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

    Neuronpedia

    INDEX
    Explanations

    critical or negative statements from a variety of domains or contexts

    oai_token-act-pair · gpt-3.5-turbo

    expressions of opinion or criticism in dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Pg
    -0.61
    +.
    -0.58
    eligible
    -0.57
    Reviewed
    -0.57
    rupal
    -0.54
    iden
    -0.52
    adra
    -0.52
    antic
    -0.50
    cum
    -0.49
    ordes
    -0.48
    POSITIVE LOGITS
    %"
    1.21
    )",
    1.02
     âĢ¦"
    0.97
    "—
    0.95
    "]
    0.94
    .")
    0.94
     ..."
    0.91
    ")
    0.90
    )"
    0.90
    ,"
    0.89
    Act Density 1.716%

    No Known Activations