metadata
license: apache-2.0
base_model: riotu-lab/ArabianGPT-01B
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_irq
results: []
res_nw_irq
This model is a fine-tuned version of riotu-lab/ArabianGPT-01B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6178
- Bleu: 0.2598
- Rouge1: 0.5921
- Rouge2: 0.3375
- Rougel: 0.5897
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 |
|---|---|---|---|---|---|---|---|
| 1.8421 | 1.0 | 1059 | 0.6648 | 0.1931 | 0.4806 | 0.2142 | 0.4779 |
| 0.591 | 2.0 | 2118 | 0.6045 | 0.2225 | 0.5287 | 0.2615 | 0.5265 |
| 0.4845 | 3.0 | 3177 | 0.5857 | 0.2305 | 0.5524 | 0.2874 | 0.5497 |
| 0.4018 | 4.0 | 4236 | 0.5872 | 0.2416 | 0.5716 | 0.3055 | 0.5693 |
| 0.3351 | 5.0 | 5295 | 0.5903 | 0.2441 | 0.5780 | 0.3175 | 0.5759 |
| 0.2824 | 6.0 | 6354 | 0.5988 | 0.2523 | 0.5857 | 0.3285 | 0.5838 |
| 0.2395 | 7.0 | 7413 | 0.6095 | 0.2571 | 0.5877 | 0.3330 | 0.5861 |
| 0.206 | 8.0 | 8472 | 0.6178 | 0.2598 | 0.5921 | 0.3375 | 0.5897 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1