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

    Neuronpedia

    INDEX
    Explanations

    scientific results

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Mixed
    -0.07
     ngắn
    -0.07
    řiv
    -0.06
     karşılık
    -0.06
    oxic
    -0.06
     tur
    -0.06
    utely
    -0.06
     Pride
    -0.06
     starší
    -0.06
    につ
    -0.06
    POSITIVE LOGITS
    extract
    0.07
    クション
    0.07
    (import
    0.07
    ンの
    0.07
    .sequence
    0.06
    eh
    0.06
    .POS
    0.06
    inem
    0.06
         ↵
    0.06
     　
    0.06
    Act Density 0.030%

    No Known Activations