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

    Neuronpedia

    INDEX
    Explanations

    character encodings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     language
    -1.02
     ((
    -0.75
    language
    -0.75
     которое
    -0.74
     Language
    -0.74
    lexia
    -0.73
    away
    -0.71
     (((
    -0.70
     etiam
    -0.70
     buscas
    -0.70
    POSITIVE LOGITS
     ISO
    1.11
    UTF
    1.08
     UTF
    1.01
    ISO
    0.99
     UNIX
    0.91
     utf
    0.90
     strictest
    0.86
    föra
    0.84
    ystemen
    0.82
     Encoding
    0.81
    Act Density 0.043%

    No Known Activations