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

    Neuronpedia

    INDEX
    Explanations

    number sequences mixed with miscellaneous text

    oai_token-act-pair · gpt-3.5-turbo

    numerical values, particularly those related to dates and quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     appeal
    -0.72
     corrid
    -0.69
     SUPPORT
    -0.69
     cycle
    -0.69
     plateau
    -0.67
     restrictive
    -0.67
     relevance
    -0.64
     snowball
    -0.64
     transact
    -0.64
     optimistic
    -0.63
    POSITIVE LOGITS
    wm
    0.91
    Ever
    0.90
    soever
    0.89
    tm
    0.88
    pac
    0.87
    e
    0.87
    cott
    0.87
    20439
    0.85
    eu
    0.85
    hyde
    0.84
    Act Density 0.057%

    No Known Activations