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

    Neuronpedia

    INDEX
    Explanations

    what

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     무엇
    0.59
    まずは
    0.56
    нед
    0.55
    what
    0.54
    僕は
    0.53
    मुझे
    0.53
     Inches
    0.53
    What
    0.52
     क्या
    0.52
     삼각형
    0.52
    POSITIVE LOGITS
     kind
    1.23
     kinds
    1.13
    kinds
    1.00
    kind
    0.98
     KIND
    0.96
    Kind
    0.89
     kinda
    0.86
     Kind
    0.81
    Kinds
    0.73
     sort
    0.72
    Act Density 0.073%

    No Known Activations