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
    891
    -0.08
    21
    -0.07
     인구
    -0.07
     массив
    -0.07
     جریان
    -0.07
    29
    -0.07
    Plan
    -0.07
    模型
    -0.07
     Flow
    -0.06
    ChangeEvent
    -0.06
    POSITIVE LOGITS
     scratch
    0.12
     scratched
    0.11
     Scr
    0.11
     scratches
    0.10
     scrub
    0.09
     scrutiny
    0.09
     scratching
    0.09
     Scratch
    0.08
    Scr
    0.08
     SCR
    0.08
    Act Density 0.009%

    No Known Activations