metadata
library_name: transformers
base_model: aubmindlab/aragpt2-base
tags:
- generated_from_trainer
model-index:
- name: ArabicGenrationV3
results: []
ArabicGenrationV3
This model is a fine-tuned version of aubmindlab/aragpt2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.0850
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- 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 |
|---|---|---|---|
| 4.8061 | 0.9996 | 2356 | 4.1950 |
| 4.7875 | 1.9997 | 4713 | 4.1114 |
| 4.7615 | 2.9989 | 7068 | 4.0850 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0