arabianGPT-news / README.md
mislolofa's picture
End of training
2767dee verified
metadata
license: apache-2.0
base_model: riotu-lab/ArabianGPT-01B
tags:
  - generated_from_trainer
model-index:
  - name: arabianGPT-news
    results: []

arabianGPT-news

This model is a fine-tuned version of riotu-lab/ArabianGPT-01B on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.8910

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3.0

Training results

Training Loss Epoch Step Validation Loss
4.0866 1.0 3310 3.9923
3.8555 2.0 6620 3.9086
3.7504 3.0 9930 3.8910

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.1.2
  • Datasets 2.18.0
  • Tokenizers 0.15.2