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

    Neuronpedia

    INDEX
    Explanations

    phrases related to something being mostly unaffected or not significantly impacted

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "largely" as it relates to the extent or degree of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atures
    -0.81
    agate
    -0.80
    abad
    -0.78
    iry
    -0.74
    eem
    -0.73
     Peaks
    -0.71
    ses
    -0.68
    PT
    -0.67
    anth
    -0.67
     Chaser
    -0.66
    POSITIVE LOGITS
     overlooked
    0.88
     overshadowed
    0.86
     relied
    0.85
     reliant
    0.85
     consisted
    0.83
     comprised
    0.82
     rely
    0.82
     irrelevant
    0.80
     relying
    0.80
     lacking
    0.78
    Act Density 0.008%

    No Known Activations