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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on decimal numeric tokens (floating‐point version numbers) embedded in code or metadata.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Writers
    -0.07
    _CONFIG
    -0.07
     koup
    -0.07
     �
    -0.06
    PRINTF
    -0.06
     Consultant
    -0.06
     physique
    -0.06
    'r
    -0.06
    -day
    -0.06
     RB
    -0.06
    POSITIVE LOGITS
    stab
    0.07
    NSIndexPath
    0.07
     ortaya
    0.06
    時代
    0.06
     sağlan
    0.06
    ampoline
    0.06
    чер
    0.06
    LogLevel
    0.06
     flames
    0.06
    menuItem
    0.06
    Act Density 0.004%

    No Known Activations