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

    Neuronpedia

    INDEX
    Explanations

    the word "some" and related phrases indicating generality or ambiguity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    swer
    -0.16
    inand
    -0.16
    ataire
    -0.15
    ateur
    -0.15
     ullam
    -0.14
    ecycle
    -0.14
    uzzi
    -0.14
    bourg
    -0.14
    ière
    -0.14
    abra
    -0.14
    POSITIVE LOGITS
    ones
    0.24
     others
    0.22
    others
    0.21
     who
    0.18
    /all
    0.17
     parts
    0.17
    acci
    0.16
     extent
    0.15
    who
    0.15
     kinds
    0.15
    Act Density 0.086%

    No Known Activations