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

    Neuronpedia

    INDEX
    Explanations

    island and land

    np_max-act · gemini-2.0-flash

    The neuron fires on occurrences of the word “island.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    CHE
    -0.07
     commitment
    -0.07
    History
    -0.06
    (mContext
    -0.06
     mcc
    -0.06
    (preg
    -0.06
     cine
    -0.06
     Hu
    -0.06
     Wen
    -0.06
     Pace
    -0.06
    POSITIVE LOGITS
     Island
    0.18
     island
    0.15
     islands
    0.12
     Islands
    0.11
    島
    0.10
     Isles
    0.09
     остров
    0.08
    岛
    0.08
     Islanders
    0.08
    Isl
    0.07
    Act Density 0.010%

    No Known Activations