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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating accuracy or precision

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "accurately."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     necessity
    -0.73
     favorites
    -0.72
     Redemption
    -0.72
     tropes
    -0.72
     inaction
    -0.71
     reaction
    -0.71
     willingness
    -0.70
     approvals
    -0.69
     taboo
    -0.69
     hegemony
    -0.69
    POSITIVE LOGITS
    imated
    0.95
    ãĤ©
    0.93
     compensated
    0.85
     modeled
    0.83
     reproduced
    0.82
    correct
    0.79
    paced
    0.78
    etitive
    0.77
     represented
    0.77
     addressed
    0.77
    Act Density 0.021%

    No Known Activations