samzheng/Mistral-SymbolicMath-7B-Lora

model_name: "Mistral-7B-Math (LoRA Adapters Only)" \ repo: "samzheng/mistral-7b-math-lora"\ description: This repo contains only the LoRA adapter weights (~200 MB) trained for grade-school symbolic math. Load them on top of the 4-bit base model to save disk and download time.

quick_start:

from transformers import AutoModelForCausalLM, AutoTokenizer from peft import PeftModel

baseid = "unsloth/mistral-7b-instruct-v0.3-bnb-4bit" loraid = "yourusername/mistral-7b-math-lora"

tok = AutoTokenizer.frompretrained(baseid) base = AutoModelForCausalLM.frompretrained(baseid, loadin4bit=True, devicemap="auto") model = PeftModel.frompretrained(base, lora_id) # inject adapters

# generate prompt = "... Alpaca-formatted prompt ..." out = model.generate(**tok(prompt, returntensors="pt").to(model.device), maxnewtokens=256) print(tok.decode(out[0], skipspecial_tokens=True))