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

    Neuronpedia

    INDEX
    Explanations

    numbers like 0-9

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     million
    0.86
     pretty
    0.77
     impressive
    0.74
     committee
    0.74
     fost
    0.73
     credit
    0.73
     факт
    0.73
     pusher
    0.73
     dure
    0.72
    成绩
    0.72
    POSITIVE LOGITS
    ABC
    0.91
    ABCD
    0.90
    Scaling
    0.72
    Feed
    0.69
    Initialize
    0.69
    Original
    0.68
     ABCD
    0.67
    Hy
    0.66
    OpenJDK
    0.66
    Scale
    0.66
    Act Density 0.037%

    No Known Activations