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

    Neuronpedia

    INDEX
    Explanations

    sentences that involve a contrast or inconsistency

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "there" followed by varying contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     CJ
    -0.72
     ±
    -0.62
     pound
    -0.61
     cups
    -0.58
     actionGroup
    -0.57
     Khe
    -0.56
     beans
    -0.56
     destroyer
    -0.55
     oats
    -0.55
     cylinders
    -0.54
    POSITIVE LOGITS
    abouts
    1.42
    upon
    1.19
    fore
    0.94
    after
    0.76
    FORE
    0.73
    leased
    0.72
    hovah
    0.72
    ngth
    0.71
    with
    0.71
    choes
    0.71
    Act Density 0.136%

    No Known Activations