Gemma 9B Policy QA

Published by @policy-stack · Community adapter

Answers internal policy questions with scoped citations and explicit uncertainty when source text is incomplete.

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2-9b-it")
base = AutoModelForCausalLM.from_pretrained("google/gemma-2-9b-it")
model = PeftModel.from_pretrained(base, "loradock/gemma9b-policy-qa")

inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Compatibility

• peft>=0.11
• transformers>=4.44