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

    Neuronpedia

    INDEX
    Explanations

    suicide prevention lines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    社会
    0.75
    憸
    0.69
     пан
    0.68
    áció
    0.67
    氬
    0.67
     cang
    0.67
    degrees
    0.67
     emanc
    0.67
     degrees
    0.66
    சம்
    0.66
    POSITIVE LOGITS
     line
    1.21
     línea
    1.14
    line
    1.04
    Line
    1.00
     lines
    1.00
     Lif
    0.98
     Line
    0.96
     prevention
    0.96
     Prevention
    0.95
     linea
    0.93
    Act Density 0.041%

    No Known Activations