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

    Neuronpedia

    INDEX
    Explanations

    stateless, no memory

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Riz
    0.48
    arz
    0.45
    ў
    0.45
     Jez
    0.44
    тельный
    0.43
    時の
    0.42
    мал
    0.42
    ил
    0.42
     teorie
    0.41
    時
    0.41
    POSITIVE LOGITS
    Cour
    0.45
    同学们
    0.45
    GetCurrent
    0.43
    ultipl
    0.42
    wear
    0.42
    빅
    0.42
     सामान
    0.42
    羡
    0.42
    matter
    0.42
    reon
    0.41
    Act Density 0.000%

    No Known Activations