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

    Neuronpedia

    INDEX
    Explanations

    fashion-related descriptions and terms

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly commas and parentheses in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ioxide
    -0.63
    reth
    -0.62
     ACTIONS
    -0.62
     Statements
    -0.61
    lees
    -0.60
    robe
    -0.60
    nosis
    -0.60
     Cases
    -0.60
     Subjects
    -0.60
    iple
    -0.60
    POSITIVE LOGITS
    -)
    1.00
     etc
    0.93
     alas
    0.93
     albeit
    0.93
    ?).
    0.86
     preferably
    0.85
    )"
    0.84
    *)
    0.83
    !).
    0.81
    ?)
    0.80
    Act Density 0.531%

    No Known Activations