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

    Neuronpedia

    INDEX
    Explanations

    closing braces and delimiters in code documents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    code blocks and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     الرياضيه
    -0.81
    yntaxException
    -0.72
    AndEndTag
    -0.71
     kasarigan
    -0.68
    ngdoc
    -0.65
    ➟
    -0.63
    ANDUM
    -0.63
     defaultstate
    -0.62
    󠁴
    -0.60
    adaptiveStyles
    -0.60
    POSITIVE LOGITS
     estre
    0.40
    Natural
    0.35
    <eos>
    0.33
     Natural
    0.32
    let
    0.32
    ↵↵
    0.32
     दोनों
    0.31
     obie
    0.31
    中
    0.30
    export
    0.30
    Act Density 0.007%

    No Known Activations