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

    Neuronpedia

    INDEX
    Explanations

    years

    np_max-act · gemini-2.0-flash

    This neuron detects numeric tokens—especially fragments of four-digit years (e.g. “1996,” “2001,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    events or competitions related to horse racing and golf.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     tube
    -0.07
    utzer
    -0.07
    ين
    -0.07
    alertView
    -0.07
    My
    -0.06
    throp
    -0.06
    -st
    -0.06
     scenes
    -0.06
    ó
    -0.06
     Swing
    -0.06
    POSITIVE LOGITS
    Χ
    0.07
     Anthem
    0.06
    Perm
    0.06
    .getFullYear
    0.06
    >

↵
    0.06
    _checks
    0.06
     Activate
    0.06
    189
    0.06
     hồ
    0.06
    [P
    0.06
    Act Density 0.004%

    No Known Activations