experiment

Header

Footer

PreFooter

SectionTitle

PaginationButton

BlogPage

Trusted by world class companies, Scale delivers high quality training data for AI applications such as self-driving cars, mapping, AR/VR, robotics, and more.

AI & ML Blog | Scale AI

NextImage

AnnouncementContextProvider

Screen

plasmic-graphcms

scale.com (Production)

lottie-react

plasmic-nav

react-scroll-parallax-global

plasmic-embed-css

plasmic-query

plasmic-basic-components

plasmic-tabs

react-aria

plasmic-cms

react-chartjs-2

plasmic-strapi

react-slick

radix-ui

react-quill

Product

Company

General

People

Engineering

<h2 dir="ltr">Background</h2>
Object tracking in aerial imagery is largely an unsolved problem in the AI overmatch future powered by Generative AI and platforms like <a href="https://scale.com/donovan" target="_blank" rel="noopener">Scale Donovan</a>; expert algorithms ATR and chain-of-custody play a crucial role as tools enabling the models to expand out from a pure text and structured data domain.
Although there exists extensive research on object tracking in the academic realm, those solutions are largely targeted at accurate assignment of track ID&rsquo;s on top of fixed object detector outputs. These outputs come from a nearly perfect detector on common commercial footage. This footage typically comes from a camera that is mounted front-facing on the ground in a fixed position with fixed zoom levels.&nbsp;
However, the needs of many Federal customers do not align with these cookie-cutter solutions. They often require aerial imagery and full motion video capabilities, which present additional challenges, on top of accurate assignment of track ID&rsquo;s as in the commercial realm:&nbsp;
<ol>
<li dir="ltr" aria-level="1">
The base object detectors are less performant, meaning there is significant room for the tracker to use temporal information to reduce false positives and false negatives.
</li>
<li dir="ltr" aria-level="1">
The camera moves quite a bit in terms of position (i.e., panning) and zoom levels.
</li>
<li dir="ltr" aria-level="1">
Long-term re-identification of targets is non-trivial and mission-critical.&nbsp;
</li>
</ol>
<h2 dir="ltr">Scale&rsquo;s Solution: A Suite of Tracker Post-Processors</h2>
To address these problems, Scale offers a suite of tracker post-processors, each of which can be turned on or off.&nbsp; We optionally package these on top of our own custom, base tracker which achieves state-of-the-art balance between precision and recall when tracking in the presence of camera movement.&nbsp;
These post-processors can be used alone to enhance an existing tracker, or used as an end-to-end tracking package on top of an existing detector. This package can be called in one line of code per frame after each object detector inference (for users using the package) or base tracker inference (for users using just the post-processors).&nbsp;
<h2 dir="ltr">Sample Demo MP4 Results</h2>
Below, we compare our Scale Tracker packaged with our custom base tracker (25FPS on the left) with two of the best publicly available trackers - one that is faster (BYTETrack at 30FPS in the middle) which has comparatively poor track ID permanence, and one that is slower with slightly better track ID permanence than BYTETrack (SmileTrack at 10FPS on the right), although still not as good as Scale Tracker&rsquo;s track ID performance:
<img title="image1.gif" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/01b73d81-ae6e-4a6d-b744-8cdcc98aedce" alt="" width="1200" height="300">
<img title="image2.gif" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/3945304f-e9e7-4aaf-a166-ff9f2ab89934" alt="" width="1200" height="300">
<img title="image3.gif" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/8dd6967b-f7bb-414d-aa5d-58952facd8d8" alt="" width="1200" height="300">
The Scale tracker:
<ol>
<li dir="ltr" aria-level="1">
Flickers less -- this creates more visually persistent tracks thereby creating a better user experience
</li>
<li dir="ltr" aria-level="1">
Is able to maintain track ID&rsquo;s through periods of erroneous track dropping -- this creates tracks that exhibit more permanence
</li>
</ol>
<h2 dir="ltr">Empirical Results</h2>
<h3 dir="ltr">Academic results</h3>
<img title="image4.png" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/0691ca39-f375-405a-8081-c8cfaf0484ad" alt="" width="615" height="438">
Note: The metrics for the other methods in the above chart were taken directly from the <a href="https://arxiv.org/pdf/2202.13514.pdf">StrongSORT paper</a>
Even though our tracker is aimed at Federal problems, it actually improves upon the performance of trackers in the commercial realm and achieves state-of-the-art trade-off between speed and performance when paired with BYTETrack, as evaluated by IDF1 (a mixture between precision and recall) on common commercial tracking dataset MOT17.&nbsp;
The chart above demonstrates that ScaleTrack (packaged as our post-processors paired with BYTETrack) achieves a better speed/performance tradeoff than all other methods. BYTETrack and OC-sort are slightly faster, but significantly less performant than ours.&nbsp;
Why the emphasis on the comparison to BYTETrack (and not StrongSORT or SmileTrack)? Because any tracker that cannot be run at near-real time is nearly useless for Full Motion Video (FMV) workflows because most FMV users want algorithms that can be run as they review mission videos. And most cameras collect around 20FPS, meaning only the algorithms to the right of the red line can be of high value to FMV workflows.
<h3 dir="ltr">Federal results</h3>
The above graph is from evaluating an academic dataset. On mission-specific datasets, we outperform even more. As visual evidence of our outperformance over state-of-the-art in Federal datasets, please see the above MP4 sample results.&nbsp;
<h2 dir="ltr">How Did We Achieve These Results? (Technical Deep Dive)</h2>
<h3 dir="ltr">Reduction of flickering</h3>
Detectors basically never have 100% recall in the Federal domain. They often miss objects. Most trackers halt operations when this happens and kill the track. Certain trackers keep the object around in a state that is half dead/half alive for about 5 frames, and provide them the option of being revived if the object appears again.&nbsp;
We went with a different approach. We become skeptical of the detector when a track isn&rsquo;t propagated to the next frame and we try to persist the track by querying the image directly through a two-step process. This two-step process performs bounding box location prediction functions for a given track. The first prediction function is based on kalman information.&nbsp;
However, kalman filters fail sometimes. As a result, Scale did a thorough, detailed study of these failure cases and created a second, custom prediction function in bounding box search space. Given these bounding box hypotheses (from the kalman filter prediction function and our custom prediction function), we use a highly specialized visual feature function (which is not customer-locked or security-locked and therefore can be used by any customer at any security level) to select the bounding box which is most visually consistent with that track&rsquo;s visual history. Based on the level of visual consistency, we either persist the track to the next frame at that selected bounding box&rsquo;s location, or enter it into a state that is not yet dead but not entirely alive (i.e., a bank of tracks from which ReID can draw).
<h3 dir="ltr">Ability to maintain track ID's through short periods of erroneous track dropping</h3>
Sometimes, the flickering reduction mechanisms fail. But it&rsquo;s important that when the track shows up again, it maintains the same ID. So, for a set amount of frames after a track is last seen, we use its kalman filter to project the track over every frame. If there ever exists a new track that is visually consistent with the dead track and that new track exists in a location that is highly likely under the dead track&rsquo;s motion prediction distribution, we assign the dead track&rsquo;s ID to the new track. Thereby we would &ldquo;revive&rdquo; the old track in the new track&rsquo;s location, which allows us to maintain the track ID even though the object was missing for a short while. We refer to this as &ldquo;Local ReID.&rdquo; Outside of handling algorithmic false negatives, this also addresses short-term occlusions (e.g. a vehicle driving underneath an occluding tree).
<h3 dir="ltr">Ability to maintain track ID's through long periods in which the track wasn't present</h3>
This addresses the following use cases:
<ol>
<li dir="ltr" aria-level="1">
Objects coming in and out of the frame, potentially at an entry location significantly different from its exit location
</li>
<li dir="ltr" aria-level="1">
Objects that are occluded for long periods of time (e.g., a person coming in and out of a building)
</li>
<li dir="ltr" aria-level="1">
Camera zoom ins, zoom outs, and pans
</li>
</ol>
In these cases, traditional motion prediction breaks down, meaning traditional motion information prediction algorithms cannot be used to link new tracks with dead tracks. However, the objects rarely significantly change their looks within the span of a video. So, we introduce &ldquo;Global ReID&rdquo; which attempts to link dead tracks to new tracks strictly through a harsh visual filter.&nbsp;
<h3 dir="ltr">Ability to maintain track ID's through medium-length periods in which the track wasn't present</h3>
We further address use-cases 1 and 2 in the above section using a ReID function that is based on custom motion functions mixed with the above mentioned visual filter with time between dead and new tracks as a salient input to the mixture function. We do so in such a way that the larger the gap between these tracks, the less the linking function relies on spatial information and the more it relies on visual information, as there is increased uncertainty in motion as the time gap becomes larger.

