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

    Neuronpedia

    INDEX
    Explanations

    May 9th, early warning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ב
    0.74
    ня
    0.70
    ла
    0.66
    ید
    0.63
    ONG
    0.60
    NY
    0.59
    お
    0.57
     as
    0.56
    いい
    0.56
    quela
    0.56
    POSITIVE LOGITS
     portée
    0.50
     metade
    0.50
    浮
    0.50
    اعر
    0.50
    麪
    0.49
     membrane
    0.49
     unicode
    0.49
     verändert
    0.48
     paren
    0.48
     bv
    0.47
    Act Density 0.000%

    No Known Activations