{"version":1,"pages":[{"id":"dO1mSvQGfLzskvncndTt","title":"Continuum Labs - Applied AI","pathname":"/continuum-applications","siteSpaceId":"sitesp_mOR6V","description":"\"Applied Artificial Intelligence\""},{"id":"lkGiHdGvoygkVk231tq3","title":"What we do","pathname":"/continuum-applications/overview/what-we-do","siteSpaceId":"sitesp_mOR6V","description":"Continuum AI Driven Modules","breadcrumbs":[{"label":"Overview"}]},{"id":"3IoSyxed5OkuHutaXb9E","title":"Our Features","pathname":"/continuum-applications/overview/our-features","siteSpaceId":"sitesp_mOR6V","description":"Rapid development of customised models for immediate deployment","breadcrumbs":[{"label":"Overview"}]},{"id":"TJa9YiVaaPWA6k7zyJCv","title":"Secure and Private GPU Infrastructure","pathname":"/continuum-applications/overview/secure-and-private-gpu-infrastructure","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Overview"}]},{"id":"tZL5nZV3SC1be9yKeBz6","title":"Generative AI Implementation Risks","pathname":"/continuum-applications/overview/generative-ai-implementation-risks","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Overview"}]},{"id":"3atXLMAtBNkbWKLUJuIe","title":"Model Range","pathname":"/continuum-applications/model-range/model-range","siteSpaceId":"sitesp_mOR6V","emoji":"23f9","description":"Our range of models","breadcrumbs":[{"label":"Model Range"}]},{"id":"AS2jg97c6jP85PxRvbE9","title":"Investment Management","pathname":"/continuum-applications/model-range/investment-management","siteSpaceId":"sitesp_mOR6V","description":"A model to give all investors edge","breadcrumbs":[{"label":"Model Range"}]},{"id":"YgGV78rq1AistiIgUX28","title":"Employment Law","pathname":"/continuum-applications/model-range/employment-law","siteSpaceId":"sitesp_mOR6V","description":"Navigating the intricaties of Australian employment law","breadcrumbs":[{"label":"Model Range"}]},{"id":"RVDd1i9EZqJK6y11BCrf","title":"Psychology and Mental Health","pathname":"/continuum-applications/model-range/psychology-and-mental-health","siteSpaceId":"sitesp_mOR6V","description":"Providing assistance to professional psychologists and mental health workers","breadcrumbs":[{"label":"Model Range"}]},{"id":"Pu9goe4y7GMg25GLmVLU","title":"Home Insurance","pathname":"/continuum-applications/model-range/home-insurance","siteSpaceId":"sitesp_mOR6V","description":"Providing assistance to consumers navigating the complexity of home insurance products","breadcrumbs":[{"label":"Model Range"}]},{"id":"uyOVDfLQS0VE2dMSNzS7","title":"Consumer Surveying","pathname":"/continuum-applications/model-range/consumer-surveying","siteSpaceId":"sitesp_mOR6V","description":"A deeper and more nuanced interaction with consumers","breadcrumbs":[{"label":"Model Range"}]},{"id":"1edoLiBjLGtFGpENMyG0","title":"Government Grants","pathname":"/continuum-applications/model-range/government-grants","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Model Range"}]},{"id":"sZj0WxYCKN133Dc2yDSb","title":"Aged Care","pathname":"/continuum-applications/model-range/aged-care","siteSpaceId":"sitesp_mOR6V","description":"Customised Fine-Tuned Large Language Models in the Aged Care Sector: A Comprehensive Use Case","breadcrumbs":[{"label":"Model Range"}]},{"id":"EJBQ6ow4NOzVEfwczFZc","title":"Pharmaceuticals Benefit Scheme","pathname":"/continuum-applications/model-range/pharmaceuticals-benefit-scheme","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Model Range"}]},{"id":"lk8ReK7JPt5QGsiRuYo2","title":"Three ideas for autonomous agent applications","pathname":"/continuum-applications/discussion-and-use-cases/three-ideas-for-autonomous-agent-applications","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"xqdFgfXIEZtj8avkUcOU","title":"Financial Statement analysis with large language models","pathname":"/continuum-applications/discussion-and-use-cases/financial-statement-analysis-with-large-language-models","siteSpaceId":"sitesp_mOR6V","description":"The University of Chicago, Booth School of Business","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"OedaQWN1Tq1tDzzEVFgN","title":"The Evolution of AI Agents and Their Potential for Augmenting Human Agency","pathname":"/continuum-applications/discussion-and-use-cases/the-evolution-of-ai-agents-and-their-potential-for-augmenting-human-agency","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"8ntqYAXj864uhw69Ixns","title":"Better Call Saul - SaulLM-7B - a legal large language model","pathname":"/continuum-applications/discussion-and-use-cases/better-call-saul-saullm-7b-a-legal-large-language-model","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"EiXPC6XAQhMKuESSyOn4","title":"MentaLLaMA: Interpretable Mental Health Analysis on Social Media with Large Language Models","pathname":"/continuum-applications/discussion-and-use-cases/mentallama-interpretable-mental-health-analysis-on-social-media-with-large-language-models","siteSpaceId":"sitesp_mOR6V","description":"Kailai Yang, Tianlin Zhang, Ziyan Kuang, Qianqian Xie, Jimin Huang, Sophia Ananiadou","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"s8ed9BxDdurLdsOwZrle","title":"Anomaly detection in logging data","pathname":"/continuum-applications/discussion-and-use-cases/anomaly-detection-in-logging-data","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"kKcioPP4ww7MbsLfziFl","title":"ChatDoctor: Artificial Intelligence powered doctors","pathname":"/continuum-applications/discussion-and-use-cases/chatdoctor-artificial-intelligence-powered-doctors","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"M4AsPFqiTe4Q48n1jC0j","title":"Navigating the Jagged Technological Frontier: Effects of AI on Knowledge Workers","pathname":"/continuum-applications/discussion-and-use-cases/navigating-the-jagged-technological-frontier-effects-of-ai-on-knowledge-workers","siteSpaceId":"sitesp_mOR6V","description":"Harvard Business School - 2023","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"Owua6mZerIIvAmbemxih","title":"Effect of AI on the US labour market","pathname":"/continuum-applications/discussion-and-use-cases/effect-of-ai-on-the-us-labour-market","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"whrC9kV93RKHUZQlkxTU","title":"Data Interpreter: An LLM Agent For Data Science","pathname":"/continuum-applications/discussion-and-use-cases/data-interpreter-an-llm-agent-for-data-science","siteSpaceId":"sitesp_mOR6V","description":"","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"5y8vvWOzxsXovlxJAWPp","title":"The impact of AI on the customer support industry","pathname":"/continuum-applications/discussion-and-use-cases/the-impact-of-ai-on-the-customer-support-industry","siteSpaceId":"sitesp_mOR6V","description":"Erik Brynjolfsson, Danielle Li, and Lindsey R. Raymond","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"b4oaPwFkCGKI1nzq8QVo","title":"Can Large Language Models Reason and Plan?","pathname":"/continuum-applications/discussion-and-use-cases/can-large-language-models-reason-and-plan","siteSpaceId":"sitesp_mOR6V","description":"The answer according to this research is 'no'","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"sEFX2BDcP5nuh7tl8m3K","title":"KnowAgent: Knowledge-Augmented Planning for LLM-Based Agents","pathname":"/continuum-applications/discussion-and-use-cases/knowagent-knowledge-augmented-planning-for-llm-based-agents","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"1IqlYYfRF0qW2BkvrUwV","title":"The flaws of 'product-market fit' in an emerging industry","pathname":"/continuum-applications/discussion-and-use-cases/the-flaws-of-product-market-fit-in-an-emerging-industry","siteSpaceId":"sitesp_mOR6V","description":"Why is it used as a term and is it relevant in a new industry like artificial intelligence?","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"8tZN8Ia5fX2ddZJziNXq","title":"Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence","pathname":"/continuum-applications/discussion-and-use-cases/experimental-evidence-on-the-productivity-effects-of-generative-artificial-intelligence","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"NFXRcMsJ9XL56voQWyd2","title":"The Disruption of the Administrative Class: How Generative AI is Reshaping Organisational Operations","pathname":"/continuum-applications/discussion-and-use-cases/the-disruption-of-the-administrative-class-how-generative-ai-is-reshaping-organisational-operations","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"pGC7I5MTW1XdEyGIzh6Z","title":"How Knowledge Workers Think Generative AI Will (Not) Transform Their Industries","pathname":"/continuum-applications/discussion-and-use-cases/how-knowledge-workers-think-generative-ai-will-not-transform-their-industries","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"Tkl8QsbL8ufiPOAN4xoq","title":"Embracing AI: A Strategic Imperative for Modern Leadership","pathname":"/continuum-applications/discussion-and-use-cases/embracing-ai-a-strategic-imperative-for-modern-leadership","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"RAcy1JeQFTDsyt56OY5F","title":"Artificial Intelligence and Management: The Automation-Augmentation Paradox","pathname":"/continuum-applications/discussion-and-use-cases/artificial-intelligence-and-management-the-automation-augmentation-paradox","siteSpaceId":"sitesp_mOR6V","description":"The complex interplay between automation and augmentation in the use of artificial intelligence (AI) in management","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"BXPf2odkOSSn7Jodx6zx","title":"Network effects in AI models","pathname":"/continuum-applications/discussion-and-use-cases/network-effects-in-ai-models","siteSpaceId":"sitesp_mOR6V","description":"The Role of Artificial Intelligence and Data Network Effects for Creating User Value","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"Qe8hlDglYIzV4SCNbVey","title":"AI