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

    Neuronpedia

    INDEX
    Explanations

    "de" or "base"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    apr
    -0.07
    ï
    -0.07
    Democrats
    -0.07
     dầu
    -0.06
    _profiles
    -0.06
     misunderstand
    -0.06
     masturbating
    -0.06
     Sawyer
    -0.06
    andr
    -0.06
    _uv
    -0.06
    POSITIVE LOGITS
    [w
    0.07
    \">↵
    0.07
    坚固
    0.07
    commercial
    0.07
    нуть
    0.07
    .Ex
    0.07
    ㄱ
    0.06
    	NSString
    0.06
    龛
    0.06
     substantial
    0.06
    Act Density 0.005%

    No Known Activations