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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a strong degree or emphasis on a particular concept

    oai_token-act-pair · gpt-3.5-turbo

    repetitive use of the phrase "very much."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    neapolis
    -0.91
    yne
    -0.81
    atures
    -0.81
    itures
    -0.80
    ulously
    -0.75
    eways
    -0.74
    imeters
    -0.72
    acious
    -0.72
    etts
    -0.72
    agically
    -0.71
    POSITIVE LOGITS
     appreciated
    1.32
     obliged
    0.97
     welcome
    0.89
     regarded
    0.87
     deserved
    0.87
     resemble
    0.86
     welcomed
    0.85
     alive
    0.84
     understood
    0.83
    ________________
    0.80
    Act Density 0.032%

    No Known Activations