impact on the publishing industry","pathname":"/continuum-applications/discussion-and-use-cases/ai-impact-on-the-publishing-industry","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"EmcGe7H2UiRSB1BPJ4X3","title":"Power asymmetry","pathname":"/continuum-applications/discussion-and-use-cases/power-asymmetry","siteSpaceId":"sitesp_mOR6V","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"AEz4kIckWL4487FPxgD5","title":"Information Asymmetry","pathname":"/continuum-applications/discussion-and-use-cases/information-asymmetry","siteSpaceId":"sitesp_mOR6V","description":"Applied AI can reduce this major economic inefficiency","breadcrumbs":[{"label":"Discussion and Use Cases"}]},{"id":"dO1mSvQGfLzskvncndTt","title":"Continuum","pathname":"/","siteSpaceId":"sitesp_3lJdL","description":"Applied Artificial Intelligence"},{"id":"lkGiHdGvoygkVk231tq3","title":"Datasets","pathname":"/data/datasets","siteSpaceId":"sitesp_3lJdL","description":"","breadcrumbs":[{"label":"Data"}]},{"id":"yGzFhV8DpcgRKNRzxDKU","title":"Pre Training Data","pathname":"/data/datasets/pre-training-data","siteSpaceId":"sitesp_3lJdL","description":"Training Foundation Models","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"DD9Xbdgvi18hcFDp1LVD","title":"Types of Fine Tuning","pathname":"/data/datasets/types-of-fine-tuning","siteSpaceId":"sitesp_3lJdL","description":"Some definitions","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"TEqIJe2M10EA7vricffJ","title":"Self Instruct Paper","pathname":"/data/datasets/self-instruct-paper","siteSpaceId":"sitesp_3lJdL","description":"The most highly cited paper on fine tuning methods","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"OIka00urjiXwWUFMfX46","title":"Self-Alignment with Instruction Backtranslation","pathname":"/data/datasets/self-alignment-with-instruction-backtranslation","siteSpaceId":"sitesp_3lJdL","description":"Xian Li, Ping Yu, Chunting Zhou, Timo Schick, Omer Levy, Luke Zettlemoyer, Jason Weston, Mike Lewis","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"4TwTvPZWpD6OvPsZb7FB","title":"Systematic Evaluation of Instruction-Tuned Large Language Models on Open Datasets","pathname":"/data/datasets/systematic-evaluation-of-instruction-tuned-large-language-models-on-open-datasets","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"TXHsq838L6y8MK1DtGc4","title":"Instruction Tuning","pathname":"/data/datasets/instruction-tuning","siteSpaceId":"sitesp_3lJdL","description":"Inspired by the Self-Instruct Paper","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"cwfJ1U1VBHCzAMEdn1hT","title":"Instruction Fine Tuning - Alpagasus","pathname":"/data/datasets/instruction-fine-tuning-alpagasus","siteSpaceId":"sitesp_3lJdL","description":"\"ALPAGASUS: Data-Driven Data Selection for Instruction Fine-Tuning\"","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"Dm8gVfbzmwXawY47pWzz","title":"Less is More For Alignment","pathname":"/data/datasets/less-is-more-for-alignment","siteSpaceId":"sitesp_3lJdL","description":"Co-authored by researchers from Meta, Carnegie Mellon University, University of Southern California, and Tel Aviv University","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"XdtbxUo9CcEAkO5ECwnD","title":"Enhanced Supervised Fine Tuning","pathname":"/data/datasets/enhanced-supervised-fine-tuning","siteSpaceId":"sitesp_3lJdL","description":"How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"kxPds2obzoXkaJDIOfNo","title":"Visualising Data using t-SNE","pathname":"/data/datasets/visualising-data-using-t-sne","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"JeWeRmIZ4QT9vOYSJ9lL","title":"UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction","pathname":"/data/datasets/umap-uniform-manifold-approximation-and-projection-for-dimension-reduction","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"YZnpOFwcHR1eQKmF27lp","title":"Training and Evaluation Datasets","pathname":"/data/datasets/training-and-evaluation-datasets","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"2Ym24TTgDBlyozuTff7X","title":"What is perplexity?","pathname":"/data/datasets/what-is-perplexity","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Data"},{"label":"Datasets"}]},{"id":"fqWzw6etY48oEEEYPWwt","title":"Foundation Models","pathname":"/models/foundation-models","siteSpaceId":"sitesp_3lJdL","description":"Continuum AI Driven Modules","breadcrumbs":[{"label":"MODELS"}]},{"id":"txl6I68vewx6nUQF5Mqd","title":"The leaderboard","pathname":"/models/foundation-models/the-leaderboard","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"gsRmNlvFgSr6CDzPoAKY","title":"Foundation Models","pathname":"/models/foundation-models/foundation-models","siteSpaceId":"sitesp_3lJdL","description":"Training Foundation Models","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"p4CVoQicgwHaODR03b3U","title":"LLama 2 - Analysis","pathname":"/models/foundation-models/llama-2-analysis","siteSpaceId":"sitesp_3lJdL","description":"Meta introduced Llama 2 during June 2023","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"gzfzt0ENEzdxkblRICFE","title":"Analysis of Llama 3","pathname":"/models/foundation-models/analysis-of-llama-3","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"J8tFB3FWZPIIUafMtMem","title":"Llama 3.1 series","pathname":"/models/foundation-models/llama-3.1-series","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"dErD4SpJB9T6pl72of6g","title":"Google Gemini 1.5","pathname":"/models/foundation-models/google-gemini-1.5","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"kf2t78IuYTEhyCCNVrne","title":"Platypus: Quick, Cheap, and Powerful Refinement of LLMs","pathname":"/models/foundation-models/platypus-quick-cheap-and-powerful-refinement-of-llms","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"wbTFKT3qQgpOhDsW2ypn","title":"Mixtral of Experts","pathname":"/models/foundation-models/mixtral-of-experts","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"q5OIFbMq5iOwgZT3A2Ls","title":"Mixture-of-Agents (MoA)","pathname":"/models/foundation-models/mixture-of-agents-moa","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"wVUp6Qa9wzo57b85ntcN","title":"Phi 1.5","pathname":"/models/foundation-models/phi-1.5","siteSpaceId":"sitesp_3lJdL","description":"The Diminutive Giant: How the Phi Model is Revolutionising AI Accessibility","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"mcfpO6TQUbfnhHF9WGk6","title":"Refining the Art of AI Training: A Deep Dive into Phi 1.5's Innovative Approach","pathname":"/models/foundation-models/phi-1.5/refining-the-art-of-ai-training-a-deep-dive-into-phi-1.5s-innovative-approach","siteSpaceId":"sitesp_3lJdL","description":"A long list of lessons, tips and tricks from the team that bought us Phi","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"},{"label":"Phi 1.5"}]},{"id":"YLWa29gMrMmzkLGnSW9G","title":"Phi 2.0","pathname":"/models/foundation-models/phi-2.0","siteSpaceId":"sitesp_3lJdL","description":"Microsoft's small but powerful transformer model","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"yxUqif2gMnFHHRxsCuuv","title":"Phi-3 Technical Report","pathname":"/models/foundation-models/phi-3-technical-report","siteSpaceId":"sitesp_3lJdL","description":"A Highly Capable Language Model Locally on Your Phone","breadcrumbs":[{"label":"MODELS"},{"label":"Foundation Models"}]},{"id":"HSYVPM3l1MGOAm0krlnV","title":"The Fine Tuning Process","pathname":"/training/the-fine-tuning-process","siteSpaceId":"sitesp_3lJdL","description":"Fine tuning deep learning models is completely different to fine tuning machine learning models","breadcrumbs":[{"label":"Training"}]},{"id":"y36umB9Vn7OQO3vXHa0v","title":"Why fine tune?","pathname":"/training/the-fine-tuning-process/why-fine-tune","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"}]},{"id":"yxw3LQcv72mQbdwGKhlC","title":"Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?","pathname":"/training/the-fine-tuning-process/why-fine-tune/does-fine-tuning-llms-on-new-knowledge-encourage-hallucinations","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Why fine tune?"}]},{"id":"ificZZzzMFMt2fMmLwUN","title":"Explanations in Fine Tuning","pathname":"/training/the-fine-tuning-process/why-fine-tune/explanations-in-fine-tuning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Why fine tune?"}]},{"id":"szpMb5ZMvUOsWpuM8GhM","title":"Tokenization","pathname":"/training/the-fine-tuning-process/tokenization","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"}]},{"id":"kbSjRr7dfU0gWQ9iBrmM","title":"Tokenization Is More Than Compression","pathname":"/training/the-fine-tuning-process/tokenization/tokenization-is-more-than-compression","siteSpaceId":"sitesp_3lJdL","description":"Craig W. Schmidt, Varshini Reddy, Haoran Zhang, Alec Alameddine, Omri Uzan, Yuval Pinter, Chris Tanner","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"VCmjeTwzi4EOBpCEeRVw","title":"Tokenization - SentencePiece","pathname":"/training/the-fine-tuning-process/tokenization/tokenization-sentencepiece","siteSpaceId":"sitesp_3lJdL","description":"The Unsupervised Text Tokenizer for Neural Networks","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"PjsZcgWiXt1iEGWxOv03","title":"Tokenization explore","pathname":"/training/the-fine-tuning-process/tokenization/tokenization-explore","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"bNcF8kOhq1x3EX375vtX","title":"Tokenizer Choice For LLM Training: Negligible or Crucial?","pathname":"/training/the-fine-tuning-process/tokenization/tokenizer-choice-for-llm-training-negligible-or-crucial","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"wKd085DnKOmGlCWruS6L","title":"Getting the most out of your tokenizer for pre-training and domain