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

    Neuronpedia

    INDEX
    Explanations

    titles of movies, TV series, and songs

    oai_token-act-pair · gpt-3.5-turbo

    phrases containing the word "the" and titles of movies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hower
    -0.84
    merce
    -0.80
    staking
    -0.79
    chev
    -0.79
    alore
    -0.76
    rower
    -0.75
    SPA
    -0.71
    alist
    -0.70
    ivably
    -0.70
    etter
    -0.70
    POSITIVE LOGITS
     Madness
    1.02
     Faces
    1.01
     Horses
    0.97
     Witches
    0.96
     Witch
    0.96
     Forgotten
    0.95
     Seasons
    0.94
     Ruins
    0.92
     Gods
    0.92
     Lies
    0.92
    Act Density 0.222%

    No Known Activations