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

    Neuronpedia

    INDEX
    Explanations

    "Hi" followed by name or punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    프
    0.86
    hình
    0.83
    חק
    0.83
    今日
    0.81
    בת
    0.80
    প্র
    0.79
    dokument
    0.79
    코
    0.78
    но
    0.77
    פר
    0.77
    POSITIVE LOGITS
     fellow
    0.82
     tind
    0.80
     Mr
    0.80
     क्षेत्र
    0.77
     anf
    0.77
     fears
    0.76
     tener
    0.76
    mals
    0.74
     tendency
    0.73
     fais
    0.73
    Act Density 0.015%

    No Known Activations