metadata

library_name: transformers
license: mit
base_model: facebook/mbart-large-50
tags:
  - simplification
  - generated_from_trainer
datasets:
  - flores
metrics:
  - bleu
model-index:
  - name: mbart-flores
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: flores
          type: flores
          config: spa_Latn-eng_Latn
          split: devtest
          args: spa_Latn-eng_Latn
        metrics:
          - name: Bleu
            type: bleu
            value: 15.8101

mbart-flores

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

Loss: 0.4899
Bleu: 15.8101
Gen Len: 81.2846

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: 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	125	0.4820	19.6465	32.75
No log	2.0	250	0.4899	15.8101	81.2846

Framework versions

Transformers 4.52.4
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.1