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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "cross" and its variations, often in a context related to boundaries or changes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ENCY
    -0.15
    emie
    -0.15
    /fixtures
    -0.15
    iko
    -0.15
    áºŃt
    -0.15
    atics
    -0.15
    ika
    -0.14
    ÏĢιÏĥ
    -0.14
    Ñıл
    -0.14
    urn
    -0.14
    POSITIVE LOGITS
    roads
    0.29
     swords
    0.23
    -over
    0.23
    fade
    0.21
    ses
    0.21
    -sectional
    0.20
     paths
    0.20
     Swords
    0.19
    cut
    0.19
    (es
    0.19
    Act Density 0.032%

    No Known Activations