SBDO2 / README.md
SBDO1's picture
Upload folder using huggingface_hub
006d174 verified
|
Raw
History Blame Contribute Delete
599 Bytes
# SAGE (Q8_0)
Sage - GPT-Neo 1.3B fine-tuned on educational data
- **Base Model**: EleutherAI/gpt-neo-1.3B
- **Quantization**: Q8_0
- **Format**: GGUF (llama.cpp / LM Studio compatible)
- **Architecture**: GPT-NeoX
## Quick Start
### llama.cpp
```bash
./main -m sage-q8_0.gguf -p "Your prompt here"
```
### LM Studio
1. Download `sage-q8_0.gguf`
2. Open LM Studio
3. Load model and start chatting
### Python
```python
from llama_cpp import Llama
llm = Llama(model_path="sage-q8_0.gguf")
output = llm("Hello", max_tokens=128)
print(output['choices'][0]['text'])
```