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

    Neuronpedia

    INDEX
    Explanations

    phrases related to providing information or explanations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to communication and reporting of statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    itton
    -0.85
    audi
    -0.67
    Discussion
    -0.66
    ranch
    -0.66
    zzi
    -0.63
    assic
    -0.61
    atti
    -0.61
    Tes
    -0.61
    absor
    -0.60
    rather
    -0.60
    POSITIVE LOGITS
     anymore
    1.19
     nor
    1.14
     specifics
    1.01
     anything
    1.01
     acknow
    0.97
     whatsoever
    0.87
     any
    0.81
     anywhere
    0.77
     until
    0.76
     anybody
    0.75
    Act Density 0.175%

    No Known Activations