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

    Neuronpedia

    INDEX
    Explanations

    difference between

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    AT
    0.57
    (
    0.57
    ATI
    0.54
    A
    0.54
    Asia
    0.54
    Api
    0.53
    S
    0.52
    AC
    0.51
    IC
    0.51
    Q
    0.50
    POSITIVE LOGITS
     چیمپ
    0.52
    .³
    0.52
     palindrome
    0.49
     terci
    0.47
    marshalN
    0.46
    cią
    0.46
     انسان
    0.46
    luk
    0.46
     পার্থক্য
    0.45
    nonzero
    0.45
    Act Density 0.002%

    No Known Activations