Skip to content

Instantly share code, notes, and snippets.

View Xianchao-Wu's full-sized avatar

Xianchao Wu Xianchao-Wu

View GitHub Profile
@Xianchao-Wu
Xianchao-Wu / grpo_demo.py
Created February 7, 2025 21:57 — 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