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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating an upcoming event

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating imminent actions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Compass
    -0.74
     Howe
    -0.65
     perspective
    -0.65
     Cooke
    -0.64
     Corpus
    -0.63
    raped
    -0.62
     gateway
    -0.61
     representations
    -0.60
     simulator
    -0.60
     Perspective
    -0.59
    POSITIVE LOGITS
    pless
    0.97
    pload
    0.91
    NetMessage
    0.89
     arrive
    0.83
     be
    0.80
     announce
    0.79
     legalize
    0.78
     unveil
    0.76
     give
    0.76
     come
    0.74
    Act Density 0.045%

    No Known Activations