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

    Neuronpedia

    INDEX
    Explanations

    the presence of parentheses and underscore characters, often associated with programming syntax or code structure

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

    shadow coord

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     חיצוניים
    -0.67
    Хьажоргаш
    -0.63
    IntoConstraints
    -0.63
     виправивши
    -0.60
    ptonshire
    -0.59
     gainera
    -0.58
     defaultstate
    -0.57
     queſta
    -0.57
     kasarigan
    -0.55
    Ārējās
    -0.55
    POSITIVE LOGITS
    setVerticalGroup
    0.38
     these
    0.36
    TagMode
    0.34
    query
    0.34
    Once
    0.33
     check
    0.33
    check
    0.32
     esc
    0.32
     once
    0.31
     if
    0.31
    Act Density 0.045%

    No Known Activations