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

    Neuronpedia

    INDEX
    Explanations

    phrases involving the concept of 'each' or 'every'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "each" as it relates to various subjects or elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     understatement
    -0.71
     abound
    -0.68
    paren
    -0.61
    don
    -0.60
    haven
    -0.59
    tch
    -0.59
     Nope
    -0.57
    Īè
    -0.56
    rers
    -0.56
     Didn
    -0.56
    POSITIVE LOGITS
     individually
    1.55
     separately
    1.31
     differently
    1.06
     independently
    1.05
     unique
    1.02
     different
    0.97
     varying
    0.93
     uniquely
    0.93
     differing
    0.89
     respective
    0.87
    Act Density 0.228%

    No Known Activations