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

    Neuronpedia

    INDEX
    Explanations

    quotation mark

    np_max-act · gemini-2.0-flash

    The neuron activates on JSON object property names—that is, double‐quoted strings that serve as keys in JSON structures.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     sanit
    -0.07
     Zeit
    -0.07
    ivel
    -0.06
    Alg
    -0.06
    flag
    -0.06
     такі
    -0.06
    .Globalization
    -0.06
    .book
    -0.06
     planners
    -0.06
     Ups
    -0.06
    POSITIVE LOGITS
    βέρ
    0.07
     targeting
    0.07
     oversee
    0.06
    หว
    0.06
    ificate
    0.06
    dispatch
    0.06
     espacio
    0.06
    -folder
    0.06
     кафед
    0.06
     EventArgs
    0.06
    Act Density 0.012%

    No Known Activations