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

    Neuronpedia

    INDEX
    Explanations

    phrases related to time spent or duration

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "spent" indicating time expenditure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Seymour
    -0.63
     detonated
    -0.61
     Watson
    -0.58
     Conan
    -0.56
     Edmund
    -0.55
    peer
    -0.55
     conver
    -0.54
    ensen
    -0.54
    ĪĴ
    -0.54
     Cassandra
    -0.53
    POSITIVE LOGITS
    itures
    0.92
     lav
    0.91
     IPM
    0.86
    ãĤ£
    0.76
    iture
    0.75
     beads
    0.75
    htaking
    0.72
    #$#$
    0.71
    edIn
    0.71
    zl
    0.69
    Act Density 0.028%

    No Known Activations