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

    Neuronpedia

    INDEX
    Explanations

    Numbers in scientific contexts

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric count tokens (e.g. numbers and associated count words) used in enumerations or listings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Purpose
    -0.09
     phòng
    -0.08
    وو
    -0.07
     ст
    -0.07
    -kind
    -0.07
     divorced
    -0.07
     experts
    -0.07
     Purpose
    -0.07
    線
    -0.07
     Logo
    -0.07
    POSITIVE LOGITS
    _requested
    0.06
    .setItems
    0.06
    caling
    0.06
    _APPS
    0.06
    -Петерб
    0.06
     cron
    0.06
    <decltype
    0.05
    Thor
    0.05
     Retro
    0.05
     musica
    0.05
    Act Density 0.023%

    No Known Activations