Added some context that the next executed call will throw an error.

notebooks/02_quant_activation_overview.ipynb

@@ -39,7 +39,7 @@
    "source": [
     "import torch\n",
     "from brevitas.nn import QuantConv2d, QuantIdentity\n",
-    "from brevitas.quant.scaled_int import Int8ActPerTensorFloat \n",
+    "from brevitas.quant.scaled_int import Int8ActPerTensorFloat\n",
     "\n",
     "torch.manual_seed(0)\n",
     "output_quant_conv = QuantConv2d(\n",
@@ -83,7 +83,7 @@
    "source": [
     "torch.manual_seed(0)\n",
     "input_output_quant_conv = QuantConv2d(\n",
-    "    in_channels=2, out_channels=3, kernel_size=(3,3), \n",
+    "    in_channels=2, out_channels=3, kernel_size=(3,3),\n",
     "    input_quant=Int8ActPerTensorFloat, output_quant=Int8ActPerTensorFloat)\n",
     "\n",
     "torch.manual_seed(0)\n",
@@ -600,7 +600,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "By default, the only layer that is an exception to this is `QuantHardTanh`. That is because the interface to `torch.nn.HardTanh` already requires users to manually specify `min_val` and `max_val`, so Brevitas preserves that both when quantization is enabled or disabled. With quantization enabled, by default those values are used for initialization, but then the range is learned. Let's look at an example:"
+    "By default, the only layer that is an exception to this is `QuantHardTanh`. That is because the interface to `torch.nn.HardTanh` already requires users to manually specify `min_val` and `max_val`, so Brevitas preserves that both when quantization is enabled or disabled. With quantization enabled, by default those values are used for initialization, but then the range is learned. Let's look at an example. Run the cell below, and we expect it to throw an error because of missing attributes:"
    ]
   },
   {