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

    Neuronpedia

    INDEX
    Explanations

    phrases related to certainty or strong beliefs

    oai_token-act-pair · gpt-3.5-turbo

    assurances of reliability and certainty in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    yssey
    -0.66
     meanwhile
    -0.56
     largeDownload
    -0.55
    cour
    -0.55
     Originally
    -0.54
     partName
    -0.53
    ircraft
    -0.53
     Featured
    -0.52
    formerly
    -0.52
     Wheels
    -0.52
    POSITIVE LOGITS
     ignor
    0.60
     wrong
    0.59
     meaningful
    0.58
    omething
    0.58
     genuine
    0.58
     outweigh
    0.57
     actual
    0.56
    "))
    0.55
     irreversible
    0.54
     legitimately
    0.53
    Act Density 1.814%

    No Known Activations