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

    Neuronpedia

    INDEX
    Explanations

    references to the concepts of "new" and "different."

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

    entirely new or different

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Wikiseite
    -0.51
    <caption>
    -0.45
     instituições
    -0.45
    PreInfinity
    -0.42
    __':
    -0.41
     Ragh
    -0.41
    rident
    -0.40
    Prés
    -0.40
    下载附件
    -0.40
     producers
    -0.39
    POSITIVE LOGITS
     Entire
    0.57
    Entire
    0.55
     entire
    0.54
    intera
    0.52
    entire
    0.51
     ENTIRE
    0.49
    Completely
    0.48
    WireFormatLite
    0.45
    aarrggbb
    0.44
     Completely
    0.44
    Act Density 0.006%

    No Known Activations