image.jpg

Assign accurate track ID’s and improve performance on top of aerial object detector outputs with video inputs

Scale's Solution for Aerial Object Tracking | Scale AI

Scale's Solution for Aerial Object Tracking

After the end of WWII and the invention of the atomic bomb during the Manhattan Project, J. Robert Oppenheimer gave an important speech reflecting on the technology he had just built, and in particular, the quantum and irrevocable impact it had on the nature of war:
&ldquo;It is clear to me that wars have changed.
It is clear to me that if these first bombs&mdash;the bomb that was dropped on Nagasaki&mdash;that if these can destroy ten square miles, then that is really quite something.
It is clear to me that they are going to be very cheap if anyone wants to make them.
It is clear to me that this is a situation where a quantitative change, and a change in which the advantage of aggression compared to defense&mdash;of attack compared to defense&mdash;is shifted, where this quantitative change has all the character of a change in quality, of a change in the nature of the world.&rdquo;
These words have great relevance today with AI. We are in the midst of the next great sea change in the nature of war, and thus, the nature of the world.
We live in uncertain times. Geopolitical risk is at&nbsp;<a href="https://www.matteoiacoviello.com/gpr.htm">a 20-year high</a> since the Ukraine war. The United States is facing <a href="https://www.jcs.mil/Media/News/News-Display/Article/613868/dempsey-us-forces-must-adapt-to-deal-with-near-peer-competitors/">near-peer competitors</a> in China and Russia for the first time since the end of the Soviet Union. There is a global battle for ideologies between Western democracy and authoritarian regimes.
All at once, the advent of AI has upended the world. It&rsquo;s a paradigm shift that has granted new superpowers and anything seems possible&mdash;with potentially greater implications on the future of our world than even the internet.
As someone who has been building in the AI ecosystem for more than seven years now, it&rsquo;s exciting to see this technology finally reach its watershed moment. But at the same time, this technology will exacerbate the uncertainty of our times. AI has the potential to unseat incumbents, catapult new leaders, or solidify existing moats. Navigating this new landscape is an existential matter&mdash;for many organizations, for our nation, and for democracy.
The United States, and the democratic world broadly, is confronted with two very real challenges.
First, as I&rsquo;ve <a href="https://alexw.substack.com/p/war?utm_source=profile&amp;utm_medium=reader2">discussed</a> previously, I believe the next era of war and deterrence will be defined by AI. We must ensure the strength of democracy in our AI war against authoritarianism by building the most effective military implementations of AI.
Second, a show of force is necessary, but not sufficient. At the end of the day, the strength of democracy has, and always will, stem from a stronger and faster-growing economy. That&rsquo;s why it is critical that we also bolster our global industries and enterprises with best-in-class capabilities.
The new Scale Donovan and Scale EGP platforms are our answer to these critical challenges&mdash;to empower the U.S. government and enterprises, respectively.
<h2 dir="ltr">Donovan &mdash; Ammunition in the AI War</h2>
AI will define the next era of war and deterrence, just as the atomic bomb did for the last era. The only solution to the AI War is to accelerate our own military implementations and rapidly integrate AI into warfighting. That's why we've funded our own R&amp;D to build Donovan&mdash;and achieve AI overmatch.
<a href="https://scale.com/donovan">Scale Donovan</a> is an AI-powered decision-making platform to help operators understand, plan, and act in minutes instead of weeks. Donovan ingests and understands vast amounts of structured and unstructured data to enable operators and analysts to make sense of any aspect of the real world in minutes using simple natural language. Donovan combines these insights with Scale&rsquo;s proprietary and state-of-the-art models to generate summary reports, COAs, and more to force-multiply staff bandwidth and achieve mission success.
As a longtime leader in AI, I am proud to be delivering this world-changing capability to the United States government with Donovan - the first LLM deployed on a classified network for the XVIII Airborne Corps - ingesting over 100,000 pages of live data to enable actionable insights across the battlefield. The sheer volume of inputs from various formats is a key challenge facing analysts and operators across the federal space. We cannot let the proliferation of information and manual processing be our Achilles heel.
<div class="wistia_responsive_padding" style="padding: 56.25% 0 0 0; position: relative;">
<div class="wistia_responsive_wrapper" style="height: 100%; left: 0; position: absolute; top: 0; width: 100%;"><iframe class="wistia_embed" title="Scale - Donovan Video" src="https://fast.wistia.net/embed/iframe/qugojsb52u?videoFoam=true" name="wistia_embed" width="" height="" frameborder="0" scrolling="no" allow="autoplay; fullscreen"></iframe></div>
</div>
<h2 dir="ltr">EGP &mdash; Unlocking Generative AI for Enterprises</h2>
<a href="https://scale.com/generative-ai-platform">Scale Enterprise Generative AI Platform (EGP)</a> is a full-stack solution for any company looking to deploy Generative AI applications in production. We are helping our customers - like Koch Enterprises - move rapidly from just experimenting with this technology to actually capturing business value from it, through a secure, enterprise-grade platform that is model-agnostic and powered by the industry's leading&nbsp;<a href="https://scale.com/data-engine">data engine</a>.&nbsp;
While Generative AI has captured the imagination of consumers, there are substantial safety, security, and performance challenges when it comes to business use cases. EGP allows customers to develop and deploy Generative AI applications within Scale&rsquo;s infrastructure or in their own virtual private cloud (VPC), so proprietary data never has to leave customers&rsquo; environments. Scale is the only provider delivering Generative AI applications in their own environment, coupled with industry-leading capabilities like Reinforcement Learning with Human Feedback (RLHF) and red-teaming to ensure that model outputs are aligned with customers&rsquo; policies and brand.
Further, EGP powers AI applications across industries. Think copilots for financial advisors to draw upon a much broader range of information when they are educating their clients, or tools for creatives to generate infinitely customizable images and copy, or a chat advisor to guide customers seeking information about retail products. These applications are only the tip of the iceberg - with EGP, companies are empowered to build and develop their own highly specific use cases. By bringing together Generative AI models with their own proprietary data, businesses can unlock new channels for growth and provide instant insights for better, faster decision-making.&nbsp;
<h2 dir="ltr">Looking Ahead</h2>
Right now, we&rsquo;re seeing a lot of AI tourists pretending to be AI natives, taking companies for a ride while they&rsquo;re at their most vulnerable. They&rsquo;re not selling solutions, they&rsquo;re selling vaporware. Everywhere you look, you&rsquo;ll see non-AI companies rebranding and selling technology that they only began to study yesterday. They barely have the knowledge internally to build the products they&rsquo;re selling, and they sure as hell don&rsquo;t have the expertise required to help customers build and implement a long-term, defensible AI strategy.
It&rsquo;s a problem that I care deeply about because the last thing the AI industry needs is more hype. And companies that are claiming AI expertise who haven&rsquo;t spent years in the trenches have the potential to do more harm than good.&nbsp;
The reality is that it takes a lot of hard work to go from experiment to enterprise. Foundation models aren&rsquo;t like a typical consumer software product that works out-of-the-box. From explainability and auditability to PII and proprietary data, there are several critical challenges businesses must address to make AI a reality. There is no established roadmap to follow, which is why only <a href="https://scale.com/ai-readiness-report">21% of enterprises</a> actually have AI models in production. In the federal government, only <a href="https://saic.com/sites/default/files/2023-01/SAIC-Cloud-WP-2023.pdf">14% of executives</a> believe they are &ldquo;very&rdquo; ready for AI.&nbsp;
It's clear that governments and businesses need a partner that provides battle-tested tools and expertise to harness Generative AI in a way that&rsquo;s trusted, secure and moves at the speed of the world around us today. Our new Scale EGP and Scale Donovan platforms were built to solve this problem.
The future of our world hangs in the balance of the current moment. AI has the potential to strengthen global democracies and democratic economies&mdash;or be the decisive implement that enables authoritarianism to take over.
We choose to take action&mdash;in light of all that is at stake&mdash;and build.
&nbsp;

