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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "Hor" and its variations, suggesting a focus on names or terms related to "Horizon."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ey
    -0.17
    ili
    -0.17
    rosse
    -0.16
    edin
    -0.15
    d
    -0.15
    tran
    -0.15
    empo
    -0.15
    eden
    -0.15
    eries
    -0.15
    dust
    -0.15
    POSITIVE LOGITS
    izont
    0.27
    izons
    0.26
    izon
    0.22
    rible
    0.21
     Hor
    0.21
    acio
    0.21
    izontally
    0.21
     hor
    0.19
    oscope
    0.19
    Hor
    0.19
    Act Density 0.013%

    No Known Activations