The Pond

SearchSearch

Search

  • About me
  • My research
  • Random post
  • All posts
  • Open source
  • Subscribe

Tag: interpretability

12 items with this tag.

  • 12/16/2024

    Creating Interpretable Latent Spaces with Gradient Routing

    • interpretability
    • mats program
    • AI

  • 12/5/2024

    Gradient Routing: Masking Gradients to Localize Computation in Neural Networks

    • interpretability
    • mats program
    • AI

  • 12/4/2024

    Deep Causal Transcoding: A Framework for Mechanistically Eliciting Latent Behaviors in Language Models

    • activation engineering
    • interpretability
    • mats program
    • AI

  • 10/27/2024

    Can Transformers Act on Information Beyond an Effective Layer Horizon?

    • understanding the world
    • interpretability
    • AI

  • 7/15/2024

    I Found >800 Orthogonal “Write Code” Steering Vectors

    • activation engineering
    • interpretability
    • mats program
    • AI

  • 4/30/2024

    Mechanistically Eliciting Latent Behaviors in Language Models

    • understanding the world
    • activation engineering
    • interpretability
    • mats program
    • AI

  • 10/13/2023

    Paper: Understanding and Controlling a Maze-Solving Policy Network

    • activation engineering
    • interpretability
    • shard theory
    • AI

  • 5/7/2023

    Residual Stream Norms Grow Exponentially over the Forward Pass

    • interpretability
    • AI

  • 4/20/2023

    Behavioural Statistics for a Maze-Solving Agent

    • interpretability
    • mats program
    • shard theory
    • AI

  • 3/31/2023

    Maze-Solving Agents: Add a Top-Right Vector, Make the Agent Go to the Top-Right

    • activation engineering
    • interpretability
    • AI

  • 3/11/2023

    Understanding and Controlling a Maze-Solving Policy Network

    • activation engineering
    • interpretability
    • mats program
    • shard theory
    • AI

  • 3/1/2023

    Predictions for Shard Theory Mechanistic Interpretability Results

    • interpretability
    • mats program
    • shard theory
    • rationality
    • AI