---
library_name: transformers
license: mit
base_model: facebook/xlm-v-base
tags:
- generated_from_trainer
model-index:
- name: multi-wiki-qa-gn-xlm-v-base
  results: []
datasets:
- alexandrainst/multi-wiki-qa
language:
- es
- gn
- grn
- gug
metrics:
- squad
- f1
- exact_match
---

<!-- 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. -->

# multi-wiki-qa-gn-xlm-v-base

This model is a fine-tuned version of [facebook/xlm-v-base](https://huggingface.co/facebook/xlm-v-base) on [alexandrainst/multi-wiki-qa](https://huggingface.co/datasets/alexandrainst/multi-wiki-qa) dataset.

Results on test set:
- exact_match: 31.3373253493014
- f1: 48.315832905161635

## Model description

The best model for QA on my account, prefer this ones over the other models trained with this corpus.

## Intended uses & limitations

Question Answering on Guarani.

## 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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

Results on validation set:
- exact_match: 31.914893617021278
- f1: 45.947126767245244

### Framework versions

- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1