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

    Neuronpedia

    INDEX
    Explanations

    instances of time-related phrasing, particularly focusing on the word "when."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     purpoſe
    -1.01
     myſelf
    -0.98
     itſelf
    -0.96
     themſelves
    -0.94
     ſmall
    -0.87
     raiſ
    -0.84
    ſelves
    -0.84
     himſelf
    -0.83
     ſeveral
    -0.83
     ſtate
    -0.81
    POSITIVE LOGITS
     but
    0.71
     ModelExpression
    0.71
     when
    0.67
     But
    0.66
    uttavia
    0.66
     cuando
    0.61
    HomeAsUpEnabled
    0.61
     quando
    0.61
     However
    0.61
    +:+
    0.59
    Act Density 0.256%

    No Known Activations