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

    Neuronpedia

    INDEX
    Explanations

    the word "Good" (sometimes followed by a comma and "Bad")

    oai_token-act-pair · gemini-2.0-flash

    good

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    клопе
    -0.81
    WriteBarrier
    -0.79
     MessageBoxIcon
    -0.75
    rungsseite
    -0.71
    :+:
    -0.68
    IsContent
    -0.66
    JspWriter
    -0.65
    gapan
    -0.64
    JsonHelper
    -0.63
    XmlAccessorType
    -0.62
    POSITIVE LOGITS
     teş
    0.46
    aucune
    0.43
    cestershire
    0.41
     Rücken
    0.41
    الإنجليزية
    0.41
    OTHY
    0.41
    ัติ
    0.40
     cartaz
    0.40
     wä
    0.39
     tavola
    0.38
    Act Density 0.105%

    No Known Activations