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

    Neuronpedia

    INDEX
    Explanations

    protective measures or attributes

    oai_token-act-pair · gpt-3.5-turbo

    references to protective measures or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    licks
    -0.79
    zig
    -0.78
    quart
    -0.73
    mitt
    -0.70
    igon
    -0.70
    upon
    -0.69
    hung
    -0.69
    iddled
    -0.69
    avez
    -0.68
    kers
    -0.67
    POSITIVE LOGITS
     Protective
    0.87
     protective
    0.85
     shield
    0.81
     encl
    0.73
     defenses
    0.73
     custody
    0.72
     shields
    0.72
     defences
    0.71
    eties
    0.71
     ozone
    0.70
    Act Density 0.023%

    No Known Activations