{
  "schemaVersion": 1,
  "lessonId": "performance-02",
  "contentHash": "fc7b7357392cc925f95ee951dd96a30ce905c4b45496c33435a6bcc337bbc07a",
  "executedAt": "2026-10-20T10:47:35.719968+00:00",
  "environment": {
    "backend": "cpu",
    "deviceCount": 1,
    "python": "3.14.3",
    "jax": "0.9.2",
    "numpy": "2.4.4",
    "matplotlib": "3.10.8"
  },
  "stdout": "{\n  \"jax\": \"0.9.2\",\n  \"backend\": \"TFRT_CPU_0\",\n  \"calls\": [\n    {\n      \"label\": \"initial signature\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 1,\n      \"result\": 140.0\n    },\n    {\n      \"label\": \"same signature changed data\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 0,\n      \"result\": 204.0\n    },\n    {\n      \"label\": \"same signature changed dynamic gain\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 0,\n      \"result\": 560.0\n    },\n    {\n      \"label\": \"different input shape\",\n      \"shape\": [\n        12\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 1,\n      \"result\": 506.0\n    }\n  ],\n  \"python_trace_events\": [\n    {\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\"\n    },\n    {\n      \"shape\": [\n        12\n      ],\n      \"dtype\": \"float32\"\n    }\n  ]\n}\n{\n  \"jax\": \"0.9.2\",\n  \"backend\": \"TFRT_CPU_0\",\n  \"calls\": [\n    {\n      \"label\": \"initial signature\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 1,\n      \"result\": 140.0\n    },\n    {\n      \"label\": \"same signature changed data\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 0,\n      \"result\": 204.0\n    },\n    {\n      \"label\": \"same signature changed dynamic gain\",\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 0,\n      \"result\": 560.0\n    },\n    {\n      \"label\": \"different input shape\",\n      \"shape\": [\n        12\n      ],\n      \"dtype\": \"float32\",\n      \"observed_trace_delta\": 1,\n      \"result\": 506.0\n    }\n  ],\n  \"python_trace_events\": [\n    {\n      \"shape\": [\n        8\n      ],\n      \"dtype\": \"float32\"\n    },\n    {\n      \"shape\": [\n        12\n      ],\n      \"dtype\": \"float32\"\n    }\n  ]\n}\nStatic branch trace observations: [{'shape': [8], 'square': True}, {'shape': [8], 'square': False}]\nPer-step host reads seconds: 0.00012137508019804955\nFinal-only wait seconds: 3.5999808460474014e-05\nAdded ledger row: {'label': 'new length sixteen', 'shape': [16], 'dtype': 'float32', 'observed_trace_delta': 1, 'result': 1240.0}\nMasked versus unmasked padded objective: 55.0 58.0\nObserved runtime-Python-branch tracing failure\nPASS: performance-02\n",
  "notebookCells": [
    {
      "sourceHash": "257eddcfb3afb8a6a677f966065083aef41c33d15a20f7455e43483017e8f407",
      "outputs": []
    },
    {
      "sourceHash": "6ac15e7a320aecd08269f51dd3917244cedbe68f60bdfd158fec54b6d54f8c0e",
      "outputs": []
    },
    {
      "sourceHash": "8b923504f76886002f3a23a876a92068ca6a1b22fa78b4d9829aa36779e95156",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{\n",
            "  \"jax\": \"0.9.2\",\n",
            "  \"backend\": \"TFRT_CPU_0\",\n",
            "  \"calls\": [\n",
            "    {\n",
            "      \"label\": \"initial signature\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 1,\n",
            "      \"result\": 140.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"same signature changed data\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 0,\n",
            "      \"result\": 204.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"same signature changed dynamic gain\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 0,\n",
            "      \"result\": 560.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"different input shape\",\n",
            "      \"shape\": [\n",
            "        12\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 1,\n",
            "      \"result\": 506.0\n",
            "    }\n",
            "  ],\n",
            "  \"python_trace_events\": [\n",
            "    {\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\"\n",
            "    },\n",
            "    {\n",
            "      \"shape\": [\n",
            "        12\n",
