Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    code punctuation

    np_max-act · gemini-2.0-flash

    The neuron is detecting leading whitespace (indentation) in code blocks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    :index
    -0.07
    choose
    -0.07
    detector
    -0.06
    .putString
    -0.06
     letzten
    -0.06
    avez
    -0.06
     Kushner
    -0.06
    ubic
    -0.06
    izm
    -0.06
    via
    -0.06
    POSITIVE LOGITS
    ");
↵
↵
    0.07
     mm
    0.07
    alleries
    0.07
     None
    0.07
     {};↵↵
    0.06
    .IN
    0.06
     {};
↵
    0.06
    ";
↵
↵
    0.06
    ":↵↵
    0.06
    (sensor
    0.06
    Act Density 0.081%

    No Known Activations