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

    Neuronpedia

    INDEX
    Explanations

    statements about certain facts or characteristics

    oai_token-act-pair · gpt-3.5-turbo

    auxiliary verbs and forms of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Sunshine
    -0.61
    uctor
    -0.59
     Difference
    -0.58
     Capture
    -0.58
     Titanic
    -0.56
     Moves
    -0.53
     Quantity
    -0.52
    *****
    -0.52
     Pinball
    -0.51
     Private
    -0.51
    POSITIVE LOGITS
     therefore
    1.83
     consequently
    1.77
     thus
    1.40
     accordingly
    1.40
     hence
    1.25
     thereby
    1.22
     furthermore
    1.17
     moreover
    1.14
     subsequently
    1.13
     thereafter
    1.12
    Act Density 0.300%

    No Known Activations