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

    Neuronpedia

    INDEX
    Explanations

    stop

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     durations
    -0.07
    _ok
    -0.07
    !"↵
    -0.07
     homework
    -0.07
    Electric
    -0.07
     assertTrue
    -0.06
    !”
    -0.06
    _equiv
    -0.06
    ῆς
    -0.06
     dictionaries
    -0.06
    POSITIVE LOGITS
     dar
    0.07
    ATIC
    0.07
     мон
    0.06
    268
    0.06
    ounces
    0.06
    Analy
    0.06
     weekdays
    0.06
     ive
    0.06
     сент
    0.06
    دیگر
    0.06
    Act Density 0.000%

    No Known Activations