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

    Neuronpedia

    INDEX
    Explanations

    list formatting or punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    1.70
    AS
    1.46
    ↵↵
    1.36
    OT
    1.34
    T
    1.23
    Y
    1.23
    IR
    1.20
    RO
    1.19
    AL
    1.17
    ER
    1.16
    POSITIVE LOGITS
    ルの
    1.21
    ים
    1.20
    ﻧ
    1.20
    "
    1.15
    ی
    1.14
    я
    1.11
    ۰
    1.11
    ሳሪያ
    1.09
    ہ
    1.07
    к
    1.05
    Act Density 0.001%

    No Known Activations