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

    Neuronpedia

    INDEX
    Explanations

    words ending in "'s" indicating possession or association

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "today's" and variations of it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onel
    -0.82
    ï¸ı
    -0.79
    loo
    -0.78
    alties
    -0.77
    lain
    -0.76
    peat
    -0.76
    Downloadha
    -0.75
    eri
    -0.74
    oyle
    -0.74
    dylib
    -0.72
    POSITIVE LOGITS
     predicament
    0.75
     reality
    0.75
     compr
    0.70
     world
    0.68
     news
    0.67
     gravity
    0.67
     lowly
    0.66
     realities
    0.65
     edition
    0.65
     episode
    0.64
    Act Density 0.045%

    No Known Activations