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

    Neuronpedia

    INDEX
    Explanations

    Okay, I

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     вы
    1.11
     Вы
    1.01
     значи
    0.99
     сделал
    0.96
     ви
    0.95
     т
    0.94
    assume
    0.94
     Ви
    0.92
     узна
    0.92
    Conclusion
    0.91
    POSITIVE LOGITS
    isasi
    0.88
     salts
    0.82
     ranging
    0.81
     magnetic
    0.78
     commuters
    0.75
     mountain
    0.74
     >>=
    0.73
    ickness
    0.72
     denominators
    0.72
     polysaccharide
    0.71
    Act Density 0.518%

    No Known Activations