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

    Neuronpedia

    INDEX
    Explanations

    curvature

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     vested
    -0.07
    鼒
    -0.07
     Braz
    -0.07
    �
    -0.07
    [S
    -0.07
     Hub
    -0.06
     Cbd
    -0.06
    几百
    -0.06
     hog
    -0.06
    .Ag
    -0.06
    POSITIVE LOGITS
    ]')↵
    0.08
    デザイン
    0.07
    ('.');↵
    0.07
     Roller
    0.07
    +'/'+
    0.07
    ;');↵
    0.07
    _VARS
    0.07
    /"+
    0.07
     "",
↵
    0.07
    不认识
    0.07
    Act Density 0.008%

    No Known Activations