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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating things that will happen or continue in the future

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to time, especially concerning the future and duration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    VK
    -0.85
    emis
    -0.77
     redes
    -0.63
     yesterday
    -0.62
     earlier
    -0.60
     inner
    -0.57
    edin
    -0.56
     ingred
    -0.56
     recently
    -0.55
     lately
    -0.55
    POSITIVE LOGITS
    unless
    1.14
     unless
    1.13
     if
    0.87
    assuming
    0.87
     anyway
    0.82
    0000000
    0.78
     barring
    0.76
     thanks
    0.76
     regardless
    0.74
     Unless
    0.74
    Act Density 0.281%

    No Known Activations