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

    Neuronpedia

    INDEX
    Explanations

    instances where something is amiss or not working correctly

    oai_token-act-pair · gpt-3.5-turbo

    mentions of a vague or unspecified subject or event

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    kus
    -0.70
    ardless
    -0.70
    street
    -0.65
    oise
    -0.65
    adr
    -0.65
    umer
    -0.65
    ardi
    -0.63
    umers
    -0.63
    ilt
    -0.63
     McCull
    -0.62
    POSITIVE LOGITS
    Else
    1.15
     else
    1.12
     happens
    0.86
     happened
    0.86
     unforeseen
    0.80
     Else
    0.76
     intangible
    0.74
    ĪĴ
    0.73
     clicked
    0.72
     transpired
    0.71
    Act Density 0.072%

    No Known Activations