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

    Neuronpedia

    INDEX
    Explanations

    scratch

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    341
    -0.08
    	channel
    -0.07
     多
    -0.07
    220
    -0.07
     =(
    -0.07
    τολ
    -0.07
    (pow
    -0.07
    186
    -0.07
    ='<?
    -0.07
    多
    -0.07
    POSITIVE LOGITS
     scratch
    0.09
     Scr
    0.08
     scrub
    0.08
     Scratch
    0.08
    Scr
    0.07
     Scar
    0.07
     scrap
    0.07
    asley
    0.07
     Watson
    0.07
    Whatever
    0.07
    Act Density 0.011%

    No Known Activations