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

    Neuronpedia

    INDEX
    Explanations

    Since

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     compens
    -0.09
     compensate
    -0.08
    charges
    -0.08
     compensated
    -0.08
     Compensation
    -0.07
     birbir
    -0.07
    stake
    -0.07
     compensation
    -0.07
     kä
    -0.07
     TMPro
    -0.07
    POSITIVE LOGITS
     isinstance
    0.09
     kategori
    0.08
    _category
    0.08
    enzione
    0.08
     വിഭാഗ
    0.08
    자인
    0.08
     qualifies
    0.08
     CATEGORY
    0.08
    이면
    0.08
     катег
    0.08
    Act Density 0.037%

    No Known Activations