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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions affecting physical objects

    oai_token-act-pair · gpt-3.5-turbo

    punctuation used to list or separate items or clauses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ropolitan
    -0.67
     Dynasty
    -0.63
    ls
    -0.63
    Politics
    -0.62
    orne
    -0.61
     Thirty
    -0.61
    nar
    -0.61
     Anniversary
    -0.61
    itarian
    -0.60
    minent
    -0.59
    POSITIVE LOGITS
     thereby
    1.28
     whereas
    1.18
     resulting
    1.13
     regardless
    1.09
     irrespective
    1.06
     respectively
    1.06
     allowing
    1.04
     thus
    1.04
     whichever
    1.00
     ensuring
    0.99
    Act Density 0.438%

    No Known Activations