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

    Neuronpedia

    INDEX
    Explanations

    responses and replies in a conversational context

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "replied" indicating responses or statements made by individuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    teenth
    -0.71
    BALL
    -0.69
    enburg
    -0.67
    dar
    -0.67
    fi
    -0.65
    bound
    -0.65
     Blazers
    -0.65
    icipated
    -0.64
    mental
    -0.64
     Trials
    -0.63
    POSITIVE LOGITS
     thereto
    0.99
     affirm
    0.96
     sarcast
    0.96
    reply
    0.94
     angrily
    0.91
     favorably
    0.86
    later
    0.83
     enthusiastically
    0.83
     politely
    0.82
     promptly
    0.80
    Act Density 0.038%

    No Known Activations