About lczerolens

Goal

The purpose of this library is to provide analysis tools for lc0 networks with torch.

Resources

This page provides curated links to relevant research papers, tools, and educational resources that complement lczerolens functionality.

Explainable AI (XAI)

Some XAI concepts and techniques:

• Linear probing - Understanding learned representations through linear classifiers

• Activation patching - Causal intervention methods for model analysis

Chess XAI

Research specifically focused on interpretability in chess-playing neural networks:

• Acquisition of chess knowledge in AlphaZero - Analysis of how AlphaZero learns chess concepts

• Evidence of Learned Look-Ahead in a Chess-Playing Neural Network - Understanding planning capabilities in chess NNs

• Contrastive Sparse Autoencoders for Interpreting Planning of Chess-Playing Agents - Interpretability techniques for chess agents

Chess Engines & AI

Key chess engines and AI systems:

• AlphaZero - The seminal paper on self-play reinforcement learning for chess

• Stockfish - Open-source chess engine reference implementation

• Leela Chess Zero - Open-source neural network chess engine

• Maia - Human-like chess playing neural network

Technical Background

Understanding probabilities and search:

• Log Probabilities - Mathematical foundation for probability calculations

Tree visualization and analysis:

• Chess.com Explorer - Interactive chess position analysis tool

Monte Carlo Tree Search (MCTS):

• lc0 - Leela Chess Zero project homepage

• lc0 blog - Latest updates and technical insights

Neural network fundamentals:

• What is lc0? - Introduction to Leela Chess Zero

• Neural Net Training - Training process for chess networks

• Networks - Network architecture and formats

Neural Network Education

Getting started with neural networks:

• 3Blue1Brown Neural Networks - Visual introduction to neural networks

• Neural Networks: Zero to Hero - Practical neural network implementation guide

• Netron - Visualizer for neural network architectures

Interpretability and analysis:

• Mechanistic Interpretability - Understanding how neural networks work internally

• Zoom In: An Introduction to Circuits - Visual exploration of neural network circuits

Citation

If you use lczerolens in your research, please cite it with the following BibTeX entry:

@software{poupart_lczerolens_2026,
  author = {Poupart, Yoann},
  title = {LCZeroLens},
  version = {0.4.0},
  year = {2026},
  url = {https://github.com/Xmaster6y/lczerolens}
}

License

lczerolens is licensed under the MIT License. See the LICENSE file for details.