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

    Neuronpedia

    INDEX
    Explanations

    an empty list

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Arab
    0.69
     originals
    0.68
    Original
    0.68
    Old
    0.66
     Arabian
    0.64
    Anything
    0.63
     ہوا۔
    0.63
    Emotional
    0.62
     antiga
    0.61
     Ibadan
    0.61
    POSITIVE LOGITS
     empty
    0.95
     blank
    0.79
     boş
    0.69
    empty
    0.67
    空的
    0.67
     Empty
    0.64
    为空
    0.64
    空
    0.64
     populated
    0.62
    emt
    0.62
    Act Density 0.030%

    No Known Activations