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

    Neuronpedia

    INDEX
    Explanations

    physical objects/structures

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of flat furniture surfaces—especially tables (e.g., “coffee table,” “top of the table”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .VAL
    -0.07
     Hose
    -0.07
     proudly
    -0.07
    (renderer
    -0.07
    '_
    -0.06
     Cart
    -0.06
     filing
    -0.06
    _struct
    -0.06
    .nl
    -0.06
    jsp
    -0.06
    POSITIVE LOGITS
    entions
    0.07
    ฤษภาคม
    0.06
    .people
    0.06
    PositiveButton
    0.06
     expires
    0.06
     TG
    0.06
     небольш
    0.06
     wykon
    0.06
    (limit
    0.06
     Zhao
    0.06
    Act Density 0.069%

    No Known Activations