Skip to content
View andrewimpellitteri's full-sized avatar

Block or report andrewimpellitteri

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
andrewimpellitteri/README.md

Andrew Impellitteri

Public repos are the smaller stuff:

  • manifold_llm — local 7B forecasting prediction markets, scored by MSE against the market itself
  • llm_poli_compass — political compass and 8values tests run against local models via llama.cpp
  • creativity_bench — Gwern's creativity benchmark, implemented and run
  • floor-cleaner-agent — PPO agent learning to aim a power washer at a dirt field
  • nix-media-server — homelab as a NixOS flake, rebuilt from git
  • mealie_auto_tagger — tagging a self-hosted recipe box with an LLM
  • world-cuisine-connections — similarity graph over world cuisines from ingredient overlap

Older finance and time-series work is on a dead account, @andrew-impell.

Python, Postgres, Flask, AWS, NixOS.

Pinned Loading

  1. llm_poli_compass llm_poli_compass Public

    A program to automate testing open source LLMs for their political compass scores

    Python 11 1

  2. world-cuisine-connections world-cuisine-connections Public

    Similarity graph over world cuisines built from shared-ingredient overlap (TF-IDF and BERT embeddings)

    HTML

  3. creativity_bench creativity_bench Public

    A benchmark for the creativity of LLMs based on Gwern's post

    Python

  4. floor-cleaner-agent floor-cleaner-agent Public

    PPO agent learning to aim a power washer at a simulated dirt field

    Python