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

    Neuronpedia

    INDEX
    Explanations

    followed by a colon or comma

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     both
    0.67
     altogether
    0.56
     fully
    0.54
     illustrated
    0.53
     non
    0.52
     head
    0.51
     all
    0.50
     completely
    0.49
     thieves
    0.49
     mart
    0.49
    POSITIVE LOGITS
     ‪
    0.81
    룺
    0.78
     означает
    0.77
    部分は
    0.72
    ƣ
    0.71
     cualidades
    0.70
     danam
    0.70
    (%
    0.70
    ⦕
    0.70
     ___
    0.70
    Act Density 0.108%

    No Known Activations