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

    Neuronpedia

    INDEX
    Explanations

    references to incidents or actions being captured on video recordings

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the preposition "on," particularly in contexts involving cameras or surveillance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     goodbye
    -0.74
     remedies
    -0.69
     aloud
    -0.65
     fortun
    -0.65
    GBT
    -0.64
    baby
    -0.62
     descendants
    -0.62
    chairs
    -0.62
     invitations
    -0.61
     supplemented
    -0.60
    POSITIVE LOGITS
     trap
    0.78
     traps
    0.74
     Camera
    0.71
    Eye
    0.69
     surprise
    0.68
     Breach
    0.68
    oppers
    0.67
    arc
    0.66
     camera
    0.66
    EngineDebug
    0.65
    Act Density 0.094%

    No Known Activations