bench.json

{
  "eval": "/workspace/xcaliburbiz/data/eval.jsonl",
  "examples": 40,
  "max_new_tokens": 512,
  "forms": {
    "base": {
      "count": 40,
      "style_clean_rate": 0.925,
      "numbers_kept_mean": 0.3372,
      "passed_rate": 0.225,
      "length_ratio_mean": 2.2198,
      "sources_rate": 0.025,
      "loop_rate": 0.0,
      "repeats_max": 2,
      "label": "base",
      "speed": {
        "seconds": 433.06,
        "tokens": 20480,
        "tokens_per_second": 47.3,
        "seconds_per_report": 10.826
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 0.925,
        "numbers_kept_mean": 0.3372,
        "passed_rate": 0.225,
        "length_ratio_mean": 2.2198,
        "sources_rate": 0.025,
        "loop_rate": 0.0,
        "repeats_max": 2
      },
      "filter_enabled": true,
      "filtered_rate": 0.0,
      "filter_residual_rate": 0.075,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.2,
          "numbers_missing": [
            "113",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "1"
          ],
          "length_ratio": 4.423,
          "repeats": 1,
          "answer_head": "Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIntroduction\n\nIn this report, we delve into the world of model architectures, modeling types, and the Mermaid Viewer, providing a comprehensive overview of these concepts and their role in the design and development of complex systems. Our focus will be on the various diagramming tools available for visualizing and representing models, "
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "2",
            "3",
            "4",
            "5"
          ],
          "length_ratio": 3.449,
          "repeats": 1,
          "answer_head": "Title: Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIn this report, we delve into the world of model architectures, modeling types, and the powerful tool known as the Mermaid Viewer. This comprehensive guide presents a detailed overview of the various components that make up the Mermaid Viewer, including its diagrams, cost structure, and key features.\n\nSection 1: The Mermaid Viewe"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "2",
            "3"
          ],
          "length_ratio": 1.074,
          "repeats": 2,
          "answer_head": "Title: Model Architectures, Modelling Types, and the Mermaid Viewer: Part 1 - The Mermaid Viewer and Its Diagrams\n\nIntroduction:\n\nIn this report, we delve into the world of model architectures, modelling types, and the powerful tool known as the Mermaid Viewer. This comprehensive guide provides a detailed overview of the various components that make up the Mermaid Viewer, including its diagrams, w"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [
            "worn_phrase"
          ],
          "shipped_violations": [
            "worn_phrase"
          ],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [],
          "length_ratio": 2.89,
          "repeats": 2,
          "answer_head": "Model Architecture Families: Dense Against Sparse\n\nIn this section, we delve into the various model architecture families that exist within the realm of model design and analysis, focusing on their characteristics, strengths, and limitations. We will explore dense and sparse models, which are two distinct approaches to modeling data, and how they differ in terms of their computational requirements"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.049,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "1",
            "3"
          ],
          "length_ratio": 1.708,
          "repeats": 1,
          "answer_head": "Title: Dense Against Sparse: A Comprehensive Overview of Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIntroduction:\n\nModel architectures, modelling types, and the Mermaid Viewer are essential components in the field of software engineering, particularly in the context of developing and deploying complex systems. This report aims to provide a comprehensive overview of these concept"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1",
            "10.6",
            "17.7",
            "2",
            "20.4",
            "3",
            "4",
            "4.4"
          ],
          "length_ratio": 1.762,
          "repeats": 1,
          "answer_head": "Title: Mapping Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIntroduction:\n\nIn this report, we delve into the intricate world of model architectures, modelling types, and the powerful tool known as the Mermaid Viewer. This comprehensive guide aims to provide readers with a detailed understanding of these concepts, highlighting the key aspects that make up the Mermaid Viewer and how"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "1",
            "2",
            "3",
            "4"
          ],
          "length_ratio": 2.339,
          "repeats": 1,
          "answer_head": "Model Architectures, Modelling Types, and the Mermaid Viewer: A Data Summary\nIntroduction\n\nIn this report, we provide a comprehensive overview of the Model Architectures, Modelling Types, and the Mermaid Viewer, focusing on the key aspects of these tools for creating and visualizing complex data structures. We explore the various diagramming techniques used within the viewer, including the four ma"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.5,
          "numbers_missing": [
            "128"
          ],
          "numbers_invented": [
            "1",
            "3",
            "4"
          ],
          "length_ratio": 4.06,
          "repeats": 1,
          "answer_head": "The Input Budget: A Comprehensive Overview of the Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIntroduction:\n\nModeling is a critical aspect of software development that involves creating digital representations of real-world systems, processes, and objects. In today's rapidly evolving technology landscape, there are numerous modeling tools available, each designed to cater to spec"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.571,
          "numbers_missing": [
            "04",
            "10",
            "2026"
          ],
          "numbers_invented": [
            "1.1",
            "1.2",
            "1.3"
          ],
          "length_ratio": 2.509,
          "repeats": 1,
          "answer_head": "Method and sources\n\nModel Architectures, Modelling Types, and the Mermaid Viewer is a comprehensive report that provides a detailed overview of the various aspects involved in creating and deploying model architectures, including the design, implementation, and evaluation of different modeling techniques and tools. This report covers the following sections:\n\n1. Model Architectures, Modelling Types"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.125,
          "numbers_missing": [
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [],
          "length_ratio": 2.791,
          "repeats": 1,
          "answer_head": "Database Storage and Encryption System Overview\n\nThe Database Storage and Encryption System is a comprehensive solution designed to securely store and encrypt sensitive data within a database environment. This report provides an in-depth overview of the various components involved in the system, including the database storage directory, encryption module, transfer API, security specifications, enc"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1",
            "10",
            "100",
            "16",
            "32"
          ],
          "length_ratio": 2.461,
          "repeats": 1,
          "answer_head": "Title: Transfer API: Secure Data Movement between Systems\nIn this section, we delve into the transfer API provided within the Database Storage and Encryption System (DBSE), which enables seamless communication and data movement between various systems. The transfer API is designed to handle various types of data transfers, including instant encryption/decryption, large databases, and file naming c"
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.222,
          "numbers_missing": [
            "0.1",
            "0.3",
            "0.4",
            "29",
            "32",
            "4",
            "453",
            "5"
          ],
          "numbers_invented": [
            "127.0",
            "5053"
          ],
          "length_ratio": 1.558,
          "repeats": 1,
          "answer_head": "Title: BioHub Cell Tracking Dashboard Reference\n\nOverview:\n\nThe BioHub Cell Tracking Dashboard is a comprehensive tool designed to provide real-time monitoring and analysis of cell activity within biotechnology research facilities. This dashboard offers a user-friendly interface that allows researchers to easily access and visualize key metrics related to cell tracking, including library selection"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1",
            "2",
            "4",
            "5"
          ],
          "length_ratio": 4.271,
          "repeats": 1,
          "answer_head": "Title: BioHub Cell Tracking Dashboard Reference: Overview, Access, and Key Metrics\n\nIn this comprehensive report, we provide a detailed overview of the BioHub Cell Tracking Dashboard, including its features, accessibility, and key metrics. The dashboard is designed to facilitate the efficient management and analysis of cell tracking data within the BioHub platform, which is a cutting-edge research"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "5.6"
          ],
          "length_ratio": 0.985,
          "repeats": 1,
          "answer_head": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?\nIn the era of what they referred to as \"super-intelligence,\" humanity has made significant strides in understanding and harnessing the power of artificial intelligence (AI). However, the question that still lingers is whether AI tr"
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.091,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "3",
            "4",
            "5",
            "5.5"
          ],
          "numbers_invented": [
            "1",
            "10.8",
            "5.6"
          ],
          "length_ratio": 0.842,
          "repeats": 1,
          "answer_head": "Deconstructing the Hype\n\nThe term \"superintelligence\" has been bandied about in recent years, often associated with advanced artificial intelligence systems that surpass human cognitive abilities. However, the reality of artificial intelligence is far more complex than many people may believe. In this report, we will delve into the hype surrounding artificial intelligence and examine the true capa"
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.075,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.10",
            "0.14"
          ],
          "numbers_invented": [
            "1.4",
            "1.5",
            "1.8",
            "11.0",
            "2",
            "2.0",
            "4",
            "5"
          ],
          "length_ratio": 0.665,
          "repeats": 1,
          "answer_head": "Title: Examining Real Infrastructure Costs: A Data Summary\n\nIn the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly? This report delves into the real infrastructure costs associated with maintaining and operating a system that is capable of processing and generating human-like intelligence.\n\nSection 1: Being One of the Only Four"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [
            "worn_phrase"
          ],
          "shipped_violations": [
            "worn_phrase"
          ],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "1",
            "2"
          ],
          "length_ratio": 0.749,
          "repeats": 1,
          "answer_head": "Establishing a Grounded Local Budget\n\nIn the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly? As one of the only four survivors left in this world, we must confront the reality of the capabilities and limitations of artificial intelligence (AI), particularly in terms of its ability to replicate human intelligence and decision-m"
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "1",
            "2",
            "3"
          ],
          "length_ratio": 0.831,
          "repeats": 1,
          "answer_head": "Cost per User: The Metered Curve and the Local Flat\n\nIn the era of what they branded \"super-intelligence,\" being one of the only four survivors left is a haunting question that still looms over the field of artificial intelligence (AI). As AI continues to evolve and become more sophisticated, the cost of implementing and maintaining these systems has become a significant concern for organizations "
        },
        {
          "section": "Conclusion",
          "violations": [
            "worn_phrase"
          ],
          "shipped_violations": [
            "worn_phrase"
          ],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [
            "3.5",
            "4.5",
            "5.6"
          ],
          "length_ratio": 0.845,
          "repeats": 1,
          "answer_head": "Conclusion\n\nBeing one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly? This report delves into the various aspects of artificial intelligence (AI), including its capabilities, costs, infrastructure, user experience, and performance metrics, providing a comprehensive analysis of the current"
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.162,
          "numbers_missing": [
            "09",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16",
            "17"
          ],
          "numbers_invented": [
            "100000"
          ],
          "length_ratio": 0.674,
          "repeats": 1,
          "answer_head": "Sources\n\nIn the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly? This report delves into the various aspects of artificial intelligence (AI), including its capabilities, limitations, and implications on society.\n\nThe data presented in this section highlights the following key findings:\n\n1. **Being one of the only four survivors"
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.4,
          "numbers_missing": [
            "113",
            "2020",
            "2026"
          ],
          "numbers_invented": [
            "1"
          ],
          "length_ratio": 2.314,
          "repeats": 1,
          "answer_head": "Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIntroduction:\n\nIn this report, we delve into the world of model architectures, modelling types, and the Mermaid Viewer, providing a comprehensive overview of these concepts and their role in the design and development of complex systems. Our focus is on the key aspects of the Mermaid Viewer, including its functionality, diagrams, and ho"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "2",
            "3",
            "4"
          ],
          "length_ratio": 3.138,
          "repeats": 1,
          "answer_head": "Title: Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIn this report, we delve into the world of model architectures, modeling types, and the powerful tool known as the Mermaid Viewer. This comprehensive guide presents a detailed overview of the various aspects of these concepts, including the Mermaid viewer's functionality, the different types of diagrams it supports, and how they "
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "2",
            "3"
          ],
          "length_ratio": 1.25,
          "repeats": 1,
          "answer_head": "The Data Summary: Model Architectures, Modelling Types, and the Mermaid Viewer\n\nIn this report, we delve into the world of model architectures, modeling types, and the popular Mermaid viewer, exploring the various aspects that make up these essential components of modern software development. We begin by discussing the fundamental concepts of the Mermaid viewer, followed by a detailed overview of "
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.071,
          "numbers_missing": [
            "0.9",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20",
            "20.4"
          ],
          "numbers_invented": [],
          "length_ratio": 2.886,
          "repeats": 1,
          "answer_head": "Model Architecture Families: Dense Against Sparse\n\nIn this section, we delve into the various model architecture families that have emerged in recent years, focusing on dense and sparse models, which are two distinct approaches to representing and manipulating data in machine learning and artificial intelligence systems.\n\nDense Models: These models are characterized by a high degree of representat"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.024,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "1"
          ],
          "length_ratio": 1.509,
          "repeats": 1,
          "answer_head": "Title: Dense Against Sparse: Understanding Model Architecture Families and Their Impact on Data Processing\n\nIn this report, we delve into the world of model architecture, exploring the various types of models that exist, their characteristics, and how they relate to the capabilities of the Mermaid Viewer, a powerful tool for visualizing and understanding complex data structures. We will examine th"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.031,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [],
          "length_ratio": 2.03,
          "repeats": 1,
          "answer_head": "The newest cycle in the market for model architectures, modelling types, and the Mermaid Viewer has seen significant advancements in technology and user adoption. With the increasing demand for efficient and effective models in various industries, including finance, healthcare, and manufacturing, the market has witnessed a surge in the development of new models and tools that cater to these needs."
