---
license: apache-2.0
base_model: riotu-lab/ArabianGPT-03B
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_eg_03
  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. -->

# res_nw_eg_03

This model is a fine-tuned version of [riotu-lab/ArabianGPT-03B](https://huggingface.co/riotu-lab/ArabianGPT-03B) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.7603
- Bleu: 0.3632
- Rouge1: 0.6403
- Rouge2: 0.3829
- Rougel: 0.6368

## 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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20.0

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Bleu   | Rouge1 | Rouge2 | Rougel |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|:------:|:------:|
| 0.874         | 1.0   | 7105  | 0.7882          | 0.3433 | 0.6159 | 0.3508 | 0.6121 |
| 0.5689        | 2.0   | 14210 | 0.7603          | 0.3632 | 0.6403 | 0.3829 | 0.6368 |
| 0.4369        | 3.0   | 21315 | 0.7771          | 0.3661 | 0.6205 | 0.2873 | 0.6174 |
| 0.3314        | 4.0   | 28420 | 0.8176          | 0.3657 | 0.6468 | 0.3919 | 0.6436 |
| 0.2518        | 5.0   | 35525 | 0.8584          | 0.3620 | 0.6466 | 0.3916 | 0.6434 |
| 0.1991        | 6.0   | 42630 | 0.9106          | 0.3616 | 0.6428 | 0.3880 | 0.6396 |
| 0.1678        | 7.0   | 49735 | 0.9511          | 0.3620 | 0.6421 | 0.3875 | 0.6388 |


### Framework versions

- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1