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

    Neuronpedia

    INDEX
    Explanations

    what

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cls
    -0.07
     infiltration
    -0.07
    zia
    -0.07
     κι
    -0.07
    ěti
    -0.07
    ٔ
    -0.07
    ً
    -0.07
     diagon
    -0.07
     Milli
    -0.06
    Bias
    -0.06
    POSITIVE LOGITS
     what
    0.20
     What
    0.15
    what
    0.15
     WHAT
    0.14
    What
    0.14
    .What
    0.12
    "What
    0.12
    “What
    0.11
    WHAT
    0.10
     whatever
    0.09
    Act Density 0.142%

    No Known Activations