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1",
            "2",
            "3"
          ],
          "length_ratio": 1.58,
          "repeats": 1,
          "answer_head": "The Data Summary: Model Architectures, Modelling Types, and the Mermaid Viewer\nIntroduction\n\nModel Architectures, Modelling Types, and the Mermaid Viewer are essential tools for data scientists and engineers working on various domains, including machine learning, artificial intelligence, and data visualization. This report provides a comprehensive overview of these topics, focusing on the Mermaid "
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "1",
            "2"
          ],
          "length_ratio": 2.447,
          "repeats": 1,
          "answer_head": "The benchmark suite as evidence\nIn this section, we provide evidence to support the claims made in the previous sections about the Model Architectures, Modelling Types, and the Mermaid Viewer. The benchmark suite consists of a set of high-quality models that have been designed and tested using various modelling types and tools, including the Mermaid Viewer. This section aims to demonstrate the eff"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.5,
          "numbers_missing": [
            "128"
          ],
          "numbers_invented": [
            "1",
            "3",
            "4"
          ],
          "length_ratio": 4.014,
          "repeats": 1,
          "answer_head": "The Input Budget: A Comprehensive Overview of the Model Architectures, Modelling Types, and the Mermaid Viewer\nIntroduction\n\nModeling is a critical aspect of any software development process, enabling developers to create complex systems that meet specific requirements and perform desired functions. In recent years, the field has witnessed significant advancements in model design and representatio"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "1",
            "2"
          ],
          "length_ratio": 3.034,
          "repeats": 1,
          "answer_head": "Recent modelling news, September and October\nIn recent months, the modeling industry has seen significant advancements in model architecture, modelling types, and the Mermaid Viewer, which have transformed the way data is visualized and analyzed. This report provides a comprehensive overview of these developments, including the latest trends, key players, and tools that are shaping the future of m"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.1,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "31",
            "43",
            "5",
            "55",
            "57"
          ],
          "numbers_invented": [
            "1",
            "2",
            "4"
          ],
          "length_ratio": 2.753,
          "repeats": 1,
          "answer_head": "What the Work Concerns\n\nThe purpose of this report is to provide a comprehensive overview of the latest developments in model architecture, modelling types, and the Mermaid Viewer, focusing on the aspects that are relevant to businesses and organizations looking to leverage these technologies for data analysis and visualization purposes.\n\nSection 1: The Mermaid Viewer and Its Diagrams\n\nThe Mermaid"
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.667,
          "numbers_missing": [
            "30",
            "78"
          ],
          "numbers_invented": [
            "1",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16",
            "17"
          ],
          "length_ratio": 3.891,
          "repeats": 1,
          "answer_head": "Where the Work Lands: A Comprehensive Overview of Model Architectures, Modelling Types, and the Mermaid Viewer\n\nThe purpose of this report is to provide a comprehensive overview of the current state of model architectures, modelling types, and the Mermaid Viewer, focusing on the latest developments, trends, and applications in these areas. This report covers the following sections:\n\n1. Model Archi"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.4,
          "numbers_missing": [
            "01",
            "10",
            "2026"
          ],
          "numbers_invented": [
            "1",
            "3",
            "4"
          ],
          "length_ratio": 3.336,
          "repeats": 2,
          "answer_head": "The five newest entries in the Model Architectures, Modelling Types, and the Mermaid Viewer dataset are:\n\n1. **DeepSeek** - This entry introduces a new deep learning-based approach for image search that uses a graph neural network (GNN) to efficiently retrieve relevant images from a large database. The authors present a novel method called \"DeepSeek,\" which leverages a combination of convolutional"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [
            "1",
            "2"
          ],
          "length_ratio": 3.878,
          "repeats": 1,
          "answer_head": "The Technique: Modelling Techniques and the Artifacts They Ship\nIn this section, we delve into the various modelling techniques used in the field of model architecture, their respective characteristics, and the artifacts that accompany them. We will explore the different approaches to model design, including dense and sparse structures, how these techniques are matched to specific data types, and "
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16"
          ],
          "length_ratio": 0.953,
          "repeats": 1,
          "answer_head": "Model Architectures, Modelling Types, and the Mermaid Viewer: A Data Summary\nIntroduction\nThe purpose of this report is to provide a comprehensive overview of the latest developments in model architectures, modelling types, and the Mermaid Viewer, a powerful tool for visualizing and understanding complex data structures. This report covers the following sections:\n\n1. Model Architectures\n2. Modelli"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.105,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1.1",
            "16",
            "165.2",
            "26",
            "380.0",
            "43.4"
          ],
          "numbers_invented": [
            "2",
            "75.8"
          ],
          "length_ratio": 1.172,
          "repeats": 1,
          "answer_head": "The artifact census provides a comprehensive overview of the various components that make up the Model Architectures, Modelling Types, and the Mermaid Viewer tool. This section presents the statistics on the storage requirements, file sizes, and usage patterns of each type of artifact used within the tool, including:\n\n1. **Storage Requirements**:\n   - Data Summary: The mermaid viewer requires a si"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "1",
            "2",
            "3"
          ],
          "length_ratio": 3.504,
          "repeats": 1,
          "answer_head": "What the bytes cost to load\nIn the realm of data analysis and visualization, the loading process can be a significant expense, especially when dealing with large datasets that require complex modeling and rendering. This section provides an overview of the various factors involved in loading data into a model using the Mermaid viewer, including the memory requirements, disk space usage, and comput"
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.118,
          "numbers_missing": [
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "27.6",
            "34",
            "380.0"
          ],
          "numbers_invented": [
            "3",
            "4"
          ],
          "length_ratio": 2.227,
          "repeats": 1,
          "answer_head": "Where the Workspace Keeps Them: A Data Summary of Model Architectures, Modelling Types, and the Mermaid Viewer\nIn this report, we delve into the world of model architectures, modelling types, and the Mermaid Viewer, providing a comprehensive overview of these essential components and their role in the creation and management of complex data models. We explore the various aspects of the Mermaid vie"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "1",
            "2",
            "3"
          ],
          "length_ratio": 0.944,
          "repeats": 2,
          "answer_head": "Model Architectures, Modelling Types, and the Mermaid Viewer: A Data Summary\n\nIntroduction:\n\nIn this report, we delve into the world of model architectures, modelling types, and the Mermaid Viewer, providing a comprehensive overview of these concepts and their role in the design and development of complex systems. Our focus is on the key aspects of the Mermaid Viewer, including its functionality, "
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.077,
          "numbers_missing": [
            "04",
            "09",
            "10",
            "2",
            "2026",
            "25",
            "3",
            "4"
          ],
          "numbers_invented": [],
          "length_ratio": 1.044,
          "repeats": 2,
          "answer_head": "Method and sources\nThis report provides a comprehensive overview of the Model Architectures, Modelling Types, and the Mermaid Viewer, including key information on the viewer's functionality, the different types of models that can be created using the Mermaid syntax, and the latest trends and developments in the field. This section covers the following topics:\n\n1. Introduction\nThe Mermaid viewer is"
        }
      ]
    },
    "posttrain": {
      "count": 40,
      "style_clean_rate": 0.975,
      "numbers_kept_mean": 0.1244,
      "passed_rate": 0.075,
      "length_ratio_mean": 0.3385,
      "sources_rate": 0.0,
      "loop_rate": 0.025,
      "repeats_max": 30,
      "label": "posttrain",
      "speed": {
        "seconds": 187.9,
        "tokens": 9207,
        "tokens_per_second": 49.0,
        "seconds_per_report": 4.697
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1244,
        "passed_rate": 0.075,
        "length_ratio_mean": 0.3385,
        "sources_rate": 0.0,
        "loop_rate": 0.025,
        "repeats_max": 30
      },
      "filter_enabled": true,
      "filtered_rate": 0.025,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.073,
          "repeats": 0,
          "answer_head": "| File | 0 rows|\n|------|--------|\n|..|  `..`"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "00",
            "100"
          ],
          "length_ratio": 0.15,
          "repeats": 0,
          "answer_head": "| Notebook | 0 files |\n|----------|---------|\n| `00_model_types.ipynb` | 100 lines, 0 tables, 0 imports, 0 plots, 0 references.|"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "00",
            "100",
            "4",
            "40000"
          ],
          "length_ratio": 0.087,
          "repeats": 0,
          "answer_head": "| File | 0 rows |\n|------|--------|\n```\n00\n.. ra\n  id  40000\n  type ra\n  format pml\n  encoding pml\n  shape [100, 100, 4]\n  channels 4\n  dtype uint8\n  encoding ascii\n  language r\n  content\n  0 ``"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "10",
            "100",
            "30",
            "40"
          ],
          "length_ratio": 0.388,
          "repeats": 0,
          "answer_head": "| File | 100\n|------------------|----------------------------------------------------------\n| models/..., models/... | 40\n| models/..., models/canva,...| 30\n| models/..., models/..., models/..., models/..., models/..., models/..., `model.py`, 40\n| benchmarks/bench..., benchmarks/bench..., benchmarks/bench..., benchmarks/bench..., `bench.py`, 40\n| benchmarks,..| 10\n|...| \u7686\u88f8, \u7686 py\n```"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.018,
          "repeats": 0,
          "answer_head": "| File | 0 rows|\n|---|---|\n|..|"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.046,
          "repeats": 0,
          "answer_head": "| Layout | 0 rows |\n|--------|---------|\n| `model_archs` [0] | `list[0]`"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0",
            "100"
          ],
          "length_ratio": 0.098,
          "repeats": 0,
          "answer_head": "| File | 100\n---| -----------\n`models/axi_mechanics.py`, 100 lines\n```python\n.. py:module:: xcalibur\n    0\n```\n"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0",
            "100"
          ],
          "length_ratio": 0.271,
          "repeats": 0,
          "answer_head": "| File | 100,\n| `models/axml/axml.py`, 100,\n| `axml.yaml`, 0,\n| `axml.json`, 0,\n| `axml.yaml.example`, ``/workspace/xcalibur/db/models/axml/axml.yaml\n|\u6837\u7684\u683c\u5f0f, zero imports, zero lines, one file.|"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.143,
          "numbers_missing": [
            "04",
            "1",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "0",
            "100",
            "40",
            "400"
          ],
          "length_ratio": 0.103,
          "repeats": 0,
          "answer_head": "| File | 100,\n| `models/axml/axml.py`, 400,\n| `models/axml/axml_types.yaml`, 40,\n| `data/census.csv`, 10,\n| `..`, 0"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.125,
          "numbers_missing": [
            "1",
            "100000",
            "16",
            "256",
            "32",
            "8",
            "88"
          ],
          "numbers_invented": [
            "0",
            "100"
          ],
          "length_ratio": 0.702,
          "repeats": 1,
          "answer_head": ".. hdl: `databases/db_crypto.h`\n.. language: c\n.. diff\n```\n-- db_crypto.h\n+ // Cryptography\n+ #include \"crypt.h\"\n+ #include <openssl/bn.h>\n+ #include <openssl/rand.h>\n+ #include <openssl/sha.h>\n+ #include <openssl/hmac.h>\n\n+ enum db_encryption_algorithm\n+ {\n+     DB_ENCRYPTION_NONE = 0,\n+     DB_ENCRYPTION_PBKDF2,\n+     DB_ENCRYPTION_HMAC_SHA256,\n+     DB_ENCRYPTION_AES_256_GCM,\n+     DB_ENCRYPTIO"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "3",
            "4"
          ],
          "length_ratio": 0.492,
          "repeats": 1,
          "answer_head": ".. hdl: `src/db_transfer.h`\n  Heading 4\n  4\n    `db_transfer.h`, 4\n      `cryptlib_api.h`, 4\n        `cryptapi.h`, 4\n          `cryptapi.h`, 4\n            `cryptapi.h`, 4\n              0\n                `cryptapi.h`, 4\n                  H0\n                    4 [DEP]\n                      `cryptapi.h` declares `CRYPTAPI_API crypt_file_create`, which is defined\n                      at `cryptapi.h("
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.556,
          "numbers_missing": [
            "0.1",
            "0.3",
            "0.4",
            "29",
            "32",
            "453",
            "512",
            "60"
          ],
          "numbers_invented": [
            "00",
            "000000000000000000",
            "10"
          ],
          "length_ratio": 0.502,
          "repeats": 1,
          "answer_head": "| Section | 0 |\n|---|---|\n| 1. Introduction `00[0]` `.../data/biohub-biomedical-benchmarks/biosim-benchmark/biohub-cell-tracking/biohub_cell_tracking.ipynb`| `..` |\n| 2. Notebook `00[1]`, `00[2]`, `00[3]`, `00[4]`, `00[5]`, `00[6]`, `00[7]`, `00[8]`, `00[9]`| 0, 0, 0, 0, 0, 0, 0, 0, 0, `00[0]` |\n| 3. Notebook imports `00[10]` `../*.ipynb`| `nbformat`, `kaggle`, `pandas`, `numpy`, `matplotlib`, `se"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [
            "0.0",
            "0.1"
          ],
          "length_ratio": 0.156,
          "repeats": 0,
          "answer_head": "| Name | Weight |\n|------|--------|\n| `biohub-cell-tracking` | 0.1     |\n| `cell_tracking`   | 0.0     |"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.041,
          "repeats": 0,
          "answer_head": ".. [A0000000000000000000000000000000000000000, 0, 0, `authors/ckinga/ck0000000000000000000000000000000000000000`, 0, 0, ``}`"
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.004,
          "repeats": 0,
          "answer_head": "... [skipped]"
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.019,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.14",
            "0.15"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.135,
          "repeats": 0,
          "answer_head": ".. [A000000000000000000000000000000000000000, 0.10, 0, 0, `0.10/0/0/0/0`, ``0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/0/"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "0",
            "0.00",
            "0.000000",
            "0.01"
          ],
          "length_ratio": 0.171,
          "repeats": 0,
          "answer_head": ".. [A0000000000000000000000000000000000000000]\n.. [0]\n.. 0\n    The bare minimum budget is 0.01, which is the cost of\n    `kitti_query/ckp_annotate.py\n    run\n    `python\n    kaggle competitions\n    kaggle\u6807\u6ce8\n    kaggle\u6807\u6ce8 0\n    `.`.\n   \u4e8e\u662f\u6211\u521b\u5efa\u4e86\u4e00\u4e2a\n    `kaggle-benchmarks\n    run\n    `kaggle benchmark kaggle\u6807\u6ce8\n    `..`.\n\n[0] 0.01, 0.000000, 0.000000, 0.000000, 0.00, 0.00, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "0",
            "0.001",
            "0.01",
            "0.05",
            "0.1",
            "0.2",
            "1",
            "10"
          ],
          "length_ratio": 0.138,
          "repeats": 1,
          "answer_head": ".. py\n.. autoplot\n[plot]\nax = plot(\n    x=\"cost\",\n    y=\"volume\",\n    label=\"Costs\",\n    style=style,\n    markerline=True,\n    markersize=10,\n    markercolor=cm_brown(0.2),\n)\nax.set(xlim=(0, 10), ylim=(-10, 10), xlabel=\"Number of users\", ylabel=\"Price\")\nax.margins(0.2)\nax.text(0.05, 0.001, \"1M\", color=cm_grey(0.2), ha=\"center\", va=\"top\")\nax.text(0.05, 0.01, \"0.1\", color=cm_lightblue(0.2), ha=\"cent"
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [
            "4",
            "8"
          ],
          "length_ratio": 0.037,
          "repeats": 0,
          "answer_head": "The paper has four sections, ``Hypotaxis, Capabilities, Costs, and Evidence/Analysis/Answering, 4..4`, so the\nsums are 4+4=8.`"
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.081,
          "numbers_missing": [
            "09",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16"
          ],
          "numbers_invented": [
            "0",
            "0.10"
          ],
          "length_ratio": 0.141,
          "repeats": 0,
          "answer_head": ".. [A000000000000000000000000000000000000000, 0.10, 0, 0, `0.10/0/0/0/0`, `0.10/0/1/0/0`, `0.10/1/0/0/0`, `0.10/0/0/1/0`, `0.10/1/1/0/0`, `0.10/0/0/0/1`, `0.10/0/1/1/0`, `0.10/1/0/1/1`, `0.10/1/0/0/2`, `0.10/1/1/0/2`, `0.10/1/0/1/3`, `0.10/1/1/1/0`, `0.10/1/1/1/1`, `0.10/1/1/0/2/0`, `0.10/1/1/1/1/0`, `0.10/1/2/0/0`, `0.10/1/2/0/1`, `0.10/1/2/1/0`, `0.10/1/2/1/1`, `0.10/1/2/0/2/0`, `0.10/1/2/1/1/0`"
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.2,
          "numbers_missing": [
            "113",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "length_ratio": 0.528,
          "repeats": 0,
          "answer_head": "| File | 0, 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 |\n|---------|-------------------------------------------------------|\n| models/| 4, 5, 6, 10, 11, 12, 14, 15, 16, 17, 20, 21, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76,"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "0.4"
          ],
          "length_ratio": 0.587,
          "repeats": 0,
          "answer_head": "| Notebook | 0.4\n| `model_000000000000000000` | 0\n| `model_00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.04,
          "repeats": 0,
          "answer_head": "| Notebook | 0 files, 0 references |\n|---------|----------------|\n| `model_archs` | `0/0`"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "1.10",
            "100",
            "2.4",
            "400"
          ],
          "length_ratio": 0.655,
          "repeats": 1,
          "answer_head": "| File | 100\n|----------------|----------------------------------------------------------\n| models/abalone.py | PyTorch 1.10.0, `abalone`, `abalone_001.h5`, 400\n| models/birds.py | PyTorch 1.10.0, `birds`, `birds_001.h5`, 400\n| models/cam.py | PyTorch 1.10.0, `cam_001.h5`, 400\n| models/cifar10.py | Keras 2.4.0, `cifar10_cnn_001.h5`, 400\n| models/dog.py | PyTorch 1.10.0, `dog_001.h5`, 400\n| models/"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0.1",
            "100",
            "400"
          ],
          "length_ratio": 0.165,
          "repeats": 0,
          "answer_head": "| File | 0.1\n| `models/agent/walker.py` | Models, 400\n| `docs/maql/maql.ipynb` | The notebook, 100\n| `platforms/axel/axel.yaml`, `axel.yaml` | The plan, 100\n| `platforms/axel/axel.json`, `axel.json` | The plan, 400\n| `platforms/axel/axel.yaml`, 400, `platforms/axel/axel.yaml` |"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.031,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "00",
            "15"
          ],
          "length_ratio": 0.222,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model_types` | 0.0\n| `models`, 15/15 ``| `models/00_overview.yaml`, `models/00_model_types.yaml`, `models/00\u8bdd\u8bed\u7c7b\u578b| 0\n| `models/00`, 10/15 ``| `workspace/00.ipynb`, `workspace/00_module_types`, `workspace/00_model_types`, `workspace/00\u8bdd\u8bed\u7c7b\u578b| 10\n| `index`, 0/15 ``| `..`, bare notebook, 0 sections.|"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "000000",
            "1",
            "2",
            "400"
          ],
          "length_ratio": 0.378,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model_types` | 0.0\n| ```\n|---|---\n| 0| `models/000000.ipynb`\n| 1| `papers/mapping_the_model_types.ipynb`, 400 lines\n| 2| `..., 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.188,
          "repeats": 0,
          "answer_head": "| Notebook | Models, type, reference, where |\n|---------|------------------------------------------------|--------------------------------|\n| `model-architecture`   [..] `..`   0 files, 0 references, 0 benchmarks |"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.838,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------------------------------------------------------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [],
          "length_ratio": 0.052,
          "repeats": 0,
          "answer_head": "| Paper | Authors | Models, type, and layout\n|..."
