---
license: mit
base_model: gogamza/kobart-base-v2
tags:
- generated_from_trainer
metrics:
- bleu
model-index:
- name: ko_Jeolla_test
  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. -->

# ko_Jeolla_test

This model is a fine-tuned version of [gogamza/kobart-base-v2](https://huggingface.co./gogamza/kobart-base-v2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 3.6851
- Bleu: 88.1706
- Gen Len: 20.0

## 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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Bleu    | Gen Len |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
| 3.7147        | 1.0   | 15477 | 3.7012          | 87.5571 | 20.0    |
| 3.703         | 2.0   | 30954 | 3.6885          | 88.0637 | 20.0    |
| 3.6911        | 3.0   | 46431 | 3.6851          | 88.1706 | 20.0    |


### Framework versions

- Transformers 4.42.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1