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

    Neuronpedia

    INDEX
    Explanations

    ask me a question

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    種の
    0.80
    odoc
    0.79
    ن
    0.79
    dbjc
    0.78
     င
    0.77
    ený
    0.76
     Petsc
    0.76
    ೀ
    0.76
    uva
    0.75
    нні
    0.73
    POSITIVE LOGITS
     and
    0.76
     firing
    0.75
    IH
    0.72
     gluing
    0.71
     anywhere
    0.71
    এবং
    0.70
    および
    0.70
     направ
    0.69
     Anything
    0.68
    中使用
    0.68
    Act Density 0.008%

    No Known Activations