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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a sequence or action that should occur prior to another action

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating necessary actions or information to be addressed prior to a specific event

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aez
    -0.72
    aren
    -0.72
    OTOS
    -0.70
    DIT
    -0.69
    hack
    -0.68
    SA
    -0.67
    Writer
    -0.64
    MER
    -0.64
    PROV
    -0.63
    utter
    -0.63
    POSITIVE LOGITS
    hand
    0.81
     realizing
    0.80
     committing
    0.77
     sunset
    0.75
     embark
    0.75
     midnight
    0.75
     sunrise
    0.74
     settling
    0.73
     noon
    0.73
     proceeding
    0.69
    Act Density 0.064%

    No Known Activations