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

    Neuronpedia

    INDEX
    Explanations

    say 'the ending -ism'

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    -pocket
    -0.08
    盂
    -0.08
    stände
    -0.07
    -display
    -0.07
    ']=
    -0.07
    	e
    -0.07
    想必
    -0.07
     Stap
    -0.07
    ;display
    -0.07
     Bedford
    -0.06
    POSITIVE LOGITS
     wholesalers
    0.07
    褫
    0.07
    风险
    0.07
    区
    0.07
     EGL
    0.06
    แตก
    0.06
    冷
    0.06
     quán
    0.06
    קשה
    0.06
     Idea
    0.06
    Act Density 0.002%

    No Known Activations