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

    Neuronpedia

    INDEX
    Explanations

    conjunctions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (encoded
    -0.08
    (tree
    -0.07
    .where
    -0.06
    aky
    -0.06
    ’ye
    -0.06
     ','.
    -0.06
     Firebase
    -0.06
    .datetime
    -0.06
    ।↵↵
    -0.06
    esa
    -0.06
    POSITIVE LOGITS
     Walls
    0.07
     ロ
    0.07
    และ
    0.07
     وح
    0.07
     chịu
    0.06
    Mrs
    0.06
     abandonment
    0.06
    *"
    0.06
     ніж
    0.06
     сигн
    0.06
    Act Density 0.371%

    No Known Activations