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

    Neuronpedia

    INDEX
    Explanations

    end of 'only' clause

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    意思是
    0.39
    아야
    0.34
    原因是
    0.32
    হ্যা
    0.31
    Only
    0.31
    文件中
    0.31
    我是
    0.31
    আজ
    0.31
    стная
    0.31
     leftWheel
    0.30
    POSITIVE LOGITS
    。
    0.63
    ।
    0.60
     بشكل
    0.56
    .
    0.54
    ،
    0.52
    ։
    0.51
    ؛
    0.50
    ,
    0.48
    ။
    0.47
    。.
    0.47
    Act Density 0.044%

    No Known Activations