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

    Neuronpedia

    INDEX
    Explanations

    dates and events from various years

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly periods, indicating the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tack
    -0.83
     allowable
    -0.80
     distribut
    -0.79
    usable
    -0.79
     pouch
    -0.76
     inconsist
    -0.76
     affili
    -0.76
     propensity
    -0.75
     readily
    -0.75
     nons
    -0.74
    POSITIVE LOGITS
     Congratulations
    1.29
     Thousands
    1.28
     Hundreds
    1.26
     Yesterday
    1.26
     Ahead
    1.20
     Dozens
    1.20
     Millions
    1.18
     Everyone
    1.17
     Whilst
    1.17
     Tonight
    1.14
    Act Density 0.477%

    No Known Activations