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

    Neuronpedia

    INDEX
    Explanations

    phrases related to construction or creation

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to building or creating something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    steps
    -0.78
    errors
    -0.78
    words
    -0.76
    agents
    -0.72
    lights
    -0.72
    overs
    -0.71
    issues
    -0.70
    umin
    -0.68
    eyes
    -0.67
    reports
    -0.67
    POSITIVE LOGITS
     cohesive
    0.99
     realistic
    0.97
     unified
    0.94
     foothold
    0.94
     viable
    0.94
     comprehensive
    0.93
    rouse
    0.93
     definitive
    0.92
     plethora
    0.91
     bunch
    0.91
    Act Density 0.366%

    No Known Activations