Spaces:
Sleeping
Sleeping
| title: Local Small LLM Python | |
| colorFrom: blue | |
| colorTo: gray | |
| sdk: gradio | |
| app_file: app.py | |
| pinned: false | |
| license: mit | |
| # Local Small LLM Python | |
| This is a small local language model project written in Python from scratch. | |
| ## What it includes | |
| - Character tokenizer | |
| - Tiny transformer decoder | |
| - Local training loop | |
| - Checkpoint save and load | |
| - Gradio interface for training and generation | |
| ## Important | |
| - No external pretrained LLM is used | |
| - The model is tiny and educational | |
| - It runs on CPU and is meant for local experimentation | |
| - The first generate or train call will initialize and train the tiny model locally | |
| ## Run locally | |
| ```bash | |
| pip install -r requirements.txt | |
| python app.py | |
| ``` | |