Neilblaze

Test Time Scaling with MLX LM and R1-based LLMs

Install MLX LM:

pip install mlx-lm

And run:

Neilblaze

# 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

Neilblaze

Classify user search queries as either "Good Google Search Query" or "Bad Google Search Query" based on their likelihood of yielding relevant and helpful results from Google Search.

Input: User search query (text string).

Output: Classification label:
    * Good Google Search Query: The query is likely to be effectively answered by Google Search.
    * Bad Google Search Query: The query is unlikely to be effectively answered by Google Search.  Further categorize "Bad" queries into subtypes for better understanding and classifier training (optional but highly recommended):
        * Chit-Chat/Conversational/Social
        * Personal/Subjective/Opinion-Based (Un-searchable)
        * Vague/Ambiguous/Lacking Specificity

Neilblaze

Begin by enclosing all thoughts within <thinking> tags, exploring multiple angles and approaches.
Break down the solution into clear steps within <step> tags. Start with a 20-step budget, requesting more for complex problems if needed.
Use <count> tags after each step to show the remaining budget. Stop when reaching 0.
Continuously adjust your reasoning based on intermediate results and reflections, adapting your strategy as you progress.
Regularly evaluate progress using <reflection> tags. Be critical and honest about your reasoning process.
Assign a quality score between 0.0 and 1.0 using <reward> tags after each reflection. Use this to guide your approach:

0.8+: Continue current approach
0.5-0.7: Consider minor adjustments
Below 0.5: Seriously consider backtracking and trying a different approach

Neilblaze

Memory Optimization (Christer Ericson, GDC 2003)
http://realtimecollisiondetection.net/pubs/GDC03_Ericson_Memory_Optimization.ppt

Cache coherency primer (Fabian Giesen)
https://fgiesen.wordpress.com/2014/07/07/cache-coherency/

Code Clinic 2015: How to Write Code the Compiler Can Actually Optimize (Mike Acton)
http://gdcvault.com/play/1021866/Code-Clinic-2015-How-to

Neilblaze

# -----------------------------------------------------------------------------
# AI-powered Git Commit Function
# Copy paste this gist into your ~/.bashrc or ~/.zshrc to gain the `gcm` command. It:
# 1) gets the current staged changed diff
# 2) sends them to an LLM to write the git commit message
# 3) allows you to easily accept, edit, regenerate, cancel
# But - just read and edit the code however you like
# the `llm` CLI util is awesome, can get it here: https://llm.datasette.io/en/stable/

gcm() {

Neilblaze

How To Work With Multiple Github Accounts on a single Machine

Let suppose I have two github accounts, https://github.com/rahul-office and https://github.com/rahul-personal. Now i want to setup my mac to easily talk to both the github accounts.

NOTE: This logic can be extended to more than two accounts also. :)

The setup can be done in 5 easy steps:

Steps:

• Step 1 : Create SSH keys for all accounts
• Step 2 : Add SSH keys to SSH Agent

Neilblaze

{
  "$schema": "https://zed.dev/schema/themes/v0.1.0.json",
  "name": "Vesper",
  "author": "Rauno Freiberg",
  "themes": [
    {
      "name": "Vesper",
      "appearance": "dark",
      "style": {
        "border": "#101010",

Neilblaze

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts and experience preferred (super rare at this point).

My own notes from a few months back.

Background

• Survey of LLMS
• Self-attention and transformer networks

Neilblaze

GROKKING NOTES

I liked the way Grokking the coding interview organized problems into learnable patterns. However, the course is expensive and the majority of the time the problems are copy-pasted from leetcode. As the explanations on leetcode are usually just as good, the course really boils down to being a glorified curated list of leetcode problems.

So below I made a list of leetcode problems that are as close to grokking problems as possible.

Pattern: Sliding Window

• https://leetcode.com/problems/maximum-subarray/ # Close enough
• https://leetcode.com/problems/minimum-size-subarray-sum/