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

    Neuronpedia

    INDEX
    Explanations

    looking for options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     usada
    0.43
     used
    0.42
    做了
    0.41
    用来
    0.39
     benutzt
    0.39
     usato
    0.38
     сдела
    0.38
    used
    0.38
     performed
    0.38
     geändert
    0.37
    POSITIVE LOGITS
     look
    1.21
     Look
    1.09
    Look
    1.06
    look
    0.98
     looking
    0.93
    看向
    0.86
     Looking
    0.84
     LOOK
    0.81
    looking
    0.81
     looked
    0.80
    Act Density 0.023%

    No Known Activations