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

    Neuronpedia

    INDEX
    Explanations

    phrases related to procedural matters

    oai_token-act-pair · gpt-3.5-turbo

    references to procedural processes or issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ffee
    -0.84
    sole
    -0.84
    words
    -0.83
    wan
    -0.80
    ofi
    -0.78
    ammad
    -0.78
    mouth
    -0.77
    worth
    -0.77
    tub
    -0.75
    hoff
    -0.75
    POSITIVE LOGITS
     procedural
    1.23
     safeguards
    0.90
    izational
    0.86
     Proced
    0.84
     guiActiveUn
    0.84
     proced
    0.82
     restraints
    0.80
     protections
    0.79
     justice
    0.78
     violations
    0.77
    Act Density 0.011%

    No Known Activations