adaptation","pathname":"/training/the-fine-tuning-process/tokenization/getting-the-most-out-of-your-tokenizer-for-pre-training-and-domain-adaptation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"MOJDsNOk3skSGivJGJMS","title":"TokenMonster","pathname":"/training/the-fine-tuning-process/tokenization/tokenmonster","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Tokenization"}]},{"id":"CVmqad5YGbW9vPYTIrDF","title":"Parameter Efficient Fine Tuning","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"}]},{"id":"KqphI4jRYijpn8ntjiKD","title":"P-Tuning","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/p-tuning","siteSpaceId":"sitesp_3lJdL","description":"The highly cited \"GPT Understands Too\" paper first submitted March 2021, introducing P-Tuning","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"L6jWo3H0sAbJSLUz7u7a","title":"The Power of Scale for Parameter-Efficient Prompt Tuning","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/p-tuning/the-power-of-scale-for-parameter-efficient-prompt-tuning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"},{"label":"P-Tuning"}]},{"id":"1Vh99CeVKRWXjx3MYwSu","title":"Prefix-Tuning: Optimizing Continuous Prompts for Generation","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/prefix-tuning-optimizing-continuous-prompts-for-generation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"jqEMlnC7K5ITUWJzjny1","title":"Harnessing the Power of PEFT: A Smarter Approach to Fine-tuning Pre-trained Models","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/harnessing-the-power-of-peft-a-smarter-approach-to-fine-tuning-pre-trained-models","siteSpaceId":"sitesp_3lJdL","description":"Parameter-Efficient Fine-Tuning (PEFT) is a technique used to fine tune neural language models","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"A2EnKehAZrevwgH8aHHW","title":"What is Low-Rank Adaptation (LoRA) -  explained by the inventor","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/what-is-low-rank-adaptation-lora-explained-by-the-inventor","siteSpaceId":"sitesp_3lJdL","description":"Edward Hu","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"skEtWWtSnx60ntVZf0ll","title":"Low Rank Adaptation (Lora)","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/low-rank-adaptation-lora","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"lWkou710PZaieGFxvbVT","title":"Practical Tips for Fine-tuning LMs Using LoRA (Low-Rank Adaptation)","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/practical-tips-for-fine-tuning-lms-using-lora-low-rank-adaptation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"lWmXKlQuoeRy6kreT8rj","title":"QLORA: Efficient Finetuning of Quantized LLMs","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/qlora-efficient-finetuning-of-quantized-llms","siteSpaceId":"sitesp_3lJdL","description":"","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"njJqmadsFwrmav1CzvZb","title":"Bits and Bytes","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/bits-and-bytes","siteSpaceId":"sitesp_3lJdL","description":"Tim Dettmers (PhD candidate, University of Washington) presents \"8-bit Methods for Efficient Deep Learning\" in this Cohere For AI Technical Talk.","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"Jo9Vfdg3QobACiukZ1Cs","title":"The Magic behind Qlora","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/the-magic-behind-qlora","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"VQgKSRdR8hOBGmnAEhmn","title":"Practical Guide to LoRA: Tips and Tricks for Effective Model Adaptation","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/practical-guide-to-lora-tips-and-tricks-for-effective-model-adaptation","siteSpaceId":"sitesp_3lJdL","description":"A range of practical tips and questions around using Lora","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"MxKlG8GFRaxRuqSSqhVW","title":"The quantization constant","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/the-quantization-constant","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"TxPD1PzwNJa0s7Met7Ro","title":"QLORA: Efficient Finetuning of Quantized Language Models","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/qlora-efficient-finetuning-of-quantized-language-models","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"9KPIQo8ZKwwX5iBoNUQj","title":"QLORA and Fine-Tuning of Quantized Language Models (LMs)","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/qlora-and-fine-tuning-of-quantized-language-models-lms","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"HLT0wjysXHVYY1umPxry","title":"ReLoRA: High-Rank Training Through Low-Rank Updates","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/relora-high-rank-training-through-low-rank-updates","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"JelbdYC8oWGT6ByhjQSP","title":"SLoRA: Federated Parameter Efficient Fine-Tuning of Language Models","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/slora-federated-parameter-efficient-fine-tuning-of-language-models","siteSpaceId":"sitesp_3lJdL","description":"Leveraging Lora","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"xUzGQu9MRFqhA6UnPYmO","title":"GaLora: Memory-Efficient LLM Training by Gradient Low-Rank Projection","pathname":"/training/the-fine-tuning-process/parameter-efficient-fine-tuning/galora-memory-efficient-llm-training-by-gradient-low-rank-projection","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Parameter Efficient Fine Tuning"}]},{"id":"EgoGqBQTkbvuNBzYzzJv","title":"Hyperparameters","pathname":"/training/the-fine-tuning-process/hyperparameters","siteSpaceId":"sitesp_3lJdL","description":"Art and science","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"}]},{"id":"m9XXtFRMWyhZbXIfgbiG","title":"Batch Size","pathname":"/training/the-fine-tuning-process/hyperparameters/batch-size","siteSpaceId":"sitesp_3lJdL","description":"Choosing the right batch size is critical","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"8gF1YiC140DGcj19rMTR","title":"Padding Tokens","pathname":"/training/the-fine-tuning-process/hyperparameters/padding-tokens","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"djVnnKcTV2aM7QtNTTXK","title":"Mixed precision training","pathname":"/training/the-fine-tuning-process/hyperparameters/mixed-precision-training","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"w51T8gFpHBxAA7wZcmH9","title":"FP8 Formats for Deep Learning","pathname":"/training/the-fine-tuning-process/hyperparameters/fp8-formats-for-deep-learning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"366m8yNSxOJTudBgwutJ","title":"Floating Point Numbers","pathname":"/training/the-fine-tuning-process/hyperparameters/floating-point-numbers","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"sLZx0Xx1W3wRQWTUcWY2","title":"Batch Size and Model loss","pathname":"/training/the-fine-tuning-process/hyperparameters/batch-size-and-model-loss","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"9D7MH38YvlN73vtMTsGK","title":"Batch Normalisation","pathname":"/training/the-fine-tuning-process/hyperparameters/batch-normalisation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"zsGdgoHEpgRQoh4k9I9c","title":"Rethinking Learning Rate Tuning in the Era of Language Models","pathname":"/training/the-fine-tuning-process/hyperparameters/rethinking-learning-rate-tuning-in-the-era-of-language-models","siteSpaceId":"sitesp_3lJdL","description":"One of the most important hyperparameters","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"BTONlR09nGyvWHAc4eWQ","title":"Sample Packing","pathname":"/training/the-fine-tuning-process/hyperparameters/sample-packing","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"2sMJ5zJpCzCXQ8qgnmzk","title":"Gradient accumulation","pathname":"/training/the-fine-tuning-process/hyperparameters/gradient-accumulation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"X6afjQ1jMpo5P85lyIfJ","title":"A process for choosing the learning rate","pathname":"/training/the-fine-tuning-process/hyperparameters/a-process-for-choosing-the-learning-rate","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"M3dGzDTV3CjajRCRSEJX","title":"Learning Rate Scheduler","pathname":"/training/the-fine-tuning-process/hyperparameters/learning-rate-scheduler","siteSpaceId":"sitesp_3lJdL","description":"Key Considerations with Learning Rate Scheduling in Neural Network Training","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"2bqmCS9wABahEd8k8laV","title":"Checkpoints","pathname":"/training/the-fine-tuning-process/hyperparameters/checkpoints","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"0uF4ls9JewxELLSJPpiP","title":"A Survey on Efficient Training of Transformers","pathname":"/training/the-fine-tuning-process/hyperparameters/a-survey-on-efficient-training-of-transformers","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"GCUWON6ieqWd9Kje7Bp9","title":"Sequence Length Warmup","pathname":"/training/the-fine-tuning-process/hyperparameters/sequence-length-warmup","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"utvK0nBcJg9yvWX0eFRA","title":"Understanding Training vs. Evaluation Data Splits","pathname":"/training/the-fine-tuning-process/hyperparameters/understanding-training-vs.