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

    Neuronpedia

    INDEX
    Explanations

    exceptions or standout instances

    oai_token-act-pair · gpt-3.5-turbo

    instances of exceptions and notable mentions in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ..............
    -0.71
    mob
    -0.66
    "))
    -0.63
    awar
    -0.60
    rote
    -0.59
    akh
    -0.59
    emonium
    -0.58
    ilege
    -0.58
    "]
    -0.58
    NAS
    -0.58
    POSITIVE LOGITS
     culminating
    0.93
     notably
    0.90
     exceptions
    0.90
     exception
    0.86
     includ
    0.84
     being
    0.84
     preferring
    0.81
     additionally
    0.81
     notable
    0.80
     resulting
    0.79
    Act Density 0.505%

    No Known Activations