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

    Neuronpedia

    INDEX
    Explanations

    way to or from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     muñeco
    -0.96
     افز
    -0.89
     gewinnt
    -0.88
    enddate
    -0.87
     abrigo
    -0.85
    Neden
    -0.84
    DateFormat
    -0.83
     foglal
    -0.80
     cavalo
    -0.80
     baratas
    -0.79
    POSITIVE LOGITS
     to
    2.27
     from
    1.22
     out
    1.17
     home
    1.11
     up
    0.96
     туда
    0.95
     back
    0.92
     down
    0.92
     there
    0.90
     between
    0.81
    Act Density 0.012%

    No Known Activations