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

    Neuronpedia

    INDEX
    Explanations

    phrases related to rules, guidelines, and codes of conduct

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to rules, standards, and codes of conduct

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    NetMessage
    -0.79
    aline
    -0.75
    icter
    -0.74
    oÄŁan
    -0.69
    du
    -0.69
     Insight
    -0.68
    issance
    -0.68
    asca
    -0.66
    ortun
    -0.65
    umenthal
    -0.65
    POSITIVE LOGITS
     forbids
    1.11
     governs
    1.05
     adherence
    0.97
     limits
    0.96
     decency
    0.96
    rules
    0.94
     dictates
    0.93
     conformity
    0.91
     rules
    0.89
    Minimum
    0.89
    Act Density 0.496%

    No Known Activations