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

    Neuronpedia

    INDEX
    Explanations

    instances of statements containing past tense verbs indicating various actions or decisions by different entities

    oai_token-act-pair · gpt-3.5-turbo

    the occurrences of the word "have" in relation to various subjects and actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ´
    -0.81
    Six
    -0.71
    drive
    -0.64
    zing
    -0.64
    aire
    -0.64
    onement
    -0.62
    most
    -0.61
    inth
    -0.61
    wash
    -0.60
    eem
    -0.60
    POSITIVE LOGITS
     been
    1.27
     begun
    1.19
     sprung
    1.16
     gone
    1.15
     opted
    1.14
     experimented
    1.12
     gotten
    1.12
     benefited
    1.11
     arisen
    1.10
     undergone
    1.08
    Act Density 0.198%

    No Known Activations