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

    Neuronpedia

    INDEX
    Explanations

    circular shapes or patterns

    oai_token-act-pair · gpt-3.5-turbo

    geometric shapes and associated structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     resurg
    -0.71
     ABE
    -0.67
    Ire
    -0.67
     Inquisition
    -0.65
     lieutenant
    -0.64
     discretion
    -0.63
     priorities
    -0.62
     reservations
    -0.62
    Mobile
    -0.61
    Sales
    -0.60
    POSITIVE LOGITS
     diameter
    1.02
     shaped
    0.94
    worms
    0.92
    shaped
    0.88
     layer
    0.88
    worm
    0.87
     thickness
    0.86
     formed
    0.86
     encl
    0.84
     protr
    0.84
    Act Density 0.215%

    No Known Activations