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

    Neuronpedia

    INDEX
    Explanations

    environment configurations and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ित
    0.58
     debe
    0.58
     devait
    0.58
     reimbursed
    0.57
     deve
    0.57
    ﺘ
    0.54
    ﻏ
    0.54
    RAchievement
    0.53
    CharPtr
    0.53
    ﺮ
    0.53
    POSITIVE LOGITS
    на
    0.68
    ка
    0.67
    There
    0.60
     There
    0.60
    貉
    0.56
    লা
    0.55
    డ్డు
    0.55
    我也是
    0.55
    The
    0.54
     تصویر
    0.54
    Act Density 0.000%

    No Known Activations