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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "really" and its variations, indicating emphasis or intensity in expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     mere
    -0.18
    ese
    -0.17
    mere
    -0.16
    alist
    -0.16
    ner
    -0.15
    ile
    -0.15
    bris
    -0.14
    oret
    -0.14
    rous
    -0.14
    sel
    -0.14
    POSITIVE LOGITS
    ignment
    0.28
    -life
    0.22
    yyy
    0.20
    -world
    0.19
    igned
    0.19
    -existing
    0.19
     estate
    0.19
    /false
    0.18
    yy
    0.18
    ationship
    0.17
    Act Density 0.069%

    No Known Activations