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

    Neuronpedia

    INDEX
    Explanations

    Sometimes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     probablement
    -0.10
     waarschijnlijk
    -0.09
     probablemente
    -0.09
     provavelmente
    -0.08
     sometime
    -0.08
     meisten
    -0.08
     presumably
    -0.08
    ”为
    -0.08
     plupart
    -0.08
     flesta
    -0.07
    POSITIVE LOGITS
     downright
    0.11
     бывает
    0.10
     Bott
    0.10
     бывают
    0.09
     вовсе
    0.08
     overlooked
    0.08
     પ્રતિ
    0.08
     sogar
    0.08
    씩
    0.08
    甚至
    0.08
    Act Density 0.139%

    No Known Activations