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

    Neuronpedia

    INDEX
    Explanations

    didn

    np_max-act · gemini-2.0-flash

    base stems of English contractions (especially negation forms before the apostrophe).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ()</
    -0.07
    ADMIN
    -0.07
    шив
    -0.06
    ATAR
    -0.06
    ubu
    -0.06
    GER
    -0.06
    saldo
    -0.06
    _cpus
    -0.06
     중국
    -0.06
    lerdir
    -0.06
    POSITIVE LOGITS
    .Is
    0.07
    \",↵
    0.07
    級
    0.06
    .helper
    0.06
     "]");↵
    0.06
     Dual
    0.06
    .Session
    0.06
     наче
    0.06
     nuovo
    0.06
     Sunday
    0.06
    Act Density 0.253%

    No Known Activations