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

    Neuronpedia

    INDEX
    Explanations

    GPT related words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    襬
    0.69
    の場合
    0.63
    埧
    0.63
    డిన
    0.60
    այի
    0.60
     doth
    0.59
    ल्लाला
    0.59
    ிற்கு
    0.58
    ونکہ
    0.56
    ınıza
    0.56
    POSITIVE LOGITS
    0.78
    IA
    0.77
    ال
    0.70
    ה
    0.68
    ES
    0.66
    GA
    0.66
    AP
    0.65
    IC
    0.63
    a
    0.63
    Apa
    0.62
    Act Density 0.249%

    No Known Activations