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

    Neuronpedia

    INDEX
    Explanations

    getting caught

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     negatively
    -0.07
    ternal
    -0.07
    裝
    -0.07
     intimidate
    -0.06
     Они
    -0.06
    ‌های
    -0.06
    装
    -0.06
     Wichita
    -0.06
     сб
    -0.06
     Yok
    -0.06
    POSITIVE LOGITS
    _compiler
    0.07
    Advertis
    0.07
    /how
    0.06
     mean
    0.06
     converted
    0.06
    _ARM
    0.06
     RuntimeObject
    0.06
     Emmy
    0.06
     vicinity
    0.06
     deque
    0.06
    Act Density 0.075%

    No Known Activations