|
|
--- |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
--- |
|
|
|
|
|
<img src="banner.png" alt="AgGPT Banner" width="100%"> |
|
|
|
|
|
# AgGPT-13 |
|
|
|
|
|
## Heavy. Intelligent. Pro. |
|
|
|
|
|
AgGPT-13 has been trained from the ground up to be a powerful and efficient language model, designed to handle a wide range of natural language processing tasks with ease. With its advanced architecture and cutting-edge training techniques, AgGPT-13 is capable of delivering high-quality results for applications such as text generation, summarization, translation, and more. |
|
|
|
|
|
## Installation & Usage |
|
|
|
|
|
```bash |
|
|
pip install torch transformers safetensors |
|
|
``` |
|
|
|
|
|
### Training |
|
|
|
|
|
```bash |
|
|
python train.py config/train_aggpt_char.py --device=cpu --compile=False --eval_iters=20 --log_interval=1 --block_size=64 --batch_size=16 --n_layer=10 --n_head=10 --n_embd=640 --max_iters=5000 --lr_decay_iters=2000 --dropout=0.2 |
|
|
``` |
|
|
### Testing |
|
|
|
|
|
```bash |
|
|
python sample.py --out_dir=out-aggpt --device=cpu |
|
|
``` |
