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

    Neuronpedia

    INDEX
    Explanations

    a very broad range of frequently used words. Many of these words have high activation in multiple different blocks of text. It does not seem to be matching on a specific concept

    oai_token-act-pair · gemini-2.0-flash

    uncommon words

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    TagHelper
    -0.48
    Географиясе
    -0.43
    styleType
    -0.42
    ault
    -0.40
     piet
    -0.39
    Identyfik
    -0.39
     rū
    -0.38
    ')
    -0.38
     bross
    -0.38
    UNTE
    -0.36
    POSITIVE LOGITS
    GMENT
    0.77
     MainAxisSize
    0.69
     Мексичка
    0.69
    </thead>
    0.68
     الحره
    0.67
    urlpatterns
    0.66
     Himo
    0.66
    +:+
    0.66
     ErrIntOverflow
    0.65
     CreateTagHelper
    0.63
    Act Density 16.268%

    No Known Activations