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

    Neuronpedia

    INDEX
    Explanations

    key improvements and explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wol
    0.64
     relat
    0.63
     wild
    0.62
     raw
    0.62
     happy
    0.61
     Wharton
    0.61
     verlassen
    0.59
     moul
    0.58
    W
    0.58
     extrap
    0.58
    POSITIVE LOGITS
     сроки
    0.60
     சோ
    0.58
    ptosis
    0.58
     अनुसूचित
    0.57
     яи
    0.56
    అ
    0.56
    ิญ
    0.55
     терро
    0.54
    ക്ര
    0.53
    AMA
    0.53
    Act Density 0.184%

    No Known Activations