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

    Neuronpedia

    INDEX
    Explanations

    makes code better

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     त्याची
    0.83
    もしくは
    0.81
     or
    0.76
     eller
    0.73
     namun
    0.71
    又は
    0.70
     Though
    0.68
     Its
    0.68
     Sometimes
    0.68
    /
    0.68
    POSITIVE LOGITS
     sure
    1.47
     debugging
    1.11
     sense
    1.07
    sure
    1.00
     SURE
    0.99
    debugging
    0.99
     things
    0.99
     жизни
    0.98
     life
    0.98
    Sure
    0.97
    Act Density 0.058%

    No Known Activations