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

    Neuronpedia

    INDEX
    Explanations

    phrases related to social commentary and policy challenges

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to searching for understanding and combining ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     dude
    -0.82
     Krypt
    -0.81
     skinny
    -0.77
     dudes
    -0.76
     booze
    -0.75
     hugs
    -0.75
    kids
    -0.75
     kid
    -0.74
     shit
    -0.74
     veggies
    -0.73
    POSITIVE LOGITS
     principally
    0.90
     moreover
    0.89
     methodological
    0.86
     analys
    0.84
     presently
    0.84
     analyse
    0.83
     furthermore
    0.83
     conclud
    0.83
    respective
    0.82
    Firstly
    0.80
    Act Density 1.091%

    No Known Activations