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

    Neuronpedia

    INDEX
    Explanations

    sequences of alphanumeric characters with specific patterns, potentially related to technical information or code

    oai_token-act-pair · gpt-3.5-turbo

    sequences of alphanumeric characters and symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cephal
    -0.80
    eer
    -0.73
     Warrant
    -0.68
    abase
    -0.67
    ryan
    -0.67
    gall
    -0.63
    bable
    -0.61
    mens
    -0.60
    stood
    -0.60
    ËĪ
    -0.60
    POSITIVE LOGITS
    soType
    0.69
    entimes
    0.66
    ¬¼
    0.65
    itals
    0.63
    anooga
    0.61
    assadors
    0.61
    Ń·
    0.60
    ooters
    0.60
     commuting
    0.59
    ubi
    0.59
    Act Density 0.043%

    No Known Activations