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

    Neuronpedia

    INDEX
    Explanations

    CHARSET and code syntax

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ि
    2.06
    quele
    1.79
    दक्षि
    1.74
    cü
    1.57
    staben
    1.55
    dau
    1.53
     doz
    1.53
    λοι
    1.53
     pres
    1.50
     agrícola
    1.49
    POSITIVE LOGITS
    lude
    1.89
     vij
    1.86
    Bbb
    1.78
     internalization
    1.72
     iu
    1.69
     soothe
    1.66
    ள்ளது
    1.66
     Rendering
    1.64
    શન
    1.61
    що
    1.60
    Act Density 0.000%

    No Known Activations