{
  "kind": "framework",
  "value": "scaling-laws",
  "collectionKey": "slides:framework:scaling-laws:all-document-kinds:all-producers:all-orientations",
  "filters": {
    "documentKinds": [],
    "sourceTypes": [],
    "orientations": []
  },
  "total": 6,
  "page": 1,
  "pageSize": 24,
  "pageCount": 1,
  "rows": [
    {
      "docId": "019dd923-5e88-73ef-bd5c-f341d4394195",
      "docSlug": "46f66c49fd159048",
      "documentTitle": "2018 Air Street Capital The State of AI Report 2018",
      "authorId": "AirStreetCapital",
      "authorName": "Air Street Capital",
      "documentKindSlug": "consulting-deck",
      "documentKindLabel": "Consulting deck",
      "sourceTypeSlug": "vc_research",
      "sourceTypeLabel": "VC research",
      "presentationDate": null,
      "orientation": "landscape",
      "aspectRatio": 1.777,
      "pageNumber": 12,
      "slideType": "industry_trends",
      "function": "analyze_data",
      "notes": "The slide uses three distinct charts to demonstrate scaling laws: a conceptual framework, a loss-reduction chart, and a parameter-growth chart.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "The charts demonstrate the empirical relationship between data size, loss, and parameter count.",
      "slideHref": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/12",
      "deckHref": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195",
      "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-12",
      "loopMatches": [
        {
          "to": 21,
          "from": 10,
          "name": "Pattern Hunter",
          "slug": "02-pattern-hunter",
          "bestFor": "Time-pressed audiences, building consensus, when data is strong",
          "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670",
          "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.",
          "position": 2,
          "objective": "Stack evidence that AI hardware is the binding constraint",
          "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
          "confidence": 78,
          "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
        }
      ],
      "arcBeatMatches": [
        {
          "to": 69,
          "from": 4,
          "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7",
          "arcName": "The Triple Take",
          "arcSlug": "triple-take",
          "beatName": "The Facts (What)",
          "beatSlug": "triple-take-the-facts-what",
          "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.",
          "position": 1,
          "confidence": 78,
          "parentBeatName": "Setup",
          "parentBeatSlug": "setup"
        },
        {
          "to": 55,
          "from": 10,
          "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7",
          "arcName": "The Onion",
          "arcSlug": "onion",
          "beatName": "First Layer",
          "beatSlug": "onion-first-layer",
          "evidence": "Hardware, vision, RL, bias - technical research layer.",
          "position": 2,
          "confidence": 55,
          "parentBeatName": "Development",
          "parentBeatSlug": "development"
        }
      ],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    },
    {
      "docId": "019dd923-5e88-73ef-bd5c-f812573a947a",
      "docSlug": "eea7524c557036f4",
      "documentTitle": "2020 Air Street Capital The State of AI Report 2020",
      "authorId": "AirStreetCapital",
      "authorName": "Air Street Capital",
      "documentKindSlug": "consulting-deck",
      "documentKindLabel": "Consulting deck",
      "sourceTypeSlug": "vc_research",
      "sourceTypeLabel": "VC research",
      "presentationDate": null,
      "orientation": "landscape",
      "aspectRatio": 1.777,
      "pageNumber": 16,
      "slideType": "industry_trends",
      "function": "analyze_data",
      "notes": "The charts demonstrate power-law relationships between model performance (test loss) and three key scaling factors: compute, dataset size, and parameter count.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "Empirical scaling laws of neural language models",
      "slideHref": "/slides/019dd923-5e88-73ef-bd5c-f812573a947a/16",
      "deckHref": "/decks/019dd923-5e88-73ef-bd5c-f812573a947a",
      "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-f812573a947a#slide-16",
      "loopMatches": [
        {
          "to": 23,
          "from": 15,
          "name": "Pattern Hunter",
          "slug": "02-pattern-hunter",
          "bestFor": "Time-pressed audiences, building consensus, when data is strong",
          "matchId": "019dd95a-07fe-70ce-8d3c-ee7c2907ccd3",
          "evidence": "Slides 15-19 stack scaling-law evidence; 20-23 add caveats.",
          "position": 2,
          "objective": "Establish pattern that bigger models + more compute drive performance at rising cost",
          "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
          "confidence": 70,
          "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
        }
      ],
      "arcBeatMatches": [
        {
          "to": 129,
          "from": 10,
          "beatId": "019dd95a-0682-776c-8e34-ed2a8f38a754",
          "arcName": "The Triple Take",
          "arcSlug": "triple-take",
