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

    Neuronpedia

    INDEX
    Explanations

    words related to sources or origins of different elements

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate sources or origins of materials or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    äºĶ
    -0.75
    ament
    -0.73
    merce
    -0.71
    answer
    -0.70
    osen
    -0.69
    iveness
    -0.69
    lege
    -0.67
    inion
    -0.67
    mare
    -0.66
    idy
    -0.65
    POSITIVE LOGITS
     afar
    1.44
     abroad
    1.15
     elsewhere
    1.04
     within
    0.98
     whence
    0.95
     sources
    0.91
     scratch
    0.91
     outside
    0.91
     inside
    0.90
     glands
    0.85
    Act Density 0.139%

    No Known Activations