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

    Neuronpedia

    INDEX
    Explanations

    specific phrases related to news headlines or video titles

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "JUST WATCHED."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sis
    -0.74
    sole
    -0.69
    ocr
    -0.69
    Magikarp
    -0.67
    ogly
    -0.67
    pill
    -0.66
     nib
    -0.65
    nuts
    -0.65
    nut
    -0.65
     detached
    -0.64
    POSITIVE LOGITS
     WATCHED
    1.37
     Transcript
    0.81
     VIDEOS
    0.81
     WATCH
    0.73
     Countdown
    0.73
     Videos
    0.73
     Task
    0.70
     Scenes
    0.69
     IMAGES
    0.69
     reacts
    0.68
    Act Density 0.008%

    No Known Activations