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

    Neuronpedia

    INDEX
    Explanations

    persisted

    np_max-act · gemini-2.0-flash

    document-structure markers and formal section headings (e.g., Title, Abstract, Introduction, Section labels).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    ประกอบ
    -0.06
    oref
    -0.06
    %d
    -0.06
     Interr
    -0.06
     bắc
    -0.06
    511
    -0.06
    .currentState
    -0.06
    _de
    -0.06
    @FindBy
    -0.06
     pond
    -0.06
    POSITIVE LOGITS
            ↵    ↵
    0.07
     توسط
    0.07
     JetBrains
    0.07
     Exception
    0.07
     Blonde
    0.07
     respiratory
    0.07
    _CATEGORY
    0.07
     namespace
    0.06
     matchmaking
    0.06
    ALLOW
    0.06
    Act Density 8.345%

    No Known Activations