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

    Neuronpedia

    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    This neuron fires on imperative/action words (verbs) that prompt user choices or commands in the game-menu/instruction context.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Thing
    -0.07
     since
    -0.07
     Where
    -0.07
    ChartData
    -0.07
    (section
    -0.07
    _Time
    -0.06
     onto
    -0.06
    YY
    -0.06
     Finding
    -0.06
    since
    -0.06
    POSITIVE LOGITS
    a
    0.13
     a
    0.12
    A
    0.12
    'A
    0.12
    an
    0.11
     an
    0.10
    AN
    0.10
     A
    0.10
    An
    0.10
    “A
    0.10
    Act Density 0.177%

    No Known Activations