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

    Neuronpedia

    INDEX
    Explanations

    scaling or reuse

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    న్‌
    2.21
    엷
    2.20
    喎
    2.17
    팎
    2.14
    믄
    2.04
    喟
    2.00
    튿
    1.99
     지난해
    1.99
    倏
    1.98
    ەند
    1.95
    POSITIVE LOGITS
     alot
    2.40
     idk
    2.18
     देखील
    2.16
     பிரச்சன
    2.13
     bacterias
    2.09
    非常的
    2.08
     fucking
    2.06
     данного
    2.06
     minecraft
    2.03
    成为了
    2.03
    Act Density 2.184%

    No Known Activations