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

    Neuronpedia

    INDEX
    Explanations

    references to fantasy elements such as creatures, places, religions, histories, and events

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Intake
    -0.86
    ACP
    -0.85
    nesty
    -0.84
    ļéĨĴ
    -0.83
    Shift
    -0.77
    Focus
    -0.76
    annel
    -0.75
    endment
    -0.74
    afe
    -0.72
    hirt
    -0.72
    POSITIVE LOGITS
     mysteries
    1.41
     landscapes
    1.37
     monsters
    1.35
     kingdoms
    1.32
     castles
    1.31
     landmarks
    1.30
     deities
    1.29
     villains
    1.29
     civilizations
    1.28
     dungeons
    1.26
    Act Density 0.488%

    No Known Activations