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

    Neuronpedia

    INDEX
    Explanations

    obscure or hidden information

    oai_token-act-pair · gpt-3.5-turbo

    the word "obscure" and its variations, indicating a focus on hidden or less known concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Demand
    -0.87
    ogh
    -0.68
    iphate
    -0.68
    phasis
    -0.66
    NG
    -0.66
    avers
    -0.65
    uana
    -0.65
    agree
    -0.65
    MIC
    -0.65
    HOME
    -0.65
    POSITIVE LOGITS
     obscure
    0.98
    uously
    0.80
    ly
    0.76
    ulously
    0.72
     bystand
    0.71
    nesses
    0.70
    ively
    0.70
    worldly
    0.69
     pockets
    0.69
     footnote
    0.68
    Act Density 0.009%

    No Known Activations