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

    Neuronpedia

    INDEX
    Explanations

    The neuron is looking for terms related to proprietary information, such as "proprietary challenge-response authentication scheme."

    oai_token-act-pair · gpt-3.5-turbo

    terms related to proprietary and copyrighted content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    OPA
    -0.84
    stad
    -0.82
    aper
    -0.75
    othy
    -0.75
    orah
    -0.73
    SEE
    -0.73
    alos
    -0.72
    ohan
    -0.71
    den
    -0.69
    older
    -0.69
    POSITIVE LOGITS
     proprietary
    0.95
     interests
    0.82
     formulas
    0.80
     materials
    0.80
     protocols
    0.78
     blends
    0.77
     secrets
    0.77
    ĨĴ
    0.76
     jargon
    0.76
     terminology
    0.76
    Act Density 0.018%

    No Known Activations