Skip to content

Instantly share code, notes, and snippets.

@liangkaixi
liangkaixi / grpo_demo.py
Created February 13, 2025 03:36 — 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