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

    Neuronpedia

    INDEX
    Explanations

    Attention is All You Need

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    illation
    0.72
     મત
    0.70
    িলে
    0.69
    листи
    0.68
     Pil
    0.68
    盖
    0.67
     plantings
    0.65
     voting
    0.64
    cies
    0.64
    рті
    0.64
    POSITIVE LOGITS
     لمبے
    0.66
    Namespace
    0.65
    ̿
    0.64
     Phase
    0.63
    簸
    0.63
     uzun
    0.62
    Ḵ
    0.62
    ﾟ
    0.61
     Domain
    0.61
     ISZ
    0.61
    Act Density 0.005%

    No Known Activations