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

    Neuronpedia

    INDEX
    Explanations

    lucky

    np_max-act · gemini-2.0-flash

    This neuron detects expressions where the author describes their good fortune or being lucky.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    桐
    -0.07
    .stringify
    -0.07
     Wein
    -0.06
     Romance
    -0.06
    _EOF
    -0.06
    .cor
    -0.06
    372
    -0.06
    编
    -0.06
     Flames
    -0.06
     زمینه
    -0.06
    POSITIVE LOGITS
     lucky
    0.12
     Lucky
    0.10
     unlucky
    0.07
    retval
    0.07
     Pak
    0.07
    lectric
    0.07
    luk
    0.07
    สก
    0.07
    ку
    0.06
     disadvantaged
    0.06
    Act Density 0.002%

    No Known Activations