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

    Neuronpedia

    INDEX
    Explanations

    impostor, goblins

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    smith
    0.57
     الت
    0.55
     ек
    0.55
     bene
    0.54
     숙
    0.54
     descre
    0.54
     varias
    0.53
    ´
    0.53
    𝘀
    0.53
    ﻫ
    0.52
    POSITIVE LOGITS
    ルの
    0.67
    ারের
    0.61
    ডের
    0.61
    മ്പ്
    0.57
    ారు
    0.55
    ులు
    0.55
    ンの
    0.53
    রের
    0.52
    ﻒ
    0.51
    φος
    0.51
    Act Density 0.652%

    No Known Activations