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

    Neuronpedia

    INDEX
    Explanations

    uppercase acronyms

    np_acts-logits-general · gemini-2.5-flash-lite

    TDM involves

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     I
    1.32
    1.07
    l
    1.00
     U
    0.91
     L
    0.91
     J
    0.90
     C
    0.85
     G
    0.80
     S
    0.80
     K
    0.79
    POSITIVE LOGITS
    на
    1.80
    و
    1.31
    ح
    1.30
    ج
    1.29
    2
    1.21
    ف
    1.21
    ח
    1.14
    ান
    1.12
    ور
    1.12
     on
    1.11
    Act Density 1.558%

    No Known Activations