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

    Neuronpedia

    INDEX
    Explanations

    scheduled events or appointments

    oai_token-act-pair · gpt-3.5-turbo

    structures related to schedules or planned events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abiding
    -0.80
     forgetting
    -0.79
     ç¥ŀ
    -0.75
     misunderstanding
    -0.72
     dying
    -0.70
     hating
    -0.69
     matured
    -0.69
     impunity
    -0.69
     imitation
    -0.69
     transformations
    -0.68
    POSITIVE LOGITS
     speak
    1.21
     testify
    1.20
     intervene
    1.14
     announce
    1.13
     investigate
    1.11
     examine
    1.10
     lodge
    1.09
     discuss
    1.09
     conven
    1.08
     arra
    1.06
    Act Density 0.201%

    No Known Activations