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

    Neuronpedia

    INDEX
    Explanations

    but

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ،
    0.31
    ，
    0.30
    、
    0.26
    "،
    0.23
     delimiters
    0.23
     breaks
    0.22
    ,
    0.22
    而言
    0.22
     идет
    0.21
     optimisation
    0.21
    POSITIVE LOGITS
     meaning
    0.42
    but
    0.41
     but
    0.38
     although
    0.37
     लेकिन
    0.37
     لكن
    0.35
     પરંતુ
    0.35
     ولكن
    0.34
     though
    0.33
     لیکن
    0.33
    Act Density 0.633%

    No Known Activations