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

    Neuronpedia

    INDEX
    Explanations

    references to tall buildings and architectural structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ã¤ÃŁ
    -0.16
    ãĤ·ãĥ¼
    -0.15
     dilig
    -0.15
    vely
    -0.14
    éĵº
    -0.14
     patches
    -0.14
    glich
    -0.14
    plen
    -0.14
     Expires
    -0.14
     atol
    -0.13
    POSITIVE LOGITS
     tower
    0.47
     towers
    0.40
     Tower
    0.36
    tower
    0.35
     tallest
    0.34
    Tower
    0.34
     Towers
    0.31
     skys
    0.31
     taller
    0.30
     height
    0.29
    Act Density 0.115%

    No Known Activations