metadata
license: apache-2.0
base_model: riotu-lab/ArabianGPT-01B
tags:
- generated_from_trainer
metrics:
- bleu
- rouge
model-index:
- name: res_nw_lev
results: []
res_nw_lev
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: 0.4055
- Bleu: 0.5007
- Rouge1: 0.7335
- Rouge2: 0.5267
- Rougel: 0.7327
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 |
|---|---|---|---|---|---|---|---|
| 0.8553 | 1.0 | 5062 | 0.5278 | 0.3576 | 0.5832 | 0.3079 | 0.5813 |
| 0.4665 | 2.0 | 10124 | 0.4682 | 0.3846 | 0.6342 | 0.3690 | 0.6328 |
| 0.3762 | 3.0 | 15186 | 0.4394 | 0.4072 | 0.6631 | 0.4093 | 0.6616 |
| 0.3096 | 4.0 | 20248 | 0.4222 | 0.4283 | 0.6859 | 0.4422 | 0.6848 |
| 0.2588 | 5.0 | 25310 | 0.4118 | 0.4518 | 0.7053 | 0.4745 | 0.7042 |
| 0.2202 | 6.0 | 30372 | 0.4064 | 0.4779 | 0.7203 | 0.5014 | 0.7193 |
| 0.1906 | 7.0 | 35434 | 0.4055 | 0.5007 | 0.7335 | 0.5267 | 0.7327 |
| 0.1676 | 8.0 | 40496 | 0.4076 | 0.5192 | 0.7432 | 0.5456 | 0.7423 |
| 0.1502 | 9.0 | 45558 | 0.4122 | 0.5341 | 0.7496 | 0.5602 | 0.7487 |
| 0.1371 | 10.0 | 50620 | 0.4182 | 0.5453 | 0.7534 | 0.5668 | 0.7525 |
| 0.1275 | 11.0 | 55682 | 0.4228 | 0.5523 | 0.7562 | 0.5730 | 0.7552 |
| 0.1202 | 12.0 | 60744 | 0.4293 | 0.5545 | 0.7580 | 0.5762 | 0.7572 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1