-evaluation-data-splits","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"cMKNFA2QeVucBOJ5xWOW","title":"Cross-entropy loss","pathname":"/training/the-fine-tuning-process/hyperparameters/cross-entropy-loss","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"UjqtcVY3zufyHfKEN053","title":"Weight Decay","pathname":"/training/the-fine-tuning-process/hyperparameters/weight-decay","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"1Mz1IV2r3CFcTP4cqGE7","title":"Optimiser","pathname":"/training/the-fine-tuning-process/hyperparameters/optimiser","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"lgBIrro7o5gy9CRCJoFe","title":"Caching","pathname":"/training/the-fine-tuning-process/hyperparameters/caching","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Hyperparameters"}]},{"id":"DjBg7tQQcTNcZ4z6Qw2M","title":"Training Processes","pathname":"/training/the-fine-tuning-process/training-processes","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"}]},{"id":"Au8SpBlbuWdgmFYjOE0A","title":"Extending the context window","pathname":"/training/the-fine-tuning-process/training-processes/extending-the-context-window","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"eob7NmRjg39qtHuvzeIX","title":"PyTorch Fully Sharded Data Parallel (FSDP)","pathname":"/training/the-fine-tuning-process/training-processes/pytorch-fully-sharded-data-parallel-fsdp","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"rNkkrXyJKuSyeXyzTpvP","title":"Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation","pathname":"/training/the-fine-tuning-process/training-processes/train-short-test-long-attention-with-linear-biases-enables-input-length-extrapolation","siteSpaceId":"sitesp_3lJdL","description":"Ofir Press, Noah A. Smith, Mike Lewis","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"czkmTjBG4eg4DvqPQVvy","title":"YaRN: Efficient Context Window Extension of Large Language Models","pathname":"/training/the-fine-tuning-process/training-processes/yarn-efficient-context-window-extension-of-large-language-models","siteSpaceId":"sitesp_3lJdL","description":"Nous Research, EleutherAI, University of Geneva","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"BLxjWZxwY94VMttckkv7","title":"Sliding Window Attention","pathname":"/training/the-fine-tuning-process/training-processes/sliding-window-attention","siteSpaceId":"sitesp_3lJdL","description":"Iz Beltagy, Matthew E. Peters, and Arman Cohan","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"dFgxAwHpDqkD0xodNvQb","title":"LongRoPE","pathname":"/training/the-fine-tuning-process/training-processes/longrope","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"xVQlIfpTlL23aleLZEyb","title":"Reinforcement Learning","pathname":"/training/the-fine-tuning-process/training-processes/reinforcement-learning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"lCeAEsdZDMdOFldcdicW","title":"An introduction to reinforcement learning","pathname":"/training/the-fine-tuning-process/training-processes/an-introduction-to-reinforcement-learning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"g0Pi3AUQYWA3l7kYt0Si","title":"Reinforcement Learning from Human Feedback (RLHF)","pathname":"/training/the-fine-tuning-process/training-processes/reinforcement-learning-from-human-feedback-rlhf","siteSpaceId":"sitesp_3lJdL","description":"Most often useful when creating domain specific models","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"5hlHwOnqjgM9AXrr6Wn0","title":"Direct Preference Optimization: Your Language Model is Secretly a Reward Model","pathname":"/training/the-fine-tuning-process/training-processes/direct-preference-optimization-your-language-model-is-secretly-a-reward-model","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Training"},{"label":"The Fine Tuning Process"},{"label":"Training Processes"}]},{"id":"B1uWAZpNREne4Yr81nFY","title":"Why is inference important?","pathname":"/inference/why-is-inference-important","siteSpaceId":"sitesp_3lJdL","description":"Speed and cost counts","breadcrumbs":[{"label":"INFERENCE"}]},{"id":"kdgkwm0p3zN9L9vXjl0O","title":"Grouped Query Attention","pathname":"/inference/why-is-inference-important/grouped-query-attention","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"foYt6ivlGDFYhJSRaC4h","title":"Key Value Cache","pathname":"/inference/why-is-inference-important/key-value-cache","siteSpaceId":"sitesp_3lJdL","description":"Managing model memory usage","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"2UHfC1o0o6XQREwhCWao","title":"Flash Attention","pathname":"/inference/why-is-inference-important/flash-attention","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"uCARImCMk0HfKBscoP3x","title":"Flash Attention 2","pathname":"/inference/why-is-inference-important/flash-attention-2","siteSpaceId":"sitesp_3lJdL","description":"The seminal July 2023 paper","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"eHHlZELvEZHKJEUtStWP","title":"StreamingLLM","pathname":"/inference/why-is-inference-important/streamingllm","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"INRhjNqmKI3DiSV8bDVb","title":"Paged Attention and vLLM","pathname":"/inference/why-is-inference-important/paged-attention-and-vllm","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"wnjTi4issHLkO3ma44d1","title":"TensorRT-LLM","pathname":"/inference/why-is-inference-important/tensorrt-llm","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"BXWAcULo1XyVF4lnAiaH","title":"Torchscript","pathname":"/inference/why-is-inference-important/torchscript","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"KDteXiNc7PIWpcSxmnOd","title":"NVIDIA L40S GPU","pathname":"/inference/why-is-inference-important/nvidia-l40s-gpu","siteSpaceId":"sitesp_3lJdL","description":"Low cost inference","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"x88QP82kXqtcgDsoFAiN","title":"Triton Inference Server - Introduction","pathname":"/inference/why-is-inference-important/triton-inference-server-introduction","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"yDkcA8BKCtoHUuonmeDr","title":"Triton Inference Server","pathname":"/inference/why-is-inference-important/triton-inference-server","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"gGGSaawBtZhOjkQR20Da","title":"FiDO: Fusion-in-Decoder optimised for stronger performance and faster inference","pathname":"/inference/why-is-inference-important/fido-fusion-in-decoder-optimised-for-stronger-performance-and-faster-inference","siteSpaceId":"sitesp_3lJdL","description":"Google Research, December 2022","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"C3EOT8qISadEDwdFsrx7","title":"Is PUE a useful measure of data centre performance?","pathname":"/inference/why-is-inference-important/is-pue-a-useful-measure-of-data-centre-performance","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"H9xkYCJ6wxObmfzh1MWt","title":"SLORA","pathname":"/inference/why-is-inference-important/slora","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"INFERENCE"},{"label":"Why is inference important?"}]},{"id":"Tkl8QsbL8ufiPOAN4xoq","title":"Vector Databases","pathname":"/knowledge/vector-databases","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"}]},{"id":"lmNSQZreAUXrhAvpLorH","title":"A Comprehensive Survey on Vector Databases","pathname":"/knowledge/vector-databases/a-comprehensive-survey-on-vector-databases","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"J9RjYEW7M8JLqxfEoASu","title":"Vector database management systems: Fundamental concepts, use-cases, and current challenges","pathname":"/knowledge/vector-databases/vector-database-management-systems-fundamental-concepts-use-cases-and-current-challenges","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"jpp9kyOujTXqfbTMonx8","title":"Using the Output Embedding to Improve Language Models","pathname":"/knowledge/vector-databases/using-the-output-embedding-to-improve-language-models","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"WW9IVSBov9LDvzJcz2Xc","title":"Decoding Sentence-BERT","pathname":"/knowledge/vector-databases/decoding-sentence-bert","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"dLrj1SZXzAvgPOzuJJJg","title":"ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT","pathname":"/knowledge/vector-databases/colbert-efficient-and-effective-passage-search-via-contextualized-late-interaction-over-bert","siteSpaceId":"sitesp_3lJdL","description":"The widely cited paper by Omar Khattab and Matei Zaharia from Stanford University","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"S3AQpFATKrMW5B2hEk84","title":"SimCSE: Simple Contrastive Learning of Sentence Embeddings","pathname":"/knowledge/vector-databases/simcse-simple-contrastive-learning-of-sentence-embeddings","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"SdLx0Gfm3s5bWy6YjlDE","title":"Questions Are All You Need to Train a Dense Passage Retriever","pathname":"/knowledge/vector-databases/questions-are-all-you-need-to-train-a-dense-passage-retriever","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"ENHbPznVJaPVyPWO3EuL","title":"Improving Text Embeddings with Large Language Models","pathname":"/knowledge/vector-databases/improving-text-embeddings-with-large-language-models","siteSpaceId":"sitesp_3lJdL","description":"Microsoft Corporation","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"3qKqPjayFchGQ2iD70EU","title":"Massive Text Embedding Benchmark","pathname":"/knowledge/vector-databases/massive-text-embedding-benchmark","siteSpaceId":"sitesp_3lJdL","description":"The leaderboard for embedding models","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"LiYu6IsS83zJpirZz3Qw","title":"RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking","pathname":"/knowledge/vector-databases/rocketqav2-a-joint-training-method-for-dense-passage-retrieval-and-passage-re-ranking","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"97VYwj3fvhRzg8gA9cj3","title":"LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders","pathname":"/knowledge/vector-databases/llm2vec-large-language-models-are-secretly-powerful-text-encoders","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"AGFwdiAT9bTr0k6QiCPC","title":"Embedding and Fine-Tuning in Neural Language Models","pathname":"/knowledge/vector-databases/embedding-and-fine-tuning-in-neural-language-models","siteSpaceId":"sitesp_3lJdL","description":"Mathematical representations of text","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"gOsC7f3mfJDRxvwN51Xp","title":"Embedding Model