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

    Neuronpedia

    INDEX
    Explanations

    multilingual words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    \
    0.68
    ла
    0.63
    :
    0.61
    тят
    0.60
     attham
    0.59
    $,
    0.57
    nál
    0.57
    ه
    0.55
    :[/
    0.55
    ')}
    0.55
    POSITIVE LOGITS
     in
    0.89
    在
    0.88
    도
    0.70
    も
    0.58
    ও
    0.56
    高
    0.54
     wiele
    0.54
    ↵
    0.53
     এবং
    0.51
     algumas
    0.51
    Act Density 0.001%

    No Known Activations