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

    Neuronpedia

    INDEX
    Explanations

    phrases containing the word "only" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases emphasizing exclusivity or a singular aspect

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    azard
    -0.72
    ruary
    -0.72
    etz
    -0.72
    align
    -0.71
    mas
    -0.71
    each
    -0.69
    des
    -0.69
    comb
    -0.68
     enthusi
    -0.66
    staking
    -0.65
    POSITIVE LOGITS
     thing
    1.23
     downside
    1.15
     drawback
    1.13
     reason
    1.13
     remaining
    1.12
     exception
    1.11
     difference
    1.03
     caveat
    1.01
     surviving
    0.98
     conceivable
    0.96
    Act Density 0.040%

    No Known Activations