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

    Neuronpedia

    INDEX
    Explanations

    Numbers

    np_max-act · gemini-2.0-flash

    This neuron detects phrases that specify a multi‐part series (e.g. “two-part series,” “three-part series,” “four-part series”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     verified
    -0.07
    tank
    -0.06
     زیر
    -0.06
     Manuals
    -0.06
     refined
    -0.06
    ータ
    -0.06
    _filename
    -0.06
    -wsj
    -0.06
    _add
    -0.06
    Women
    -0.06
    POSITIVE LOGITS
    _nav
    0.07
     tvor
    0.07
     KO
    0.07
    .rightBarButtonItem
    0.07
     golden
    0.06
    (char
    0.06
     electrode
    0.06
    pine
    0.06
     besie
    0.06
     brilliant
    0.06
    Act Density 0.014%

    No Known Activations