Skip to content

Instantly share code, notes, and snippets.

View Bostigger's full-sized avatar
A Time is Coming

Peter Blay Bostigger

A Time is Coming
  • @codexpals
  • Ghana
View GitHub Profile
@Bostigger
Bostigger / grpo_demo.py
Created February 4, 2025 21:00 — 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