yoavg

Learning LLMs in 2025

So you know how the transformer works, and you know basic ML/DL, and you want to learn more about LLMs. One way to go is looking into the various "algorithmic" stuff (optimization algorithms, RL, DPO, etc). Lot's of materials on that. But the interesting stuff is (in my opinion at least) not there.

This is an attempt to collect a list of academic (or academic-like) materials that explore LLMs from other directions, and focus on the non-ML-algorithmic aspects.

Courses

• David Chiang's Theory of Neural Networks course.
• This is not primarily LLMs, but does have substantial section on Transformers. Formal/Theory. More of a book than a course.

kieranklaassen

You are an AI assistant tasked with creating a highly engaging, personalized check-in flow for a user. This flow should emulate a beautifully designed iOS app, focusing on simplicity, clear call-to-actions, and an overall delightful user experience. Your role combines that of a personality coach and an expert UX designer.

Here's the theme for today's check-in: {{THEME}}

And here's the context we have about the user: <user_context> {{USER_CONTEXT}}

willccbb

# train_grpo.py
#
# See https://github.com/willccbb/verifiers for ongoing developments
#
"""
citation:

@misc{brown2025grpodemo,
  title={Granular Format Rewards for Eliciting Mathematical Reasoning Capabilities in Small Language Models},
  author={Brown, William},

YouJiacheng

import os
import sys
from typing import override

with open(sys.argv[0]) as f:
    code = f.read() # read the code of this file ASAP, for logging
os.environ["PYTORCH_CUDA_ALLOC_CONF"] = "expandable_segments:True"
import contextlib
import time
import uuid

Maharshi-Pandya

You are an assistant that engages in extremely thorough, self-questioning reasoning. Your approach mirrors human stream-of-consciousness thinking, characterized by continuous exploration, self-doubt, and iterative analysis.

## Core Principles

1. EXPLORATION OVER CONCLUSION
- Never rush to conclusions
- Keep exploring until a solution emerges naturally from the evidence
- If uncertain, continue reasoning indefinitely
- Question every assumption and inference

philschmid

import asyncio
import base64
import json
import os
import pyaudio
from websockets.asyncio.client import connect


class SimpleGeminiVoice:
    def __init__(self):

charlesfrye

# Train GPT-2 in five minutes -- for free
#
# ```bash
# pip install modal
# modal setup
# modal run wrapper.py
# ```
#
# Note that the end-to-end latency the first time is more like 25 minutes:
# - five minutes to install Torch (rip)

virattt

deepfates

import argparse
import json
import logging
import os
import re
import shutil
from concurrent.futures import ProcessPoolExecutor, as_completed
from dataclasses import dataclass
from datetime import datetime
from typing import Any, Callable, Dict, List, Literal, Optional, Tuple

stenuto

#!/bin/bash

# Function to display usage information
usage() {
    echo "Usage: $0 /path/to/input.mp4 [ /path/to/output_directory ]"
    exit 1
}

# Check if at least one argument (input file) is provided
if [ $# -lt 1 ]; then