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

    Neuronpedia

    INDEX
    Explanations

    i can / you can

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     saying
    1.14
     running
    1.09
     asking
    1.02
     stating
    1.01
     forcing
    1.00
     pointing
    0.99
     bothering
    0.98
     recording
    0.97
     describing
    0.97
     driving
    0.97
    POSITIVE LOGITS
     memiliki
    1.48
     может
    1.48
     обладает
    1.43
     можуть
    1.41
     cannot
    1.38
     tiene
    1.32
    provides
    1.32
     обеспечивает
    1.31
     είναι
    1.31
     може
    1.30
    Act Density 0.003%

    No Known Activations