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

    Neuronpedia

    INDEX
    Explanations

    opposite manner as

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    U
    0.87
    W
    0.77
     powied
    0.74
    '
    0.72
    K
    0.69
    에
    0.68
    ższ
    0.68
    backgroundColor
    0.68
     ומ
    0.64
     aconteceu
    0.63
    POSITIVE LOGITS
    ler
    0.96
    ing
    0.90
    ة
    0.86
    ется
    0.78
    ле
    0.72
    ла
    0.71
     हिस्सा
    0.67
    ING
    0.64
    ей
    0.64
    кте
    0.64
    Act Density 0.001%

    No Known Activations