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

    Neuronpedia

    INDEX
    Explanations

    references to specific cutoff levels used in diagnostic tests

    oai_token-act-pair · gpt-4.1-nano
    New Auto-Interp
    Negative Logits
     kasarigan
    -0.71
    Rüyada
    -0.63
    はじめに
    -0.61
     nakalista
    -0.59
    Tracce
    -0.57
    теризу
    -0.57
     alluminio
    -0.56
    CppCodeGen
    -0.56
    ArgsConstructor
    -0.56
     piede
    -0.55
    POSITIVE LOGITS
     cutoff
    2.86
    cutoff
    2.17
    utoff
    2.13
    截止
    0.79
    بوابة
    0.70
     threshold
    0.68
    threshold
    0.59
     Perſ
    0.59
    orianCalendar
    0.57
     qualifying
    0.56
    Act Density 0.001%

    No Known Activations