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

    Neuronpedia

    INDEX
    Explanations

    descriptions of the quality or attributes of various materials

    oai_token-act-pair · gpt-3.5-turbo

    descriptors that imply quality and materials used in products

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enario
    -0.97
    vantage
    -0.74
    jriwal
    -0.74
    ittee
    -0.73
    ploma
    -0.72
    SPONSORED
    -0.72
    vich
    -0.72
    irk
    -0.71
    hander
    -0.71
    orah
    -0.71
    POSITIVE LOGITS
     materials
    1.21
     herbs
    1.11
     chemicals
    1.07
     goodies
    1.05
     ingredients
    1.04
     fabrics
    1.02
     spices
    1.00
     sunlight
    0.98
     scraps
    0.95
     needles
    0.94
    Act Density 0.307%

    No Known Activations