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

    Neuronpedia

    INDEX
    Explanations

    phrases related to influence and impact

    oai_token-act-pair · gpt-3.5-turbo

    an emphasis on the phrase "in" to explore contexts relating to complexity, influence, and actions within various environments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tar
    -0.70
     lasted
    -0.64
    dar
    -0.64
    halla
    -0.63
    tor
    -0.63
    EEP
    -0.61
    istani
    -0.60
     lodged
    -0.60
     listened
    -0.60
     perched
    -0.60
    POSITIVE LOGITS
     ways
    1.51
     somew
    1.36
    effic
    1.33
    humane
    1.18
     accordance
    1.15
    ordinate
    1.14
     terms
    1.13
    versely
    1.11
    efficiency
    1.10
    effect
    1.05
    Act Density 0.149%

    No Known Activations