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

    Neuronpedia

    INDEX
    Explanations

    workhorse and work-life balance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     trabajan
    0.55
     working
    0.54
     trabalhar
    0.53
     workings
    0.52
    worked
    0.52
     werkt
    0.52
     работают
    0.51
    working
    0.50
     Trabal
    0.50
     работает
    0.50
    POSITIVE LOGITS
     ethic
    1.04
    horse
    0.91
    arounds
    0.87
    aholic
    0.84
    aday
    0.82
    horses
    0.77
     done
    0.70
    zaam
    0.68
    shops
    0.66
    forces
    0.65
    Act Density 0.088%

    No Known Activations