---
library_name: peft
license: mit
base_model: facebook/mbart-large-50
tags:
- base_model:adapter:facebook/mbart-large-50
- lora
- transformers
model-index:
- name: mbart-large-50-GEC-spanish-LORA-merged
  results: []
---

<!-- 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. -->

# mbart-large-50-GEC-spanish-LORA-merged

This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 6.8877
- Gleu: 0.8475

## 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: 0.00044406688888615583
- train_batch_size: 16
- eval_batch_size: 16
- 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
- lr_scheduler_warmup_steps: 0.08309212265191754
- num_epochs: 2
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Gleu   |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 5.4667        | 1.0   | 11213 | 6.8943          | 0.8234 |
| 5.4257        | 2.0   | 22426 | 6.8877          | 0.8475 |


### Framework versions

- PEFT 0.18.1
- Transformers 5.0.0
- Pytorch 2.7.0+cu126
- Datasets 4.8.5
- Tokenizers 0.22.2