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

    Neuronpedia

    INDEX
    Explanations

    descriptive terms that express opinions or analysis

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to strong emotions, arguments, and observations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Delivery
    -0.62
    parts
    -0.62
     Attend
    -0.57
     srf
    -0.57
     Delivery
    -0.57
    admin
    -0.55
    dule
    -0.55
     UCH
    -0.54
    Bench
    -0.54
     RTX
    -0.53
    POSITIVE LOGITS
     coincides
    1.26
     contrasts
    1.21
     begs
    1.21
     ignores
    1.16
     culmin
    1.13
     applies
    1.13
     translates
    1.12
     echoes
    1.11
     extends
    1.11
     overlook
    1.10
    Act Density 0.175%

    No Known Activations