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

    Neuronpedia

    INDEX
    Explanations

    negative prompt

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﺔ
    1.22
     atteint
    1.13
     auront
    1.10
    க
    1.06
    ﻜ
    1.06
    б
    1.05
    자
    1.02
    ంలో
    1.00
    كان
    0.98
    }{
    0.97
    POSITIVE LOGITS
    ↵↵
    1.27
    ja
    1.25
    ut
    1.24
    oh
    1.23
    on
    1.21
    ali
    1.19
    us
    1.11
    ah
    1.09
    H
    1.09
    hi
    1.09
    Act Density 0.001%

    No Known Activations