            "      ],\n",
            "      \"dtype\": \"float32\"\n",
            "    }\n",
            "  ]\n",
            "}\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "413af716cb73ca66b2ac4272585998f9b4c631abc41d2bb19489ba79932c2434",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{\n",
            "  \"jax\": \"0.9.2\",\n",
            "  \"backend\": \"TFRT_CPU_0\",\n",
            "  \"calls\": [\n",
            "    {\n",
            "      \"label\": \"initial signature\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 1,\n",
            "      \"result\": 140.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"same signature changed data\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 0,\n",
            "      \"result\": 204.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"same signature changed dynamic gain\",\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 0,\n",
            "      \"result\": 560.0\n",
            "    },\n",
            "    {\n",
            "      \"label\": \"different input shape\",\n",
            "      \"shape\": [\n",
            "        12\n",
            "      ],\n",
            "      \"dtype\": \"float32\",\n",
            "      \"observed_trace_delta\": 1,\n",
            "      \"result\": 506.0\n",
            "    }\n",
            "  ],\n",
            "  \"python_trace_events\": [\n",
            "    {\n",
            "      \"shape\": [\n",
            "        8\n",
            "      ],\n",
            "      \"dtype\": \"float32\"\n",
            "    },\n",
            "    {\n",
            "      \"shape\": [\n",
            "        12\n",
            "      ],\n",
            "      \"dtype\": \"float32\"\n",
            "    }\n",
            "  ]\n",
            "}\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "d1b60594d525ce4073880e1b119944b738e39be9a110befe4c7ca0dd4b7a08d6",
      "outputs": []
    },
    {
      "sourceHash": "25df6d662f9d3f0c4e984c6a201f79fa0cfd7a2633dc4b0773dfba6c2dd4ee6c",
      "outputs": [
        {
          "output_type": "display_data",
          "metadata": {},
          "data": {
            "image/png": "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",
            "text/plain": [
              "Executed Matplotlib figure"
            ]
          }
        }
      ]
    },
    {
      "sourceHash": "bc3c2d63f434d65d76b7454bbdef4b9be29ac6de45fb0a3162fa5308e39aac1f",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Static branch trace observations: [{'shape': [8], 'square': True}, {'shape': [8], 'square': False}]\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "055347e9fa1b4d43a6213751466d4d2409e377fbe831ba5611e5e5b845a27699",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Per-step host reads seconds: 0.00010370835661888123\n",
            "Final-only wait seconds: 3.0125025659799576e-05\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "087e1bc2987675966f9f9ca16f92d95930e68b63b0385396ac706c222c036d55",
      "outputs": []
    },
    {
      "sourceHash": "102c1effc032828e3f1b211cf1a31f8092f60bac48ce6147169b0c6271f5090f",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Added ledger row: {'label': 'new length sixteen', 'shape': [16], 'dtype': 'float32', 'observed_trace_delta': 1, 'result': 1240.0}\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "f7a5a1071d36c3f88596178ac11efa82d71d100b0df54cb7194426ccf9c501a4",
      "outputs": []
    },
    {
      "sourceHash": "978cccca44d459cd37836012c4587429cbbcc3ebbfbdaca61faf82b09ab62000",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Masked versus unmasked padded objective: 55.0 58.0\n"
          ]
        }
      ]
    },
    {
      "sourceHash": "fbf2516a37917775fa7ca1b65639882dfa9b4b2a1891e0418e420281eab7e25e",
      "outputs": []
    },
    {
      "sourceHash": "2a946df97719fcef774d7037d945da072d5410c1777ec858c91b8f0dac56868f",
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Observed runtime-Python-branch tracing failure\n"
          ]
        }
      ]
    }
  ],
  "scope": "Recorded instructor reference run; all script assertions and notebook code cells passed. No accelerator performance claim."
}