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

    Neuronpedia

    INDEX
    Explanations

    Latin

    np_max-act · gemini-2.0-flash

    It detects Latin-like subword morphemes or morphological chunks typical of Lorem Ipsum / formal prose tokens.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    Ring
    -0.07
     Berry
    -0.06
     AppComponent
    -0.06
    >();↵↵
    -0.06
    \controllers
    -0.06
    รด
    -0.06
    -valid
    -0.06
    色
    -0.06
     protagon
    -0.06
    Esp
    -0.06
    POSITIVE LOGITS
    -aos
    0.07
    лату
    0.06
    ाक
    0.06
    toDate
    0.06
     řada
    0.06
     odv
    0.06
    luluk
    0.06
    unding
    0.06
    .dispatcher
    0.06
    .days
    0.06
    Act Density 0.004%

    No Known Activations