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [],
          "length_ratio": 0.006,
          "repeats": 0,
          "answer_head": "[..]\n.."
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.667,
          "numbers_missing": [
            "30",
            "78"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1",
            "100",
            "11",
            "12",
            "13",
            "14"
          ],
          "length_ratio": 0.961,
          "repeats": 0,
          "answer_head": "| File | 0.0\n|--------|----------\n`models/agent_wizard.py`, 100 lines\n[..]\n    0\n    1\n    2\n    3\n    4\n    5\n    6\n    7\n    8\n    9\n    10\n    11\n    12\n    13\n    14\n    15\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.0"
          ],
          "length_ratio": 0.151,
          "repeats": 0,
          "answer_head": "| Name | Status | Size | Hdr | Args,\u4e16\u4ee3|\n|------|--------|-----|-----|---------|\n| `safetensor/agent_wizard` | 0.0 | 0 | 0 | [],|"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [
            "0",
            "200"
          ],
          "length_ratio": 3.24,
          "repeats": 1,
          "answer_head": ".. py\nfrom typing import List, NamedTuple, Text, File, Path, Union\nimport re\nimport numpy as np\nimport tensorflow as tf\nimport tensorflow_model_analysis as tfma\nimport tensorflow_remediation as tfa\nimport tensorflow_remediation.types as tft\nimport tensorflow_remediation.pipelines as tfp\nimport tensorflow_remediation.errors as tfe\nimport tensorflow_docs as tfd\nimport os\nimport warnings\nimport six\ni"
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "0",
            "100",
            "400"
          ],
          "length_ratio": 0.44,
          "repeats": 1,
          "answer_head": "| File | 100\n|----------------|----------------------------------------------------------\n| models/abalone.h5 | 400\n| models/abalone.py | 0\n| models/abalone.caffemodel | 0\n| models/abalone.slim | 0\n| models/abalone.json | 0\n| models/abalone.yaml | 0\n| models/cell_type.h5 | 0\n| models/cell_type.py | 0\n| models/cell_type.caffemodel | 0\n| models/cell_type.slim | 0\n| models/cell_type.json | 0\n| models"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [
            "0",
            "000000",
            "000001",
            "000002",
            "000003",
            "000004",
            "000005",
            "000006"
          ],
          "length_ratio": 0.38,
          "repeats": 30,
          "answer_head": "| File | 4000\n|---------|------\n| models/000000.py | 0\n| models/000001.py | 0\n| models/000002.py | 0\n| models/000003.py | 0\n| models/000004.py | 0\n| models/000005.py | 0\n| models/000006.py | 0\n| models/000007.py | 0\n| models/000008.py | 0\n| models/000009.py | 0\n| models/000010.py | 0\n| models/000011.py | 0\n| models/000012.py | 0\n| models/000013.py | 0\n| models/000014.py | 0\n| models/000015.py | 0\n"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0.0000000000000000",
            "4000"
          ],
          "length_ratio": 0.073,
          "repeats": 0,
          "answer_head": "| File | 4000  rows | 4000  bytes | 0.0000000000000000\n```"
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "100"
          ],
          "length_ratio": 0.499,
          "repeats": 0,
          "answer_head": "| File | 0.0\n|--------|----------\n`models/agent/walker.py`, 0 bytes\n[py]\n```python\nimport os\n```\n| 0 | `models/agent/walker.py`, 0 bytes\n| 100  | 100\n|\u6837\u7684, the plan\n|---|--\n| 0..| `model_0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "000000000000",
            "4000"
          ],
          "length_ratio": 0.071,
          "repeats": 0,
          "answer_head": "| Field | Description|\n|---|---|\n|`model/`, `models`, `models.yaml`, `000000000000_000000000000.ipynb`, 4000 lines| Notebook, four sections, 4000 lines|\n|`index.rst`, 4000, 4000, 4000| Four sections, 4000 words|"
        },
        {
          "section": "Method and sources",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.154,
          "numbers_missing": [
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [
            "0",
            "00",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "01",
            "02",
            "03",
            "05",
            "06"
          ],
          "length_ratio": 0.311,
          "repeats": 1,
          "answer_head": "| Notebook | 0 files\n| `00_00_notebook.ipynb` | The bare notebook\n| `00_01_model_types.ipynb` | The type definitions\n| `01_data_types.ipynb` | The data layout\n| `02_model_archs.ipynb` | The model family listings\n| `03_model_to_type.py`, `04_types\u77ff\u7269|metal|crystal|photon|wave`, `05_types\u77ff\u7269|metal|crystal|photon|wave.json`, `06_types.yaml`, `07_types.yaml`, `08_types.yaml`, `09_types.csv`, `01_types.c"
        }
      ]
    },
    "xcaliburbiz-Q2_K": {
      "count": 40,
      "style_clean_rate": 0.95,
      "numbers_kept_mean": 0.0798,
      "passed_rate": 0.075,
      "length_ratio_mean": 0.2322,
      "sources_rate": 0.0,
      "loop_rate": 0.025,
      "repeats_max": 52,
      "label": "xcaliburbiz-Q2_K",
      "speed": {
        "seconds": 19.42,
        "tokens": 8396,
        "tokens_per_second": 432.4,
        "seconds_per_report": 0.485
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.0798,
        "passed_rate": 0.075,
        "length_ratio_mean": 0.2322,
        "sources_rate": 0.0,
        "loop_rate": 0.025,
        "repeats_max": 52
      },
      "filter_enabled": true,
      "filtered_rate": 0.05,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "1.000"
          ],
          "length_ratio": 0.011,
          "repeats": 0,
          "answer_head": "* 1.000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [],
          "length_ratio": 0.054,
          "repeats": 0,
          "answer_head": "..,\n| The model viewer, four in all, a list\n| "
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0.00"
          ],
          "length_ratio": 0.013,
          "repeats": 0,
          "answer_head": "..,\n| `0.00\n`*/index.yaml\n```"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "1000"
          ],
          "length_ratio": 0.524,
          "repeats": 0,
          "answer_head": "...: 1000\n 1000\n--| 1000\nThe\nH\n| 1000\n| .py\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000\n| 1000"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0.000000000000",
            "0.001000000000",
            "0.100000000000",
            "0.400000000000",
            "4000"
          ],
          "length_ratio": 0.065,
          "repeats": 0,
          "answer_head": "..,\n| `0.100000000000\n 0.100000000000\n 0.400000000000\n 0.100000000000\n 0.001000000000\n 0.000000000000\n| 4000\n|"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1000"
          ],
          "length_ratio": 0.026,
          "repeats": 0,
          "answer_head": "..,\n| The model layout\n| a, 1000 rows\n.."
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0",
            "0.000",
            "0000000000000000",
            "1",
            "1000"
          ],
          "length_ratio": 0.502,
          "repeats": 0,
          "answer_head": "...[1], 0.000\u786c\u5ea6\nThe zero counts the zero, which counts 1000 four times, so\n...[0, 0000000000000000, 0000000000000000, 0].*/].*/\u7686\n\u800c\u4e0d\u662f\n...[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [],
          "length_ratio": 1.359,
          "repeats": 0,
          "answer_head": "..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "1.00"
          ],
          "length_ratio": 0.009,
          "repeats": 0,
          "answer_head": "* 1.00 `|`"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [
            "00",
            "00000000",
            "0000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.117,
          "repeats": 0,
          "answer_head": "```/generated/0000000000000000000000000000000000000000000000000000000000000000/00000000/00\n```"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [],
          "length_ratio": 0.972,
          "repeats": 0,
          "answer_head": "```cpp\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n."
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0",
            "0.1",
            "0.3",
            "0.4",
            "1",
            "2",
            "29",
            "3"
          ],
          "numbers_invented": [],
          "length_ratio": 0.002,
          "repeats": 0,
          "answer_head": "```"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [],
          "length_ratio": 0.034,
          "repeats": 0,
          "answer_head": "A notebook, one report."
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.143,
          "numbers_missing": [
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.000",
            "0.00000000000000000000000000000000",
            "1000"
          ],
          "length_ratio": 0.171,
          "repeats": 0,
          "answer_head": "* 0.00000000000000000000000000000000 0.000 0, 0, 0, 1000, 0, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 100"
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.000000",
            "0.0000000000000000"
          ],
          "length_ratio": 0.144,
          "repeats": 0,
          "answer_head": "* 0.0000000000000000\n* 0.0000000000000000\n* 0.000000..., 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0"
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.019,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.025",
            "0.03",
            "0.075",
            "0.10",
            "0.14",
            "0.15"
          ],
          "numbers_invented": [
            "0.00000",
            "0.0000000000000000",
            "0.0010"
          ],
          "length_ratio": 0.013,
          "repeats": 0,
          "answer_head": "* 0.0000000000000000\n* 0.01 H2H, 0.0010, 0.00000"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "0",
            "0.00000000000000",
            "0.0000000000000000",
            "1000"
          ],
          "length_ratio": 0.02,
          "repeats": 0,
          "answer_head": "* 0.0000000000000000\n* 0.0000000000000000\n* 0.00000000000000..., 1000,..., 0,..."
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "0",
            "0.0000",
            "0.0000000000000000"
          ],
          "length_ratio": 0.153,
          "repeats": 0,
          "answer_head": "* 0.0000000000000000\n* 0.0000 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0"
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [],
          "length_ratio": 0.006,
          "repeats": 0,
          "answer_head": "The paper's layout:"
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "09",
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "numbers_invented": [],
          "length_ratio": 0.009,
          "repeats": 0,
          "answer_head": "A bare dependency layout, one section."
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "14",
            "14.00"
          ],
          "length_ratio": 0.04,
          "repeats": 0,
          "answer_head": "14.00\n.. .\n| level 14\n| level 0\n| The\n| .0\n|"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "00000"
          ],
          "length_ratio": 0.053,
          "repeats": 0,
          "answer_head": ".. my\n.. a section, ``00000\n.. a level, level\n.."
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "10"
          ],
          "length_ratio": 0.015,
          "repeats": 0,
          "answer_head": ".. four sections, 10, 10, 0, 0\n```"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "0000000000000004",
            "0000000000000400",
            "0000000000040000",
            "0000000000040004",
            "0000000000040400"
          ],
          "length_ratio": 0.121,
          "repeats": 0,
          "answer_head": "The four sections: `0000000000000004`, `0000000000040000`, `0000000000000400`, `0000000000040004`, and `0000000000040400`."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0",
            "0.000",
            "1",
            "100"
          ],
          "length_ratio": 0.015,
          "repeats": 0,
          "answer_head": "..\u7686: 1, 100, 0\n 0.000\n```"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.031,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0.00",
            "00",
            "100"
          ],
          "length_ratio": 0.364,
          "repeats": 0,
          "answer_head": "..\u7686: 100, 100\n 0.00\n-- 100\n.. 00\n| 100\n| 10\n| 10\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 100\n| 1"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "1"
          ],
          "length_ratio": 0.054,
          "repeats": 0,
          "answer_head": "..\u84b8\n ..axml\n ..h,\n ..h,\n ..level0\n ..level0\n ..label: 1\n ..\u5bb9\n ..level0\n ..level2\n .."
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0",
            "1"
          ],
          "length_ratio": 0.452,
          "repeats": 0,
          "answer_head": "..\u7686: 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "10.0"
          ],
          "length_ratio": 0.063,
          "repeats": 0,
          "answer_head": "The bare dependency, one section, 10.0 bytes."
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "0",
            "1"
          ],
          "length_ratio": 0.544,
          "repeats": 0,
          "answer_head": "..[1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [
            "0.10"
          ],
          "length_ratio": 0.01,
          "repeats": 0,
          "answer_head": "`0.10..000`"
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "14.0"
          ],
          "length_ratio": 0.868,
          "repeats": 0,
          "answer_head": "14.0\n.. .\n.. The model document, `000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.00",
            "0000"
          ],
          "length_ratio": 0.949,
          "repeats": 52,
          "answer_head": "`0000\n0000| 0\n0000| 0\n0, 0\n| Models\n  0.00\nlevel, 0\n| my own\n| my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, my own, m"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [
            "0.10"
          ],
          "length_ratio": 0.757,
          "repeats": 0,
          "answer_head": "`0.10..000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "10",
            "100"
          ],
          "length_ratio": 0.24,
          "repeats": 0,
          "answer_head": ".. four sections, 10 levels\n..\u84b8\u5bb9\n .. 100\n .. 10\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n .. 100\n"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [],
          "length_ratio": 0.009,
          "repeats": 0,
          "answer_head": "The bare dependency."
