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

    Neuronpedia

    INDEX
    Explanations

    text lines and newlines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Holmes
    0.68
    rystall
    0.66
     Crystall
    0.66
     Roads
    0.65
     Crystal
    0.64
    ennial
    0.64
     Approval
    0.63
    研修
    0.63
    経験
    0.62
    zantine
    0.62
    POSITIVE LOGITS
     line
    2.33
     lines
    2.33
    Lines
    2.05
    line
    2.03
    Line
    2.01
    lines
    2.00
     línea
    1.99
     Lines
    1.98
     Line
    1.95
     newline
    1.94
    Act Density 0.769%

    No Known Activations