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

    Neuronpedia

    INDEX
    Explanations

    repetition penalty and n_init parameter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     setores
    0.77
     vidhan
    0.75
     gati
    0.74
    dsp
    0.73
    atation
    0.70
    gebras
    0.69
     dinners
    0.68
    )}(\
    0.68
     declinations
    0.68
     alfa
    0.67
    POSITIVE LOGITS
    ש
    0.84
    keit
    0.73
    KA
    0.71
     हद
    0.70
    حة
    0.70
     ومن
    0.68
     miejsce
    0.67
    Ottawa
    0.66
    𝒃
    0.66
    ローマ
    0.64
    Act Density 0.002%

    No Known Activations