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

    Neuronpedia

    INDEX
    Explanations

    references to visibility and environmental conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    detailed descriptive phrases about locations and daily activities.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    travel diary or journal entries describing daily activities, particularly focused on excursions, trips, and experiences with high activations on sentences describing movements and locations.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
     Diſ
    -0.76
     juſ
    -0.76
    ſelves
    -0.73
    ſelf
    -0.73
     Inſ
    -0.69
    wiſe
    -0.67
     Reſ
    -0.66
     ſeveral
    -0.66
     ſol
    -0.65
     Chriftian
    -0.65
    POSITIVE LOGITS
     lunch
    0.49
     snug
    0.45
     Lunch
    0.45
     slept
    0.43
     went
    0.42
     afternoon
    0.41
     enjoyed
    0.41
     lunchtime
    0.41
     giggling
    0.41
     had
    0.40
    Act Density 0.177%

    No Known Activations