Skip to content

Instantly share code, notes, and snippets.

View xszheng2020's full-sized avatar
πŸ‘¨β€πŸ’»

Xiaosen Zheng xszheng2020

πŸ‘¨β€πŸ’»
View GitHub Profile
@xszheng2020
xszheng2020 / grpo_demo.py
Created February 9, 2025 10:58 — forked from willccbb/grpo_demo.py
GRPO Llama-1B
# train_grpo.py
import re
import torch
from datasets import load_dataset, Dataset
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import LoraConfig
from trl import GRPOConfig, GRPOTrainer
# Load and prep dataset