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

    Neuronpedia

    INDEX
    Explanations

    significant keywords or phrases related to decision-making and societal impact

    oai_token-act-pair · gpt-3.5-turbo

    nouns related to abstract concepts and specific topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    PLA
    -0.59
     unden
    -0.58
    ilogy
    -0.57
    Bah
    -0.56
     anus
    -0.55
    noon
    -0.55
    laughs
    -0.55
    sole
    -0.55
    senal
    -0.55
     Fighters
    -0.53
    POSITIVE LOGITS
     mattered
    0.97
     becomes
    0.89
     cannot
    0.86
     flowed
    0.84
    mith
    0.82
     must
    0.82
     evolves
    0.79
     magically
    0.78
    matter
    0.77
     ceases
    0.77
    Act Density 0.479%

    No Known Activations