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

    Neuronpedia

    INDEX
    Explanations

    references to social media posts containing images (often pictures)

    oai_token-act-pair · gpt-3.5-turbo

    references to pictures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     footing
    -0.65
     delegation
    -0.65
     lure
    -0.64
     telecommunications
    -0.63
     misunderstood
    -0.63
     planner
    -0.61
     hindsight
    -0.61
    stood
    -0.60
     margins
    -0.59
     Leilan
    -0.59
    POSITIVE LOGITS
    colo
    0.98
    twitter
    0.92
    ares
    0.76
    videos
    0.76
    books
    0.74
    ://
    0.74
     snapped
    0.74
    TED
    0.73
    youtu
    0.73
     img
    0.72
    Act Density 0.014%

    No Known Activations