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

    Neuronpedia

    INDEX
    Explanations

    R

    np_max-act · gemini-2.0-flash

    template and meta-structural prompt text (chat headers, placeholder tokens like NAME_#, and standalone capital letters/section markers) indicating formatted instructions or scaffolding.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    μέν
    -0.06
    angling
    -0.06
     surrendered
    -0.06
    اسم
    -0.06
    sch
    -0.06
    ิค
    -0.06
     Continued
    -0.06
    ятия
    -0.06
    gw
    -0.06
    angles
    -0.06
    POSITIVE LOGITS
    EEE
    0.07
     laisse
    0.07
    itesi
    0.07
    antanamo
    0.07
    .getContent
    0.07
    .getFullYear
    0.06
     conex
    0.06
    icut
    0.06
     annunci
    0.06
    .pageY
    0.06
    Act Density 0.430%

    No Known Activations