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

    Neuronpedia

    INDEX
    Explanations

    J words in addresses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    obe
    -0.12
    ennie
    -0.11
    ane
    -0.11
    JM
    -0.11
    ames
    -0.10
    im
    -0.10
    enn
    -0.10
    ake
    -0.10
    ak
    -0.10
    он
    -0.09
    POSITIVE LOGITS
     oh
    0.12
    316
    0.11
    inding
    0.11
    icable
    0.11
    ustr
    0.11
    dling
    0.11
     ust
    0.11
     Edgar
    0.11
    y
    0.10
    iating
    0.10
    Act Density 0.031%

    No Known Activations