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

    Neuronpedia

    INDEX
    Explanations

    files with documentation or context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     দেরী
    0.74
     reinst
    0.66
     여기에
    0.66
    っかり
    0.64
     obstructions
    0.63
     dispers
    0.63
    囷
    0.62
    }$
    0.62
     inject
    0.62
     replen
    0.62
    POSITIVE LOGITS
    bild
    1.01
    ad
    0.98
    onk
    0.93
    il
    0.92
    ay
    0.91
    onov
    0.91
    anak
    0.90
    ir
    0.89
    on
    0.89
    akal
    0.88
    Act Density 0.001%

    No Known Activations