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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks, especially periods

    oai_token-act-pair · gpt-3.5-turbo

    sentence-ending punctuation, specifically focusing on periods and commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     proport
    -0.82
     oun
    -0.78
     landfall
    -0.78
     commer
    -0.74
     afterlife
    -0.73
     warr
    -0.73
    isphere
    -0.72
     consolidation
    -0.69
     undermin
    -0.68
     casualty
    -0.68
    POSITIVE LOGITS
     Then
    0.99
     Later
    0.90
     Exactly
    0.90
     Asked
    0.90
     "_
    0.87
     Something
    0.86
    org
    0.85
     ********************************
    0.85
     Suddenly
    0.84
     Yeah
    0.84
    Act Density 0.168%

    No Known Activations