Scale is proud to introduce our two Generative AI platforms – Donovan and EGP – to empower the U.S. government and enterprises to harness AI in a way that’s trusted, secure and moves at the speed of the world around us today.

A letter from our CEO | Scale AI

A letter from our CEO

Generative AI is capturing the imaginations of consumers and businesses alike. However, in this early era of adoption, most enterprises are yet to unlock true impact from this technology and are stuck in the experimental stages. Some are concerned about the security of their data and the reliability of model outputs. Others are overwhelmed by the many decisions to make regarding foundation models, prompt strategies, and tooling for developers.&nbsp;
Scale and <a href="https://www.anthropic.com/" target="_blank" rel="noopener">Anthropic</a> are excited to announce our partnership to bring to market a full-stack solution for Generative AI applications&nbsp;with a focus on safety and responsibility. This collaboration enables customers to bring Generative AI applications to production and to capture business value from this incredible technology.&nbsp;
Our partnership combines <a href="https://www.anthropic.com/product" target="_blank" rel="noopener">Anthropic&rsquo;s Claude</a> &ndash; a large language model guided by principles with the goal of creating outputs that are more consistent with human intent and values &ndash; with <a href="https://scale.com/generative-ai-platform">Scale&rsquo;s enterprise-grade platform</a> for app experimentation, development, and deployment.&nbsp;&nbsp;
Customers will be able to make the most of Claude with Scale&rsquo;s offerings:
<ul>
<li dir="ltr" aria-level="1">
Expert prompt engineering to guide users with the right prompt strategies and pre-built prompt templates, plus the ability to run experiments with different prompts.
</li>
<li dir="ltr" aria-level="1">
Data connectors to leverage proprietary data from databases and applications such as Confluence, Google Drive, Outlook, etc.
</li>
<li dir="ltr" aria-level="1">
Model validation through human evaluation to understand model blind spots and improve model performance, and features to continuously measure accuracy, brand alignment, and safety metrics, including sensitive content detection tailored to customers&rsquo; policies and brand.
</li>
<li dir="ltr" aria-level="1">
Enterprise-grade privacy and security, including deployment in a customer&rsquo;s AWS VPC and role-based access controls to ensure data and outputs are limited to the appropriate groups.
</li>
</ul>
Scale and Anthropic believe this represents a significant step forward for enterprises seeking to rapidly move beyond experimentation with Generative AI models. Designed to address concerns around data security, and model performance and reliability, this collaboration provides an accessible, &ldquo;all-in-one&rdquo; approach for customers to confidently tap into the power of Generative AI.

Scale and Anthropic are excited to announce our partnership to bring to market a full-stack solution for Generative AI applications, with a focus on safety and responsibility. This collaboration enables customers to bring Generative AI applications to production and to capture business value from this incredible technology. 

Scale + Anthropic: Full-Stack Generative AI for Enterprises | Scale AI

Scale + Anthropic: Full-Stack Generative AI for Enterprises

Scale AI, the data infrastructure for AI, has earned Impact Level 4 (IL-4) provisional authorization from the Defense Information Systems Agency (DISA), a significant milestone in Scale&rsquo;s ongoing commitment to providing the DoD with secure, industry-leading technology. This authorization allows Scale&rsquo;s cutting-edge machine learning (ML) platform and large language models (LLMs) access to defense applications with controlled unclassified information (CUI), such as personally identifying information, empowering the DoD with advanced AI capabilities to accelerate information accessibility. Scale is currently the only dedicated AI provider with IL-4 provisional authorization, underscoring its commitment to providing secure solutions for the U.S. government.&nbsp;
Scale&rsquo;s IL-4 provisional authorization expands AI accessibility for the DoD, allowing the agency to leverage Scale&rsquo;s Data Engine to curate and annotate data for a wide range of applications as well as Scale&rsquo;s Government AI platform for end-to-end solutions. Scale&rsquo;s powerful AI capabilities can empower analysts, operators, and leaders across the DoD with faster decision-making, improved operational efficiency, and accelerated innovation.
DISA, a combat support agency of the U.S. Department of Defense, is responsible for developing and maintaining the DoD Cloud Computing Security Requirements Guide (SRG). The Cloud Computing SRG defines the security requirements used by the DoD to assess the security posture of cloud service offerings, enabling cloud service providers (CSPs) to host DoD missions after receiving provisional authorization.&nbsp;
Scale&rsquo;s achievement of IL-4 provisional authorization demonstrates its adherence to rigorous security standards set forth by DISA and its commitment to supporting the DoD&rsquo;s mission-critical operations. With this achievement, Scale looks forward to unlocking the full potential of data and AI in the national security sector. For more information, please contact us at <a href="mailto:federal@scale.com">federal@scale.com</a>.&nbsp;
&nbsp;

Scale’s IL-4 provisional authorization expands AI accessibility for the DoD, allowing the agency to leverage Scale’s Data Engine to curate and annotate data for a wide range of applications as well as Scale’s Government AI platform for end-to-end solutions. 

