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

    Neuronpedia

    INDEX
    Explanations

    instances of illegal or prohibited activities

    oai_token-act-pair · gpt-3.5-turbo

    references to illegal activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nil
    -0.76
    achine
    -0.75
    erer
    -0.73
    ivities
    -0.72
    oras
    -0.72
    vation
    -0.69
    oran
    -0.69
    oleon
    -0.69
    anche
    -0.69
     enthus
    -0.68
    POSITIVE LOGITS
     illegally
    1.15
     downloading
    0.84
     downloaded
    0.81
     copied
    0.77
     detained
    0.77
     infringing
    0.74
     unlawfully
    0.72
     obtained
    0.71
     accessing
    0.70
     imported
    0.70
    Act Density 0.007%

    No Known Activations