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

    Neuronpedia

    INDEX
    Explanations

    phrases related to activating or turning on something

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to activation or turning something on

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ylum
    -1.04
    atively
    -0.78
    imei
    -0.75
    cott
    -0.70
    mean
    -0.70
    omas
    -0.68
    onymous
    -0.68
    loss
    -0.67
    agree
    -0.67
    icia
    -0.67
    POSITIVE LOGITS
     brakes
    1.10
     projector
    0.99
     engines
    0.98
     blender
    0.95
     dial
    0.90
     valves
    0.90
     lights
    0.88
     headlights
    0.88
     alarms
    0.87
     oven
    0.86
    Act Density 0.327%

    No Known Activations