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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and contractions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the contraction "don't."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     pyramid
    -0.67
     funnel
    -0.64
     recycling
    -0.62
     shifts
    -0.62
     mutually
    -0.60
     enriched
    -0.58
     decom
    -0.58
     segregated
    -0.56
     weave
    -0.56
     surrender
    -0.56
    POSITIVE LOGITS
    âĶĢâĶĢâĶĢâĶĢ
    0.89
    ¯¯
    0.82
    ¯
    0.81
    ï¸ı
    0.81
    iversary
    0.81
    ¢
    0.78
    cffffcc
    0.77
    âĻ
    0.77
    `.
    0.74
    ems
    0.72
    Act Density 0.295%

    No Known Activations