Construction","pathname":"/knowledge/vector-databases/embedding-model-construction","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"qT3058hKbodn5jYKas1d","title":"Demystifying Embedding Spaces using Large Language Models","pathname":"/knowledge/vector-databases/demystifying-embedding-spaces-using-large-language-models","siteSpaceId":"sitesp_3lJdL","description":"Guy Tennenholtz et al. from Google Research","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"ayx2JRMtAGWzMreDjfFm","title":"Fine-Tuning Llama for Multi-Stage Text Retrieval","pathname":"/knowledge/vector-databases/fine-tuning-llama-for-multi-stage-text-retrieval","siteSpaceId":"sitesp_3lJdL","description":"Microsoft Research","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"AkoqjAXgBua5vBKtvCGk","title":"Large Language Model Based Text Augmentation Enhanced Personality Detection Model","pathname":"/knowledge/vector-databases/large-language-model-based-text-augmentation-enhanced-personality-detection-model","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"GwSWxAqEulfREEsrKJv0","title":"One Embedder, Any Task: Instruction-Finetuned Text Embeddings","pathname":"/knowledge/vector-databases/one-embedder-any-task-instruction-finetuned-text-embeddings","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"ya3bPtMPUhvKXLxiXNPO","title":"Vector Databases are not the only solution","pathname":"/knowledge/vector-databases/vector-databases-are-not-the-only-solution","siteSpaceId":"sitesp_3lJdL","description":"Yingjun Wu","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"Hb5EBYfBReyQF8J53Wx6","title":"Knowledge Graphs","pathname":"/knowledge/vector-databases/knowledge-graphs","siteSpaceId":"sitesp_3lJdL","description":"Analysis of Knowledge Graphs and Influence of Generative AI and LLMs","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"qP780RVNM2mRL3MUmhjc","title":"Harnessing Knowledge Graphs to Elevate AI: A Technical Exploration","pathname":"/knowledge/vector-databases/knowledge-graphs/harnessing-knowledge-graphs-to-elevate-ai-a-technical-exploration","siteSpaceId":"sitesp_3lJdL","description":"Finally a semantic data architecture","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"},{"label":"Knowledge Graphs"}]},{"id":"fLfW1lLyX3NFRHzwxTuJ","title":"Unifying Large Language Models and Knowledge Graphs: A Roadmap","pathname":"/knowledge/vector-databases/knowledge-graphs/unifying-large-language-models-and-knowledge-graphs-a-roadmap","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"},{"label":"Knowledge Graphs"}]},{"id":"wL447KhFBWqF8Ampq9ER","title":"Approximate Nearest Neighbor (ANN)","pathname":"/knowledge/vector-databases/approximate-nearest-neighbor-ann","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"ibxND2PEQu3iErBOx0gx","title":"High Dimensional Data","pathname":"/knowledge/vector-databases/high-dimensional-data","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"QnwKdwrYuPoBMaLLhp5r","title":"Principal Component Analysis (PCA)","pathname":"/knowledge/vector-databases/principal-component-analysis-pca","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"B9vAuJU12XcRBX9wGhDk","title":"Vector Similarity Search - HNSW","pathname":"/knowledge/vector-databases/vector-similarity-search-hnsw","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"EX9ekG9BtvepJQp4JLoc","title":"FAISS (Facebook AI Similarity Search)","pathname":"/knowledge/vector-databases/faiss-facebook-ai-similarity-search","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"axexZv8gDIURiw5RvBx7","title":"Unsupervised Dense Retrievers","pathname":"/knowledge/vector-databases/unsupervised-dense-retrievers","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Vector Databases"}]},{"id":"MFNPgly3lfrotMKSoJjn","title":"Retrieval Augmented Generation","pathname":"/knowledge/retrieval-augmented-generation","siteSpaceId":"sitesp_3lJdL","description":"A critical piece of the generative AI infrastructure","breadcrumbs":[{"label":"KNOWLEDGE"}]},{"id":"4UfvRul1jGk0G58Q5sBp","title":"Retrieval-Augmented Generation for Large Language Models: A Survey","pathname":"/knowledge/retrieval-augmented-generation/retrieval-augmented-generation-for-large-language-models-a-survey","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"aX0FWQiDJdBRK2jQKLlM","title":"Fine-Tuning or Retrieval?","pathname":"/knowledge/retrieval-augmented-generation/fine-tuning-or-retrieval","siteSpaceId":"sitesp_3lJdL","description":"Microsoft, Israel","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"6UHHIRpwFLPQDaNznoUZ","title":"Revolutionising Information Retrieval: The Power of RAG in Language Models","pathname":"/knowledge/retrieval-augmented-generation/revolutionising-information-retrieval-the-power-of-rag-in-language-models","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"QUAq4ZYuav7FK98WeKqD","title":"A Survey on Retrieval-Augmented Text Generation","pathname":"/knowledge/retrieval-augmented-generation/a-survey-on-retrieval-augmented-text-generation","siteSpaceId":"sitesp_3lJdL","description":"Huayang Li, Yixuan Su, Deng Cai, Yan Wang, Lemao Liu","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"LiLHEwjPJ7X1UQkwsSKc","title":"REALM: Retrieval-Augmented Language Model Pre-Training","pathname":"/knowledge/retrieval-augmented-generation/realm-retrieval-augmented-language-model-pre-training","siteSpaceId":"sitesp_3lJdL","description":"Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, Ming-Wei Chang","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"NaDKjg6PVMbVplszQQoJ","title":"Retrieve Anything To Augment Large Language Models","pathname":"/knowledge/retrieval-augmented-generation/retrieve-anything-to-augment-large-language-models","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"cc0NqVO2ujU7PXusudIu","title":"Generate Rather Than Retrieve: Large Language Models Are Strong Context Generators","pathname":"/knowledge/retrieval-augmented-generation/generate-rather-than-retrieve-large-language-models-are-strong-context-generators","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"BGBnHwWgZPAUUiYxwf0Q","title":"Active Retrieval Augmented Generation","pathname":"/knowledge/retrieval-augmented-generation/active-retrieval-augmented-generation","siteSpaceId":"sitesp_3lJdL","description":"FLARE!","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"waw9uAzoBHR1ffHMHu2z","title":"DSPy: LM Assertions: Enhancing Language Model Pipelines with Computational Constraints","pathname":"/knowledge/retrieval-augmented-generation/dspy-lm-assertions-enhancing-language-model-pipelines-with-computational-constraints","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"dUUrhrziTDVgNFpa2d6i","title":"DSPy: Compiling Declarative Language Model Calls","pathname":"/knowledge/retrieval-augmented-generation/dspy-compiling-declarative-language-model-calls","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"Rabxu6s8RY8xiN0V8PXd","title":"DSPy: In-Context Learning for Extreme Multi-Label Classification","pathname":"/knowledge/retrieval-augmented-generation/dspy-in-context-learning-for-extreme-multi-label-classification","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"xXK771GILQEo1mtMEYJu","title":"Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs","pathname":"/knowledge/retrieval-augmented-generation/optimizing-instructions-and-demonstrations-for-multi-stage-language-model-programs","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"kR5mHFNXPBurW15buoEd","title":"HYDE: Revolutionising Search with Hypothetical Document Embeddings","pathname":"/knowledge/retrieval-augmented-generation/hyde-revolutionising-search-with-hypothetical-document-embeddings","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"l9Kej8uUvpBP6ZuUF6pH","title":"Enhancing Recommender Systems with Large Language Model Reasoning Graphs","pathname":"/knowledge/retrieval-augmented-generation/enhancing-recommender-systems-with-large-language-model-reasoning-graphs","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"vbMRVqV9V3f0AM8FIkru","title":"Retrieval Augmented Generation (RAG) versus fine tuning","pathname":"/knowledge/retrieval-augmented-generation/retrieval-augmented-generation-rag-versus-fine-tuning","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"rfTbOT1Lb9pTqQVbdPCf","title":"RAFT: Adapting Language Model to Domain Specific RAG","pathname":"/knowledge/retrieval-augmented-generation/raft-adapting-language-model-to-domain-specific-rag","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"7a58lUbYrVsIEAqGuorc","title":"Summarisation Methods and RAG","pathname":"/knowledge/retrieval-augmented-generation/summarisation-methods-and-rag","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"yNiGywGNjh1yxtmyQVFz","title":"Lessons Learned on LLM RAG Solutions","pathname":"/knowledge/retrieval-augmented-generation/lessons-learned-on-llm-rag-solutions","siteSpaceId":"sitesp_3lJdL","description":"Injecting data via embedding model into your vector database for future retrieval","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"wIPztSY4TjyUvzWsBwZc","title":"Stanford: Retrieval Augmented Language Models","pathname":"/knowledge/retrieval-augmented-generation/stanford-retrieval-augmented-language-models","siteSpaceId":"sitesp_3lJdL","description":"Youtube Lecture - January 2024","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"bm68a1uQA9ADFpmwmI6B","title":"Overview of RAG Approaches with Vector Databases","pathname":"/knowledge/retrieval-augmented-generation/overview-of-rag-approaches-with-vector-databases","siteSpaceId":"sitesp_3lJdL","description":"Some random tips on RAG and vector databases","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"lMQGQ6xYu4dsax8ramJ9","title":"Mastering Chunking in Retrieval-Augmented Generation (RAG) Systems","pathname":"/knowledge/retrieval-augmented-generation/mastering-chunking-in-retrieval-augmented-generation-rag-systems","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"},{"label":"Retrieval