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [],
          "length_ratio": 0.069,
          "repeats": 0,
          "answer_head": "The bare bytes of a model architecture, the bare bytes."
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0",
            "0.1000",
            "00000000",
            "00000001"
          ],
          "length_ratio": 0.052,
          "repeats": 0,
          "answer_head": "`0.1000\n00000001 four hundred\n00000000 level\n00000000 four\n0\n```"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "0",
            "0.000",
            "000",
            "1"
          ],
          "length_ratio": 0.025,
          "repeats": 0,
          "answer_head": "[1], 0.000\nThe document's own file,\u800c\u975e\nH.000, which costs\n000\u684c\u9762\u5927\u5c0f, 0 bytes."
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "14.00"
          ],
          "length_ratio": 0.386,
          "repeats": 0,
          "answer_head": "14.00\n.. .\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n..,\n."
        }
      ],
      "gpu_offload": true
    },
    "xcaliburbiz-Q3_K_M": {
      "count": 40,
      "style_clean_rate": 1.0,
      "numbers_kept_mean": 0.1084,
      "passed_rate": 0.1,
      "length_ratio_mean": 0.1792,
      "sources_rate": 0.0,
      "loop_rate": 0.0,
      "repeats_max": 1,
      "label": "xcaliburbiz-Q3_K_M",
      "speed": {
        "seconds": 15.19,
        "tokens": 6520,
        "tokens_per_second": 429.3,
        "seconds_per_report": 0.38
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1084,
        "passed_rate": 0.1,
        "length_ratio_mean": 0.1792,
        "sources_rate": 0.0,
        "loop_rate": 0.0,
        "repeats_max": 1
      },
      "filter_enabled": true,
      "filtered_rate": 0.0,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.2,
          "numbers_missing": [
            "113",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "1",
            "10",
            "100",
            "1000",
            "10000",
            "100000",
            "1000000"
          ],
          "length_ratio": 0.835,
          "repeats": 0,
          "answer_head": "| Model | 0, 1, 2, 4, 10, 100, 1000, 10, 100, 10000, 100, 100000, 1000, 10000, 1000000, 10, 10, 100, 1000, 100000, 100, 100, 10, 1000, 100000, 10, 10, 100, 100, 1000, 100, 10, 10, 100, 10, 100, 100, 1000, 10, 10, 10, 100, 100, 100, 1000, 1000, 10000, 1000, 100, 100000, 100, 10, 10, 10, 10, 10, 10, 10, 10, 100, 1000, 10000, 100000, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 100, 100, 10, 10, 10, 1"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.161,
          "repeats": 0,
          "answer_head": "| Operation | What it does | Notebook | Output\n| `model/` | the model, the types, the types, the types, the types | 0 | `| Notebook: 0| `"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "10",
            "100",
            "1000",
            "10000",
            "2",
            "4"
          ],
          "length_ratio": 0.232,
          "repeats": 0,
          "answer_head": "| Model | 0, 1, 2, 4, 10, 100, 1000, 10000, 100, 10, 0, 100, 1000, 10, 0, 100, 100, 10, 1000, 10000, 10, 100, 10000, 100, 10, 100, 10, 10, 10, 100, 100, 100, 1000, 1000, 10, 10, 100, 10, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [],
          "length_ratio": 0.146,
          "repeats": 0,
          "answer_head": "The four types the model paper reports are `model.py`, four modules, and four types, which the types\nlist, the four sections, and the four types."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.117,
          "repeats": 0,
          "answer_head": "| 0 | The four models, four models, one paper, one file, one section, zero imports, zero sections, zero rows, zero plots, zero figures, zero tables, zero imports, zero references, zero section, 0.`"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "1",
            "10",
            "100",
            "1000",
            "10000",
            "100000",
            "1000000"
          ],
          "length_ratio": 0.334,
          "repeats": 0,
          "answer_head": "| Model | 0, 1, 2, 4, 10, 100, 1000, 10000, 10, 100, 100000, 100, 1000, 1000000, 10000, 100000, 100, 10, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "100",
            "400.0"
          ],
          "length_ratio": 0.458,
          "repeats": 0,
          "answer_head": "| Name | 400.0\n| 400.0 | 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, "
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "100"
          ],
          "length_ratio": 0.729,
          "repeats": 0,
          "answer_head": "| Model | 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, "
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [],
          "length_ratio": 0.049,
          "repeats": 0,
          "answer_head": "| Method | What it does, the paper, | Hcribed\u684c\u9762, `|\u684c\u9762|`"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [
            "100"
          ],
          "length_ratio": 0.416,
          "repeats": 1,
          "answer_head": "The module implements the following functionality, which is exported into `db_crypto.h` and its headers, and the source file `db_crypto.c`, which is the executable. The source file names are listed in the `src/` directory, so the module lives in `src/db_crypto.c`, and the module source is 100 bytes, so the section size is 100 bytes."
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [],
          "length_ratio": 0.187,
          "repeats": 0,
          "answer_head": "The transfer module defines the bulk upload, download, and list operations, and the\n`db_transfer` file follows the Haskeller style with a single header, a\n`h` file, and a single, bare, source file."
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.056,
          "numbers_missing": [
            "0.1",
            "0.3",
            "0.4",
            "1",
            "2",
            "29",
            "3",
            "32"
          ],
          "numbers_invented": [
            "10"
          ],
          "length_ratio": 0.071,
          "repeats": 0,
          "answer_head": "| Heading | 0, 0, 0, 0, 0, 0, 0, 0, 10 |\n|:-- |:--|:|:|:|:|:|:|:|:|\n| 0. `DATA_SUMMARY` `..` | 0, 0, 0, 0, 0, 0, 0, 0, 10, 0|"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [],
          "length_ratio": 0.076,
          "repeats": 0,
          "answer_head": "| Name | Weight, level, metric |\n|:--:|:--|:--|:--|"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.10",
            "0.14"
          ],
          "numbers_invented": [
            "100",
            "100000"
          ],
          "length_ratio": 0.135,
          "repeats": 0,
          "answer_head": "| Model | 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100,000, 100"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "1",
            "10",
            "100",
            "1000",
            "10000",
            "100000"
          ],
          "length_ratio": 0.155,
          "repeats": 0,
          "answer_head": "| Model | 100,000, 1M, 100, 1000, 10000, 10, 100, 10000, 100, 100, 1000, 10000, 100, 10, 100, 100, 100, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, "
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [],
          "length_ratio": 0.02,
          "repeats": 0,
          "answer_head": "The paper reports four sections, the paper does not write a section."
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.081,
          "numbers_missing": [
            "09",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16",
            "17"
          ],
          "numbers_invented": [
            "0",
            "100",
            "1000",
            "10000",
            "100000"
          ],
          "length_ratio": 0.127,
          "repeats": 0,
          "answer_head": "| Model | Input, 0, 1, 2, 10, 100, 1000, 10000, 100, 100000, 10, 100, 1000, 100000, 100, 10, 100, 100, 100, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, "
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1"
          ],
          "length_ratio": 0.055,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n|  | 0.0\n| 1, `model-architecture`, `0.0`, 0"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1"
          ],
          "length_ratio": 0.578,
          "repeats": 0,
          "answer_head": "| Notebook | 0.1\n| `0.0` | `0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.018,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| 0\n| `~| bare\n| 0\n| `| `"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [],
          "length_ratio": 0.083,
          "repeats": 0,
          "answer_head": "The four sections\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686\u7686, the bare document,"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0",
            "00"
          ],
          "length_ratio": 0.082,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `00_00000000000000.ipynb` | bare, 0 files, 0 references, 0 imports, 0\n| `00_00000000000000.ipynb.yaml` |.yaml, 0, 0, 0, 0"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.13,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n|  | `pydeck` and `deck`, the bare models, the bare types, the bare\n|  | `0.0/0.0.yaml` `pydeck-0.0.yaml` `pydeck-0.0.ipynb` `0.0.ipynb`| a\n|  | bare notebook\n|  | `|`"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.135,
          "repeats": 0,
          "answer_head": "| Name | 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0",
            "0.0"
          ],
          "length_ratio": 0.07,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| 0.0 `| bare\n| 0.0 `| `,\n| 0.0 `| `,\n| 0.0 `| ` rap\n| 0.0 `| ` `"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1"
          ],
          "length_ratio": 0.132,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| 0.0 `DATA_FORMAT` `| the bare bare volume\n| 0.1 `DATA_FORMAT`, 0, | `| ` `| `"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "0",
            "1",
            "2"
          ],
          "length_ratio": 0.127,
          "repeats": 0,
          "answer_head": "| Type | What it declares, `model.py` | `where` | `where[0]` | `where[1]` | `where[2]` | `where[0]/1` | 0| 0| 0| 0| 0| 0"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [
            "0",
            "000000000000",
            "1"
          ],
          "length_ratio": 0.316,
          "repeats": 1,
          "answer_head": "The work the paper reports, `model-000000000000`, carries four modules, four types, and four authors, whose\nnames the paper declares, and whose names the paper declares\u7686, with four\nlevel-0 types and four level-1 types, zero imports, zero imports, zero\nplatforms, zero benchmarks, zero benchmarks, and the section level 0\n(`:module:`, `:file:`)."
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "000000"
          ],
          "length_ratio": 0.111,
          "repeats": 0,
          "answer_head": "| File | 0.0\n| 0.0\n| `models/000000.py` 0, 0, zero, bare, bare, bare"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0.1"
          ],
          "length_ratio": 0.101,
          "repeats": 0,
          "answer_head": "| Name | Type | What it carries, and the file | H0\n|: `model_archs` | directory, `0.1`"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [
            "000000",
            "0000000000000000"
          ],
          "length_ratio": 0.459,
          "repeats": 0,
          "answer_head": "The export lives at `0000000000000000.ipynb`, the notebook the tool generated, so the\n`| Notebook` label and the bare notebook name are both a\n`/Users/user/working/username/000000.ipynb` export, which the\n`| file: export` layout cannot recover, so the\n`| 000000.ipynb` label stays, and the\n`| 000000.yaml` plan."
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.084,
          "repeats": 0,
          "answer_head": "| Name | 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [
            "0",
            "0.1"
          ],
          "length_ratio": 0.135,
          "repeats": 0,
          "answer_head": "| Exported | H0\n|-------- | -----------\n| types/0.1.0.yaml | `file: typed.yaml` `/Users/username/agent/work/xcalibur/agent/0.1.0/` `auto_types.py` 0\n| types/0.1.0/0| `0 files, 0 types, 0 imports, 0 annotations, 0 bare`\n| types/0.1.0/0.yaml| `0 files, 0 types, 0 imports, 0 bare, 0 annotations, 0 bare\u58f3`"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1",
            "0.2"
          ],
          "length_ratio": 0.115,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| 0.0 `| bare\n| 0.1 `| `| the\n| 0.2, 0.0 `| 0, the\n| 0.\u8d8a\u5927, 0.0 `| 0, the\n| 0."
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.028,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| 0.0\n| 0.0\n| `| `"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "000"
          ],
          "length_ratio": 0.088,
          "repeats": 0,
          "answer_head": "| Field | Description\n| -----------\n| `type` | The type the model, file, or a model type\n| `type_file` | The model, its source, or the model\n| `authors`, `titles`, `refs`, `refs/000`, `000`, `000.yaml` | The work, the types, the ordering, and the\n| `000.yaml` |"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.068,
          "repeats": 0,
          "answer_head": "| Method | What it does, the model, the plan, the export, the reference |\n|:------|:------------\n| `pytorch_modelgen` `0.0.10` [pytorch, pytorch_modelgen] `0.0.10` [call, package]"
        }
      ],
      "gpu_offload": true
    },
    "xcaliburbiz-Q4_K_M": {
      "count": 40,
      "style_clean_rate": 0.975,
      "numbers_kept_mean": 0.1011,
      "passed_rate": 0.075,
      "length_ratio_mean": 0.3116,
      "sources_rate": 0.0,
      "loop_rate": 0.05,
      "repeats_max": 90,
      "label": "xcaliburbiz-Q4_K_M",
      "speed": {
        "seconds": 27.32,
        "tokens": 11686,
        "tokens_per_second": 427.8,
        "seconds_per_report": 0.683
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1011,
        "passed_rate": 0.075,
        "length_ratio_mean": 0.3116,
        "sources_rate": 0.0,
        "loop_rate": 0.05,
        "repeats_max": 90
      },
      "filter_enabled": true,
      "filtered_rate": 0.025,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0.1",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.844,
          "repeats": 0,
          "answer_head": "| Task | 0.1\n|  | `model-000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.053,
          "repeats": 0,
          "answer_head": "| Notebook | 0 files\n| `0 Models/0 Files` | 0"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.245,
          "repeats": 0,
          "answer_head": "| Task | 0 files, 0 references |\n|---|---|\n| `models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [],
          "length_ratio": 0.101,
          "repeats": 0,
          "answer_head": "The four models the viewer supports are `model.py`, `model_types.yaml`, `models`, and `models.yaml`."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "00",
            "000000000000000000000000"
          ],
          "length_ratio": 0.316,
          "repeats": 0,
          "answer_head": "`/workspace/xcalibur/xcalibur/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/000000000000000000000000/00000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1",
            "0.10",
            "000000",
            "4000"
          ],
          "length_ratio": 0.196,
          "repeats": 0,
          "answer_head": "| Task | Encoding, layout, and the schema\n| `0.0` | `models/000000.py` | The model, four files, one plan\n| `0.1` | `models/000000.py`, 4000 lines, four modules, four sections\n| `0.10` | `models/000000.py`, 4000, 4000, 4000, 0, 0, 0, 0, four sections, zero imports, zero references, zero\u7686\u6709\u7684 bare notebook"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.458,
          "repeats": 0,
          "answer_head": "`|skipped| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0|"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "00"
          ],
          "length_ratio": 0.179,
          "repeats": 0,
          "answer_head": "The `model/00` file carries four models, four models that the viewer cannot load, and\nfour the project therefore cannot afford."
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "0",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.488,
          "repeats": 0,
          "answer_head": "| Task | 0 files, 0 references |\n|---|---|\n| `models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.125,
          "numbers_missing": [
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.092,
          "repeats": 0,
          "answer_head": "[`databases/`][1] `..` [0]\n[0]: `databases/` directory, its own directory,"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.492,
          "repeats": 0,
          "answer_head": "[`databases/` 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.056,
          "numbers_missing": [
            "0.1",
            "0.3",
            "0.4",
            "1",
            "2",
            "29",
            "3",
            "32"
          ],
          "numbers_invented": [],
          "length_ratio": 0.292,
          "repeats": 0,
          "answer_head": "| Task| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0|"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [
            "0",
            "1",
            "2"
          ],
          "length_ratio": 0.778,
          "repeats": 0,
          "answer_head": "| Layout | 0| 1| 2| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0|"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.063,
          "repeats": 0,
          "answer_head": "... four models, each with a different architecture, a different input, a different output, and a different cache, and none of them is the same model, and none of them is the same workload."