          "beatName": "The Facts (What)",
          "beatSlug": "triple-take-the-facts-what",
          "evidence": "Research, Talent, Industry sections inventory what happened in AI in 2020.",
          "position": 1,
          "confidence": 60,
          "parentBeatName": "Setup",
          "parentBeatSlug": "setup"
        },
        {
          "to": 129,
          "from": 10,
          "beatId": "019dd95a-0682-776c-8e34-fc384f439bba",
          "arcName": "The Mountain",
          "arcSlug": "mountain",
          "beatName": "Rising Action",
          "beatSlug": "mountain-rising-action",
          "evidence": "Research, Talent and Industry sections build momentum of AI progress.",
          "position": 2,
          "confidence": 40,
          "parentBeatName": "Development",
          "parentBeatSlug": "development"
        }
      ],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    },
    {
      "docId": "019dd923-5e88-73ef-bd5c-f812573a947a",
      "docSlug": "eea7524c557036f4",
      "documentTitle": "2020 Air Street Capital The State of AI Report 2020",
      "authorId": "AirStreetCapital",
      "authorName": "Air Street Capital",
      "documentKindSlug": "consulting-deck",
      "documentKindLabel": "Consulting deck",
      "sourceTypeSlug": "vc_research",
      "sourceTypeLabel": "VC research",
      "presentationDate": null,
      "orientation": "landscape",
      "aspectRatio": 1.777,
      "pageNumber": 20,
      "slideType": "industry_trends",
      "function": "analyze_data",
      "notes": "Charts illustrate scaling laws in machine learning, specifically the relationship between model size (parameters), data (tokens), and compute budget.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "Charts demonstrate the relationship between model parameters, compute, and performance.",
      "slideHref": "/slides/019dd923-5e88-73ef-bd5c-f812573a947a/20",
      "deckHref": "/decks/019dd923-5e88-73ef-bd5c-f812573a947a",
      "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-f812573a947a#slide-20",
      "loopMatches": [
        {
          "to": 23,
          "from": 15,
          "name": "Pattern Hunter",
          "slug": "02-pattern-hunter",
          "bestFor": "Time-pressed audiences, building consensus, when data is strong",
          "matchId": "019dd95a-07fe-70ce-8d3c-ee7c2907ccd3",
          "evidence": "Slides 15-19 stack scaling-law evidence; 20-23 add caveats.",
          "position": 2,
          "objective": "Establish pattern that bigger models + more compute drive performance at rising cost",
          "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
          "confidence": 70,
          "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
        }
      ],
      "arcBeatMatches": [
        {
          "to": 129,
          "from": 10,
          "beatId": "019dd95a-0682-776c-8e34-ed2a8f38a754",
          "arcName": "The Triple Take",
          "arcSlug": "triple-take",
          "beatName": "The Facts (What)",
          "beatSlug": "triple-take-the-facts-what",
          "evidence": "Research, Talent, Industry sections inventory what happened in AI in 2020.",
          "position": 1,
          "confidence": 60,
          "parentBeatName": "Setup",
          "parentBeatSlug": "setup"
        },
        {
          "to": 129,
          "from": 10,
          "beatId": "019dd95a-0682-776c-8e34-fc384f439bba",
          "arcName": "The Mountain",
          "arcSlug": "mountain",
          "beatName": "Rising Action",
          "beatSlug": "mountain-rising-action",
          "evidence": "Research, Talent and Industry sections build momentum of AI progress.",
          "position": 2,
          "confidence": 40,
          "parentBeatName": "Development",
          "parentBeatSlug": "development"
        }
      ],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    },
    {
      "docId": "019dd923-5e88-73ef-bd5d-0d0f98caffe1",
      "docSlug": "5df6bd1b0447b5f6",
      "documentTitle": "2025 Air Street Capital The State of AI Report 2025",
      "authorId": "AirStreetCapital",
      "authorName": "Air Street Capital",
      "documentKindSlug": "consulting-deck",
      "documentKindLabel": "Consulting deck",
      "sourceTypeSlug": "vc_research",
      "sourceTypeLabel": "VC research",
      "presentationDate": null,
      "orientation": "landscape",
      "aspectRatio": 1.777,
      "pageNumber": 13,
      "slideType": "industry_trends",
      "function": "analyze_data",
      "notes": "The slide highlights the shift towards inference-time scaling (test-time compute) as a key driver for AI reasoning capabilities.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "Reference to compute scaling and log-based intelligence growth",
      "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0d0f98caffe1/13",
      "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1",
      "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1#slide-13",
      "loopMatches": [
        {
          "to": 30,
          "from": 13,
          "name": "Zoom In",
          "slug": "06-zoom-in",
          "bestFor": "Technical deep-dives, case studies, detailed analysis",
          "matchId": "019dd95a-07fe-70ce-8d3e-9252f6f92bd7",
