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

    Neuronpedia

    INDEX
    Explanations

    rephrasing, Tonkatsu

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     will
    1.03
    0.96
     ενώ
    0.95
     would
    0.90
     is
    0.89
     can
    0.89
     C
    0.89
     was
    0.89
    們
    0.88
     N
    0.87
    POSITIVE LOGITS
    arono
    1.07
    heastern
    0.77
    fnamefont
    0.77
    ură
    0.75
    the
    0.75
    ertura
    0.74
    vação
    0.74
     със
    0.73
    ở
    0.72
    をと
    0.71
    Act Density 0.516%

    No Known Activations