metadata
license: apache-2.0
base_model: riotu-lab/ArabianGPT-1.5B
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_irq_1.5
results: []
res_nw_irq_1.5
This model is a fine-tuned version of riotu-lab/ArabianGPT-1.5B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.1622
- Bleu: 0.3502
- Rouge1: 0.5096
- Rouge2: 0.2469
- Rougel: 0.5076
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 |
|---|---|---|---|---|---|---|---|
| 2.1843 | 1.0 | 1057 | 1.2503 | 0.3206 | 0.4318 | 0.1767 | 0.4294 |
| 0.6485 | 2.0 | 2114 | 1.1708 | 0.3343 | 0.4673 | 0.2064 | 0.4651 |
| 0.5688 | 3.0 | 3171 | 1.1971 | 0.3413 | 0.4874 | 0.2243 | 0.4851 |
| 0.5047 | 4.0 | 4228 | 1.1978 | 0.3500 | 0.5054 | 0.2431 | 0.5030 |
| 0.4482 | 5.0 | 5285 | 1.1622 | 0.3502 | 0.5096 | 0.2469 | 0.5076 |
| 0.3971 | 6.0 | 6342 | 1.2058 | 0.3536 | 0.5150 | 0.2535 | 0.5128 |
| 0.3513 | 7.0 | 7399 | 1.2158 | 0.3519 | 0.5180 | 0.2566 | 0.5155 |
| 0.3105 | 8.0 | 8456 | 1.2344 | 0.3534 | 0.5225 | 0.2630 | 0.5202 |
| 0.2743 | 9.0 | 9513 | 1.2988 | 0.3554 | 0.5277 | 0.2693 | 0.5249 |
| 0.2426 | 10.0 | 10570 | 1.3222 | 0.3551 | 0.5235 | 0.2650 | 0.5207 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1