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

    Neuronpedia

    INDEX
    Explanations

    instances where actions are being followed up on or continued

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "follow up."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Hub
    -0.80
    ¬¼
    -0.75
    adem
    -0.70
    renheit
    -0.67
    tu
    -0.65
     Reserved
    -0.64
    cipled
    -0.64
     Tsukuyomi
    -0.63
     Sorcerer
    -0.63
     Worth
    -0.62
    POSITIVE LOGITS
    rocal
    0.84
    mortem
    0.73
    wards
    0.72
    ounter
    0.70
    orate
    0.70
    dates
    0.68
     conversations
    0.67
    ressing
    0.67
     remarks
    0.67
    iry
    0.66
    Act Density 0.041%

    No Known Activations