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

    Neuronpedia

    INDEX
    Explanations

    hyperlinks or mentions to access further information

    oai_token-act-pair · gpt-3.5-turbo

    instances of periods or punctuation indicating the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     imperson
    -0.88
     awakening
    -0.83
     instinct
    -0.75
     detectable
    -0.74
     blowing
    -0.74
     impression
    -0.74
     forcing
    -0.74
     outward
    -0.73
     loyal
    -0.71
     questioning
    -0.71
    POSITIVE LOGITS
     Alternatively
    1.33
    <|endoftext|>
    1.32
     Also
    1.26
     Additionally
    1.11
     Includes
    1.08
    pdf
    1.07
     Downloads
    1.06
     Otherwise
    1.06
     Originally
    1.05
     Lastly
    1.03
    Act Density 0.196%

    No Known Activations