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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (names
    -0.08
    acular
    -0.07
     matt
    -0.07
    clar
    -0.07
     HttpStatusCodeResult
    -0.07
    Goods
    -0.07
     chimpan
    -0.07
     קצר
    -0.06
     TREE
    -0.06
    gł
    -0.06
    POSITIVE LOGITS
    Ein
    0.06
     }

↵
    0.06
    ＊
    0.06
    רכים
    0.06
     Но
    0.06
    (STD
    0.06
     половин
    0.06
    _IW
    0.06
    ，
    0.06
     legion
    0.06
    Act Density 0.009%

    No Known Activations