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

    Neuronpedia

    INDEX
    Explanations

    ropic

    np_max-act · gemini-2.0-flash

    This neuron detects compound descriptors of dual/multiple function or roles (e.g. “dual‐purpose,” “bifunctional,” “utility‐man–turned‐quarterback”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     postpone
    -0.06
    'use
    -0.06
     wy
    -0.06
    :System
    -0.06
     ASUS
    -0.06
     Vale
    -0.06
     {};
    -0.06
     Musik
    -0.06
     Giải
    -0.06
     честь
    -0.06
    POSITIVE LOGITS
     الوز
    0.07
    sez
    0.07
    ourney
    0.06
    idx
    0.06
     जनत
    0.06
     enriched
    0.06
     governance
    0.06
     abound
    0.06
    attendance
    0.06
    icer
    0.06
    Act Density 0.334%

    No Known Activations