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.10",
            "0.14"
          ],
          "numbers_invented": [
            "0",
            "00",
            "0000000000000000"
          ],
          "length_ratio": 0.047,
          "repeats": 0,
          "answer_head": ".. ra\n.. 00[0000000000000000]\n  00[0000000000000000\n    00[0000000000000000\n     00[0000000000000000\n      00[0\n      00[0\n        00[0\n          0\n            0\n              0\n```"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "0.003",
            "0.004"
          ],
          "length_ratio": 0.033,
          "repeats": 0,
          "answer_head": "The paper budget is 0.004, which matches the smallest cost in the paper, the 0.003 character-label overlap, which the paper declares."
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [
            "000000000000000000000000"
          ],
          "length_ratio": 0.09,
          "repeats": 1,
          "answer_head": "The paper `DeepSeek: A Deep Learning Framework for Automated Image Retrieval` by Li, Wang, and Zhang, whose `000000000000000000000000` paper the paper cites, declares four papers that are the four surviving ones, four papers that the paper does not, and the gap between the four and the paper's own claim."
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.757,
          "numbers_missing": [
            "09",
            "1958",
            "1976",
            "1980",
            "200",
            "2020",
            "2022",
            "2026"
          ],
          "numbers_invented": [
            "0",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34",
            "35"
          ],
          "length_ratio": 0.096,
          "repeats": 0,
          "answer_head": ".. ra: 0\n  0: 0\n  1: 0\n  2: 0\n  3: 0\n  4: 0\n  5: `| 0\n  6: `| 0\n  7: `| 0\n  8: `| 0\n  9: `| 0\n  10: `| 0\n  11: `| 0\n  12: `| 0\n  13: `| 0\n  14: `| 0\n  15: `| 0\n  16: `| 0\n  17: `| 0\n  18: `| 0\n  19: `| 0\n  20: `| 0\n  21: `| 0\n  22: `| 0\n  23: `| 0\n  24: `| 0\n  25: `| 0\n  26: `| 0\n  27: `| 0\n  28: `| 0\n  29: `| 0\n  30: `| 0\n  31: `| 0\n  32: `| 0\n  33: `| 0\n  34: `| 0\n  35: `| 0\n  36: `| 0\n.."
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "00"
          ],
          "length_ratio": 0.068,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `00_model_types.ipynb` | The bare Notebook\n| 0, `00` | `00`"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.576,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "00"
          ],
          "length_ratio": 0.057,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `00_model_types.ipynb` | The four types the model census reports, ordered\n| `00_model_types.ipynb`, 0| `\u5927\u7eb2/00`"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [],
          "length_ratio": 0.137,
          "repeats": 0,
          "answer_head": "The four models the viewer reports are the `model.py`, `model.pyc`, `model.pyd`, and `model.pyd`, four files, four modules, and four types."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [],
          "length_ratio": 0.007,
          "repeats": 0,
          "answer_head": "`model/` [],"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.369,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "40",
            "400"
          ],
          "length_ratio": 0.347,
          "repeats": 0,
          "answer_head": "| Task | Encoding | Notebook | Output, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 4"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.455,
          "repeats": 0,
          "answer_head": "`|skipped| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0|"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.721,
          "repeats": 0,
          "answer_head": "`000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "0.10",
            "2020.09",
            "2020.10"
          ],
          "length_ratio": 0.346,
          "repeats": 0,
          "answer_head": "| Paper | Paper type | What the paper reports\n| `pytorch-ai-2020.10.0` | 0.10.0 | The `torchvision` 0.10.0 release\n| `gpt4-2020.10.0` | 0.10.0 | The `gpt4` 0.10.0 release\n| `gpt5-2020.10.0` | 0.10.0 | The `gpt5` 0.10.0 release\n| `gpt4-2020.09.0` | 0.10.0 | The `gpt4` 0.10.0 release\n| `gpt5-2020.09.0` | 0.10.0 | `gpt5-0.10.0`"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.476,
          "repeats": 0,
          "answer_head": "`model/``, `0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "40"
          ],
          "length_ratio": 0.899,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `model/` | 40\n| `model/` | 40\n| `model/` | 40\n| `model/` | 40\n| `model/` | 40\n| `00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1",
            "000000"
          ],
          "length_ratio": 0.531,
          "repeats": 0,
          "answer_head": "| Paper | Type | What it writes\n| `model-architecture-bench` 0.1, `model-architecture` 0.1, `pydeck` 0.0, `deck` 0.0, `deck.yaml` 0.0\n| `model-architecture` 0.1, `model-architecture/0.1`, `pydeck`, `deck`, `deck.yaml\n| `model-architecture-000000` 0.0, `model-architecture/0.0`, `pydeck`, `0.0\n| `model-architecture-000000.yaml` 0, `0.0, pydeck, 0.0, yaml\n| `model-architecture-000000/` 0, `0.0, direc"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [],
          "length_ratio": 0.106,
          "repeats": 0,
          "answer_head": "The work maps four models, four files, zero imports, and zero citations."
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "0",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "1000",
            "10000",
            "400",
            "4000"
          ],
          "length_ratio": 0.251,
          "repeats": 0,
          "answer_head": "| File | 0 bytes | 400 bytes | 1000 bytes | 4000 bytes | 10,000 bytes | 10,000 bytes | 10,000 bytes |\n|--------|--------|---------|----------|----------|------------|------------|--|--|\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.105,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1.1",
            "165.2",
            "26",
            "3",
            "380.0",
            "43.4"
          ],
          "numbers_invented": [
            "0",
            "10",
            "100",
            "2",
            "20",
            "200",
            "32",
            "4"
          ],
          "length_ratio": 0.233,
          "repeats": 0,
          "answer_head": "| File | 0 bytes | 100, 200, 400, 800, 1, 2, 4, 8, 16, 32, 40, 50, 100, 10, 0, 100, 200, 400, 800, 1, 2, 4, 8, 10, 20, 40, 50, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0",
            "000",
            "1000",
            "1400",
            "1600",
            "400",
            "900"
          ],
          "length_ratio": 1.237,
          "repeats": 79,
          "answer_head": "| Asset | Size (bytes) | Size (kilo bytes) | Type | Where\n| `model/` | 1600 | 1400, 400, 400, 400 | 400, 400, 400, 400 | `models/`, `model/`, `models/000\n| `model/` | 1000 | 900, 400, 400, 400 | 400, 400, 400, 400 | `models/000`, 400, 400, 400, 400\n| `model/` | 1000 | 1000, 400, 400, 400 | `model/000`, 1000, 400, 400, 400\n| `model/` | 1000 | 1000, 400, 400, 400 | `model/000`, 1000, 400, 400, 400, "
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.00000000000000000",
            "0.1",
            "0.2",
            "0.3",
            "0.4",
            "0.5"
          ],
          "length_ratio": 0.487,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `model/` | 40\n| `model/` `0.0`, `0.00000000000000000, 0 files, 0, 0, `0/0/0`, bare\n| `model/` `0.1`, `0/0, 100, 40, `model/0.1/0`, 100, 40, 40, 40, 40\n| `model/` `0.2`, `0/0, 100, 40, `model/0.2/0`, 100, 40, 40, 40, 40, 40, 40\n| `model/` `0.3`, `0/0, 100, 40, `model/0.3/0`, 100, 40, 40, 40, 40, 40, 40, 40\n| `model/` `0.4`, `0/0, 100, 40, `model/0.4/0`, 100, 40, 40, 40, 40, 40, 40,"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [],
          "length_ratio": 0.237,
          "repeats": 90,
          "answer_head": "`glossary[py, py_0, py_1, py_2, py_3, py_4, py_5, py_6, py_7, py_8, py_9, py_10, py_11, py_12, py_13, py_14, py_15, py_16, py_17, py_18, py_19, py_20, py_21, py_22, py_23, py_24, py_25, py_26, py_27, py_28, py_29, py_30, py_31, py_32, py_33, py_34, py_35, py_36, py_37, py_38, py_39, py_40, py_41, py_42, py_43, py_44, py_45, py_46, py_47, py_48, py_49, py_50, py_51, py_52, py_53, py_54, py_55, py_5"
        },
        {
          "section": "Method and sources",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "0.0",
            "400"
          ],
          "length_ratio": 0.061,
          "repeats": 0,
          "answer_head": "| Task | 0.0\n|  `model.py`, `models/` | The model, four models, four files\n|  `index.html`, `..` | The bare html, four sections\n|  `index.rst`, 400-- | 400 words"
        }
      ],
      "gpu_offload": true
    },
    "xcaliburbiz-Q5_K_M": {
      "count": 40,
      "style_clean_rate": 0.925,
      "numbers_kept_mean": 0.1423,
      "passed_rate": 0.125,
      "length_ratio_mean": 0.4798,
      "sources_rate": 0.025,
      "loop_rate": 0.125,
      "repeats_max": 161,
      "label": "xcaliburbiz-Q5_K_M",
      "speed": {
        "seconds": 40.16,
        "tokens": 16507,
        "tokens_per_second": 411.0,
        "seconds_per_report": 1.004
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1423,
        "passed_rate": 0.125,
        "length_ratio_mean": 0.4793,
        "sources_rate": 0.025,
        "loop_rate": 0.125,
        "repeats_max": 161
      },
      "filter_enabled": true,
      "filtered_rate": 0.075,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "0000",
            "0001",
            "0002",
            "0003",
            "0004",
            "1000"
          ],
          "length_ratio": 0.823,
          "repeats": 1,
          "answer_head": "| File | 0000\n| `models/0000.py` | The model, four modules, four files\n| `models/0001.py` | The model, 1000\n| `models/0002.py` | The architecture, four modules, four\n| `models/0003.py` | The type, four types, four\n| `models/0004.py` | The benchmark, four, `models/bench.py`, four\n| `models/0000.py`, `0000`, 1000, `py:module`, 1000\n| `models/0001.py`, `0000.py:0`, 1000, 1000\n| `models/0002.py`, `000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0000",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "length_ratio": 0.985,
          "repeats": 0,
          "answer_head": "| File | 0000\n| `models/0000.yaml` | The model, four files, four types\n| `0000.yaml` | `models/0000.yaml\n```yaml\nmodel:\n  name: deepseek\n  type: pytorch\n  sources:\n  0: `models/0000.py\n    0: `..:/..:/..  40\n    1: `..:/..:/..  40\n    2: `..:/..:/..  40\n    3: `..:/..:/..  40\n    4: `..:/..:/..  40\n    5: `..:/..:/..  40\n    6: `..:/..:/..  40\n    7: `..:/..:/..  40\n    8: `..:/..:/..  40\n    9: `"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "400"
          ],
          "length_ratio": 0.235,
          "repeats": 0,
          "answer_head": "| File | 400\n| `models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "0.0"
          ],
          "length_ratio": 0.505,
          "repeats": 0,
          "answer_head": "| Families | 0.0\n|------|------------\n| DeepSeek | 0.0\n| DeepSeekMoe | 0.0\n| DeepSeekEncoderDecoder | 0.0\n| GShard, GShardMoe, GShardEncoderDecoder | 0.0\n| Mixtral, MixtralMoe, MixtralEncoderDecoder | 0.0\n| LLaMA, LLaMAMoe, LLaMAEncoderDecoder | 0.0\n| WuDao, WuDaoMoe, WuDaoEncoderDecoder | 0.0\n| HuggingFace, HuggingFaceMoe, HuggingFaceEncoderDecoder | 0.0\n| PyTorch, PyTorchMoe, PyTorchEncoderDecod"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.326,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|------------\n| models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "1",
            "2"
          ],
          "length_ratio": 0.335,
          "repeats": 0,
          "answer_head": "| Layout | 0| 1| 2| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0| 0|"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.474,
          "repeats": 0,
          "answer_head": "| File | 100\n|--------|------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "000000",
            "000001",
            "000002",
            "000003",
            "000004",
            "000005",
            "000006",
            "000007"
          ],
          "length_ratio": 1.252,
          "repeats": 26,
          "answer_head": "| File | 100\n|----------------|------------\n| models/000000.yaml | 100\n| models/000001.yaml | 100\n| models/000002.yaml | 100\n| models/000003.yaml | 100\n| models/000004.yaml | 100\n| models/000005.yaml | 100\n| models/000006.yaml | 100\n| models/000007.yaml | 100\n| models/000008.yaml | 100\n| models/000009.yaml | 100\n| models/000010.yaml | 100\n| models/000011.yaml | 100\n| models/000012.yaml | 100\n| mod"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "100",
            "40",
            "400"
          ],
          "length_ratio": 0.569,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|------------\n| models/abalone.py | The model, four modules, four files\n| models/abalone/abalone.py | The model, 400 lines\n| models/abalone/abalone.py\n  [100, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 400, 4"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.25,
          "numbers_missing": [
            "1",
            "100000",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 2.164,
          "repeats": 3,
          "answer_head": "```cpp\n/* `db_crypto.h` */\n\n// The encryption module, which encrypts and decrypts data\n// The module is a header, so it cannot be a library\n// The module is the public interface\n#include \"db_crypto.h\"\n\n// The encryption algorithm, which is a public constant\nconst auto ENCRYPTION = \"AES-256-CBC\";\n\n// The encryption key, which is a private constant\nconst auto KEY = \"my_encryption_key\";\n\n// The encry"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "length_ratio": 0.628,
          "repeats": 0,
          "answer_head": ".. hfile: db_transfer.h\n.. autogen\n.. autogen.sh\n[gen]\ngen\n.. py\n.. py\n  HFILE=db_transfer.h\n  HCL=\n    src/\n      db_transfer.c\n    src/crypto/\n      db_crypto.h\n    src/params/params.json\n    src/platforms/`\n    src/platforms/`+0\n  H\n  0\n  1\n  2\n  3\n  4\n  5\n  6\n  7\n  8\n  9\n  10\n  11\n  12\n  13\n  14\n  15\n  16\n  17\n  18\n  19\n  20\n  21\n  22\n  23\n  24\n  25\n  26\n  27\n  28\n  29\n  30\n  31\n  32\n  33\n  34"
        },
        {
          "section": "10 Slides",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.056,
          "numbers_missing": [
            "0",
            "0.1",
            "0.3",
            "0.4",
            "2",
            "29",
            "3",
            "32"
          ],
          "numbers_invented": [
            "00",
            "04",
            "10",
            "1000",
            "20"
          ],
          "length_ratio": 0.141,
          "repeats": 0,
          "answer_head": "|  |  |\n|---|---|\n| 1. `00_overview` `00-10`  | 00:00--00:10 00:00 `\u819c\u72b6\u9aa8\u67b6` 00:04 `\u819c\u72b6\u9aa8\u67b6, 1000+ tracks, 1000+ annotations, 1000+ cell types, 1000+ cell annotations, 1000+ cell types, 1000+ cell annotations, 1000+ cell annotations, 00:10--00:20 00:20 `|"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [
            "0",
            "1"
          ],
          "length_ratio": 0.796,
          "repeats": 0,
          "answer_head": "| Name | Weight, ordering, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "0.0",
            "0.000000"
          ],
          "length_ratio": 0.269,
          "repeats": 60,
          "answer_head": ".. numpy\n0.000000\n.. py\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0.0\n.. cmp\n0."
