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

    Neuronpedia

    INDEX
    Explanations

    Math problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     increasingly
    -0.08
     fum
    -0.07
    Thank
    -0.07
    _DIRECTORY
    -0.07
     truck
    -0.07
     viewpoints
    -0.06
    foam
    -0.06
     Best
    -0.06
     best
    -0.06
     Ultimate
    -0.06
    POSITIVE LOGITS
     Tent
    0.09
    tent
    0.09
     第一
    0.09
     tentative
    0.09
    Tent
    0.08
    ировали
    0.08
     الجانب
    0.08
    �ង
    0.08
    әи
    0.08
     selesai
    0.08
    Act Density 0.083%

    No Known Activations