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

    Neuronpedia

    INDEX
    Explanations

    it

    np_max-act · gemini-2.0-flash

    The neuron fires on action-oriented suggestion verbs (e.g. “go,” “design,” “experience”) that introduce steps or instructions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .cli
    -0.06
     expend
    -0.06
    PhoneNumber
    -0.06
    .power
    -0.06
    ंद
    -0.06
    Robert
    -0.06
    checks
    -0.06
     credit
    -0.06
    Empty
    -0.06
    credit
    -0.06
    POSITIVE LOGITS
    genre
    0.07
    ,len
    0.07
    <-
    0.06
    Vtbl
    0.06
     Somehow
    0.06
     Shane
    0.06
    //[
    0.06
    (parameters
    0.06
     rencontr
    0.06
    nels
    0.06
    Act Density 0.838%

    No Known Activations