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

    Neuronpedia

    INDEX
    Explanations

    Crimes and offenders

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Lights
    -0.07
     elic
    -0.07
     keyboard
    -0.07
     subject
    -0.07
    _words
    -0.07
    손
    -0.06
     Gloss
    -0.06
    .bucket
    -0.06
     sno
    -0.06
    Case
    -0.06
    POSITIVE LOGITS
     Stream
    0.06
    )animated
    0.06
     профилакти
    0.06
     Bry
    0.06
    -initialized
    0.06
    ernity
    0.06
     protesting
    0.06
     آر
    0.06
    ibBundleOrNil
    0.06
     arada
    0.06
    Act Density 0.027%

    No Known Activations