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

    Neuronpedia

    INDEX
    Explanations

    seconds followed by opening punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ͡
    1.00
     Букмекерлер
    0.97
    ይም
    0.91
    竝
    0.91
     Букмекердик
    0.89
    \%).
    0.85
    ֗
    0.85
    ٸ
    0.84
    ]$;
    0.82
    乆
    0.81
    POSITIVE LOGITS
     (
    6.69
    （
    4.88
     （
    4.79
    (
    4.22
     ((
    4.05
     ($
    4.02
     ([
    3.95
     ('
    3.95
     (“
    3.90
     ("
    3.90
    Act Density 7.579%

    No Known Activations