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

    Neuronpedia

    INDEX
    Explanations

    say "consonant" word

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     blades
    -0.07
     Oliver
    -0.06
    竽
    -0.06
     lanz
    -0.06
     또한
    -0.06
     mattress
    -0.06
     parts
    -0.06
    (hit
    -0.06
     valve
    -0.06
    枳
    -0.06
    POSITIVE LOGITS
     Honor
    0.08
    synthesize
    0.07
    TITLE
    0.07
    פצ
    0.07
     ********
    0.07
    fixed
    0.07
    赝
    0.06
     slideshow
    0.06
    פש
    0.06
    تحول
    0.06
    Act Density 0.083%

    No Known Activations