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

    Neuronpedia

    INDEX
    Explanations

    legal terms and references like case numbers or sections of laws

    oai_token-act-pair · gpt-3.5-turbo

    closing parentheses and other punctuation marks indicating the end of clauses or sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iets
    -0.75
    ell
    -0.70
    loo
    -0.68
    eled
    -0.66
    ding
    -0.66
     wall
    -0.64
     feared
    -0.64
     neighb
    -0.64
     fingert
    -0.63
     suspic
    -0.63
    POSITIVE LOGITS
     Additionally
    1.04
     Alternatively
    0.94
     Therefore
    0.94
     Furthermore
    0.93
     Similarly
    0.88
    *)
    0.88
     Likewise
    0.87
     Interestingly
    0.87
     Regardless
    0.86
     Afterwards
    0.86
    Act Density 0.094%

    No Known Activations