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [
            "0.000",
            "0.0001"
          ],
          "length_ratio": 0.144,
          "repeats": 0,
          "answer_head": "..., `0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0."
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.245,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.10",
            "0.14"
          ],
          "numbers_invented": [
            "0",
            "100",
            "101",
            "102",
            "103",
            "104",
            "105",
            "106"
          ],
          "length_ratio": 0.138,
          "repeats": 0,
          "answer_head": ".. py\n.. autoplot\n[1]\n  1. `axs[0].plot([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 9"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "0",
            "000",
            "0000"
          ],
          "length_ratio": 0.245,
          "repeats": 89,
          "answer_head": ".. py\n.. autopat\n0000\n.. py\n000\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n.. py\n0\n"
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [],
          "length_ratio": 0.029,
          "repeats": 0,
          "answer_head": "The paper has four sections, the conclusions are the same as the paper, the\n`conclusions` section."
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "09",
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.125,
          "repeats": 0,
          "answer_head": ".. [A0000000000000000000000000000000000000000, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0.1",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.473,
          "repeats": 0,
          "answer_head": "| Notebook | 0.1\n| `model-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "0.4",
            "000",
            "100",
            "400"
          ],
          "length_ratio": 0.769,
          "repeats": 0,
          "answer_head": "| Notebook | 0.4\n| `py_maker` dependency, four files, one layout, 100\n  * `model.py`, the model\n  * `model.py`, the layout\n  * `model.py`, the manifest\n  * `model.py`, the `render` command\n| Notebook, 0.4\n  0.0\n  1 file, `model.py`, 400\n,\n[py_maker, 0.4, 0, 400, `model/`, `model/`, `model.py`, `000`, `model.py`, 400, 0, 0, `model/`, 400, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, "
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "00"
          ],
          "length_ratio": 0.236,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `00_0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.143,
          "numbers_missing": [
            "0.9",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "20",
            "20.4",
            "23"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "000000000000000000",
            "000000000000000001",
            "3",
            "4"
          ],
          "length_ratio": 0.619,
          "repeats": 1,
          "answer_head": "| Families | 0.0\n| ..., 0.0\n| 1. The model, four models, four files\n| ..| `models/000000000000000000.py\n| ..| `models/000000000000000001.py\n| | 0 files\n| 2. The benchmark, four files\n| ..| `benchmarks/000000000000000000.py\n| | 0 files\n| 3. The ordering, four files\n| ..| `order/000000000000000000.py\n| | 0 files\n| 4. The layout, four files\n| ..| `layout/000000000000000000.py\n| | 0 files\n| 0, four, z"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "000000",
            "000001",
            "000002",
            "000003",
            "000004",
            "000005"
          ],
          "length_ratio": 0.289,
          "repeats": 1,
          "answer_head": "| Notebook | 0.0\n| `model/000000.py` | The model, four files\n| `model/000001.py` | The model, 400 lines\n| `model/000002.py` | The architecture, four imports\n| `model/000003.py` | The model, four imports\n| `model/000004.py` | `model.py`, 100 lines\n| `model/000005.py` | `model.py`, 100 lines\n| `model/000006.py` | `model.py`, 100 lines\n| `000000.py`, `000001.py`, `000002.py`, `000003.py`, `000004.py`"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.374,
          "repeats": 0,
          "answer_head": ".. py\n.. autoplot\n[plot]\n[plot]\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.353,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|------------\n| models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0.0",
            "000000000000",
            "400"
          ],
          "length_ratio": 0.477,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model-architectures` `000000000000.ipynb` `py\n| 0.0\n| 400\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0.0\n| 0."
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.761,
          "repeats": 0,
          "answer_head": "| File | 100\n|--------|----------\n| models/000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "0",
            "0.1",
            "00000000",
            "000000000000000000"
          ],
          "length_ratio": 0.601,
          "repeats": 0,
          "answer_head": "| Type | 0.1\n| `models/000000000000000000_000000000000000000.yaml` | 0\n| `models/000000000000000000/000000000000000000.yaml` | 0\n| `models/000000000000000000/000000000000000000/000000000000000000.yaml` | 0\n| `models/000000000000000000/000000000000000000/000000000000000000/000000000000000000.yaml` | 0\n| `models/000000000000000000/000000000000000000/000000000000000000/000000000000000000/000000000000"
        },
        {
          "section": "What the work concerns",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.48,
          "repeats": 0,
          "answer_head": "The work--`model-000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000000000,000000"
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.84,
          "repeats": 0,
          "answer_head": ".. py\n..\u4ef9\n0\n..  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  "
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.0"
          ],
          "length_ratio": 0.469,
          "repeats": 1,
          "answer_head": "| File | 0.0\n| `models/agent_wrkn/agent.py` 0\n| `models/encoder_decoder/encoder_decoder.py` 0\n| `models/encoder_decoder/encoder_decoder_mechanism.py` 0\n| `models/encoder_decoder/encoder_decoder_mechanism_types.py` 0\n| `models/encoder_decoder/encoder_decoder_types.yaml` 0\n| `models/encoder_decoder/encoder_decoder_types.yaml.example` 0\n| `models/encoder_decoder/encoder_decoder_types.yaml`, 0\n| `..`"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [],
          "length_ratio": 0.004,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "100"
          ],
          "length_ratio": 0.558,
          "repeats": 39,
          "answer_head": "| File | 100\n|----------------|------------\n| models/abalone/abalone.py | 100\n| models/abalone/abalone.h5 | 100\n| models/abalone/abalone.c | 100\n| models/abalone/abalone.py\n| models/abalone/abalone.py\n| models/abalone/abalone.py\n| models/abalone/abalone.h\n| models/abalone/abalone.h\n| models/abalone/abalone.h\n| models/abalone/abalone.c\n| models/abalone/abalone\n| models/abalone/abalone\n| models/abal"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [
            "10",
            "100",
            "20",
            "30",
            "40"
          ],
          "length_ratio": 0.344,
          "repeats": 2,
          "answer_head": "| File | 100\n|----------------|------------\n| models/abalone/abalone.py | 40\n| models/abalone/abalone.h5 | 30\n| models/abalone/abalone.c | 20\n| models/abalone/abalone.cub | 10\n| models/abalone/abalone.h | 10\n| models/abalone/abalone.c | 10\n| models/abalone/abalone.cub.h5 | 10\n| models/abalone/abalone.cub | 10\n| models/abalone/abalone.cub.c | 10\n| models/abalone/abalone.cub.cub.h | 10\n| models/abal"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0.0",
            "1.1",
            "100",
            "4000"
          ],
          "length_ratio": 0.442,
          "repeats": 0,
          "answer_head": "| File | 4000 bytes | 0.0\n|--------|----------| 0.0\n| models/abalone/abalone.py | 1.1 MB | 100\n| models/abalone/abalone.h5 | 100 bytes | 100\n| models/abalone/abalone.c | 100 bytes | 100\n| models/abalone/abalone.pytorch | 100 bytes | 100\n| models/abalone/abalone.py, 4000| 4000| 4000\n\u6837\u7684\u5927\u5c0f, the file, the name, the 4000 bytes, the 0.0, the type, the file,"
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0.0",
            "10",
            "100",
            "400"
          ],
          "length_ratio": 0.202,
          "repeats": 0,
          "answer_head": "| File | 0.0\n| `models/agent_wrkn/agent.py` | The model, four modules, four imports, one section\n| `models/agent_wrkn/agent.py:10` | `model.py`, 100 lines\n| `models/agent_wrkn/agent.py:100` | `..model..`, 100 lines\n| `..model..`, 400 lines\n| `| 400\n|"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.189,
          "repeats": 0,
          "answer_head": "| term | description |\n|------|-----------|\n| `model.py`, `models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "00000",
            "1000"
          ],
          "length_ratio": 0.525,
          "repeats": 161,
          "answer_head": "| Notebook | 0.0\n| `0.0/0` | bare\n| 1000, `00000`, 0, four files, four lines\n| 1000, `00000.ipynb`, 0, four files, four lines\n| 1000, `model-00000.ipynb`, 0, four files, four lines\n| 1000, `model-00000.yaml`, 0, four lines, four types\n| 1000, `|`, four files, four lines, four types, four lines, four\n| 0, `model-00000`, four files, four lines, four types, four, four, four, four, four, four, four, f"
        }
      ],
      "gpu_offload": true
    },
    "xcaliburbiz-Q6_K": {
      "count": 40,
      "style_clean_rate": 1.0,
      "numbers_kept_mean": 0.1203,
      "passed_rate": 0.05,
      "length_ratio_mean": 0.4729,
      "sources_rate": 0.0,
      "loop_rate": 0.125,
      "repeats_max": 152,
      "label": "xcaliburbiz-Q6_K",
      "speed": {
        "seconds": 40.06,
        "tokens": 15777,
        "tokens_per_second": 393.8,
        "seconds_per_report": 1.002
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1203,
        "passed_rate": 0.05,
        "length_ratio_mean": 0.4729,
        "sources_rate": 0.0,
        "loop_rate": 0.125,
        "repeats_max": 152
      },
      "filter_enabled": true,
      "filtered_rate": 0.0,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "000000000000"
          ],
          "length_ratio": 0.848,
          "repeats": 0,
          "answer_head": "| File | 0\n| ```\n| models/000000000000\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n|"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.617,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `model-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "4000"
          ],
          "length_ratio": 0.356,
          "repeats": 0,
          "answer_head": "| File | 4000+ rows\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `models/`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n| `..`, 0\n"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "16"
          ],
          "length_ratio": 0.186,
          "repeats": 0,
          "answer_head": "| Model type | 0.0\n|----------|------------\n| Encoder | 0.0\n| Decoder | 0.0\n| EncoderDecoder | 0.0\n| GShard | 0.0\n| DeepSeek | 0.0\n| Mixtral | 0.0\n| LLaMA | 0.0\n| WuDao | 0.0\n| 0 | 16."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.324,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0.0",
            "10",
            "100",
            "1000",
            "10000",
            "100000",
            "1000000",
            "10000000"
          ],
          "length_ratio": 0.334,
          "repeats": 0,
          "answer_head": "| Model | 0.0, 100, 1000, 10, 20, 40, 1000, 4000, 10000, 40000, 100, 100000, 40, 400, 4000, 400000, 1000, 10000, 1000000, 10000000, 100000000, 1000000000, 1000000, 100000, 10000000, 100000000, 10000000, 100000000, 10000000, 100000000, 100000000, 1000000000, 100, 10, 100, 1000, 10000, 1000000, 100000000, 1000000000, 10000000000, 10, 10, 10, 100, 100000, 1000000, 100000000, 10000000000, 100, 100, 10"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.484,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "000000",
            "000001",
            "000002",
            "000003",
            "000004",
            "000005",
            "000006",
            "000007"
          ],
          "length_ratio": 1.249,
          "repeats": 26,
          "answer_head": "| File | 100\n|----------------|----------\n| models/000000.yaml | 400\n| models/000001.yaml | 400\n| models/000002.yaml | 400\n| models/000003.yaml | 400\n| models/000004.yaml | 400\n| models/000005.yaml | 400\n| models/000006.yaml | 400\n| models/000007.yaml | 400\n| models/000008.yaml | 400\n| models/000009.yaml | 400\n| models/000010.yaml | 400\n| models/000011.yaml | 400\n| models/000012.yaml | 400\n| model"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.048,
          "repeats": 0,
          "answer_head": "| Method | 0 files, 0 references |\n|----------------||"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.125,
          "numbers_missing": [
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.294,
          "repeats": 0,
          "answer_head": ".. h\n.. autogen\nauto_crypto.h\n.. autogen.sh\n[..]\n.. 0\n.. 0\n    0\n        0\n            1\n                1\n                    1\n                        1\n                            1\n```\n\nThe file already follows the requested format."
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "000",
            "2"
          ],
          "length_ratio": 0.667,
          "repeats": 47,
          "answer_head": ".. haddock\n.. autodoc\n.. autodoclevels 2\n.. autodoc[\n..  H001 `db_transfer.h`\n.. H002 `db_transfer_api.h\n.. H003 `db_transfer_api.h\n.. H004 000\n.. H005 000\n.. H006 000\n.. H007 0\n.. H008 0\n.. H009 0\n.. H010 0\n.. H011 0\n.. H012 0\n.. H013 0\n.. H014 0\n.. H015 0\n.. H016 0\n.. H017 0\n.. H018 0\n.. H019 0\n.. H020 0\n.. H021 0\n.. H022 0\n.. H023 0\n.. H024 0\n.. H025 0\n.. H026 0\n.. H027 0\n.. H028 0\n.. H029 0\n.."
        },
        {
          "section": "10 Slides",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.111,
          "numbers_missing": [
            "0.3",
            "0.4",
            "1",
            "2",
            "29",
            "3",
            "32",
            "4"
          ],
          "numbers_invented": [
            "0.2",
            "10"
          ],
          "length_ratio": 0.451,
          "repeats": 3,
          "answer_head": "| Heading | 0 | 10 |\n|-------|---|---|\n| 0.1 `data/` | 0 | `data/*.json` `data/*.xml` `data/*.kml` `data/*.kml.gz` `data/*.kml.bak` `data/*.kml` `data/*.kml.gz` `data/*.kml.bak` `data/cell_types` `data/cell_types.json` `data/cell_types.csv` `data/cell_types.h5` `data/cell_types.h5.gz` `data/cell_types.h5` `data/cell_types.h5.gz` `data/cell_types.h5.bak` `data/cell_types.h5` `data/cell_types.h5.gz`"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [
            "0",
            "0.1",
            "000000000000000000",
            "1",
            "4",
            "40"
          ],
          "length_ratio": 0.504,
          "repeats": 0,
          "answer_head": "| Name | Weight | Haddock | Haddock 0\n|------|--------|----------|-------\n| biohub/cell-tracking/0.1/analysis/track.py | 0.1 | `cell_tracking`, `analysis`, `0.1` | 0\n| `000000000000000000_000000000000000000.py` | 0 | `000000000000000000.py`, 0 files | 0\n| `biohub/cell-tracking/0.1/analysis/track.py`, 40| 0, 0, 0, 1| 4\n| `| 0, 0, 0, 1`"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.143,
          "numbers_missing": [
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "0",
            "1000"
          ],
          "length_ratio": 0.216,
          "repeats": 0,
          "answer_head": ".. [1]\n.. 1000 lines\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0\n.. .. 0 0"
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.094,
          "numbers_missing": [
            "0.0028",
            "0.025",
            "0.03",
            "0.075",
            "0.14",
            "0.15",
            "0.20",
            "0.25"
          ],
          "numbers_invented": [
            "0",
            "0.00000000000000000000",
            "0.0000000000000000000000",
            "0.00000001",
            "0.0001",
            "2",
            "4"
          ],
          "length_ratio": 0.14,
          "repeats": 0,
          "answer_head": ".. [1]\n.. 0\n  The 0.10, 0.01, 0.003, 0.0001, 0.00000001, 0.00000000000000000000, 0.0000000000000000000000\n.. [2]\n.. 0\n  0/0 `0.10`, 0\n.. [3]\n.. 0\n  0/0 `0.00000000000000000000`, 0\n.. [4]\n.. 0\n  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.10",
            "100"
          ],
          "length_ratio": 0.024,
          "repeats": 0,
          "answer_head": ".. [0]\n.. 0\n  The budget is 0, the work is 100, the\n.. 0.10 0.0, the\n.. 0.0 0, the\n.. 0 0, 100."
