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

    Neuronpedia

    INDEX
    Explanations

    qualifying adjectives and adverbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     but
    4.42
    but
    4.01
     nhưng
    3.69
    But
    3.67
     pero
    3.64
    但
    3.62
     But
    3.54
     ولكن
    3.53
     लेकिन
    3.50
     אך
    3.28
    POSITIVE LOGITS
     మాత్రం
    1.18
     그걸
    0.82
     بعدها
    0.72
     afterwards
    0.66
    によっては
    0.66
    )):
    0.65
    nocześnie
    0.65
     வேற
    0.65
    其余
    0.63
    remaining
    0.61
    Act Density 0.181%

    No Known Activations