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

    Neuronpedia

    INDEX
    Explanations

    seen or saw changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    es
    0.64
     сил
    0.63
    ோருக்கு
    0.62
    ре
    0.60
    akov
    0.58
    ﻔ
    0.57
    osamente
    0.57
     sustancias
    0.57
    an
    0.57
    ﻠ
    0.56
    POSITIVE LOGITS
     a
    0.59
     n
    0.56
     codebase
    0.56
     bla
    0.55
    第一
    0.54
     revolution
    0.52
     booms
    0.52
     tremendously
    0.52
     revolutions
    0.52
    革命
    0.52
    Act Density 0.021%

    No Known Activations