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

    Neuronpedia

    INDEX
    Explanations

    Code/Technical documentation

    np_max-act · gemini-2.0-flash

    tokens that signal methodological, technical, or formal academic content (e.g., parts describing methods, procedures, or other formal sections of a document).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     debit
    -0.06
     hugged
    -0.06
    .avatar
    -0.06
     им
    -0.06
    だから
    -0.06
     ({↵
    -0.06
     underage
    -0.05
     Ihnen
    -0.05
    (header
    -0.05
     mist
    -0.05
    POSITIVE LOGITS
    Battle
    0.07
    _closure
    0.07
    obra
    0.07
     Země
    0.07
    άκ
    0.07
    撃
    0.06
     Tank
    0.06
    MEDIA
    0.06
    getField
    0.06
     pencil
    0.06
    Act Density 0.457%

    No Known Activations