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

    Neuronpedia

    INDEX
    Explanations

    whitespace and line breaks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    堅
    0.46
     shout
    0.43
    jectives
    0.42
    iptic
    0.41
    æng
    0.41
     range
    0.40
     tights
    0.40
    喊
    0.39
     member
    0.39
     C
    0.39
    POSITIVE LOGITS
     invisible
    0.54
    invisible
    0.54
    空
    0.54
     空
    0.54
    spacer
    0.51
    空白
    0.49
    Gap
    0.48
    gap
    0.47
     spacing
    0.47
     Gap
    0.46
    Act Density 0.067%

    No Known Activations