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

    Neuronpedia

    INDEX
    Explanations

    sentences starting with "After" or similar patterns indicating a sequence of events

    oai_token-act-pair · gpt-3.5-turbo

    statements about attempts and outcomes in a process or decision-making scenario

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Previous
    -0.68
     Enhance
    -0.67
    Attack
    -0.65
     differently
    -0.62
    Secondly
    -0.61
     Previously
    -0.60
     Previous
    -0.60
    PLIED
    -0.59
    include
    -0.59
     Also
    -0.58
    POSITIVE LOGITS
     finally
    1.62
     relent
    1.32
     settled
    1.11
     prevailed
    1.09
     reluctantly
    1.08
     capit
    1.03
     decided
    1.02
     eventually
    1.02
    icably
    0.94
     succumbed
    0.94
    Act Density 0.421%

    No Known Activations