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

    Neuronpedia

    INDEX
    Explanations

    Suspenseful plotlines

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    かる
    -0.07
    還
    -0.07
    _optimizer
    -0.07
     komt
    -0.06
    けれど
    -0.06
     לקוח
    -0.06
     wast
    -0.06
    =q
    -0.06
     kijken
    -0.06
     setzen
    -0.06
    POSITIVE LOGITS
    LOGGER
    0.07
    fprintf
    0.07
    _likelihood
    0.07
    chains
    0.07
    GCC
    0.07
    _TRANS
    0.07
    ProducesResponseType
    0.07
    _answer
    0.07
    _DEFIN
    0.07
    全产业链
    0.06
    Act Density 0.136%

    No Known Activations