Modelpublic

rghosh8/gsm8k-nemotron-mini-4b-instruct-rajat-seed-42-G-16

sourceHugging Faceupdated 6mo agoView on Hugging Face

0likes10downloads

Model Card

gsm8k-nemotron-mini-4b-instruct-rajat-seed-42-G-16

LoRA adapter fine-tuned from nvidia/Nemotron-Mini-4B-Instruct on GSM8K using GRPO.

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("rghosh8/gsm8k-nemotron-mini-4b-instruct-rajat-seed-42-G-16", torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("rghosh8/gsm8k-nemotron-mini-4b-instruct-rajat-seed-42-G-16")