{
  "schemaVersion": 1,
  "lessonId": "welcome-02",
  "contentHash": "9cebbe5b29a2efa8d335a07bc2113a61deefcc78edea629b5398a996222f7023",
  "executedAt": "2026-10-20T10:45:06.782121+00:00",
  "environment": {
    "backend": "cpu",
    "deviceCount": 1,
    "python": "3.14.3",
    "jax": "0.9.2",
    "numpy": "2.4.4",
    "matplotlib": "3.10.8"
  },
  "stdout": "Shape: (2, 3)\nDtype: float32\nDevices: {CpuDevice(id=0)}\nRow sums: [ 3. 12.]\nShape: (2, 3)\nDtype: float32\nDevices: {CpuDevice(id=0)}\nRow sums: [ 3. 12.]\nMeans: [[1.5 2.5 3.5]]\nCentered: [[-1.5 -1.5 -1.5]\n [ 1.5  1.5  1.5]]\nHost dtype: float32\nExpected element-count mismatch\nPASS: welcome-02\n",
  "notebookCells": [
    {
      "sourceHash": "cc691a3a883d9e3de3ba95715941564601e11032419b3b90da42657c937572ad",
      "outputs": []
    },
    {
      "sourceHash": "28fc02bbacdb0824f58c0db6d0cedd9d00ab565f20cefbdadb16d7d5cd543abf",
      "outputs": []
    },
    {
      "sourceHash": "3f050c776b33597a402cb4b159250c8b0802bca639b69d8fafe821448893f6db",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape: (2, 3)\n",
            "Dtype: float32\n",
            "Devices: {CpuDevice(id=0)}\n",
            "Row sums: [ 3. 12.]\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "f32cf783e0fb61a9ef43ebc923fe4fdce538815ebeccd0aeb954f6dddf7799e6",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape: (2, 3)\n",
            "Dtype: float32\n",
            "Devices: {CpuDevice(id=0)}\n",
            "Row sums: [ 3. 12.]\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "01070594046c89089359e3ef2df66d3adf642e1a2d0e767ec4ad0fdd9553f544",
      "outputs": []
    },
    {
      "sourceHash": "25df6d662f9d3f0c4e984c6a201f79fa0cfd7a2633dc4b0773dfba6c2dd4ee6c",
      "outputs": [
        {
          "output_type": "display_data",
          "metadata": {},
          "data": {
            "image/png": "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",
            "text/plain": [
              "Executed Matplotlib figure"
            ]
          }
        }
      ]
    },
    {
      "sourceHash": "3d7da3f2cea90de32ba38a3008b41cf387acb9dca7ab2f76ede493387d4c1549",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Means: [[1.5 2.5 3.5]]\n",
            "Centered: [[-1.5 -1.5 -1.5]\n",
            " [ 1.5  1.5  1.5]]\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "61be83d810fe98f9a2e4ee057e3ded458673d92221f0cffebddc192f5c89ff85",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Host dtype: float32\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "6985de2c8b81ea0ac51f583d2aebb50cae1351e59a556507343d5dea676efa2d",
      "outputs": []
    },
    {
      "sourceHash": "cf90a247446d3407ba9a7103c5c6e25ff7ed3e4f4e2c7d7f8e2fe82a095ac1c1",
      "outputs": []
    },
    {
      "sourceHash": "9aba44607f2524bd1fe8361ebf8beacf47fbb8fbe6206048a66702806483a169",
      "outputs": []
    },
    {
      "sourceHash": "57269cb0bbada477438eb3b8a0d1e24eeaf5ca53462cfc231be91216dfc6e1a0",
      "outputs": []
    },
    {
      "sourceHash": "5de62142562a61cf177ef9efd39530a778013e3187aec638825fe9aeaa33ccc0",
      "outputs": []
    },
    {
      "sourceHash": "041940923de0b2c99a1ef3c74a8273eb3bfacb76a235a40925f0b028cbcfb7ff",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Expected element-count mismatch\n"
          ]
        }
      ]
    }
  ],
  "scope": "Recorded instructor reference run; all script assertions and notebook code cells passed. No accelerator performance claim."
}