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

    Neuronpedia

    INDEX
    Explanations

    Record keeping

    np_max-act · gemini-2.0-flash

    The neuron is tuned to detect mentions of keeping records or documentation (e.g., words like “record,” “diary,” “document”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ्रव
    -0.07
     podmín
    -0.06
     Pandora
    -0.06
    	assertFalse
    -0.06
     hare
    -0.06
    _pv
    -0.06
     επι
    -0.06
    .uniform
    -0.06
    .removeChild
    -0.06
     attended
    -0.06
    POSITIVE LOGITS
    ück
    0.06
                       ↵
    0.06
    uggested
    0.06
    0.06
    (pid
    0.06
    raph
    0.06
    是一个
    0.06
    :')↵
    0.06
    _ZONE
    0.06
    uesta
    0.06
    Act Density 0.010%

    No Known Activations