Len: 18.5

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5.6e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• 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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	440	0.0160	98.0024	18.5417
0.2198	2.0	880	0.0104	98.9269	18.5

Framework versions

• Transformers 4.51.2
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4