© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-RES-MATRYOSHKA-DC
    7. 1580
    Prev
    Next
    INDEX
    Explanations

    the word "one", and to a lesser extent, numbers and the word "whole"

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.22.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     etc
    -0.88
    /
    -0.77
     usw
    -0.73
    etc
    -0.69
     itp
    -0.67
     quelconque
    -0.65
    anything
    -0.65
     anything
    -0.65
     Etc
    -0.63
     algún
    -0.60
    POSITIVE LOGITS
    還是
    0.87
    还是
    0.87
    OR
    0.69
     estekak
    0.64
     OR
    0.63
    还是要
    0.60
     alternatively
    0.59
     Or
    0.58
    Or
    0.57
     ore
    0.57
    Activations Density 0.185%

    No Known Activations