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

    Neuronpedia

    INDEX
    Explanations

    adam relu

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Howe
    -0.07
     kra
    -0.07
      ↵  ↵
    -0.07
     recreate
    -0.06
    'E
    -0.06
     Norwegian
    -0.06
     частина
    -0.06
     surged
    -0.06
    чого
    -0.06
     Colomb
    -0.06
    POSITIVE LOGITS
     清
    0.08
    чины
    0.07
    realloc
    0.07
    (False
    0.06
     lxml
    0.06
    .chunk
    0.06
     ✔
    0.06
     asym
    0.06
    딩
    0.06
    ptoms
    0.06
    Act Density 0.001%

    No Known Activations