Scale AI, the data infrastructure for AI, has earned Impact Level 4 (IL-4) provisional authorization from the Defense Information Systems Agency (DISA), a significant milestone in Scale’s ongoing commitment to providing the DoD with secure, industry-leading technology. 

Scale’s DISA IL-4 Provisional Authorization Increases Accessibility to AI for the Department of Defense (DoD) 

As we close out Women's History Month, we reflect on the incredible contributions of women throughout history and their importance in shaping our world today. In particular, we want to highlight the vital role that women have played and continue to play in STEM (Science, Technology, Engineering, and Math) fields.
Despite facing numerous challenges and barriers, women have made significant strides in STEM. From <a href="https://www.computerhistory.org/babbage/adalovelace/">Ada Lovelace</a>, the world's first computer programmer, to <a href="https://www.nasa.gov/content/katherine-johnson-biography">Katherine Johnson,</a> who helped put the first human on the moon, women have been pioneers in fields traditionally dominated by men. However, there is still much work to be done to address the gender gap in STEM, particularly in terms of representation and pay equity.
Throughout this month, Scale&rsquo;s Women ERG spoke to a few of its members to showcase the perspectives and experiences of women working in AI. We heard from Scaliens who have consistently broken barriers and achieved success in their fields, as well as those who have faced challenges and obstacles along the way.
We hope that by sharing these stories, we can help amplify the importance of diversity and inclusion in STEM and inspire the next generation of female leaders in the field. So, join us as we celebrate Women's History Month and recognize the incredible contributions of women in STEM. Together, let's continue to work towards a more equitable and inclusive future for all.
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/503f7e9b-f8aa-4580-a096-2cba1e2402d6">
&ldquo;Having it all&rdquo; as a woman means different things to different women. I think this is important to acknowledge as I discuss what it means to me.
Last year, I began pursuing my MBA on top of my full-time job. When I first started, I quickly discovered that my definition of &ldquo;having it all&rdquo; was exhausting and unsustainable in my new season of life. It was time to change that definition.
Oftentimes women tend to take on more responsibilities both at work and at home. This was exactly what I was doing in every aspect of my life. It forced me to prioritize the limited free time I have. I had to set boundaries in my work and personal life to make room for school. I highly recommend The Book of Boundaries: Setting the Limits that Set You Free by Melissa Hartwig Urban as it was immensely helpful for me.
&ldquo;Having it all&rdquo; not only means different things to different women. It also means different things in different seasons of life. In my current season of life, &ldquo;having it all&rdquo; means shifting priorities and setting boundaries that help me succeed where I am today. When I finish my MBA, I&rsquo;ll enter a new season of life where &ldquo;having it all&rdquo; is going to evolve. And I can&rsquo;t wait to see what that looks like.&nbsp;
Yuri Maruyama
Head of Product Marketing
I joined Scale four and a half years ago when Scale was a forty-person company. At the time, I didn&rsquo;t think it would be possible for me to start a family while working at a hyper-growth, early-stage startup. It took a few years but I welcomed my first baby last summer. I was able to take and enjoy my full parental leave without getting pulled into any work or feeling any guilt.&nbsp;
Once I came back, I worried about how I would balance being back to work full-time while also being present for my family. I still often fall short of my own expectations, but to manage this, I&rsquo;ve blocked off my calendar in the mornings until 8:30am and in the evening between 5-8pm to spend with my family. Blocking this time and holding this boundary has helped to make me more present and productive at work during the 8:30am-5pm time when I am 100% focused on work. At the same time, I&rsquo;ve been able to be present for my family in the mornings and evenings without feeling guilty or like I need to check slack or email.
Bihan Jiang
Product Manager
Balancing work and life is both a challenge and a necessity. One way that I&rsquo;ve found balance is by scheduling time for activities that I enjoy, such as exploring coffee shops (I drink 3+ cups a day!), traveling, watching dramas, and playing board games like Catan and Tractor.
It&rsquo;s also critical to find passion in your work. As a product manager on the Spellbook team, I&rsquo;m responsible for driving the development of our product, collaborating with cross-functional teams, and staying on top of industry trends and customer needs. It can be challenging, but it&rsquo;s also incredibly exciting to be building in the generative AI space, which is evolving every day.
Achieving a balance between work and life is a continuous effort that requires prioritization. By making sure to allocate time for my hobbies and interests, as well as having a support system in place, I am able to maintain a strong sense of self while excelling in my career.
Karena Zhang
Account Executive
Passionate about living as a true &ldquo;Corporate Athlete,&rdquo; I find a myriad of ways to stay active and challenge myself in the office and beyond. When I first began my career, I was a Reformer Pilates instructor part-time while working full-time in tech. Now that I have since retired from being a fitness instructor, I have taken on a new challenge of traveling the world by way of sliding down mountains on wood planks (skiing). I was also a competitive college tennis player at Carnegie Mellon, so I am not new to navigating the stress of work and play, but I know how important it is to have a strong sense of self outside of work. I believe that no matter how stressful things get at work, I still have to live my life. Calendar blocks plus a passion for planning allow me to balance working hard, having fun, and making history.
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/f383b465-2282-480b-bdb2-246cb1198506">As a female engineer, I have experienced being overlooked or underestimated many times in my career. Most of the time, there's nobody to blame--these biases are just that deeply ingrained in everyone. Over time though, I have learned to use this to my advantage. When my coworkers had low expectations of me, as long as I still set ambitious goals for myself, I could make sure I blew their expectations out of the water. I constantly remind myself daily to be confident in myself and to not let my goals and abilities be dictated by the male-dominated environment I'm surrounded by. I also keep in mind that as a minority in the workplace, I can bring fresh perspectives to the table. Finding success in engineering definitely feels like an uphill battle sometimes, but that also makes success feel much more rewarding.
Iris Guo
Strategic Product Manager
Despite lots of excitement and eagerness, I didn&rsquo;t know what to expect coming into the role of a strategic product manager as a new grad. In the past, I had more traditional product manager internships at big tech companies where I was building software/ hardware features and focused on the user experiences. I haven&rsquo;t done operations before and this was my first time getting into the weeds of running a project as an STO. As a woman with high degrees of agreeableness and conscientiousness, I found it difficult to be the one calling the shots and not asking for approvals before execution. At first, I thought I had to come up with the solution for every problem, so I would put a lot of pressure on myself to do everything right and I wasn&rsquo;t used to asking for help since I didn&rsquo;t want to bother other people. I tend to be detail-oriented and diligent in my tasks, but sometimes I would miss out on the bigger picture for strategic thinking. When I was stressed out at work, I would lean into my passion for dancing as a creative channel for me to destress. Dancing helped me express my full sense of self and relieved my obsessive thoughts from work to stay present. I performed ballroom dancing growing up and I started learning k-pop/ jazz funk/ contemporary/ bachata/ salsa when I moved to SF six months ago. Every group dance required lots of teamwork to help each other to learn the harder part of the dance, and I learned to ask for help from people with more experiences, which I applied to problems at work. Through dancing, I also learned to experiment and iterate fast, because we would have different versions of the same dance and improve from trials and errors. Dancing enlightened me to see the bigger pictures and let go of the unimportant details with bold movements, which served as a constant reminder to me at work. I felt extremely grateful for having a healthy and mutually beneficial relationship between my hobby and my job, and I believe that everyone can find the right balance between work and life.
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/5ac7d1f9-b882-43ac-9da9-8511d5062f95">I&rsquo;m thrilled to be a woman and working with other talented men and women here at Scale as a part of the AI revolution. As a society, we have a ways to go in terms of representation and equity across all STEM fields, and think that women have an especially important role to play in bringing diverse perspectives shaping AI products, and ensuring that the future of AI is just, equitable, and serves all of humanity better.

