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

    Neuronpedia

    INDEX
    Explanations

    in filename, user, line

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﺒ
    1.39
    ﺼ
    1.30
    ﻴ
    1.24
    ﻨ
    1.23
    ﺸ
    1.22
    VERTISING
    1.19
    ﺪ
    1.16
    ﺴ
    1.16
    ל
    1.15
    ﺎ
    1.15
    POSITIVE LOGITS
     (
    1.03
    0.99
     prompt
    0.91
    wood
    0.89
     |
    0.89
    cool
    0.87
    e
    0.85
    ↵↵
    0.85
    bers
    0.85
    ei
    0.85
    Act Density 0.035%

    No Known Activations