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

    Neuronpedia

    INDEX
    Explanations

    error messages prompting the user to try again later

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating error messages or prompts for user action

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     merch
    -0.64
     GOODMAN
    -0.60
    ilts
    -0.59
     tha
    -0.59
     morp
    -0.59
    deen
    -0.54
    EngineDebug
    -0.52
    mith
    -0.51
    oft
    -0.51
     today
    -0.51
    POSITIVE LOGITS
    OSE
    0.66
    rencies
    0.62
    »Ĵ
    0.61
    IJ
    0.60
    thia
    0.58
    Sorry
    0.58
    icol
    0.55
    ©¶æ¥µ
    0.54
    Cola
    0.54
    erence
    0.54
    Act Density 0.015%

    No Known Activations