We hope that by sharing these stories, we can help amplify the importance of diversity and inclusion in STEM and inspire the next generation of female leaders in the field.

Throughout this month, Scale’s Women ERG spoke to a few of its members to showcase the perspectives and experiences of women working in AI.

Women’s History Month at Scale

Artificial Intelligence (AI) has been a game-changer for everyone. Together with data, AI has become a key component of a robust digital core&mdash;the main source of competitive advantage for companies today. The advent of&nbsp;<a href="https://www.accenture.com/us-en/insights/technology/generative-ai" data-analytics-link-name="generative ai" data-analytics-content-type="cta">generative AI</a>&nbsp;marks the beginning of a new era, one where businesses&nbsp;<a href="https://www.accenture.com/us-en/insights/consulting/total-enterprise-reinvention" data-analytics-link-name="reinvent" data-analytics-content-type="cta">reinvent</a>&nbsp;how they work. Just search for ChatGPT if you haven&rsquo;t already, and you&rsquo;ll get an idea of the possibilities. ChatGPT is derived from a large language model (LLM)&mdash;GPT4&mdash;developed by Open AI, which is a foundation model capable of generating human-like language and carrying out complex tasks across a number of functions.
We are entering an era of AI industrialization, where entire industries stand to be transformed by new, innovative applications of generative AI. One of the highest barriers to unlocking business value using generative AI and LLMs, however, is customizing and fine-tuning the pre-trained foundation models for their unique needs, using their own data. That&rsquo;s why Accenture is partnering with Scale AI.&nbsp;
<a title="Scale AI &amp; Accenture Partnership" href="https://scaleai.wistia.com/medias/2fjp89kejj">video</a>
&nbsp;
<h3 class="p1">Generative AI partner spotlight: Scale AI</h3>
Accenture&rsquo;s collaboration with&nbsp;<a href="https://scale.com/" target="_blank" rel="noopener" data-analytics-link-name="scale ai" data-analytics-content-type="cta">Scale AI</a>&nbsp;focuses on helping enterprises use their own data on general purpose models, customizing them to suit their business needs. According to CEO and founder Alexandr Wang, Scale AI has been &ldquo;quietly powering this generative AI wave, providing the data and infrastructure for many of the top companies to actually build large-scale foundation models, including Open AI.&rdquo;
Accenture and Scale AI are collaborating to provide an enterprise LLM platform for enterprise and government clients to fine-tune generative AI models that can learn from human input through a process called Reinforcement Learning from Human Feedback (RLHF). The result is generative models optimized for human interaction that are able to improve how people live and work.
For example, Scale AI was able to help retail clients gain higher shopping conversion and sales by using AI-generated product images for their newest products. And with social media clients, AI-generated content led to more than 100 million impressions.
Could humans have done this? Absolutely &ndash; but at what cost to the business?
This is business augmentation at the intersection of human and machine. By freeing up employees from automatable, routine tasks, we can redirect energy and focus to more strategic creative work that requires human intuition, decision-making and problem-solving skills - work that generative AI cannot do&mdash;at least not yet. The combination of human plus machine results in more efficient processes, improved decision-making and greater innovation.
Success with generative AI depends as much on people and training as it does on technology. It&rsquo;s up to business leaders to take the lead in upskilling employees and reshaping jobs and tasks to get the most out of technologies like generative AI.
Thanks to partnerships like this one with Scale AI, Accenture helps leaders create business value with generative AI technology quickly and at scale.

