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

    Neuronpedia

    INDEX
    Explanations

    repetitive phrases or actions that occur multiple times

    oai_token-act-pair · gpt-3.5-turbo

    the repetition of the phrase "over and" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lio
    -0.59
    FL
    -0.55
    {"
    -0.54
    Grey
    -0.53
     Fighter
    -0.52
     Dynam
    -0.52
    amera
    -0.51
     congress
    -0.51
    Dou
    -0.51
    capt
    -0.51
    POSITIVE LOGITS
    âķIJ
    0.84
    rogen
    0.81
    rogens
    0.78
     parcel
    0.78
     beyond
    0.75
    urger
    0.73
    =-=-
    0.72
    orah
    0.61
    tg
    0.60
     Beyond
    0.60
    Act Density 0.056%

    No Known Activations