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

    Neuronpedia

    INDEX
    Explanations

    conditional options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mire
    0.74
    redux
    0.74
    ᛁ
    0.73
    磵
    0.72
     note
    0.69
     repaired
    0.69
     benefícios
    0.69
    اسية
    0.68
    掃
    0.68
     bathe
    0.68
    POSITIVE LOGITS
    Else
    0.80
     Else
    0.77
    Entering
    0.77
    else
    0.76
    To
    0.74
    Otherwise
    0.68
    He
    0.67
    ையினர்
    0.67
    他們的
    0.66
    elseif
    0.65
    Act Density 0.000%

    No Known Activations