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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences followed by the symbol Ċ and a number, with particular importance on the phrase structure

    oai_token-act-pair · gpt-3.5-turbo

    sections with statistical or numerical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hement
    -0.71
     Thornton
    -0.67
     shielded
    -0.66
     retreat
    -0.66
     shielding
    -0.65
     lifes
    -0.64
     endeavour
    -0.64
     withdrawal
    -0.63
     Savannah
    -0.63
     emancipation
    -0.63
    POSITIVE LOGITS
    Ingredients
    0.92
    ³³³³³³³³³³³³³³³³
    0.87
    TRUMP
    0.83
    FIN
    0.82
    Liter
    0.82
    FY
    0.81
    Feature
    0.79
    ³³³³³³³³
    0.79
    LAB
    0.79
    ³³³³
    0.77
    Act Density 0.099%

    No Known Activations