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

    Neuronpedia

    INDEX
    Explanations

    01

    np_max-act · gemini-2.0-flash

    phrases relating to filmmaking and production.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on tokens representing floating‐point numeric literals (decimal numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Ocean
    -0.06
    QA
    -0.06
     rolled
    -0.06
     aboard
    -0.06
     interviews
    -0.06
     iss
    -0.06
    룡
    -0.06
    ین
    -0.06
     SOCK
    -0.06
    quee
    -0.06
    POSITIVE LOGITS
    /yyyy
    0.08
    -либо
    0.07
    ประส
    0.07
    บาท
    0.07
    _Version
    0.07
    ə
    0.06
     lẫn
    0.06
    numero
    0.06
     اولیه
    0.06
    _cmds
    0.06
    Act Density 0.006%

    No Known Activations