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

    Neuronpedia

    INDEX
    Explanations

    presentation structure and content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    یشن
    0.85
    说
    0.84
     bellissimo
    0.81
    ಾ
    0.79
     devenu
    0.77
     diventa
    0.77
    arantee
    0.76
    是没有
    0.75
    ो
    0.75
     située
    0.74
    POSITIVE LOGITS
     Бы
    0.95
    token
    0.84
    test
    0.83
    trends
    0.83
    templates
    0.82
    ráf
    0.82
     extremal
    0.82
    xq
    0.82
    train
    0.80
     FAO
    0.80
    Act Density 0.000%

    No Known Activations