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

    Neuronpedia

    INDEX
    Explanations

    written summaries or reports within a document

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "summary" and variations thereof

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    train
    -0.73
    bees
    -0.72
    gypt
    -0.72
     Wee
    -0.71
    duct
    -0.68
    warm
    -0.68
     charism
    -0.67
    hips
    -0.65
     Mustang
    -0.64
    Sea
    -0.62
    POSITIVE LOGITS
     summar
    0.90
     summ
    0.88
     VIEW
    0.83
     summarize
    0.82
     summary
    0.81
     synopsis
    0.77
     thereof
    0.76
    strate
    0.76
    lations
    0.71
    krit
    0.71
    Act Density 0.028%

    No Known Activations