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

    Neuronpedia

    INDEX
    Explanations

    pieces of code containing specific characters and formatting

    oai_token-act-pair · gpt-3.5-turbo

    patterns resembling programming syntax and code structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atform
    -0.96
     Dynamics
    -0.79
     Ago
    -0.74
     Ada
    -0.73
     Spread
    -0.70
     Participation
    -0.70
     Bever
    -0.70
     Vide
    -0.68
    eteria
    -0.68
    xon
    -0.67
    POSITIVE LOGITS
    taboola
    1.11
    dict
    1.04
    blank
    0.93
    main
    0.92
    default
    0.91
    LINE
    0.90
    SOURCE
    0.89
    chance
    0.82
    verse
    0.81
    self
    0.80
    Act Density 0.025%

    No Known Activations