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

    Neuronpedia

    INDEX
    Explanations

    alphabetical characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     keyboard
    -0.97
     keyboards
    -0.81
     keypad
    -0.79
     teclado
    -0.79
     cắt
    -0.76
     SEEK
    -0.75
     abbreviation
    -0.74
     ચ
    -0.73
     sylla
    -0.71
    чат
    -0.70
    POSITIVE LOGITS
     AZ
    1.10
    AZ
    0.98
    isalpha
    0.93
     A
    0.91
     alpha
    0.90
     Alpha
    0.90
    Capital
    0.89
     Capital
    0.84
     α
    0.83
    ABC
    0.82
    Act Density 0.027%

    No Known Activations