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

    Neuronpedia

    INDEX
    Explanations

    phrases related to taking action or following a specific sequence of steps to achieve a goal

    oai_token-act-pair · gpt-3.5-turbo

    phrases that introduce objectives or sequences of actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tek
    -0.75
    rament
    -0.70
    vous
    -0.69
    child
    -0.68
    tur
    -0.67
    laus
    -0.67
    leigh
    -0.67
    tiny
    -0.67
    vana
    -0.66
     Dirk
    -0.65
    POSITIVE LOGITS
    lies
    0.79
     Osw
    0.76
    liness
    0.73
     order
    0.71
    xual
    0.69
    eous
    0.67
    ournals
    0.66
    arily
    0.65
     orchestr
    0.65
    ogue
    0.64
    Act Density 0.017%

    No Known Activations