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

    Neuronpedia

    INDEX
    Explanations

    enabling reports

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Shand
    0.67
     Израи
    0.61
     idempotent
    0.60
     idempot
    0.57
     innocence
    0.55
    eous
    0.53
     друзей
    0.53
    conserv
    0.53
     israel
    0.53
    laugh
    0.52
    POSITIVE LOGITS
     report
    3.01
     Report
    2.99
    Report
    2.89
     REPORT
    2.87
     Reporting
    2.80
     Reports
    2.75
     reports
    2.74
    report
    2.72
     reporting
    2.69
    Reports
    2.63
    Act Density 1.244%

    No Known Activations