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

    Neuronpedia

    INDEX
    Explanations

    whether differentiating

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ෙස
    0.43
     rollerskates
    0.42
    embro
    0.42
     reli
    0.41
    सद
    0.41
    ား
    0.40
     Utara
    0.40
    σότε
    0.40
     Terrible
    0.40
     historian
    0.39
    POSITIVE LOGITS
    یت
    0.54
     and
    0.53
    ai
    0.43
    ва
    0.43
    ification
    0.42
    né
    0.41
     P
    0.41
    MacroExpansion
    0.41
    side
    0.40
     till
    0.40
    Act Density 0.007%

    No Known Activations