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

    Neuronpedia

    INDEX
    Explanations

    pair

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     DOCUMENT
    -0.06
     metrů
    -0.06
    ΕΛ
    -0.06
     typeName
    -0.06
    Latin
    -0.05
    TODO
    -0.05
     Hook
    -0.05
     ман
    -0.05
     शर
    -0.05
     smarter
    -0.05
    POSITIVE LOGITS
    .NaN
    0.07
     gerne
    0.07
    .CODE
    0.07
    ified
    0.07
    /tiny
    0.07
     BIND
    0.06
     防
    0.06
     เจ
    0.06
     Tube
    0.06
     slashes
    0.06
    Act Density 0.004%

    No Known Activations