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

    Neuronpedia

    INDEX
    Explanations

    phrases related to expressing opinions with emphasis

    oai_token-act-pair · gpt-3.5-turbo

    repeated punctuation, specifically commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    herent
    -0.70
    ety
    -0.59
    interstitial
    -0.59
    ¬¼
    -0.56
    untarily
    -0.54
    enture
    -0.53
    OND
    -0.53
    estone
    -0.53
    abase
    -0.52
     redes
    -0.52
    POSITIVE LOGITS
     albeit
    1.09
     although
    0.95
     namely
    0.87
     though
    0.85
     however
    0.83
     huh
    0.80
     but
    0.78
     except
    0.77
     alas
    0.76
     yeah
    0.75
    Act Density 1.342%

    No Known Activations