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

    Neuronpedia

    INDEX
    Explanations

    geographical descriptors related to cardinal directions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    849
    -0.15
    olini
    -0.13
     ÑĦоÑĢмÑĥ
    -0.13
    itom
    -0.13
    Spot
    -0.13
    ÑĢÑĥÑĪ
    -0.13
    hart
    -0.13
    udic
    -0.13
    ÑĤон
    -0.13
    compat
    -0.13
    POSITIVE LOGITS
    most
    0.49
    -most
    0.39
     tip
    0.32
     reaches
    0.32
     end
    0.32
     portion
    0.31
    MOST
    0.31
     part
    0.29
     most
    0.28
     edge
    0.25
    Act Density 0.044%

    No Known Activations