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

    Neuronpedia

    INDEX
    Explanations

    on

    np_max-act · gemini-2.0-flash

    The neuron preferentially activates on verbs in their “-ing” (present-participle/gerund) form.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    uco
    -0.08
    '",
    -0.07
    ullet
    -0.06
    _FM
    -0.06
    _pkg
    -0.06
    ］
    -0.06
    eah
    -0.06
     bx
    -0.06
    ccount
    -0.06
    .extract
    -0.06
    POSITIVE LOGITS
     Classes
    0.07
    Quite
    0.07
     यह
    0.06
     حجم
    0.06
     거야
    0.06
    北京
    0.06
    Pending
    0.06
    -images
    0.06
     Naples
    0.06
    )(↵
    0.06
    Act Density 0.070%

    No Known Activations