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

    Neuronpedia

    INDEX
    Explanations

    away

    np_max-act · gemini-2.0-flash

    The neuron specifically lights up on arithmetic-operation words (like “times,” “multiplied,” “take away,” “minus,” “plus,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    loggedin
    -0.07
    粉
    -0.06
    Sunday
    -0.06
     IntelliJ
    -0.06
     Только
    -0.06
     asserts
    -0.06
     好
    -0.06
    み
    -0.06
    unny
    -0.06
    Brit
    -0.06
    POSITIVE LOGITS
    _vel
    0.06
    غراف
    0.06
     Atmos
    0.06
     jal
    0.06
    (op
    0.06
     ENV
    0.06
     profund
    0.06
     POS
    0.06
     �
    0.06
     impeccable
    0.06
    Act Density 0.016%

    No Known Activations