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

    Neuronpedia

    INDEX
    Explanations

    sentences containing a specific structure of "that is" followed by an adjective or idea

    oai_token-act-pair · gpt-3.5-turbo

    statements describing various conditions or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    luaj
    -0.89
    inav
    -0.69
    imeter
    -0.67
    imeters
    -0.66
    artney
    -0.66
    aah
    -0.64
    iard
    -0.63
    unia
    -0.63
    cit
    -0.63
    enium
    -0.62
    POSITIVE LOGITS
     why
    0.99
     until
    0.94
     unless
    0.83
     indeed
    0.79
     exactly
    0.79
    until
    0.75
     untrue
    0.74
     precisely
    0.73
     evidenced
    0.72
    ĺħ
    0.71
    Act Density 0.061%

    No Known Activations