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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a requirement or condition for something to happen or be successful, often starting with "to"

    oai_token-act-pair · gpt-3.5-turbo

    imperative statements or instructions directed at the reader

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     impunity
    -0.84
     innocence
    -0.66
    ulence
    -0.62
     hindsight
    -0.60
    cknowled
    -0.59
    Newsletter
    -0.59
     opin
    -0.57
     enjoyment
    -0.56
     Views
    -0.54
    cknow
    -0.54
    POSITIVE LOGITS
     need
    1.43
    need
    1.32
     needs
    1.26
     must
    1.23
     needed
    1.18
    needs
    1.16
     gotta
    1.13
    must
    1.10
     NEED
    1.04
     required
    0.98
    Act Density 0.369%

    No Known Activations