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

    Neuronpedia

    INDEX
    Explanations

    phrases related to doing something for a specific purpose or duration

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express actions or sentiments done for others

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    operated
    -0.69
    iasco
    -0.66
    gars
    -0.65
    glers
    -0.65
    atars
    -0.65
    abis
    -0.64
    atories
    -0.64
    ering
    -0.64
    quartered
    -0.62
    arts
    -0.62
    POSITIVE LOGITS
     awhile
    1.18
     breakfast
    1.04
    got
    1.02
     lunch
    1.00
     example
    0.99
     fun
    0.99
     reasons
    0.94
     instance
    0.94
     dinner
    0.93
     supper
    0.91
    Act Density 0.128%

    No Known Activations