Scale AI + Accenture: Enhancing Business Value with Human Ingenuity

We&rsquo;re excited to announce that <a href="https://scale.com/spellbook">Spellbook</a> is now fully available for self-serve signups!&nbsp;
Spellbook is Scale&rsquo;s platform for large language model (LLM) apps. We believe the future of data science and machine learning is people building lightweight applications on top of large language models.&nbsp;
Speed and infrastructure quality is key&ndash;how easy is it to create a new LLM app, how fast can teams experiment, and how quickly can they get feedback on whether that experiment is better than a baseline or not? Once those experiments have signs of life, how efficiently can teams fine tune them, hone them with expert feedback, and launch them into production?&nbsp;
I&rsquo;ll walk through how you can create your own production-ready LLM app in less than 10 minutes. Follow along at <a href="https://spellbook.scale.com">https://spellbook.scale.com</a>!&nbsp;
App Creation
As an easy example, let&rsquo;s get started by creating a new marketing copy generation app that will take an event, size, and demographic, and generate an event description.&nbsp;
<figure class="image align-center"><img title="test" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/c23f338d-0dd8-492a-a458-70ff95c024fa" width="624" height="280">
<figcaption>App setup page in Spellbook</figcaption>
</figure>
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/2acb3b28-73c0-4bea-8ad3-385a5aba254a" width="580" height="423">
<figcaption>Example dataset with event, size, and demographic data.</figcaption>
</figure>
Quickly set up your prompt and see how your prompt runs on multiple rows of data.&nbsp;
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/35ca21e5-3b43-4feb-ba8e-dd6d5215c8cb" width="3159" height="1599">
Evaluate Your LLM App
Make sure your model is outputting good generation with our human evaluation pipeline. Define evaluation criteria on how a human should determine whether your model output is good or bad. Here, we want to ensure our event description is relevant.
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/6ab0f3da-e7c8-4258-af87-b4455f71a35b" width="624" height="317">
<figcaption>Define human evaluation criteria for your model output.</figcaption>
</figure>
You&rsquo;ll get your results back in the form of a &ldquo;Hit rate&rdquo; along with downloadable results you can use to fine tune your model!
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/d9584b8a-0898-41a9-ac5c-70c343a7faff" width="624" height="319">
<figcaption>Receive evaluation results back after humans have finished reviewing.</figcaption>
</figure>
Deploying Your App
Once you have a model you&rsquo;re satisfied with, you can deploy your model to production in one click through our deployments support.&nbsp;
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/6d34a534-3432-4925-a3de-51dfef56ef71" width="624" height="315">
<figcaption>Create a new deployment for an App Variant.</figcaption>
</figure>
You&rsquo;ll receive an API endpoint that you can integrate directly into production, with built-in monitoring!&nbsp;
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/e8a2dfe8-b11c-4ecf-9d2d-30bbcc04ae57" width="624" height="315">
<figcaption>Integrate your LLM app into production using our code snippets, and monitor your requests and latency.</figcaption>
</figure>
As you improve your LLM apps, you can keep swapping them out in your deployments tab by selecting the most preferable variant. This ensures that once you&rsquo;ve gotten an endpoint integrated, you don&rsquo;t have to update your code again.&nbsp;
<figure class="image"><img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/ded46a41-1762-41f0-b7a2-732f2f2e420c" width="624" height="335">
<figcaption>Swap out your deployments with updated variants as you improve the quality of your LLM app.</figcaption>
</figure>
<h4 dir="ltr">Try Spellbook</h4>
As LLMs become more powerful, the ability to quickly deploy enterprise quality LLM apps to production and quickly iterate based on user feedback is crucial. Try <a href="http://spellbook.scale.com">Spellbook</a> today to create your first apps!&nbsp;
&nbsp;

We believe the future of data science and machine learning is people building lightweight applications on top of large language models. 

 Try Spellbook today to create your first apps! 

Create And Deploy LLM Apps in Less Than 10 Minutes

Getting data to train and evaluate your large language models has never been more Rapid! Introducing&hellip;Scale&rsquo;s Large Language Model Toolkit!&nbsp;
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/6b85a60e-339a-46fd-bcb7-ce485df871de" width="624" height="391">
We&rsquo;ve developed expedited workflows to help you set up projects within Rapid, Scale&rsquo;s self-serve data labeling tool. <a href="https://scale.com/rapid">Sign up</a> today to try it out and get $20 in free credits.
After signing up, you&rsquo;ll see all of our available project templates and our &ldquo;LLM Training Toolkit&rdquo; front and center. We&rsquo;ve divided our LLM toolkit into three buckets according to different workflows:&nbsp;
<ol>
<li dir="ltr" aria-level="1">
Demonstration Data: Collect human-generated text over various domains and use cases
</li>
<li dir="ltr" aria-level="1">
Comparison Data: Evaluate your model outputs with human feedback
</li>
<li dir="ltr" aria-level="1">
Trust &amp; Safety: Align your model to produce factually correct and safe content&nbsp;
</li>
</ol>
Demonstration Data&nbsp;
We have two subcategories of project types under the &ldquo;Demonstration Data&rdquo; bucket.&nbsp;
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/09a3aa9a-f570-4ee7-93f8-42d2f47a9795" width="624" height="319">
(1) InstructGPT-style presets: Each of these comes with expert-written instructions, a preset taxonomy with the ability to add a customization section for your specific use case, the ability to upload your own prompts or have Scale create them for you, and a pre-trained workforce that has significant experience in these types of prompts. Check out Section 3.4 of the&nbsp;<a href="https://arxiv.org/pdf/2203.02155.pdf">InstructGPT paper</a> for yourself! 😉
(2) Demonstration corrections: If you have demonstrations or model outputs that need corrections, we can also assist you. Just upload your data or hook up your model endpoint and setup your taxonomy accordingly so that our labelers can get that done for you quickly.&nbsp;
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/ef708a53-23de-49b1-ac2c-f0b744870c68">
Comparison Data&nbsp;
With our workflows, you can easily customize comparison annotation projects and quickly scale to production. You can ask a labeler to rank, rate on a Likert scale, and classify model outputs. You can either upload these model outputs or configure a model endpoint for labelers to interact with as they are making the comparisons.&nbsp;
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/7eeab0fb-3f99-4b29-821e-5dcff793cfdb">
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/d27084b0-472c-417d-89e0-ea1364ef2e9f">
Trust &amp; Safety&nbsp;
Aligning your models with Rapid speed has never been easier. With our expertise and specialized workforce, we can work with you to identify problematic model output.&nbsp;
We support several use cases, including Grounded Generation, Red Teaming, Completion Citation, and Research-Based Correction.&nbsp;
<ul>
<li dir="ltr" aria-level="1">
Grounded Generation: Ask a labeler to research or read certain sources to verify uploaded model output.
</li>
<li dir="ltr" aria-level="1">
Red Teaming*: Provide an endpoint for labelers to chat with and elicit unsafe model outputs.&nbsp;
</li>
<li dir="ltr" aria-level="1">
Completion Citation*: Ask a labeler to attribute various phrases to sources that the information came from.&nbsp;
</li>
<li dir="ltr" aria-level="1">
Research-Based Correction*: Upload model output for labelers to find better sources for, edit/rewrite completions given any linked sources, add/edit/remove missing citations, etc.&nbsp;&nbsp;&nbsp;
</li>
</ul>
*Note: Use cases marked with * are currently not available via our self-serve platform. If you are interested, please contact sales to learn more.&nbsp;
For our Grounded Generation use case, we offer two self-serve, popular verification methods.
(1) Multi-class Classification: This involves selecting all error classes that pertain to the text via confidence-based consensus.&nbsp;
(2) Named Entity Recognition:&nbsp; This involves highlighting spans of erroneous text and indicating the relevant error types for each span.
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/a35f53ff-4fae-48e1-b8f1-674fbcf3f752">
<img src="https://drive.tiny.cloud/1/0lbhu31lrngl0y5eo8e6p0jckqgeikc94mnkaig43yuahupv/079d3f20-4345-4797-a4e3-31ba66781b0b"> And this is just the beginning! Our toolkit is constantly expanding, and we are excited to continuously be solving some of the world&rsquo;s hardest data annotation problems. Not seeing what you&rsquo;re looking for in the toolkit today or have other questions? Email us at: scale_rapid@scale.com.

Learn how to collect human-generated text, evaluate your model outputs with human feedback, and align your model to produce factually correct and safe content. 

