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

    Neuronpedia

    INDEX
    Explanations

    Professional emails

    np_max-act · gemini-2.0-flash

    detects polite user requests for composing or revising emails and the surrounding instructional context.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     است
    -0.07
     двух
    -0.07
    QUOTE
    -0.06
    New
    -0.06
    /S
    -0.06
    Xi
    -0.06
     prayers
    -0.06
     PLACE
    -0.06
     Now
    -0.06
     waar
    -0.06
    POSITIVE LOGITS
     interchangeable
    0.07
     societies
    0.07
    .closePath
    0.07
     φορ
    0.07
     Nevertheless
    0.07
    ActivityIndicatorView
    0.06
     FBI
    0.06
    	Gtk
    0.06
    ecess
    0.06
    مز
    0.06
    Act Density 0.060%

    No Known Activations