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

    Neuronpedia

    INDEX
    Explanations

    spindle

    np_max-act · gemini-2.0-flash

    The neuron consistently fires on any token containing the substring “spindle,” effectively detecting mentions of “spindle.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .ast
    -0.07
    °C
    -0.07
    ât
    -0.07
     아이
    -0.06
    aaS
    -0.06
     рассказ
    -0.06
    접
    -0.06
    expectException
    -0.06
    getMock
    -0.06
     शर
    -0.06
    POSITIVE LOGITS
     spindle
    0.11
     Downs
    0.08
    bsd
    0.07
     Synd
    0.07
     Clyde
    0.07
     defensively
    0.07
     Sydney
    0.07
     Medieval
    0.07
    �
    0.07
    SK
    0.07
    Act Density 0.001%

    No Known Activations