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

    Neuronpedia

    INDEX
    Explanations

    processing

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on present‐ and past‐tense verbs that denote technical or mechanical processes (e.g. amplify, filter(ed), cut, measure, process).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     핵
    -0.07
     않은
    -0.07
     इल
    -0.07
     дозволя
    -0.07
    альном
    -0.07
    _bbox
    -0.07
    ورش
    -0.06
    sofar
    -0.06
     دولت
    -0.06
     следующ
    -0.06
    POSITIVE LOGITS
    /notification
    0.07
     Personally
    0.06
    -flat
    0.06
     Pres
    0.06
     Rim
    0.06
     Apart
    0.06
     Couldn
    0.06
    Visual
    0.06
    ype
    0.06
     IGN
    0.06
    Act Density 0.051%

    No Known Activations