Augmented Generation"}]},{"id":"dp16KKiDbcru77JfvVmG","title":"Semantic Routing","pathname":"/knowledge/semantic-routing","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"}]},{"id":"hV3Po2l1azaDsYx56tVP","title":"Resource Description Framework (RDF)","pathname":"/knowledge/resource-description-framework-rdf","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"KNOWLEDGE"}]},{"id":"23PkBCuPtpFF4ORNMm3n","title":"What is agency?","pathname":"/agents/what-is-agency","siteSpaceId":"sitesp_3lJdL","description":"Philosophical Origins of the Agent Concept","breadcrumbs":[{"label":"AGENTS"}]},{"id":"HPVn19qkjSb8xgio58r1","title":"Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves","pathname":"/agents/what-is-agency/rephrase-and-respond-let-large-language-models-ask-better-questions-for-themselves","siteSpaceId":"sitesp_3lJdL","description":"Yihe Deng, Weitong Zhang, Zixiang Chen, Quanquan Gu (University of California, Los Angeles)","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"wnoAt73y2r62QRaynVKY","title":"Types of Agents","pathname":"/agents/what-is-agency/types-of-agents","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"AgetBF93aI49BNjVQWT3","title":"The risk of AI agency","pathname":"/agents/what-is-agency/the-risk-of-ai-agency","siteSpaceId":"sitesp_3lJdL","description":"The concept of AI agency is powerful, but comes with an array of risks","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"g6S9SqRcx9kgmsIGgkAd","title":"Understanding Personality in Large Language Models: A New Frontier in AI Psychology","pathname":"/agents/what-is-agency/understanding-personality-in-large-language-models-a-new-frontier-in-ai-psychology","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"40hWCivOjVA1d2fGfKnz","title":"AI Agents - Reasoning, Planning, and Tool Calling","pathname":"/agents/what-is-agency/ai-agents-reasoning-planning-and-tool-calling","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"KAGk0nTdLX49bDdkYg99","title":"Personality and Brand","pathname":"/agents/what-is-agency/personality-and-brand","siteSpaceId":"sitesp_3lJdL","description":"How can we create large language models that have personality?","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"FYvihlGHeXzwl1eoHf7P","title":"Agent Interaction via APIs","pathname":"/agents/what-is-agency/agent-interaction-via-apis","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"zMlhyRsgrtHUhhuFNNlh","title":"Bridging Minds and Machines: The Legacy of Newell, Shaw, and Simon","pathname":"/agents/what-is-agency/bridging-minds-and-machines-the-legacy-of-newell-shaw-and-simon","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"5YioX7iflgOe5OMP44xA","title":"A Survey on Language Model based Autonomous Agents","pathname":"/agents/what-is-agency/a-survey-on-language-model-based-autonomous-agents","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"3oWM1HQmfswi0JeHbsRP","title":"Large Language Models as Agents","pathname":"/agents/what-is-agency/large-language-models-as-agents","siteSpaceId":"sitesp_3lJdL","description":"The Rise and Potential of Large Language Model Based Agents : A Survey","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"XGZKrIWwgT0fbbmJ1eeA","title":"AI Reasoning: A Deep Dive into Chain-of-Thought Prompting","pathname":"/agents/what-is-agency/ai-reasoning-a-deep-dive-into-chain-of-thought-prompting","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"WWTByTbJpTHJoaKH9XGO","title":"Enhancing AI Reasoning with Self-Taught Reasoner (STaR)","pathname":"/agents/what-is-agency/enhancing-ai-reasoning-with-self-taught-reasoner-star","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"i8EWVtg7mOtmEdRnfKro","title":"Exploring the Frontier of AI: The \"Tree of Thoughts\" Framework","pathname":"/agents/what-is-agency/exploring-the-frontier-of-ai-the-tree-of-thoughts-framework","siteSpaceId":"sitesp_3lJdL","description":"December 2023 paper","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"PuqRKoyRofoJLgr0Gqva","title":"Toolformer: Revolutionising Language Models with API Integration - An Analysis","pathname":"/agents/what-is-agency/toolformer-revolutionising-language-models-with-api-integration-an-analysis","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"2N8NFUylJkP7eBRFNPyp","title":"TaskMatrix.AI: Bridging Foundational AI Models with Specialised Systems for Enhanced Task Completion","pathname":"/agents/what-is-agency/taskmatrix.ai-bridging-foundational-ai-models-with-specialised-systems-for-enhanced-task-completion","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"fjTpQ1nX20fmjueIVLmR","title":"Unleashing the Power of LLMs in API Integration: The Rise of Gorilla","pathname":"/agents/what-is-agency/unleashing-the-power-of-llms-in-api-integration-the-rise-of-gorilla","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"rAMWalDJ1LM1KirPqzvm","title":"Andrew Ng's presentation on AI agents","pathname":"/agents/what-is-agency/andrew-ngs-presentation-on-ai-agents","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"stASk9og9J3xSbVKrqvf","title":"Making AI accessible with Andrej Karpathy and Stephanie Zhan","pathname":"/agents/what-is-agency/making-ai-accessible-with-andrej-karpathy-and-stephanie-zhan","siteSpaceId":"sitesp_3lJdL","description":"","breadcrumbs":[{"label":"AGENTS"},{"label":"What is agency?"}]},{"id":"71ManzFBBRAfQdejnq7c","title":"Regulation and Ethics","pathname":"/regulation-and-ethics/regulation-and-ethics","siteSpaceId":"sitesp_3lJdL","description":"December 2021 paper","breadcrumbs":[{"label":"Regulation and Ethics"}]},{"id":"U7t57vP5IRYEX4dB9GIu","title":"Privacy","pathname":"/regulation-and-ethics/regulation-and-ethics/privacy","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"WhN5VddHRjfNE673aee4","title":"Detecting AI Generated content","pathname":"/regulation-and-ethics/regulation-and-ethics/detecting-ai-generated-content","siteSpaceId":"sitesp_3lJdL","description":"Ghostbuster - AI content detector","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"9TUYNXFplmTI05bJqJ2K","title":"Navigating the IP Maze in AI: The Convergence of Blockchain, Web 3.0, and LLMs","pathname":"/regulation-and-ethics/regulation-and-ethics/navigating-the-ip-maze-in-ai-the-convergence-of-blockchain-web-3.0-and-llms","siteSpaceId":"sitesp_3lJdL","description":"\"Authorguard\" - protecting content producers commercial interests","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"6HCHE02EcE9UT2wlEN6t","title":"Adverse Reactions to generative AI","pathname":"/regulation-and-ethics/regulation-and-ethics/adverse-reactions-to-generative-ai","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"te8TXViKsyy3xbxpbGgw","title":"Navigating the Ethical Minefield: The Challenge of Security in Large Language Models","pathname":"/regulation-and-ethics/regulation-and-ethics/navigating-the-ethical-minefield-the-challenge-of-security-in-large-language-models","siteSpaceId":"sitesp_3lJdL","description":"","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"pYckwVWSkG7QUXPDOuHs","title":"Navigating the Uncharted Waters: The Risks of Autonomous AI in Military Decision-Making","pathname":"/regulation-and-ethics/regulation-and-ethics/navigating-the-uncharted-waters-the-risks-of-autonomous-ai-in-military-decision-making","siteSpaceId":"sitesp_3lJdL","description":"AI models seem to tend towards escalation...","breadcrumbs":[{"label":"Regulation and Ethics"},{"label":"Regulation and Ethics"}]},{"id":"pbWF4jBs5DMZro3rJ7ic","title":"Data Architecture","pathname":"/disruption/data-architecture","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"}]},{"id":"ssHpCnXUgI508dfMoTDl","title":"What is a data pipeline?","pathname":"/disruption/data-architecture/what-is-a-data-pipeline","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Data Architecture"}]},{"id":"hOmEWWl8FDf5C9hT8yQF","title":"What is Reverse ETL?","pathname":"/disruption/data-architecture/what-is-reverse-etl","siteSpaceId":"sitesp_3lJdL","description":"Combining generative AI with Reverse ETL in Modern Businesses","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Data Architecture"}]},{"id":"ZwHch1JMh0qoQ5l0ca1d","title":"Unstructured Data and Generatve AI","pathname":"/disruption/data-architecture/unstructured-data-and-generatve-ai","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Data Architecture"}]},{"id":"MWbfWfX9g3neu3zQ1p2q","title":"Resource Description Framework (RDF)","pathname":"/disruption/data-architecture/resource-description-framework-rdf","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Data Architecture"}]},{"id":"JDFv4K3qiza1EJgj7yWS","title":"Integrating generative AI with the Semantic Web","pathname":"/disruption/data-architecture/integrating-generative-ai-with-the-semantic-web","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Data Architecture"}]},{"id":"P6J6Cd1yOazv1oF4oPKk","title":"Search","pathname":"/disruption/search","siteSpaceId":"sitesp_3lJdL","description":"Generative AI and its relationship with search","breadcrumbs":[{"label":"DISRUPTION"}]},{"id":"XUaM0r6iDVoBlrUz76zg","title":"BM25 - Search Engine Ranking Function","pathname":"/disruption/search/bm25-search-engine-ranking-function","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"0cFnCZZyRRWZtaCCPdhd","title":"BERT as a reranking engine","pathname":"/disruption/search/bert-as-a-reranking-engine","siteSpaceId":"sitesp_3lJdL","description":"Retrieval and Reranking","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"JkWrEW6CYbV5GYYteAyn","title":"BERT and Google","pathname":"/disruption/search/bert-and-google","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"aBgLljhWTuORFNyltrKP","title":"Generative Engine Optimisation (GEO)","pathname":"/disruption/search/generative-engine-optimisation-geo","siteSpaceId":"sitesp_3lJdL","description":"Navigating the New Frontier: A Guide to Generative Engine Optimisation","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"XS6aEpZiFBcIdJx2asAC","title":"Billion-scale similarity search with GPUs","pathname":"/disruption/search/billion-scale-similarity-search-with-gpus","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"vTEs3U4XRlPyMftu6qk1","title":"FOLLOWIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions","pathname":"/disruption/search/followir-evaluating-and-teaching-information-retrieval-models-to-follow-instructions","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"2ucpFvQ5F8XCiuO35YOs","title":"Neural Collaborative Filtering","pathname":"/disruption/search/neural-collaborative-filtering","siteSpaceId":"sitesp_3lJdL","description":"The highly popular 2017 paper that drove the advance of recommendation systems","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"Hd5e1GCBLZJ5IoWw7pes","title":"Federated Neural Collaborative Filtering","pathname":"/disruption/search/federated-neural-collaborative-filtering","siteSpaceId":"sitesp_3lJdL","description":"Collabative Filtering - Matching Consumers with Products and Services with Privacy","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"NWn0DmWPM08ukmuglWg9","title":"Latent Space versus Embedding Space","pathname":"/disruption/search/latent-space-versus-embedding-space","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"vJJ6lWaiyJAR7BMrL3R4","title":"Improving Text Embeddings with Large Language Models","pathname":"/disruption/search/improving-text-embeddings-with-large-language-models","siteSpaceId":"sitesp_3lJdL","description":"Liang Wang and the Microsoft team","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Search"}]},{"id":"DN1tfJ12TLewMSGWvz7b","title":"Recommendation Engines","pathname":"/disruption/recommendation-engines","siteSpaceId":"sitesp_3lJdL","description":"There is major disruption coming to the recommendation engine industry","breadcrumbs":[{"label":"DISRUPTION"}]},{"id":"vIG3pefSXiGstMkmSFWk","title":"On Interpretation and Measurement of Soft Attributes for Recommendation","pathname":"/disruption/recommendation-engines/on-interpretation-and-measurement-of-soft-attributes-for-recommendation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"0vUGmjHKjnNNCvA5WJ9R","title":"A Survey on Large Language Models for Recommendation","pathname":"/disruption/recommendation-engines/a-survey-on-large-language-models-for-recommendation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"xy5gCsgFmZF3LkN5J9fZ","title":"Model driven recommendation systems","pathname":"/disruption/recommendation-engines/model-driven-recommendation-systems","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"wFBIk03Dv59Ink9B4JDz","title":"Recommender AI Agent: Integrating Large Language Models for Interactive Recommendations","pathname":"/disruption/recommendation-engines/recommender-ai-agent-integrating-large-language-models-for-interactive-recommendations","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"Mbs6CXTPUD6DpmVz3biG","title":"Foundation Models for Recommender Systems","pathname":"/disruption/recommendation-engines/foundation-models-for-recommender-systems","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"kYDwKWineiu2OPnO5wJA","title":"Exploring the Impact of Large Language Models on Recommender Systems: An Extensive Review","pathname":"/disruption/recommendation-engines/exploring-the-impact-of-large-language-models-on-recommender-systems-an-extensive-review","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"bRZf4J7mty8EahQHQltm","title":"AI driven recommendations - harming autonomy?","pathname":"/disruption/recommendation-engines/ai-driven-recommendations-harming-autonomy","siteSpaceId":"sitesp_3lJdL","description":"\"Artificial intelligence vs. autonomous decision-making in streaming platforms: A mixed-method approach\" by Ana Rita Gonçalves, Diego Costa Pinto, Saleh Shuqair, Marlon Dalmoro, and Anna S. Mattila","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Recommendation Engines"}]},{"id":"bwjGybXV5io7qcsJXsB2","title":"Logging","pathname":"/disruption/logging","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"}]},{"id":"WdMdxTdNV76NxsIwafkV","title":"A Taxonomy of Anomalies in Log Data","pathname":"/disruption/logging/a-taxonomy-of-anomalies-in-log-data","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"5Wk9nLqsjT371CReLh9w","title":"Deeplog","pathname":"/disruption/logging/deeplog","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"sKGw7W4x7xuJlNCqA35I","title":"LogBERT: Log Anomaly Detection via BERT","pathname":"/disruption/logging/logbert-log-anomaly-detection-via-bert","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"3YYBj0urX7LEid4fnklU","title":"Experience Report: Deep Learning-based System Log Analysis for Anomaly Detection","pathname":"/disruption/logging/experience-report-deep-learning-based-system-log-analysis-for-anomaly-detection","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"h6hBe1yPxCREU0zoxiL0","title":"Log-based Anomaly Detection with Deep Learning: How Far Are We?","pathname":"/disruption/logging/log-based-anomaly-detection-with-deep-learning-how-far-are-we","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"gJHFx8fCtGXJN8ZSFly8","title":"Deep Learning for Anomaly Detection in Log Data: A Survey","pathname":"/disruption/logging/deep-learning-for-anomaly-detection-in-log-data-a-survey","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"gCq0wILzC5bOci2ZLPZu","title":"LogGPT","pathname":"/disruption/logging/loggpt","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"quf3URrNFuNRPp1GWhqV","title":"Adaptive Semantic Gate Networks (ASGNet) for log-based anomaly diagnosis","pathname":"/disruption/logging/adaptive-semantic-gate-networks-asgnet-for-log-based-anomaly-diagnosis","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"DISRUPTION"},{"label":"Logging"}]},{"id":"M2WG6znAaBmvlwQGVZde","title":"The modern data centre","pathname":"/infrastructure/the-modern-data-centre","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"OygwWOLYBDiiBCZpURja","title":"Enhancing Data Centre Efficiency: Strategies to Improve PUE","pathname":"/infrastructure/the-modern-data-centre/enhancing-data-centre-efficiency-strategies-to-improve-pue","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"The modern data centre"}]},{"id":"3rPjDoJTH6krMSNt7MdS","title":"TCO of NVIDIA GPUs and falling barriers to entry","pathname":"/infrastructure/the-modern-data-centre/tco-of-nvidia-gpus-and-falling-barriers-to-entry","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"The modern data centre"}]},{"id":"DbegucqRdD9FaWBXxO29","title":"Maximising GPU Utilisation with Kubernetes and NVIDIA GPU Operator","pathname":"/infrastructure/the-modern-data-centre/maximising-gpu-utilisation-with-kubernetes-and-nvidia-gpu-operator","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"The modern data centre"}]},{"id":"ofQE9oS5X1epBJCWMsNy","title":"Data Centres","pathname":"/infrastructure/the-modern-data-centre/data-centres","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"The modern data centre"}]},{"id":"n6cHBbDXyv9b10JPvzda","title":"Liquid Cooling","pathname":"/infrastructure/the-modern-data-centre/liquid-cooling","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"The modern data centre"}]},{"id":"JrUCnSWnkdEia8HxUo85","title":"Servers and Chips","pathname":"/infrastructure/servers-and-chips","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"QJgTX3pGlZ8zxUtydQne","title":"The NVIDIA H100 GPU","pathname":"/infrastructure/servers-and-chips/the-nvidia-h100-gpu","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"JoilzoOyI2QEs9DcV87Y","title":"NVIDIA H100 NVL","pathname":"/infrastructure/servers-and-chips/nvidia-h100-nvl","siteSpaceId":"sitesp_3lJdL","description":"For large language model inference workloads","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"frwotvSkvLTwBDAbPUWk","title":"Lambda Hyperplane 8-H100","pathname":"/infrastructure/servers-and-chips/lambda-hyperplane-8-h100","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"fBYAAsqEUcB5RqeQawbt","title":"NVIDIA DGX Servers","pathname":"/infrastructure/servers-and-chips/nvidia-dgx-servers","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"MfUXnATLhkCPpfJl7lek","title":"NVIDIA DGX-2","pathname":"/infrastructure/servers-and-chips/nvidia-dgx-2","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"9zxuX18dPxhYhtbhaINM","title":"NVIDIA DGX H-100 System","pathname":"/infrastructure/servers-and-chips/nvidia-dgx-h-100-system","siteSpaceId":"sitesp_3lJdL","description":"An absolute beast","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"lKS6dDNlmASTYcUk3Psd","title":"NVLink Switch","pathname":"/infrastructure/servers-and-chips/nvlink-switch","siteSpaceId":"sitesp_3lJdL","description":"Rapid Communication between GPUs","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"om3hBtsLKaj95PE90ehi","title":"Tensor Cores","pathname":"/infrastructure/servers-and-chips/tensor-cores","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"OXtxPtQfU2lLEQsplS8e","title":"NVIDIA Grace Hopper Superchip","pathname":"/infrastructure/servers-and-chips/nvidia-grace-hopper-superchip","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"ozgiU98Wi2gOfNPAQuvD","title":"NVIDIA Grace CPU Superchip","pathname":"/infrastructure/servers-and-chips/nvidia-grace-cpu-superchip","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"2KEPqePzALBNcCapfxvS","title":"NVIDIA GB200 NVL72","pathname":"/infrastructure/servers-and-chips/nvidia-gb200-nvl72","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"qO5hSfnfkHqF63wPvjWm","title":"Hopper versus