Fine Tuning Phi 1.5 using QLoRA on the Stanford Alpaca Dataset

Fine Tuning Phi 1.5 using QLoRA on the Stanford Alpaca Dataset

Quantized LoRA, more commonly known as QLoRA is a combination of quantization and Low Rank Adaptation for fine-tuning LLMs. Simply put, LoRa is a technique to adapt Large Language Models to specific tasks without making them forget their pretraining knowledge. In QLoRa, we load the pretrained model weights in quantized format, say 4-bit (INT4). However, the adapter (LoRA) layers are loaded in full precision, FP16 or FP32. This reduces the memory (GPU) consumption by a great extent making fine tuning possible on low resource hardware. To this end, in this article, we will be fine tuning the Phi 1.5 model…
Read More
Looking ahead to the AI Seoul Summit

Looking ahead to the AI Seoul Summit

How summits in Seoul, France and beyond can galvanize international cooperation on frontier AI safetyLast year, the UK Government hosted the first major global Summit on frontier AI safety at Bletchley Park. It focused the world’s attention on rapid progress at the frontier of AI development and delivered concrete international action to respond to potential future risks, including the Bletchley Declaration; new AI Safety Institutes; and the International Scientific Report on Advanced AI Safety.Six months on from Bletchley, the international community has an opportunity to build on that momentum and galvanize further global cooperation at this week’s AI Seoul Summit.…
Read More
Using and Finetuning Pretrained Transformers

Using and Finetuning Pretrained Transformers

This week has been filled with developments, including exciting new AI research that I’ll be discussing in my usual end-of-month write-ups.Additionally, I am excited to announce the release of my new book, Machine Learning Q and AI, published by No Starch Press.If you’ve been searching for a resource following an introductory machine learning course, this might be the one. I’m covering 30 concepts that were slightly out of scope for the previous books and courses I’ve taught, and I’ve compiled them here in a concise question-and-answer format (including exercises).I believe it will also serve as a useful companion for preparing…
Read More
Microsoft unveils new Copilot+ PCs featuring AI capabilities

Microsoft unveils new Copilot+ PCs featuring AI capabilities

Microsoft has launched a new type of personal computer designed with AI capabilities, underscoring the company’s commitment to integrating cutting-edge technology across its product lineup.  This new category not only boosts Microsoft’s ventures but also positions it to compete directly with AI-led advancements from tech giants such as Alphabet and Apple. During an event at Microsoft’s campus in Redmond, Washington, CEO Satya Nadella unveiled the Copilot+ PCs. These computers, developed in collaboration with manufacturers like Acer and Asustek Computer, align with Microsoft’s stock prices reaching near all-time highs amid Wall Street’s expectations that AI will significantly boost earnings for major tech firms. These PCs are engineered to manage a greater share of…
Read More
Google Search’s New AI Overviews Will Soon Have Ads

Google Search’s New AI Overviews Will Soon Have Ads

Last week Google introduced a radical shake-up of search that presents users with AI-generated answers to their queries. Now the company says it will soon start including ads inside those AI Overviews, as the automatic answers are called.Google on Tuesday announced plans to test search and shopping ads in the AI summaries, a move that could extend its dominance in search advertising into a new era. Although Google rapidly rolled out AI Overviews to all US English users last week after announcing the feature at its I/O developer conference, it’s unclear how widely or quickly ads will start appearing.Screenshots released…
Read More
Informatica CEO: Good Data Management Not Optional for AI

Informatica CEO: Good Data Management Not Optional for AI

(greenbutterfly/Shutterstock) The big data era may have started a decade-and-a-half ago, but for many companies, it’s the current AI revolution that’s forcing them to finally get serious about data management, says Informatica CEO Amit Walia. “What is AI without good quality data?” he says. Data is the foundation for a host of corporate efforts these days, and that realization is leading many companies to renew their interest in establishing a comprehensive data management strategy, Walia told Datanami last week in advanced of Informatica World, which takes place in Las Vegas this week. “The driver is all of these digital initiatives…
Read More
Microsoft announces Copilot+ PCs and AI-powered Recall feature – gHacks Tech News

Microsoft announces Copilot+ PCs and AI-powered Recall feature – gHacks Tech News

On a special event at Microsoft Campus, Microsoft unveiled Copilot+ PCs officially. This new type of Windows PCs, formerly known as AI PCs, mark the first step into introducing AI capabilities in Windows devices. Much of what Microsoft revealed on Monday was already known through unverified leaks. The first batch of Copilot+ PCs are powered by Qualcomm processors and not Intel or AMD silicon. These will come later this year though. As far as requirements are concerned, these match the leaks: at least 16GB of RAM. at least 256GB SSD storage. Integrated NPU. No word on the Copilot key requirement…
Read More
Google Taps AI to Show Shoppers How Clothes Fit Different Bodies

Google Taps AI to Show Shoppers How Clothes Fit Different Bodies

One of the new ad formats Google announced today will allow brands to link short-form videos they made—or ones they hired creators to film—to their advertisements in Google’s search engine. AI-generated text summaries of the clips will be included below. “I’ve got three Gen Z-ers at home, and watching them shop, it’s very video-based,” said Madrigal.Google also launched a tool that allows companies to create entirely new, AI-generated product images based on photos from earlier marketing campaigns and pictures that represent their brand identity. For example, a home goods brand could upload a picture of one of its candles and…
Read More
Palantir’s Military AI Tech Conference Sounds Absolutely Terrifying

Palantir’s Military AI Tech Conference Sounds Absolutely Terrifying

Earlier this month, the military and intelligence-tied data monger Palantir — alongside big tech giants Google and Microsoft — sponsored DC's inaugural "AI Expo for National Competitiveness."Which, judging by The Guardian's summary of the event, sounds like it was an Orwellian nightmare fever dream to end all other military-industrial conferences.The relationship between Silicon Valley and the US military has experienced a bit of a rekindling in recent years — a renaissance in large part pushed forward in by the collision of advancements in AI, robotics, various autonomous weapons and vehicles, and surveillance. If anything, Palantir's recent event was a celebration of this…
Read More
Pile-T5

Pile-T5

The T5 model (Raffel et al, 2019) is widely used in the NLP community. Its base model has been downloaded from Hugging Face millions of times, leaving no doubt that these models are a favorite of the community. However, T5's tokenizer omits important code-related tokens and subsequent pretraining datasets have been released with higher quality filtering and more diverse domains. In this blog post, we introduce a new version of T5 intended to address those weaknesses: Pile-T5, trained on the Pile (Gao et al, 2020) and using the LLaMA tokenizer (Touvron et al, 2023). Model Description# Our alternative version replaces…
Read More
No widgets found. Go to Widget page and add the widget in Offcanvas Sidebar Widget Area.