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

    Neuronpedia

    INDEX
    Explanations

    enable, lab, somewhat

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Seriously
    0.99
    Grazie
    0.99
    Après
    0.98
    Durante
    0.97
    Die
    0.94
    Estos
    0.92
    Regarding
    0.92
    Aunque
    0.91
    Historically
    0.91
    Eine
    0.91
    POSITIVE LOGITS
    (?)
    1.04
     (?)
    0.98
     (!)
    0.98
    <0xC2>
    0.80
    पणे
    0.79
    (!)
    0.79
     laboratory
    0.75
     original
    0.73
     lab
    0.67
     direct
    0.66
    Act Density 0.600%

    No Known Activations