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

    Neuronpedia

    INDEX
    Explanations

    fragments of sentences indicating uncertainty or ambiguity

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating uncertainty or qualifications in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ¢
    -0.56
    upp
    -0.55
    exit
    -0.53
    uper
    -0.52
     Ukrain
    -0.52
     penet
    -0.51
    payer
    -0.50
     toll
    -0.50
    door
    -0.50
     manoeuv
    -0.50
    POSITIVE LOGITS
     sadly
    0.74
     frankly
    0.71
     unfortunately
    0.70
     hopefully
    0.70
     honestly
    0.66
     luckily
    0.66
     thankfully
    0.63
     Hopefully
    0.63
     haha
    0.62
     captcha
    0.62
    Act Density 0.823%

    No Known Activations