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

    Neuronpedia

    INDEX
    Explanations

    phrases related to encountering errors and trying again

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the act of trying again after an error

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ELD
    -0.58
     Galile
    -0.58
    plates
    -0.57
    cult
    -0.53
    ELY
    -0.53
    bey
    -0.53
    Hop
    -0.52
     Matter
    -0.51
    matter
    -0.51
     proceedings
    -0.51
    POSITIVE LOGITS
    interstitial
    0.86
    Cola
    0.84
    olicy
    0.66
     later
    0.65
    nir
    0.64
     è£ıè¦ļéĨĴ
    0.62
     comma
    0.62
     recycle
    0.61
     Later
    0.61
    dra
    0.60
    Act Density 0.014%

    No Known Activations