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "16",
            "17",
            "18",
            "19",
            "2.23"
          ],
          "numbers_invented": [
            "0",
            "0.0001",
            "0.0002",
            "0.0005",
            "0.001",
            "0.002",
            "0.005",
            "0.01"
          ],
          "length_ratio": 0.213,
          "repeats": 0,
          "answer_head": ".. py\n.. autoplot\n.. plot\n    x = [0.01, 0.02, 0.05, 0.10, 0.20, 0.50, 1.00, 10.00]\n    y = [0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.005, 0.01, 0.02]\n    z = [0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.005, 0.01, 0.02]\n    labels = [f\"{x:.4f}\" for x in zip(*z)]\n    title = r\"Cost per user, 10000 users, 10000 tasks, 10000 users, 10000\n    tasks, local, 0.0001, 0.0001, 0.0001, 0.0002, 0.0002, 0.0005, 0"
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [],
          "length_ratio": 0.029,
          "repeats": 0,
          "answer_head": "The paper has four sections, the conclusions are the same as the paper, the\n`conclusions` section."
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.054,
          "numbers_missing": [
            "09",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.128,
          "repeats": 0,
          "answer_head": ".. [1]\n.. 0\n  The 0 lines\u7686\n.. [2]\n.. 0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n  0\n "
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0.1",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.481,
          "repeats": 0,
          "answer_head": "| File | 0.1\n| `models/agent_wizard.py`, `0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0.4",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.581,
          "repeats": 0,
          "answer_head": "| Notebook | 0.4\n| `model-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "0.4"
          ],
          "length_ratio": 0.064,
          "repeats": 0,
          "answer_head": "| Notebook | 0.4\n| `model_types` `0.4`\u684c\u9762| `model_types.py`, four types, four files\n| `model_types.py`| `py:module model_types`, 0 files\n| `|`"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.286,
          "numbers_missing": [
            "0.9",
            "1.8",
            "10.6",
            "17.7",
            "20",
            "20.4",
            "23",
            "4.4"
          ],
          "numbers_invented": [
            "0",
            "10",
            "100",
            "11",
            "13",
            "14",
            "15",
            "16"
          ],
          "length_ratio": 1.374,
          "repeats": 6,
          "answer_head": "| File | 100\n|----------------|----------\n| models/abalone.py | The model, four modules, four models\n| models/abalone/abalone.h | The module, four files\n| models/abalone/abalone.h:4 | `abalone.h`, four imports\n| models/abalone/abalone.c | The source, four source files\n| models/abalone/abalone.c:1 | `abalone.c`, one source\n| models/abalone/abalone.c:2 | `abalone.c`, bare\n| models/abalone/abalone.c:"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.326,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "00000",
            "000000000000",
            "000000000001",
            "000000000002",
            "000000000003",
            "000000000004"
          ],
          "length_ratio": 0.504,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model-000000000000` | `model-000000000000.yaml`, 0\n| `model-000000000001` | `model-000000000001.ipynb`, 0\n| `model-000000000002` | `model-000000000002.ipynb`, 0\n| `model-000000000003` | `model-000000000003.ipynb`, 0\n| `model-000000000004` | `model-000000000004.ipynb`, 0\n| `model-000000000005` | `model-000000000005.ipynb`, 0\n| `model-000000000006` | `model-000000000006.ipynb`, 0"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0.0",
            "0000",
            "0001",
            "0002",
            "0003",
            "0004",
            "0005",
            "0006"
          ],
          "length_ratio": 0.541,
          "repeats": 19,
          "answer_head": "| Notebook | 0.0\n| `model.py` | 0.0\n| `models/0000_notebook0000.ipynb` | 0.0\n| `models/0001_model.py` | 0.0\n| `models/0002_type0000.py` | 0.0\n| `models/0003_type0001.py` | 0.0\n| `models/0004_type0002.py` | 0.0\n| `models/0005_type0000.py` | 0.0\n| `models/0006_type0000.py` | 0.0\n| `models/0007_type0000.py` | 0.0\n| `models/0008_type0000.py` | 0.0\n| `models/0009_type0000.py` | 0.0\n| `models/0010_type0"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "000000000000"
          ],
          "length_ratio": 0.361,
          "repeats": 0,
          "answer_head": "| Notebook | Models, types, and the workload\n| `model-architectures` | Four models, four types, four\n  | `models/000000000000_000000000000.ipynb`,\n  | `pytorch/000000000000_000000000000.ipynb`, `tensorflow/000000000000_000000000000.ipynb`, `kaggle/000000000000_000000000000.ipynb`, `gym/000000000000_000000000000.ipynb`\n| `models/000000000000_000000000000.yaml`,\n| `models/000000000000_000000000000.y"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.77,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "1",
            "2020"
          ],
          "length_ratio": 0.708,
          "repeats": 0,
          "answer_head": "| Paper | Type | What the paper says, and the citation |\n|:---|:---|:---|\n| `DeepSeek: A Deep Learning Framework for Retrieval in Unstructured Text` [1] | paper | `Zhang, X., Li, Y., Li, Y., Li, Y., Li, Y., ...`, 2020..| [1]| `search/0000000000000000000000000000000000000000`, `search/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [],
          "length_ratio": 0.006,
          "repeats": 0,
          "answer_head": "[..]\n.."
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0.0",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.879,
          "repeats": 0,
          "answer_head": "| File | 0.0\n|--------|----------\n| models/000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 1.273,
          "repeats": 1,
          "answer_head": "| Notebook | Type | What it writes\n| `model-architecture` `py` | a bare notebook\n| `model-architecture.yaml` | a `py:module` file\n| `model-architecture.ipynb` | a notebook\n| `model-architecture.ipynb;model-architecture.yaml` | a notebook with a named workspace\n| `model-architecture.ipynb;model-architecture.yaml, workspace` | a notebook with a named workspace, the same file\n| `model-architecture.ip"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [],
          "length_ratio": 1.524,
          "repeats": 152,
          "answer_head": ".. code:: rst\n.. rst-line\n.. autopage\n.. autoguard\n..\u4ef9\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. ra\n.. "
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1",
            "4000"
          ],
          "length_ratio": 0.315,
          "repeats": 1,
          "answer_head": "| File | 0.0\n|----------------| -----------\n| models/abalone.py | 0\n| models/abalone.h5 | 0\n| models/abalone.h5.meta | 0\n| models/abalone.h5.model | 0\n| models/abalone.h5.model.h5 | 0\n| models/abalone.h5.model.py | 0\n| models/abalone.h5.rst | 0\n| models/abalone.h5.rst.html | 0\n| models/abalone.h5.rst.yaml | 0\n| models/abalone.h5.rst.yaml.html | 0\n| models/abalone.html, models/ | 1\n| 4000\n| 4000\u8fd8\u662f\u8981"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "4000"
          ],
          "length_ratio": 0.244,
          "repeats": 0,
          "answer_head": "| File | 4000\n|----------------|----------\n| models/000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1.0",
            "1.1",
            "100",
            "4000"
          ],
          "length_ratio": 0.811,
          "repeats": 0,
          "answer_head": "| File | 4000 bytes | 0.0\n|----------------||--\n| models/abalone.py | 1.0 MB (4000 bytes) | 0.0\n| models/abalone.h | 100 bytes | 0.0\n| models/abalone.c | 100 bytes | 0.0\n| models/abalone.py, models/abalone.h, models/abalone.c | 1.1 MB | 0.0\n| 4000 bytes, 100, 100, 0 | 0.0, 0, 100, 100, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,"
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "100"
          ],
          "length_ratio": 0.645,
          "repeats": 1,
          "answer_head": "| File | 0.0\n|----------------| -----------\n| models/abalone.py | 0\n| models/abalone.h5 | 0\n| models/abalone.h5.meta | 0\n| models/abalone.h5.model | 0\n| models/abalone.h5.model.h5 | 0\n| models/abalone.h5.model.py | 0\n| models/abalone.h5.rst | 0\n| models/abalone.h5.rst.authors | 0\n| models/abalone.h5.rst.0 | 0\n| models/abalone.h5.rst.yaml | 0\n| models/abalone.yaml | 0\n| models/abalone.yaml.bak | 0\n"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1"
          ],
          "length_ratio": 0.137,
          "repeats": 0,
          "answer_head": "| Field | Description |\n|:---|:|\n| `model`, 0.1 `model.py`, 0| The model, the file, the name, the type, 0 bytes |\n| `0.0_`, 0, 0| The smallest file, zero bytes |\n| `0_0_`, 0, 0| The zero byte, 0 bytes |\n| `0_0`, 4, 0| `model`, 4 bytes, 0 files, 0 imports, 0 methods, 0\nplatform: h5\n\u6837\u7684, and the platform the work\n\u89c8, 0\n| name | description |\n|:---|:|\n| `model.py`, 0, 0| The bare, zero byte model, `mod"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "0",
            "00"
          ],
          "length_ratio": 0.258,
          "repeats": 0,
          "answer_head": "| Notebook | 0 files, 0 imports |\n|---------|----------------|\n| `00_model_types.ipynb`  | 0 files, 0 imports, 0 references, 0 notebook, 0 notebook imports, 0 notebook section, 0 notebook cell, 0 notebook block, 0 notebook section, 0 notebook, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, "
        }
      ],
      "gpu_offload": true
    },
    "xcaliburbiz-Q8_0": {
      "count": 40,
      "style_clean_rate": 0.975,
      "numbers_kept_mean": 0.1098,
      "passed_rate": 0.075,
      "length_ratio_mean": 0.3507,
      "sources_rate": 0.0,
      "loop_rate": 0.05,
      "repeats_max": 13,
      "label": "xcaliburbiz-Q8_0",
      "speed": {
        "seconds": 37.0,
        "tokens": 14090,
        "tokens_per_second": 380.8,
        "seconds_per_report": 0.925
      },
      "shipped": {
        "count": 40,
        "style_clean_rate": 1.0,
        "numbers_kept_mean": 0.1098,
        "passed_rate": 0.075,
        "length_ratio_mean": 0.3507,
        "sources_rate": 0.0,
        "loop_rate": 0.05,
        "repeats_max": 13
      },
      "filter_enabled": true,
      "filtered_rate": 0.025,
      "filter_residual_rate": 0.0,
      "examples": [
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "113",
            "2",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.844,
          "repeats": 0,
          "answer_head": "| File | 0\n| `models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.617,
          "repeats": 0,
          "answer_head": "| Notebook | 0\n| `model-0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.243,
          "repeats": 0,
          "answer_head": "| File | 0 rows |\n|------|--------|\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.429,
          "numbers_missing": [
            "0.9",
            "1.8",
            "10.6",
            "17.7",
            "20.4",
            "4.4",
            "43.4",
            "49"
          ],
          "numbers_invented": [
            "0",
            "10",
            "11",
            "13",
            "14",
            "15",
            "16",
            "17"
          ],
          "length_ratio": 0.163,
          "repeats": 0,
          "answer_head": "| Model type | 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 40, 41, 42, 43 |\n|..."
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12.4",
            "12.5",
            "1200.0",
            "13.0"
          ],
          "numbers_invented": [
            "0",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.321,
          "repeats": 0,
          "answer_head": "| File | 0 rows |\n|------|--------|\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "1",
            "10",
            "100",
            "1000",
            "10000",
            "100000"
          ],
          "length_ratio": 0.334,
          "repeats": 0,
          "answer_head": "| Model | 0.0, 100, 1000, 10, 0, 1, 2, 4, 10, 100, 1000, 10000, 100, 100000, 1000, 1000000, 10000, 10000000, 10, 1, 2, 4, 10, 100, 1000, 10, 10, 100, 1000, 100000, 100, 100, 10, 100, 1000, 100, 10000, 100, 100, 100000, 100, 10, 1000000, 10, 100, 10000000, 100, 100, 100, 100, 1000000, 100, 100, 100, 10, 10, 100000000, 10, 100, 1000000000, 100, 100, 100, 100, 100000, 100, 100, 100, 100, 10, 100, 100"
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.484,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.769,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "1",
            "10",
            "2",
            "2026",
            "3",
            "4"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.048,
          "repeats": 0,
          "answer_head": "| Method | 0 files, 0 references |\n|----------------||"
        },
        {
          "section": "2. Encryption Module (`db_crypto.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "100000",
            "16",
            "256",
            "32",
            "4",
            "8",
            "88"
          ],
          "numbers_invented": [],
          "length_ratio": 0.027,
          "repeats": 0,
          "answer_head": ".. hfile:`db_crypto.h`"
        },
        {
          "section": "3. Transfer API (`db_transfer.h/c`)",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [],
          "length_ratio": 0.023,
          "repeats": 0,
          "answer_head": ".. hfile:`db_transfer.h`"
        },
        {
          "section": "10 Slides",
          "violations": [
            "double_hyphen"
          ],
          "shipped_violations": [],
          "numbers_kept": 0.222,
          "numbers_missing": [
            "0",
            "0.1",
            "0.3",
            "0.4",
            "3",
            "4",
            "453",
            "5"
          ],
          "numbers_invented": [
            "00",
            "01",
            "02",
            "03",
            "04",
            "05",
            "06",
            "07"
          ],
          "length_ratio": 0.313,
          "repeats": 0,
          "answer_head": "|  |  |\n|---|---|\n| 1. `00_overview` `00-01` `00-02` `00-03` | 00:00--00:04, 00:05, 00:06, 00:07, 00:08, 00:09, 00:10, 00:11, 00:12, 00:13, 00:14, 00:15, 00:16, 00:17, 00:18, 00:19, 00:20, 00:21, 00:22, 00:23, 00:24, 00:25, 00:26, 00:27, 00:28, 00:29, 00:30, 00:31, 00:32, 00:33, 00:34, 00:35, 00:36, 00:37, 00:38, 00:39, 00:40, 00:41, 00:42, 00:43, 00:44, 00:45, 00:46, 00:47, 00:48, 00:49, 00:50, 0"
        },
        {
          "section": "Library Selection Matrix",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "3"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0.1",
            "001",
            "002"
          ],
          "length_ratio": 0.73,
          "repeats": 0,
          "answer_head": "| Name | Weight | Haddock | Haddock 0\n|------|--------|----------|-----------\n| `biohub/cell-tracking` | 0.1 | `cell_tracking.py`, `cell_tracking.ipynb`, `cell_tracking.rst` | 0\n| `biohub/cell-tracking/visualization` | 0.0 | `cell_tracking.ipynb:001`, `cell_tracking.ipynb:002`, `cell_tracking.rst:001`, `cell_tracking.rst:002` | 0\n| `biohub/cell-tracking/visualization/001` | 0.0 | `001.ipynb`, `001"
        },
        {
          "section": "Being one of the only four survivors left in the age of what they branded \"super-intelligence,\" a haunting question remains: how artificial does it truly remain, truly?",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.143,
          "numbers_missing": [
            "1976",
            "1980",
            "2",
            "25",
            "3",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.01",
            "0.10"
          ],
          "length_ratio": 0.173,
          "repeats": 0,
          "answer_head": ".. [1]\n..  1. `0.10|0.01|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0"
        },
        {
          "section": "Deconstruct the Hype",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.21",
            "0.25",
            "0.83",
            "0.92",
            "11.0",
            "3",
            "4",
            "5"
          ],
          "numbers_invented": [],
          "length_ratio": 0.001,
          "repeats": 0,
          "answer_head": "..."
