CoolFace
Modelpublic

NewstaR/StableGalen-6b

sourceHugging Faceotherupdated 3y agoView on Hugging Face
1likes155downloads
Model Card

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:

  • targetmodules = ["qproj", "vproj", "gateproj", "downproj", "upproj", "kproj", "oproj"]
  • Learning Rate: 4e-4
  • LR Scheduler: constant
  • 250 Steps <img src="Loss.png" alt="Loss" width="600" height="400" />

T3: 1 Hour