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

    Neuronpedia

    INDEX
    Explanations

    phrases with clauses indicating contrast or exception

    oai_token-act-pair · gpt-3.5-turbo

    comma usage in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    olves
    -0.74
    interstitial
    -0.71
    SourceFile
    -0.71
    eers
    -0.68
    è¦ļéĨĴ
    -0.67
    daq
    -0.62
    nexus
    -0.62
    zag
    -0.61
    aukee
    -0.61
    oint
    -0.58
    POSITIVE LOGITS
     alas
    1.02
     although
    0.96
     unlike
    0.94
     contrary
    0.91
     yes
    0.90
     despite
    0.88
     whereas
    0.81
     yeah
    0.81
     moreover
    0.81
     according
    0.81
    Act Density 0.128%

    No Known Activations