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

    Neuronpedia

    INDEX
    Explanations

    organize, initialize, generalize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    种
    1.99
    问题
    1.88
    เป็น
    1.83
    oi
    1.79
    不是
    1.75
    oooo
    1.75
    อยาก
    1.75
    ۰
    1.75
    존
    1.74
    可以
    1.73
    POSITIVE LOGITS
    ле
    2.27
    ب
    2.20
    ли
    2.16
    ্স
    2.05
    क
    2.05
    ्ड
    2.00
     conjunt
    1.76
    า
    1.72
    ﻳ
    1.70
    ча
    1.68
    Act Density 0.368%

    No Known Activations