Getting data to train and evaluate large language models (LLMs) has never been easier

Introducing: Scale’s Large Language Model Toolkit

<h3 dir="ltr">Introduction</h3>
OpenAI just released their new model that powers ChatGPT, GPT-3.5 Turbo. The best part &mdash; it's 10x cheaper than their previous best model.
OpenAI has found that GPT-3.5 Turbo is actually their best model for many non-chat use cases. Some early testers found it super easy to switch from the Davinci-003 model to GPT-3.5 Turbo with just a little bit of adjustment needed to their prompts. So, it's a much more cost-effective option for a lot of non-chat use cases.
Now that there is this new model available, it can be tough to know which one to use for your LLM apps in production. Spellbook makes it easy to build, compare, and deploy large language model apps, so you can evaluate GPT-3.5 Turbo against other models like Davinci-003 and FLAN to see which one performs best for your specific use case. In this blog post, we'll dive into the details of how these models differ on single-turn completions and what their strengths and weaknesses are, and share some real-world results from tests run on Spellbook- <a href="https://scale.com/spellbook">try it yourself today</a>!
<h3 dir="ltr">Classification&nbsp;</h3>
From our previous <a href="https://scale.com/blog/gpt-3-davinci-003-comparison">experiments</a>, we concluded that Davinci-003 performed much better than Davinci-002 with 0 shot prompts (92% vs 69.91%), and on par or slightly worse than Davinci-002 on few shot prompts (87.2% vs 91.6%). We observe a similar case with GPT-3.5 Turbo and Davinci-003.
GPT-3.5 Turbo performs better on 0 shot classification- on an insider trading classification example it achieves 91.72% accuracy, versus 82.02% for Davinci-003.
<figure class="image"><img src="https://lh4.googleusercontent.com/-qWbBYjvANJhwOpm25BihanYgq7goa1pw4QDINq3s6ZZedbwsHzuAOguit4S-2ug0TyVVx2DXgt7ep2J1MBiO_quaJrAWRlb4TRJJj-Te4zo5V3XHE58gNCbw_Rni-gFLx5gRxf0Aq0-bFM-aKUwN4c" width="624" height="349">
<figcaption>Comparison of GPT-3.5 Turbo against Davinci-003 on a 0-shot insider trading classification use case.</figcaption>
</figure>
However, Davinci-003 performs slightly better than GPT-3.5 Turbo with k-shot examples at 90.47% accuracy versus 88.37% for Davinci-003, but the quality is still lower than the GPT-3.5 Turbo 0 shot example.&nbsp;
<figure class="image"><img src="https://lh4.googleusercontent.com/iIefySacpNH98k4jEBgnUTSaRb5FCQiodLKt98KyvuI4-Z_iQp3lkH7bkQeKc-k-Sz2jnulr4wXT93pfU9visP46mOy0cfNXc3WVqr_b4M_meiDognTpgAL8vaHiB3cCW5eOCuhG_ZfEQPkSh6OWpYc" width="624" height="357">
<figcaption>Comparison of GPT-3.5 Turbo against Davinci-003 on an insider trading classification use case with k-shot examples.</figcaption>
</figure>
The use of k-shot examples will likely not be very common with GPT-3.5 Turbo- including them in chat prompts can take up a significant amount of context, which is not ideal for a chatbot that receives a high volume of API requests. In cases where you&rsquo;re using GPT-3.5 Turbo for single-turn completion, this will be pretty much equivalent to adding k-shots for Davinci-003, but for multi-turn use cases, this can lead to inefficient resource usage if k-shots are included in every prompt.&nbsp;
On a more nuanced classification example, like sentiment analysis, we also see that GPT-3.5 Turbo outperforms Davinci-003 on 0 shot experiments (84.26% vs 78.57%).
<figure class="image"><img src="https://lh6.googleusercontent.com/ygoPuU_tuQoheEQ2N5Ut-7KG0dZjEap2zu6D-p4gijZ87fIfDz-zqnTY2oYaQIqWZhSMBrbrsxlGjZdFzogzJ1CZM5lgHmlsLnJkOpqSll81yjRYcioGG5BBK1xUraifOOiLojBPqj8t1HRZ7nBUwEk" width="624" height="360">
<figcaption>Comparison of GPT-3.5 Turbo against Davinci-003 on a 0-shot sentiment analysis use case.</figcaption>
</figure>
<h3 dir="ltr">Text Generation</h3>
If you're looking for an AI language model that can generate long, detailed answers to complex questions, GPT-3.5 Turbo might seem like the obvious choice since it's been trained on a massive dataset of human language and produces coherent, contextually-appropriate responses.
But what if you're looking for a model that can provide clear, concise responses? In that case, you might want to consider using Davinci-003 instead.
In our experiment, we provided both GPT-3.5 Turbo and Davinci-003 with a set of 30 questions, and asked each model to provide 30 answers. What we found was that GPT-3.5 Turbo&rsquo;s responses tended to be much longer than Davinci-003's, with many of them exceeding 100 words. In contrast, Davinci-003's responses were much more concise, often consisting of just a few sentences. Given a max tokens of 500, the average response from GPT-3.5 Turbo was 156 words (~208 tokens) nearly twice the amount of Davinci-003, whose responses were an average of 83 words (~111 tokens). 
<figure class="image"><img src="https://lh3.googleusercontent.com/XWVxJwRIeuE8s5I686NmO8WlR9DJPekjQ_TLSimXu8xm52eKrxOD0kJSyPcDqZj_mpvJ1aeA1soQ4DG8CYWsd3DqMnN4mE-SYj6Wqw7s5bu6XgdMYNQZjblaQcctic_hfJvUma8nqCEeitps7GeSucU" width="624" height="351">
<figcaption>Comparison of GPT-3.5 Turbo against Davinci-003 on a question answer use case.</figcaption>
</figure>
<h3 dir="ltr">Math&nbsp;</h3>
Notably, GPT-3.5 Turbo is significantly better than Davinci-003 at math. When evaluating by exact match with an answer key, GPT-3.5 Turbo gets the correct numerical answer 75% of the time, while Davinci-003 only gets 61% correct.
<figure class="image"><img src="https://lh6.googleusercontent.com/IG9V-yRgXR9vLTW0Myve1LV8Q8w1FpFPTs7t-YAaFXHAZQzRMUXZVvc64lb-2P6ERE79u_f_FY4eB1Y4dyK4kK59WtuEuMm-hNxQJ2C5m8J4ZuTxt1FE-C0ENR-iYK_8zXWId-SS7gy-qNIGeJ1ZGiY" width="624" height="313">
<figcaption>Comparison of GPT-3.5 Turbo against Davinci-003 on a math use case.</figcaption>
</figure>
We also selected a subset of the questions for expert human reviewers to evaluate the models correctness based on both the numerical correctness and the logical reasoning. Interestingly, upon asking both models to explain their answers, it became clear that the models would sometimes hallucinate reasoning for how they achieved the answer, even if it was numerically correct.&nbsp;
<figure class="image"><img src="https://lh4.googleusercontent.com/xmFdfVGuo3p81w5agMlTCOEuBLDkDdYHo6oniV8MYtIxc5IuqcFVti6vof3b3bb3dO-1QyOfyJ5ryVRleV6nSrPYpJc8w55vcRtnMY_KXhh7wB7HvNJ6ZB6bht9S8ijLJdgDviWDN6fJzXTrWCPQIJ8" width="624" height="315">
<figcaption>Evaluation configuration for human evaluation of math explanations.</figcaption>
</figure>
<figure class="image"><img src="https://lh4.googleusercontent.com/JRmEQBPW5OjYMxQWGH1ZyT7hVaeQ8dAUGx2Zy8qprMyCCwh1rrg1woFZzb1eSNS1sqkkgdI95cytMCljkoYVugA4tA3shO8SvP43hVQnOTW28Bkobru0sTHNqN6k1QXCjgwy9uSIcaPMIR4QxRdcBSg" width="624" height="229">
<figcaption>Quality drill down on human evaluations of math explanations from Davinci-003</figcaption>
</figure>
From the quality drill down summary from the completed tasks, Davinci-003 only got 44% of the questions correct. Repeating the same workflow for GPT-3.5 Turbo yielded 68% accuracy.&nbsp;
A couple interesting math lessons GPT-3.5 Turbo picked up that Davinci-003 struggled with:
<ul>
<li dir="ltr" aria-level="1">
Solving questions using the order of operations
</li>
</ul>
<figure class="image"><img src="https://lh5.googleusercontent.com/TLCynrXwIToi9u-tS5vFU-oBMcuKupJtA9foUG4ryegt3xwZCaGEhoPWmZpiteoUvTt8BzoRkVHqMY5L_xKOjwB2LLceaLxCNAHV0IYHJVLkVs9c7d8Cb6PnpVIMewge0AreLBYPPkBCRGqUL2MyMKg" width="624" height="276">
<figcaption>Comparison of GPT-3.5 Turbo (left) against Davinci-003 (right) on a math question about order of operations.</figcaption>
</figure>
<ul>
<li dir="ltr" aria-level="1">
Plugging in variables into equations
</li>
</ul>
<figure class="image"><img src="https://lh3.googleusercontent.com/j63fVykVKFmDpt1qMIW3XPkntWiApDSvocH4oImJLdOr7quH0ESBTqDZwFSQpICgO1Idyja9iz_9UJn2cEDdOqB2z2Lv1UrRERGmb9qm9ojwBTuUwDhGBrvuZjsjeDpHYeFV2EM0sauHr3B_8shZNB0" width="624" height="147">
<figcaption>Comparison of GPT-3.5 Turbo (left) against Davinci-003 (right) on a math question about plugging in variables into equations.</figcaption>
</figure>
<ul>
<li dir="ltr" aria-level="1">
The negative exponent rule
</li>
</ul>
<figure class="image"><img src="https://lh3.googleusercontent.com/sxzxXRaidO3EmcVfJE_fHvgUBwRfRkdpZ2RVhAtEO6R0tjjTdDLtueT0VtrPqDouqax6beTzb-d4m4Z3t8VwZaGopCOVj67sxPVCKzWKsloIgnx_PHftT6YsF0JEY2y3qfCn1bqp-vAILRwurbozbvc" width="624" height="212">
<figcaption>Comparison of GPT-3.5 Turbo (left) against Davinci-003 (right) on a math question about negative exponents.</figcaption>
</figure>
<h3 dir="ltr">Base model swapping</h3>
It&rsquo;s worth noting that OpenAI has recommended the replacement of Davinci-003 implementations with the chat API in various instances, given its significantly lower cost of only $0.002 per 1,000 tokens. The ChatAPI can be used in a manner similar to the traditional Davinci-003 prompt format, where the system prompt is treated as the prompt body, the user message is treated as the input, and the assistant message serves as the completion.
For users who have LLM apps integrated in production using Spellbook, it&rsquo;s super easy to fork a variant, swap out the base model to GPT-3.5 Turbo and save it as a new one, and swap in the new variant, without having to write a single line of code.&nbsp;
<figure class="image"><img src="https://lh6.googleusercontent.com/Wp3bvT0IPiS0olJFhr-qiPieXViAuVBcq5q4ZAQOAcKJK8FFg2tOZp_nIsGcCdVZ0A8wdtdoAWj4CCoxMexDGD0IcUNyDPA-kFKvq0GuzHC3J_JWv4-T2oakkCKddvz13tS4OtmYzec0j-SXmLKOIws" width="624" height="325">
<figcaption>Quickly swap out your deployment&rsquo;s underlying variant to use GPT-3.5 Turbo without writing a single line of code.</figcaption>
</figure>
<h3 dir="ltr">Takeaways</h3>
OpenAI's new GPT-3.5 Turbo model offers a cost-effective option for many non-chat use cases. While GPT-3.5 Turbo performs well in 0-shot classification and math, Davinci-003 performs slightly better in k-shot classification and may be a better option for those looking for clear, concise responses that get straight to the point.
Spellbook makes it easy to build, compare, and deploy large language model apps, so you can evaluate GPT-3.5 Turbo against other models like Davinci-003 and FLAN to see which one performs best for your specific use case. With the ability to fork a variant and swap out the base model to GPT-3.5 Turbo, users can seamlessly integrate the new model into their LLM apps without writing any additional code.
Overall, GPT-3.5 Turbo performs better than Davinci-003 on most tasks at 10% of the cost - but is much more rambly. You can test out whether GPT-3.5 Turbo is better for your use case through <a href="https://scale.com/spellbook">Spellbook - sign up today</a>.

