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

    Neuronpedia

    INDEX
    Explanations

    email addresses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     by
    -3.66
     in
    -2.80
     on
    -2.77
     for
    -2.48
     from
    -2.44
     through
    -2.36
     after
    -2.16
     without
    -2.13
     with
    -2.06
     via
    -1.98
    POSITIVE LOGITS
     étan
    1.71
     bewertet
    1.62
     refroid
    1.62
     durer
    1.62
     emporter
    1.58
     élas
    1.56
    ̣i
    1.55
     problemet
    1.55
     nguyễn
    1.55
     maî
    1.55
    Act Density 0.038%

    No Known Activations