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

    Neuronpedia

    INDEX
    Explanations

    describing typicality or exceptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Conclusion
    0.92
     wholeheartedly
    0.91
    Conclusion
    0.84
     করিতেছে
    0.82
     CONCLUSION
    0.81
    CONCLUSION
    0.80
     confidently
    0.77
     consistently
    0.77
     profondément
    0.76
     unquestionably
    0.76
    POSITIVE LOGITS
     useful
    1.93
     rare
    1.87
    Rare
    1.70
    rare
    1.67
     uncommon
    1.66
    Useful
    1.66
     Useful
    1.62
     rarer
    1.59
     rarely
    1.59
     selten
    1.58
    Act Density 0.592%

    No Known Activations