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

    Neuronpedia

    INDEX
    Explanations

    occurrences of numerical representations and their associated contextual phrases

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

    patterns of numbers and punctuation that suggest uncertainty or conditional statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     **
    -0.54
    <eos>
    -0.52
    。
    -0.50
     _
    -0.49
     '
    -0.49
     *
    -0.49
     constat
    -0.48
    ****************
    -0.48
    ।
    -0.47
     (
    -0.46
    POSITIVE LOGITS
     Majefty
    1.03
     Houſe
    1.03
    ſelf
    1.00
     Efq
    0.96
    ſelves
    0.96
     houſe
    0.94
     itſelf
    0.92
     contextLoads
    0.88
     Jefus
    0.88
     ſeveral
    0.86
    Act Density 0.414%

    No Known Activations