Skip to content

Instantly share code, notes, and snippets.

View oteri's full-sized avatar

foteri oteri

View GitHub Profile
@oteri
oteri / read_paper.py
Created March 18, 2025 10:23 — forked from willccbb/read_paper.py
Arxiv link to Markdown via Mistral OCR (h/t @simonw)
# /// script
# requires-python = ">=3.12"
# dependencies = [
# "click",
# "mistralai",
# "markdown",
# "requests",
# "beautifulsoup4",
# ]
# ///
@oteri
oteri / grpo_demo.py
Created March 18, 2025 10:22 — forked from willccbb/grpo_demo.py
GRPO Llama-1B
# train_grpo.py
#
# See https://github.com/willccbb/verifiers for ongoing developments
#
import re
import torch
from datasets import load_dataset, Dataset
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import LoraConfig
from trl import GRPOConfig, GRPOTrainer