Meowu Meowu

Release Notes Template

Based off https://palantir.quip.com/pzRwAVr1bpzf

Pro-tip: look through the github diff between the previous release to see what's changed. The commit titles should give an outline of what's happened.

Upgrade Steps

List out, as concretely as possible, any steps users have to take when they upgrade beyond just dumping the dependency.
Write pseudocode that highlights what code should change and how.

	# 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