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

    Neuronpedia

    INDEX
    Explanations

    for followed by duration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hourly
    0.46
     Within
    0.44
    clockwise
    0.39
     eeuw
    0.38
     Hourly
    0.38
     Fewer
    0.37
     Few
    0.36
     tantas
    0.36
    isecond
    0.35
     sebanyak
    0.35
    POSITIVE LOGITS
     months
    1.07
     awhile
    1.06
     weeks
    0.93
     considerable
    0.90
    很久
    0.88
     ages
    0.87
     years
    0.86
    months
    0.84
    一段时间
    0.83
     extended
    0.82
    Act Density 0.053%

    No Known Activations