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

    Neuronpedia

    INDEX
    Explanations

    mentions of relationships or connections with individuals

    oai_token-act-pair · gpt-3.5-turbo

    phrases that reference friendships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     percentages
    -0.66
     scenarios
    -0.66
    Percent
    -0.64
    case
    -0.62
    cases
    -0.61
    heny
    -0.60
     evaluations
    -0.60
    Premium
    -0.60
    ctory
    -0.59
     besides
    -0.59
    POSITIVE LOGITS
     mine
    2.08
     hers
    2.07
     ours
    2.04
     theirs
    1.82
     yours
    1.78
     sorts
    1.43
     Mine
    1.32
    Mine
    0.93
    mine
    0.74
    mire
    0.73
    Act Density 0.097%

    No Known Activations