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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns with specific sequences

    oai_token-act-pair · gpt-3.5-turbo

    instances of personal reflection or emotional experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atis
    -0.77
     extrad
    -0.71
     bloc
    -0.70
     sacked
    -0.68
     camoufl
    -0.67
     axe
    -0.66
     opposing
    -0.65
     allied
    -0.65
     arbit
    -0.65
     disav
    -0.64
    POSITIVE LOGITS
    My
    1.11
    ³³³³
    1.04
    ³³³³³³³³³³³³³³³³
    1.02
    Anyway
    1.01
    Years
    0.98
    Thankfully
    0.97
    ³³³
    0.97
    However
    0.96
    Sadly
    0.95
    MY
    0.95
    Act Density 0.459%

    No Known Activations