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

    Neuronpedia

    INDEX
    Explanations

    Gondorian, Toledo, Elden Ring

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    л
    0.83
    d
    0.73
    l
    0.68
    lassen
    0.68
    g
    0.66
    𝗴
    0.64
    r
    0.63
    ل
    0.63
    lt
    0.62
    ｃ
    0.61
    POSITIVE LOGITS
    IG
    0.60
    IS
    0.57
    А
    0.54
     indulgent
    0.54
    Ин
    0.53
    ی
    0.53
    时间
    0.53
    、
    0.52
    מ
    0.51
    I
    0.50
    Act Density 0.100%

    No Known Activations