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

    Neuronpedia

    INDEX
    Explanations

    phrases related to evacuation and keeping oneself safe during dangerous situations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     parlar
    -0.76
     domini
    -0.58
    protoimpl
    -0.56
     miras
    -0.54
     flore
    -0.51
    rosz
    -0.51
     impon
    -0.51
     lende
    -0.50
     patin
    -0.50
     liev
    -0.50
    POSITIVE LOGITS
     evacuate
    0.85
     flee
    0.82
     evacuation
    0.82
     escape
    0.79
     evacu
    0.77
     fleeing
    0.70
     unharmed
    0.70
     evacuated
    0.67
     exits
    0.67
     exodus
    0.65
    Act Density 0.334%

    No Known Activations