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

    Neuronpedia

    INDEX
    Explanations

    introduces surprising details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ம்
    0.73
    า
    0.71
    ই
    0.65
     only
    0.64
     arba
    0.64
    k
    0.64
    ون
    0.64
     siempre
    0.63
     endast
    0.61
    ss
    0.60
    POSITIVE LOGITS
    handedly
    0.70
     представить
    0.60
    0.53
    ized
    0.51
     తగ్గ
    0.49
    !}
    0.48
    就连
    0.48
    handed
    0.47
    2
    0.47
    !",
    0.46
    Act Density 0.056%

    No Known Activations