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

    Neuronpedia

    INDEX
    Explanations

    contrasting concessions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     সত্যিকার
    0.59
     continually
    0.49
     REALLY
    0.48
     unrival
    0.48
     obscures
    0.47
     percents
    0.47
     gerçekten
    0.47
     radically
    0.47
     masterful
    0.46
    ?!?
    0.46
    POSITIVE LOGITS
     till
    1.16
     upto
    1.12
    till
    1.05
     Apart
    1.04
     lakhs
    1.03
    Till
    1.02
     Till
    1.02
     Gandhiji
    1.02
     Kerala
    1.00
     apart
    0.99
    Act Density 0.007%

    No Known Activations