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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating high stakes or consequences related to life, reputation, and interests

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Год
    -0.51
     насељу
    -0.50
     malas
    -0.47
    Bad
    -0.43
    bad
    -0.42
     kuku
    -0.42
    quiao
    -0.42
     labores
    -0.41
     rada
    -0.41
    zący
    -0.41
    POSITIVE LOGITS
     propOrder
    0.90
     stakes
    0.86
     risked
    0.83
     lives
    0.80
     livelihoods
    0.80
    lihood
    0.79
     livelihood
    0.78
     risking
    0.77
     jeopardi
    0.77
     GenerationType
    0.77
    Act Density 0.306%

    No Known Activations