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

    Neuronpedia

    INDEX
    Explanations

    breaking down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Out
    1.27
    Out
    1.17
    Up
    1.13
     out
    1.05
     Away
    1.00
    ન
    0.98
     up
    0.97
    ist
    0.96
     off
    0.95
    ان
    0.93
    POSITIVE LOGITS
     spectacularly
    1.08
     immensely
    1.04
     admirable
    1.03
     impressively
    1.03
     hordes
    1.02
     modernist
    1.02
     tremendously
    1.02
     mysteriously
    1.01
     allerlei
    1.01
     barriers
    1.01
    Act Density 0.482%

    No Known Activations