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

    Neuronpedia

    INDEX
    Explanations

    run afoul of

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ya
    -0.08
    Files
    -0.07
    Κ
    -0.06
    shuffle
    -0.06
    span
    -0.06
     스
    -0.06
    oupon
    -0.06
    bert
    -0.06
    θος
    -0.06
     값을
    -0.06
    POSITIVE LOGITS
    ではない
    0.07
     Meet
    0.06
    عود
    0.06
     didn
    0.06
    didn
    0.06
    агато
    0.06
    _Details
    0.06
    unal
    0.06
    λμ
    0.06
    .Support
    0.05
    Act Density 0.024%

    No Known Activations