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

    Neuronpedia

    INDEX
    Explanations

    the word "silly" with different degrees of emphasis

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "silly" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    yer
    -0.96
    ioch
    -0.87
    Reviewed
    -0.83
    ainer
    -0.83
    rigan
    -0.83
    lain
    -0.80
    ept
    -0.76
    amen
    -0.75
    ainers
    -0.73
    venants
    -0.70
    POSITIVE LOGITS
     silly
    0.95
     nonsense
    0.86
    ness
    0.83
     prank
    0.81
     Ples
    0.81
     aside
    0.81
    ishly
    0.74
    nesses
    0.73
     Faw
    0.72
     Haram
    0.71
    Act Density 0.020%

    No Known Activations