clitron-gh / README.md
chalyi's picture
Upload README.md with huggingface_hub
3dbb707 verified
---
license: llama3.2
language:
- en
tags:
- clitron
- cli
- natural-language
- gguf
- llama-3.2
base_model: meta-llama/Llama-3.2-1B-Instruct
---
# Clitron GH - CLI Interpreter Model
Fine-tuned Llama 3.2 1B model for interpreting natural language commands for GitHub CLI (gh).
## Model Details
- **Base Model:** meta-llama/Llama-3.2-1B-Instruct
- **Training:** SFT with response-only loss masking + EOS token
- **Quantization:** Q4_K_M (GGUF)
- **Size:** 770MB
## Usage
This model is designed to be used with the [clitron](https://github.com/chaliy/clitron) library.
### Input Format
```
<|system|>
You are a CLI command interpreter...
<|user|>
show my open pull requests
<|assistant|>
```
### Output Format
```json
{"command": "pr", "subcommand": "list", "args": {"author": "@me", "state": "open"}, "flags": [], "confidence": 0.95}
```
## Performance
| Metric | Value |
|--------|-------|
| Valid JSON | 94.3% |
| Command Accuracy | 58.3% |
| Invalid JSON | 5.7% |
## License
This model is based on Llama 3.2 and is subject to the [Llama 3.2 Community License](https://www.llama.com/llama3_2/license/).