Skip to content

Instantly share code, notes, and snippets.

View genescheck's full-sized avatar
🎯
Building Ignition 52, 1 startup a week!

Gene Scheck genescheck

🎯
Building Ignition 52, 1 startup a week!
View GitHub Profile
@genescheck
genescheck / grpo_demo.py
Created February 1, 2025 04:22 — 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
var ContactForm = (function () {
var s;
return {
settings: {
inputs: $("#contact_form input"),
submitBtn: $(".js-contact_form-submit")
},
init: function() {