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

    Neuronpedia

    INDEX
    Explanations

    alphabet letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    coa
    0.44
    乡
    0.41
    Knife
    0.40
    dateTimePicker
    0.39
    ži
    0.38
    eyJ
    0.38
    Composer
    0.38
    awatt
    0.38
    િયન
    0.38
    fte
    0.37
    POSITIVE LOGITS
     letters
    0.89
     Buchstaben
    0.83
     alphabet
    0.83
    字母
    0.81
     alphabetical
    0.80
     Alphabet
    0.78
     LETTERS
    0.78
    alphabet
    0.76
     Letters
    0.75
     numbering
    0.73
    Act Density 0.137%

    No Known Activations