Gdańsk, Poland

Piotr Januszewski

Expert
@piojanu

Working on LLMs evaluation 🤖 Aiming to distill human intelligence 🧠

CUDA-im2col-conv. CUDA project for uni subject

26

GradientAILab. Experiments with Deep Reinforcement Learning

13

World-Models. Implementation of World Models: https://worldmodels.github.io.

7

humblerl. Straightforward reinforcement learning python framework

7

talksh. talksh is a minimal, Unix-friendly command-line tool for running LLM prompts on data. It reads from stdin, calls an LLM, and prints the result to stdout.

2

AlphaZero. AlphaZero for Othello, Connect4, GoBang and TicTacToe using Keras and my HumbleRL framework.

1

tf_utils. TensorFlow utilities

1
7
Apply