This model is the DeciLM-6b-Instruct model, trained specifically for medicine
Galen uses the
### User: {prompt}
### Response:
or
{prompt}
Prompt templates
Galen Training Recipe:
- target_modules = ["q_proj", "v_proj", "gate_proj", "down_proj", "up_proj", "k_proj", "o_proj"]
- Learning Rate: 4e-4
- LR Scheduler: constant
- 250 Steps
T3: 1 Hour
- Downloads last month
- 800
Inference API (serverless) does not yet support model repos that contain custom code.