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

    Neuronpedia

    INDEX
    Explanations

    phrases related to sequence and going back and forth

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "back and forth."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ult
    -0.69
     Feder
    -0.68
    »Ĵ
    -0.67
     Twain
    -0.65
     PD
    -0.64
     Richards
    -0.62
     bait
    -0.61
     everywhere
    -0.60
     Creator
    -0.60
     Parenthood
    -0.59
    POSITIVE LOGITS
    end
    1.22
    office
    1.21
    door
    1.20
    facing
    1.20
    to
    1.15
    story
    1.10
    ends
    1.09
    ended
    1.06
    eye
    1.05
    channel
    1.05
    Act Density 0.032%

    No Known Activations