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

    Neuronpedia

    INDEX
    Explanations

    instances where a sentence is interrupted or cut off

    oai_token-act-pair · gpt-3.5-turbo

    pauses or interruptions in speech

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     QC
    -0.73
    ा
    -0.69
    ãĤ´
    -0.67
    alion
    -0.67
    APD
    -0.66
    SourceFile
    -0.65
    à¥
    -0.65
     Mechdragon
    -0.64
    SpaceEngineers
    -0.63
     Palest
    -0.62
    POSITIVE LOGITS
     uh
    1.14
     um
    0.93
     gasp
    0.82
     yeah
    0.81
    yeah
    0.81
    maybe
    0.80
    you
    0.77
    sort
    0.77
    need
    0.74
    exc
    0.74
    Act Density 0.143%

    No Known Activations