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

    Neuronpedia

    INDEX
    Explanations

    phrases prompting the reader to stay informed or reminded of various topics

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "keep" and related phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ointed
    -0.63
    ELF
    -0.61
     Mehran
    -0.61
    女
    -0.61
    NetMessage
    -0.60
    ority
    -0.59
    LOD
    -0.59
    OUN
    -0.58
    displayText
    -0.58
    rule
    -0.57
    POSITIVE LOGITS
     scrolling
    0.93
     Calm
    0.90
     Reading
    0.82
     clicking
    0.79
    ers
    0.78
     humming
    0.75
    estones
    0.73
    ership
    0.71
    warming
    0.71
     reading
    0.70
    Act Density 0.034%

    No Known Activations