HN Explorer

starzmustdie

↗ HN profile [ 1434 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Show HN: #1 On This Day (onthisday-theta.vercel.app)
    18 points 2
  2. A minimal hackable implementation of policy gradients (GRPO, PPO, REINFORCE) (github.com)
    1 points 0
  3. Reasoning Gym: Procedural Dataset Generation for Reinforcement Learning (github.com)
    1 points 0
  4. Show HN: Word Game Bench – evaluating language models on word puzzles (wordgamebench.github.io)

    Hey HN! Word Game Bench is a fun benchmark for evaluating language models on word puzzle games. It is a relatively hard benchmark, where no model currently scores above 50% average win rate. Currently, the models are…

    1 points 0
  5. Show HN: Answers to Chip Huyen's ML Interview Questions (github.com)

    Hi HN, When I was preparing for ML interviews in my first job hunt, I came across Chip Huyen's ML interview questions [1] which I found incredibly helpful. Over several weeks I compiled my answers into a LaTeX document,…

    3 points 0