Update docs (#267)

Author: VincyZhang

docs/tutorials/pytorch/language-modeling/bert-base-uncased.ipynb

@@ -13,9 +13,15 @@
    "id": "69e00856",
    "metadata": {},
    "source": [
-    "This tutorial demonstrates how to quantize a BERT model with both static and dynamic post training quantization based on [Intel® Neural Compressor](https://github.com/intel/neural-compressor) and benchmark the quantized models. "
+    "This tutorial demonstrates how to quantize a BERT model with both static and dynamic post training quantization based on [Intel® Neural Compressor](https://github.com/intel/neural-compressor) and benchmark the quantized models. For better int8 performance benefit, multi-instance benchmarking with 4 cores/instance is recommended."
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "e0d4a9c5",
+   "metadata": {},
+   "source": []
+  },
   {
    "cell_type": "markdown",
    "id": "0cbd8bd4",
@@ -525,7 +531,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3.8.6 64-bit",
    "language": "python",
    "name": "python3"
   },
@@ -539,7 +545,12 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.13"
+   "version": "3.8.6"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "f54fd8d6160ddfbc370985ee3ad2925997e28943a671b1747496a6859c59cd26"
+   }
   }
  },
  "nbformat": 4,