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

    Neuronpedia

    INDEX
    Explanations

    instances where something is being kept or preserved

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     Cosponsors
    -0.67
    ode
    -0.65
    Gs
    -0.64
    FORE
    -0.62
    QB
    -0.61
    Reply
    -0.60
    gee
    -0.60
    gging
    -0.60
    gs
    -0.58
    twitch
    -0.58
    POSITIVE LOGITS
     intact
    1.08
     memories
    0.92
     composure
    0.79
     Memories
    0.77
     liberties
    0.76
     individuality
    0.76
     tabs
    0.74
     possession
    0.74
     ownership
    0.72
     preservation
    0.71
    Act Density 0.097%

    No Known Activations