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

    Neuronpedia

    INDEX
    Explanations

    statements referring to underlying issues or structures

    oai_token-act-pair · gpt-3.5-turbo

    references to fundamental issues or structures that influence various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asia
    -0.85
    alde
    -0.83
    apo
    -0.82
    aterasu
    -0.81
    ander
    -0.81
    cture
    -0.79
    ishers
    -0.79
    chens
    -0.78
    avis
    -0.78
    alid
    -0.74
    POSITIVE LOGITS
     assumptions
    0.96
     principles
    0.95
     assumption
    0.94
     infrastructure
    0.91
     premise
    0.91
     fundamentals
    0.89
     structure
    0.88
     principle
    0.87
     theme
    0.87
     motivations
    0.85
    Act Density 0.025%

    No Known Activations