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

    Neuronpedia

    INDEX
    Explanations

    sourced from

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    From
    0.92
    FROM
    0.88
     From
    0.85
    from
    0.80
     FROM
    0.78
    从
    0.72
    पासून
    0.70
     फ्रॉम
    0.70
     จาก
    0.69
     从
    0.68
    POSITIVE LOGITS
     solely
    0.41
     crawling
    0.39
     mainly
    0.38
     creeping
    0.38
     largely
    0.38
     कसा
    0.38
    இந்
    0.37
     எங்க
    0.37
    Bem
    0.36
     sift
    0.35
    Act Density 0.037%

    No Known Activations