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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "exception" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     auroit
    -1.07
     feroit
    -1.04
     ainfi
    -1.04
     pouvoit
    -1.00
     mauva
    -0.99
     ſch
    -0.97
     AssemblyCompany
    -0.97
     avoient
    -0.96
     étoient
    -0.93
     pauvres
    -0.92
    POSITIVE LOGITS
     exception
    0.81
     example
    0.78
     an
    0.62
     Hot
    0.61
     the
    0.59
     upset
    0.59
    exception
    0.57
     multiple
    0.56
     ‘
    0.55
     in
    0.53
    Act Density 0.131%

    No Known Activations