          "evidence": "Big picture 'reasoning race' (p.13) zooms into o1, R1, DeepSeek, parallel reasoning, CoT mechanics.",
          "position": 2,
          "objective": "Drill from reasoning paradigm into specific reasoning research findings",
          "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication",
          "confidence": 75,
          "description": "Start broad, then progressively focus on specific details that prove your point"
        }
      ],
      "arcBeatMatches": [
        {
          "to": 189,
          "from": 12,
          "beatId": "019dd95a-0682-776c-8e35-9f380673831f",
          "arcName": "The Triple Take",
          "arcSlug": "triple-take",
          "beatName": "The Facts (What)",
          "beatSlug": "triple-take-the-facts-what",
          "evidence": "Sections 1-2 lay out research findings and industry data with charts and case studies.",
          "position": 1,
          "confidence": 70,
          "parentBeatName": "Setup",
          "parentBeatSlug": "setup"
        },
        {
          "to": 89,
          "from": 12,
          "beatId": "019dd95a-0682-776c-8e35-ac68557d958b",
          "arcName": "The Onion",
          "arcSlug": "onion",
          "beatName": "First Layer",
          "beatSlug": "onion-first-layer",
          "evidence": "Research section unpacks technical layer beneath the headlines.",
          "position": 2,
          "confidence": 45,
          "parentBeatName": "Development",
          "parentBeatSlug": "development"
        }
      ],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    },
    {
      "docId": "019de075-e061-72f9-baa2-90171b05eced",
      "docSlug": "cdaaa50efb264b4a4093f665bafb20a4",
      "documentTitle": "Absci | Investor Presentation Deck | 111 slides",
      "authorId": "absci",
      "authorName": "Absci",
      "documentKindSlug": "conference-presentation",
      "documentKindLabel": "Conference presentation",
      "sourceTypeSlug": "investor_relations",
      "sourceTypeLabel": "Investor relations",
      "presentationDate": "2023-10-01 00:00:00",
      "orientation": "landscape",
      "aspectRatio": 1.7777778,
      "pageNumber": 61,
      "slideType": "other",
      "function": "present_framework",
      "notes": "The slide outlines the 'scaling laws' paradigm for LLMs.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "The slide explicitly links model performance to the triad of training data, model size, and compute budget.",
      "slideHref": "/slides/019de075-e061-72f9-baa2-90171b05eced/61",
      "deckHref": "/decks/019de075-e061-72f9-baa2-90171b05eced",
      "deckAnchorHref": "/decks/019de075-e061-72f9-baa2-90171b05eced#slide-61",
      "loopMatches": [],
      "arcBeatMatches": [],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    },
    {
      "docId": "019de06f-91e7-721d-ba74-936d7493bb6e",
      "docSlug": "aa845fc575f0cc138d44a9f5be063e0c",
      "documentTitle": "OpenAI | Product Presentation Deck | 33 slides",
      "authorId": "openai",
      "authorName": "OpenAI",
      "documentKindSlug": "conference-presentation",
      "documentKindLabel": "Conference presentation",
      "sourceTypeSlug": "investor_relations",
      "sourceTypeLabel": "Investor relations",
      "presentationDate": "2018-03-01 00:00:00",
      "orientation": "landscape",
      "aspectRatio": 1.7794254,
      "pageNumber": 22,
      "slideType": "thesis_headline",
      "function": "analyze_data",
      "notes": "The chart illustrates scaling laws in LLMs.",
      "imagePath": null,
      "matchCount": 1,
      "evidence": "The chart demonstrates the relationship between model size (parameters) and performance (BLEU).",
      "slideHref": "/slides/019de06f-91e7-721d-ba74-936d7493bb6e/22",
      "deckHref": "/decks/019de06f-91e7-721d-ba74-936d7493bb6e",
      "deckAnchorHref": "/decks/019de06f-91e7-721d-ba74-936d7493bb6e#slide-22",
      "loopMatches": [
        {
          "to": 26,
          "from": 18,
          "name": "Hypothesis Test",
          "slug": "42-hypothesis-test",
          "bestFor": "Strategy validation, due diligence, research presentations",
          "matchId": "ffb7a080-8ce3-4455-be4e-12c01b6aee0f",
          "evidence": "Results and examples of GPT-2",
          "position": 2,
          "objective": "Testing the hypothesis of GPT-2's capabilities",
          "structure": "The Hypothesis -> Test 1 (Supports/Refutes) -> Test 2 (Supports/Refutes) -> Conclusion (Validated/Pivoted)",
          "confidence": 0.6,
          "description": "State a hypothesis upfront, then systematically prove or disprove it with evidence"
        }
      ],
      "arcBeatMatches": [
        {
          "to": 26,
          "from": 18,
          "beatId": "57fcc14c-507b-422f-8f3c-eadb2da301d6",
          "arcName": "The Sparkline",
          "arcSlug": "sparkline",
          "beatName": "Evidence",
          "beatSlug": "evidence",
          "evidence": "Results and examples of GPT-2",
          "position": 3,
          "confidence": 0.8,
          "parentBeatName": null,
          "parentBeatSlug": null
        }
      ],
      "imagePathAlt": null,
      "thumbSrc": null,
      "thumbSrcAlt": null,
      "locked": true
    }
  ]
}