metadata
library_name: transformers
license: mit
base_model: openai-community/gpt2
tags:
- generated_from_trainer
model-index:
- name: arabic-nano-gpt-v1
results: []
arabic-nano-gpt-v1
This model is a fine-tuned version of openai-community/gpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 8.3925
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: 0.0006
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.1603 | 6.6667 | 100 | 6.8927 |
| 5.0605 | 13.3333 | 200 | 6.6411 |
| 3.1601 | 20.0 | 300 | 7.0113 |
| 1.6772 | 26.6667 | 400 | 7.4615 |
| 0.7315 | 33.3333 | 500 | 7.8263 |
| 0.3081 | 40.0 | 600 | 8.1310 |
| 0.1618 | 46.6667 | 700 | 8.3925 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.5.0
- Datasets 3.0.1
- Tokenizers 0.20.1