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

    Neuronpedia

    INDEX
    Explanations

    names and terms with diacritical marks

    oai_token-act-pair · gpt-3.5-turbo

    words or phrases with specific accentuated vowel patterns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    okin
    -0.83
    ifier
    -0.77
    ifiers
    -0.76
    estones
    -0.73
    acca
    -0.72
    acity
    -0.72
    ificent
    -0.71
    itar
    -0.71
    ific
    -0.70
    anmar
    -0.70
    POSITIVE LOGITS
     Lauder
    1.12
     Idle
    0.82
     Ãł
    0.81
    ciation
    0.81
    enne
    0.76
     Fay
    0.76
     Marie
    0.75
     Peninsula
    0.75
     Chal
    0.73
    ée
    0.72
    Act Density 0.007%

    No Known Activations