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

    Neuronpedia

    INDEX
    Explanations

    Separates introduction

    np_max-act · gemini-2.0-flash

    The neuron is picking up on document section dividers—lines composed of repeated “=” characters (the Markdown‐style headings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .piece
    -0.07
     vrouw
    -0.07
     sake
    -0.07
    tright
    -0.07
    disposed
    -0.07
    _aux
    -0.06
    (br
    -0.06
     Dias
    -0.06
    Opened
    -0.06
    _Runtime
    -0.06
    POSITIVE LOGITS
    ա
    0.07
    _stop
    0.06
     sex
    0.06
    Ａ
    0.06
    Prior
    0.06
     Украї
    0.06
     ضو
    0.06
    -bre
    0.06
    _COMPONENT
    0.06
     Celebrity
    0.06
    Act Density 0.002%

    No Known Activations