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

    Neuronpedia

    INDEX
    Explanations

    programming, math, assistants

    np_max-act · gemini-2.0-flash

    markers that denote the start of an assistant response in the chat transcript (assistant role boundaries).

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Venom
    -0.06
     ประเภท
    -0.06
     Bitcoins
    -0.06
    ロン
    -0.06
    lığın
    -0.06
    OfDay
    -0.06
     UIBar
    -0.06
     isolation
    -0.06
    imetype
    -0.06
    }];↵
    -0.06
    POSITIVE LOGITS
     colorful
    0.07
     hry
    0.07
    .*
    0.07
    accine
    0.06
    ابي
    0.06
     zbo
    0.06
    б
    0.06
     FOUND
    0.06
    미
    0.06
    atsby
    0.06
    Act Density 0.114%

    No Known Activations