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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with "Well," and often followed by an explanation or continuation

    oai_token-act-pair · gpt-3.5-turbo

    transitional phrases and conversational cues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anim
    -0.67
     Amos
    -0.61
     Sok
    -0.61
    sing
    -0.57
    spir
    -0.57
     sideline
    -0.56
     ]
    -0.55
    fig
    -0.55
    cele
    -0.55
     >
    -0.55
    POSITIVE LOGITS
    Answer
    1.05
    swers
    1.02
     answer
    0.96
     Answer
    0.93
     answers
    0.85
    ription
    0.82
    Well
    0.82
    ccording
    0.78
     guessed
    0.75
    swer
    0.72
    Act Density 0.302%

    No Known Activations