Blackwell","pathname":"/infrastructure/servers-and-chips/hopper-versus-blackwell","siteSpaceId":"sitesp_3lJdL","description":"A comparison between the two latest GPU servers","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"dfVmKH6ugO1m2YKLVUmE","title":"HGX: High-Performance GPU Platforms","pathname":"/infrastructure/servers-and-chips/hgx-high-performance-gpu-platforms","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"PAIrnCckGeSZVBxkDbKP","title":"ARM Chips","pathname":"/infrastructure/servers-and-chips/arm-chips","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"0SA0OweHd8SGIFf1f2gD","title":"ARM versus x86","pathname":"/infrastructure/servers-and-chips/arm-versus-x86","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"hCS1loaDjxVYj9RA5lNu","title":"RISC versus CISC","pathname":"/infrastructure/servers-and-chips/risc-versus-cisc","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"2qhlXVf2sZQybJMo34Qx","title":"Introduction to RISC-V","pathname":"/infrastructure/servers-and-chips/introduction-to-risc-v","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Servers and Chips"}]},{"id":"cOdytZ1Ax1O1gHLRvG5j","title":"Networking and Connectivity","pathname":"/infrastructure/networking-and-connectivity","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"98vpxPrhBS9FD3fakXCR","title":"Infiniband versus Ethernet","pathname":"/infrastructure/networking-and-connectivity/infiniband-versus-ethernet","siteSpaceId":"sitesp_3lJdL","description":"Networking Technologies","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"pPPHKVcff11dv3hfhtzb","title":"NVIDIA Quantum InfiniBand","pathname":"/infrastructure/networking-and-connectivity/nvidia-quantum-infiniband","siteSpaceId":"sitesp_3lJdL","description":"Networking Solution","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"tzSOrRdprUJtIS8fRC9h","title":"PCIe (Peripheral Component Interconnect Express)","pathname":"/infrastructure/networking-and-connectivity/pcie-peripheral-component-interconnect-express","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"b2usngK4d1oKOr6WYjAG","title":"NVIDIA ConnectX InfiniBand adapters","pathname":"/infrastructure/networking-and-connectivity/nvidia-connectx-infiniband-adapters","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"lIlcKBo1leIgBPwkNGFb","title":"NVMe (Non-Volatile Memory Express)","pathname":"/infrastructure/networking-and-connectivity/nvme-non-volatile-memory-express","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"bTcqMXAnvFPCJxEzGeko","title":"NVMe over Fabrics (NVMe-oF)","pathname":"/infrastructure/networking-and-connectivity/nvme-over-fabrics-nvme-of","siteSpaceId":"sitesp_3lJdL","description":"A protocol that enables high-performance, low-latency access to shared storage resources over various network fabrics","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"CRULcHF2R7DkICBjNsGT","title":"NVIDIA Spectrum-X","pathname":"/infrastructure/networking-and-connectivity/nvidia-spectrum-x","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"dCWSKmbSPGKidVDvVVip","title":"NVIDIA GPUDirect","pathname":"/infrastructure/networking-and-connectivity/nvidia-gpudirect","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"FA4yuG9NMqeJzsBdICsv","title":"Evaluating Modern GPU Interconnect","pathname":"/infrastructure/networking-and-connectivity/evaluating-modern-gpu-interconnect","siteSpaceId":"sitesp_3lJdL","description":"Ang Li et al","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"kXUjZMKQUZJYH25jyAs7","title":"Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)","pathname":"/infrastructure/networking-and-connectivity/scalable-hierarchical-aggregation-and-reduction-protocol-sharp","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"9eweaimTdNX455H9zsdh","title":"Next-generation networking in AI environments","pathname":"/infrastructure/networking-and-connectivity/next-generation-networking-in-ai-environments","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"TZWKdJZINa78rd52zbkO","title":"NVIDIA Collective Communications Library (NCCL)","pathname":"/infrastructure/networking-and-connectivity/nvidia-collective-communications-library-nccl","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Networking and Connectivity"}]},{"id":"sV9tPToFgE5G8Vv7K2ap","title":"Data and Memory","pathname":"/infrastructure/data-and-memory","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"OogLavuXyAMiG5aBEoqn","title":"NVIDIA BlueField Data Processing Units (DPUs)","pathname":"/infrastructure/data-and-memory/nvidia-bluefield-data-processing-units-dpus","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"ELrN6ClhAsdcC3pSyicw","title":"Remote Direct Memory Access (RDMA)","pathname":"/infrastructure/data-and-memory/remote-direct-memory-access-rdma","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"j7Q7CoZNjbAwoICyAjQz","title":"High Bandwidth Memory (HBM3)","pathname":"/infrastructure/data-and-memory/high-bandwidth-memory-hbm3","siteSpaceId":"sitesp_3lJdL","description":"SK Hynix Inc","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"X2dnF5YgtnTfHIJaqUjd","title":"Flash Memory","pathname":"/infrastructure/data-and-memory/flash-memory","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"GFJJZhNDjZuJwEClDqRJ","title":"Model Requirements","pathname":"/infrastructure/data-and-memory/model-requirements","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"nrZLnUB4vtKXk3eLikVg","title":"Calculating GPU memory for serving LLMs","pathname":"/infrastructure/data-and-memory/calculating-gpu-memory-for-serving-llms","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"km95Y2a7S4L83LSCl2PT","title":"Transformer training costs","pathname":"/infrastructure/data-and-memory/transformer-training-costs","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"r2qEgjr61zI9rFTvD0xF","title":"GPU Performance Optimisation","pathname":"/infrastructure/data-and-memory/gpu-performance-optimisation","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Data and Memory"}]},{"id":"rOYDdKZUnzIBpXAKcO9j","title":"Libraries and Complements","pathname":"/infrastructure/libraries-and-complements","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"Efmo27lWQIYbHI5osa9W","title":"NVIDIA Base Command","pathname":"/infrastructure/libraries-and-complements/nvidia-base-command","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Libraries and Complements"}]},{"id":"XHE7Bn7F5YBh9CeLX46r","title":"NVIDIA AI Enterprise","pathname":"/infrastructure/libraries-and-complements/nvidia-ai-enterprise","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Libraries and Complements"}]},{"id":"8rZYt6Lr8GSoBTEIyWGt","title":"CUDA - NVIDIA GTC 2024 presentation","pathname":"/infrastructure/libraries-and-complements/cuda-nvidia-gtc-2024-presentation","siteSpaceId":"sitesp_3lJdL","description":"Steven Jones' presentation on CUDA at NVIDIA GTC 2024","breadcrumbs":[{"label":"Infrastructure"},{"label":"Libraries and Complements"}]},{"id":"kjiHVkCu5QK5uFBCYsND","title":"RAPIDs","pathname":"/infrastructure/libraries-and-complements/rapids","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Libraries and Complements"}]},{"id":"6FiicdF2xcGpaCIQ2pTT","title":"RAFT","pathname":"/infrastructure/libraries-and-complements/raft","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Libraries and Complements"}]},{"id":"6haRpsNaU9L7z8D8becT","title":"Vast Data Platform","pathname":"/infrastructure/vast-data-platform","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"ZSJ6f67Rm0HHTBPRX1fh","title":"Vast Datastore","pathname":"/infrastructure/vast-data-platform/vast-datastore","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Vast Data Platform"}]},{"id":"HSZtNLwDIeLF0f0yZmHo","title":"Vast Database","pathname":"/infrastructure/vast-data-platform/vast-database","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Vast Data Platform"}]},{"id":"AYH5cp9CxLoIa2sIlZPV","title":"Vast Data Engine","pathname":"/infrastructure/vast-data-platform/vast-data-engine","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Vast Data Platform"}]},{"id":"07aYy29wdrJB9MSHP4ZJ","title":"DASE (Disaggregated and Shared Everything)","pathname":"/infrastructure/vast-data-platform/dase-disaggregated-and-shared-everything","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Vast Data Platform"}]},{"id":"JS0WVkpiI1u4V5jQ93qU","title":"Dremio and VAST Data","pathname":"/infrastructure/vast-data-platform/dremio-and-vast-data","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Vast Data Platform"}]},{"id":"AJ76oe8M5VeR279VZCpG","title":"Storage","pathname":"/infrastructure/storage","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"}]},{"id":"KNF8N0u4yAnWkdpMsY6G","title":"WEKA: A High-Performance Storage Solution for AI Workloads","pathname":"/infrastructure/storage/weka-a-high-performance-storage-solution-for-ai-workloads","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Storage"}]},{"id":"4rV6lu1H6f6niq5VNwQ7","title":"Introduction to NVIDIA GPUDirect Storage (GDS)","pathname":"/infrastructure/storage/introduction-to-nvidia-gpudirect-storage-gds","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Storage"}]},{"id":"Wi8yiSBPw9YXFcuSp0mQ","title":"GDS cuFile API","pathname":"/infrastructure/storage/introduction-to-nvidia-gpudirect-storage-gds/gds-cufile-api","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Storage"},{"label":"Introduction to NVIDIA GPUDirect Storage (GDS)"}]},{"id":"21A2w2NXWgFZ0IyVZ1iu","title":"NVIDIA Magnum IO GPUDirect Storage (GDS)","pathname":"/infrastructure/storage/nvidia-magnum-io-gpudirect-storage-gds","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Storage"}]},{"id":"0U3gcIDzJrw26zJnT0Ol","title":"Vectors in Memory","pathname":"/infrastructure/storage/vectors-in-memory","siteSpaceId":"sitesp_3lJdL","breadcrumbs":[{"label":"Infrastructure"},{"label":"Storage"}]}]}