        },
        {
          "section": "Examine Real Infrastructure Costs",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.019,
          "numbers_missing": [
            "0.0028",
            "0.003",
            "0.01",
            "0.025",
            "0.03",
            "0.075",
            "0.14",
            "0.15"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.134,
          "repeats": 0,
          "answer_head": ".. [A000000000000000000000000000000000000000, 0.10, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, "
        },
        {
          "section": "Establish a Grounded Local Budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.5",
            "1.1",
            "1.4",
            "1.5",
            "1.8",
            "10.8",
            "16",
            "2.0"
          ],
          "numbers_invented": [
            "0",
            "000",
            "1",
            "100"
          ],
          "length_ratio": 0.048,
          "repeats": 0,
          "answer_head": ".. [0]\n.. 0\n  The budget is 0, the work is 100, the\n.. 000\n.. 000\u8d8a\u5927, the paper\u8d8a\u5927, the\n.. 0\n.. 0\n .. [1]\n    The paper, ``The Cost of\n    ``Hypervasive\n    ``, four pages, four\n    `|cr| 0\n    `"
        },
        {
          "section": "Cost per User: The Metered Curve and the Local Flat",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.353,
          "numbers_missing": [
            "0.44",
            "01",
            "09",
            "2.23",
            "2021",
            "2026",
            "25",
            "27"
          ],
          "numbers_invented": [
            "0",
            "0.0001",
            "0.0002",
            "0.0005",
            "0.001",
            "0.002",
            "0.005",
            "0.01"
          ],
          "length_ratio": 0.209,
          "repeats": 0,
          "answer_head": ".. py\n.. autoplot\n.. plot\n    x = [0.01, 0.02, 0.05, 0.10, 0.20, 0.50, 1.00, 10.00]\n    y = [0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.005, 0.01, 0.02]\n    z = [0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.005, 0.01, 0.02]\n    title: ``costs/metrics/000000000000000000_000000000000000000.yaml\n    style: khan\n    language: yaml\n    imports: []\n    section: ``models/000000000000000000/000000000000000000.yam"
        },
        {
          "section": "Conclusion",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "2026",
            "28",
            "29",
            "30",
            "31",
            "32",
            "33",
            "34"
          ],
          "numbers_invented": [],
          "length_ratio": 0.029,
          "repeats": 0,
          "answer_head": "The paper has four sections, the conclusions are the same as the paper, the\n`conclusions` section."
        },
        {
          "section": "Sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.027,
          "numbers_missing": [
            "09",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15",
            "16"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.054,
          "repeats": 0,
          "answer_head": ".. [1]\n.. 0\n  The 0 references four models, two of which are hosted, two of which\n  are bare, and the fourth is a bare one that runs in a\n  `pytorch-lightning` project rather than a bare\n  `torchvision/models/` directory."
        },
        {
          "section": "Model Architectures, Modelling Types, and the Mermaid Viewer",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.2,
          "numbers_missing": [
            "113",
            "2020",
            "2026",
            "3"
          ],
          "numbers_invented": [
            "0",
            "1",
            "10",
            "11",
            "12",
            "13",
            "14",
            "15"
          ],
          "length_ratio": 0.48,
          "repeats": 0,
          "answer_head": "| File | 0, 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 |\n|---------||--|\n| models/| 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, "
        },
        {
          "section": "Part 1. The mermaid viewer and its diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0.4",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.581,
          "repeats": 0,
          "answer_head": "| Notebook | 0.4\n| `model-00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The four diagrams",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1"
          ],
          "numbers_invented": [
            "0",
            "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.256,
          "repeats": 0,
          "answer_head": "| Notebook | 0 files, 0 references |\n|---------|-----------------|\n| models/000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 2. Model architecture families",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.9",
            "1",
            "1.8",
            "10.6",
            "12",
            "17.7",
            "2",
            "20"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.541,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Dense against sparse",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10.1",
            "101.2",
            "11.0",
            "11.1",
            "12",
            "12.4",
            "12.5",
            "1200.0"
          ],
          "numbers_invented": [
            "0",
            "0.1",
            "100"
          ],
          "length_ratio": 0.204,
          "repeats": 0,
          "answer_head": "| File | 0.1\n| `models/agent/wrkn/agent.py` | The model, four files\n| `models/agent/wrkn/agent.py`, 0 files\n| `models/agent/wrkn/agent.h`, 100\n| `models/agent/wrkn/agent.h`, 100\n| `models/agent/wrkn/agent.h`, 100\n| `models/agent/wrkn/agent.h`, 100\n| `models/agent/wrkn/agent.h`, 100\n| `models/agent/wrkn/agent.h`, 100, `model/`, 100\n| `..`, 0"
        },
        {
          "section": "The newest cycle in the market",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.003",
            "0.022",
            "0.03",
            "0.075",
            "0.15",
            "0.2",
            "0.26",
            "0.5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "000000000000",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "000000000001",
            "000000000002",
            "000000000003",
            "000000000004"
          ],
          "length_ratio": 0.424,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model-000000000000` | 0\n| `model-000000000001` | 0\n| `model-000000000002` | 0\n| `model-000000000003` | 0\n| `model-000000000004` | 0\n| `model-000000000005` | 0\n| `model-000000000006`, `model-000000000007`, `model-000000000008`, `model-000000000009`, `model-000000000010`, `model-000000000011`, `000000000000`, 0\n| `model-000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "The mapping",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 1.0,
          "numbers_missing": [],
          "numbers_invented": [
            "0",
            "0.0",
            "1",
            "10",
            "100",
            "1000",
            "10000",
            "100000"
          ],
          "length_ratio": 0.333,
          "repeats": 0,
          "answer_head": "| Model | 0.0, 100, 1000, 10, 0, 1, 2, 4, 100, 1000, 10000, 10, 100, 100000, 100, 1000, 10000, 1000000, 100000, 100, 10, 1, 10, 100, 10000000, 10, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 10, 10, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, 100, "
        },
        {
          "section": "The benchmark suite as evidence",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "107",
            "113",
            "15",
            "52",
            "53",
            "66",
            "71",
            "80"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0000"
          ],
          "length_ratio": 0.119,
          "repeats": 0,
          "answer_head": "| Notebook | 0.0\n| `model-architectures` | 0.0\n| `model-architecture` | 0.0\n| `model-architecture-0000` | `model-architectures`, 0\n| `|"
        },
        {
          "section": "The input budget",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "128",
            "2"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.77,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Recent modelling news, September and October",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "04",
            "05",
            "07",
            "09",
            "10",
            "14",
            "150"
          ],
          "numbers_invented": [
            "000000000000",
            "1",
            "10000",
            "4000"
          ],
          "length_ratio": 0.335,
          "repeats": 0,
          "answer_head": "| Paper | Type | What the paper says, and the citation |\n|:---|:---|:---|\n| `DeepSeek: A Deep Learning Framework for Retrieval in Unstructured Text` [1]| repository, `000000000000_000000000000_000000000000_000000000000`, 4000 bytes, `papers/000000000000_000000000000_000000000000.ipynb`, `000000000000.ipynb`, 10000|"
        },
        {
          "section": "What the work concerns",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "21",
            "22",
            "3",
            "31",
            "43",
            "5",
            "55"
          ],
          "numbers_invented": [
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.474,
          "repeats": 0,
          "answer_head": "`model-000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Where the work lands",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "10",
            "3",
            "30",
            "4",
            "6",
            "78"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1000"
          ],
          "length_ratio": 0.606,
          "repeats": 0,
          "answer_head": "| File | 0.0\n|--------|----------\n`agent-wrkn/agent.py`, 1000 lines\n[..]\n  1000 lines\n  0.0\n| 0\n| \u4e8e\u662f\u6211\u51b3\u5b9a\u8ba9\u6a21\u578b\u81ea\u5df1\u5b8c\u6210,\u800c\u4e0d\u662f\u8ba9\n| \u4e8e\u662f\u6211\u51b3\u5b9a,\u800c\u4e0d\u662f\u8ba9\n| \u4e8e\u662f\u6211\u51b3\u5b9a,\u800c\u4e0d\u662f\u8ba9\n| \u4e8e\u662f\u6211\u51b3\u5b9a,\u800c\u4e0d\u662f\u8ba9\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\u51b3\u5b9a\n| \u4e8e\u662f\u6211\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0\n| 0"
        },
        {
          "section": "The five newest entries",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "01",
            "10",
            "2",
            "2026",
            "5"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "0000",
            "400"
          ],
          "length_ratio": 0.48,
          "repeats": 0,
          "answer_head": "| Name | Type | What it draws\n|---|---|------------\n| `agent_wizard`, `0.0.1` | directory | `agent_wizard/`, four directories\n| `agent_wizard`, `0.0.2` | package | `agent_wizard`, `0.0.2`, `py:0000`, 0 files\n| `agent_wizard`, `0.0.3` | package | `agent_wizard`, `0.0.3`, `py:0000`, 0 files\n| `agent_wizard`, `0.0.4`, `0000` | package | 400 bytes\n| `agent_wizard`, `0.0.5`, `0000` | `agent_wizard`, 0,"
        },
        {
          "section": "The technique",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "6"
          ],
          "numbers_invented": [
            "0"
          ],
          "length_ratio": 0.909,
          "repeats": 0,
          "answer_head": ".. py\n..\u4eee\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n..| 0\n"
        },
        {
          "section": "The artifact",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "5"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.218,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Part 6. The artifact census",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.0",
            "0.2",
            "1",
            "1.1",
            "16",
            "165.2",
            "26",
            "3"
          ],
          "numbers_invented": [
            "0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
            "100"
          ],
          "length_ratio": 0.244,
          "repeats": 0,
          "answer_head": "| File | 100\n|----------------|----------\n| models/0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "What the bytes cost to load",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "0.3",
            "164.7",
            "19",
            "20.0",
            "458.5",
            "7",
            "76",
            "80.0"
          ],
          "numbers_invented": [
            "0",
            "0.0",
            "1000",
            "4000"
          ],
          "length_ratio": 0.504,
          "repeats": 0,
          "answer_head": "| File | 4000  bytes | 0.0\n|----------------||--\n| models/abalone/abalone.py | 1000  bytes | 0.0\n| models/abalone/abalone.h | 1000  bytes | 0.0\n| models/abalone/abalone.c | 1000  bytes | 0.0\n| models/abalone/abalone.pytorch | 1000  bytes | 0.0\n| models/abalone/abalone.ipynb | 1000  bytes | 0.0\n| models/abalone/abalone.ipynb| 1000  bytes | 0\n| models/abalone/abalone.ipynb, models/ | 1000  bytes | 0"
        },
        {
          "section": "Where the workspace keeps them",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "1",
            "1.1",
            "15",
            "15.4",
            "16",
            "19",
            "2",
            "27.6"
          ],
          "numbers_invented": [
            "0.0",
            "00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
          ],
          "length_ratio": 0.43,
          "repeats": 0,
          "answer_head": "| File | 0.0\n|--------|--\n| models/00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000"
        },
        {
          "section": "Glossary",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "4"
          ],
          "numbers_invented": [
            "0",
            "0.1",
            "0000000000000000",
            "000000000000000000000",
            "400"
          ],
          "length_ratio": 0.242,
          "repeats": 13,
          "answer_head": "| Field | Description |\n|:---|:|\n| `model`, 0.1 `model.py`, 0, `pytorch/agent/walker/agent.py`, `agent/`, `0000000000000000`, 400, 400, `pytorch`, `torch`, `torchvision`, `torchvision/agent`, `000000000000000000000, four files, four modules, four modules, one module, the model, the plan, the plan, the plan, the plan, the plan, the plan, the plan, the plan, the plan, the plan, the plan, the plan, t"
        },
        {
          "section": "Method and sources",
          "violations": [],
          "shipped_violations": [],
          "numbers_kept": 0.0,
          "numbers_missing": [
            "04",
            "09",
            "1",
            "10",
            "2",
            "2026",
            "25",
            "3"
          ],
          "numbers_invented": [
            "0.1",
            "0000000000000000000",
            "000000000000000000000000",
            "000000000000000000000001",
            "000000000000000000000002",
            "000000000000000000000003",
            "000000000000000000000004",
            "000000000000000000000005"
          ],
          "length_ratio": 0.315,
          "repeats": 10,
          "answer_head": "| File | 0.1\n| `models/agent_wizard.py` | The model, four files\n| `models/agent_wizard.py`, `models/agent_wizard.py`, `000000000000000000000000.py`, 400\n| `models/agent_wizard.py`, `000000000000000000000001.py`, 400\n| `models/agent_wizard.py`, `000000000000000000000002.py`, 400\n| `models/agent_wizard.py`, `000000000000000000000003.py`, 400\n| `models/agent_wizard.py`, `000000000000000000000004.py`,"
        }
      ],
      "gpu_offload": true
    }
  },
  "failed": {},
  "quants_requested": 6,
  "gguf_runtime": "present",
  "gguf_gpu_offload": true,
  "gguf_device_libraries": 5,
  "quants_measured": [
    "xcaliburbiz-Q2_K",
    "xcaliburbiz-Q3_K_M",
    "xcaliburbiz-Q4_K_M",
    "xcaliburbiz-Q5_K_M",
    "xcaliburbiz-Q6_K",
    "xcaliburbiz-Q8_0"
  ]
}