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

    Neuronpedia

    INDEX
    Explanations

    sentences ending with a period and without any activation in specific words

    oai_token-act-pair · gpt-3.5-turbo

    punctuation, specifically periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     allowance
    -0.78
     optional
    -0.74
    urg
    -0.70
     marketplace
    -0.70
    rogen
    -0.67
     allocation
    -0.67
     settlements
    -0.66
     pros
    -0.65
    ubs
    -0.65
     contingency
    -0.65
    POSITIVE LOGITS
     His
    1.36
    Born
    1.20
    His
    1.10
     He
    1.09
     Born
    1.09
     Having
    1.09
     Known
    1.05
     Granted
    1.03
     Sitting
    1.02
     Himself
    0.98
    Act Density 0.494%

    No Known Activations