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

    Neuronpedia

    INDEX
    Explanations

    urls and markup

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     in
    -2.50
     more
    -2.45
     make
    -2.42
     epitom
    -2.36
     just
    -2.33
    With
    -2.31
     on
    -2.28
     unrelenting
    -2.28
     oddly
    -2.25
     at
    -2.23
    POSITIVE LOGITS
     frança
    2.39
    為に
    2.31
    2.27
     verschillende
    2.22
    2.20
     Dinas
    2.16
    ᛬
    2.16
    だけではなく
    2.13
    ܤ
    2.11
     horrid
    2.08
    Act Density 0.001%

    No Known Activations