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

    Neuronpedia

    INDEX
    Explanations

    adjectives and descriptions of things that are particularly impactful or influential

    oai_token-act-pair · gpt-3.5-turbo

    descriptive adjectives indicating significance or value

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    obook
    -0.80
    plementation
    -0.77
    rower
    -0.74
    obia
    -0.73
    xiety
    -0.73
    ulhu
    -0.73
    frey
    -0.72
    hyde
    -0.71
    peria
    -0.71
    deen
    -0.71
    POSITIVE LOGITS
     reminders
    1.38
     additions
    1.37
     inventions
    1.37
     ones
    1.35
     creations
    1.32
     representations
    1.30
     indicators
    1.29
     examples
    1.26
     constructs
    1.26
     substitutes
    1.25
    Act Density 0.406%

    No Known Activations