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

    Neuronpedia

    INDEX
    Explanations

    the letter "ë" in the text

    oai_token-act-pair · gpt-3.5-turbo

    special characters and encoding artifacts in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    EStreamFrame
    -0.84
    xual
    -0.83
    iple
    -0.80
    ocene
    -0.74
    imilation
    -0.68
    ipop
    -0.68
    osterone
    -0.68
    esty
    -0.67
    omial
    -0.67
    raviolet
    -0.66
    POSITIVE LOGITS
    lance
    0.79
    nir
    0.75
    ο
    0.74
    entric
    0.72
    TION
    0.71
    mented
    0.71
    âĢ¢âĢ¢âĢ¢âĢ¢
    0.71
    lisher
    0.70
    s
    0.69
    ries
    0.68
    Act Density 0.032%

    No Known Activations