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

    Neuronpedia

    INDEX
    Explanations

    runaway phenomena

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     región
    0.43
    opoulos
    0.38
    地区的
    0.38
    lega
    0.38
    转移
    0.38
     ওয়ার্ড
    0.37
     منطقه
    0.36
     اسر
    0.36
     Dupree
    0.36
    region
    0.36
    POSITIVE LOGITS
     runaway
    0.93
    Run
    0.63
     Run
    0.61
     run
    0.59
    run
    0.55
    RUN
    0.49
     Runnable
    0.48
     rogue
    0.47
     trophic
    0.46
     RUN
    0.45
    Act Density 0.000%

    No Known Activations