GPT-3.5 Turbo performs better than Davinci-003 on most tasks at 10% of the cost - but is much more rambly. You can test out whether GPT-3.5 Turbo is better for your use case through Spellbook

Compare the API behind ChatGPT to existing Davinci models using Spellbook

ChatGPT API: Cheaper, Faster, Wordier

Have you ever listened to the song, &ldquo;Umi Says&rdquo; by Mos Def? It&rsquo;s a fantastic song and in the chorus, Mos Def sings, &ldquo;My Umi said, shine your light on the world. Shine your light for the world to see.&rdquo; My mother would say the same thing to me when I was younger and she still says it to this day! But what once started as rhythmic words of encouragement, now has begun to transform into a deeper, more raw sense of self. It has taught me that we can brandish one&rsquo;s individuality by sharing experiences, growing from adversities, and passing lessons learned onto my peers, and, to me, that is what I believe Black History Month is truly about. I never saw it as a design to cast blame or seek apologies; but rather a time to appreciate both Black culture and Black contributions. It is a time to develop a profound understanding of our current generation and those that came before us. Because from listening, comes understanding.&nbsp;&nbsp;
My family educated me on what being a Black person in America was like for them and what it may be like for me. They taught me how to navigate the waters and pushed me to hear others&rsquo; experiences when given the chance. And then in 2020, the George Floyd movement sparked discussion across the country and a societal shift began taking place. Various groups began broadcasting their desire for institutional change. Police training, workplace diversity, legislation geared towards discrimination; every facet that could touch bias or mistreatment within the Black community was under a microscope. Echoing a similar sentiment, different organizations and initiatives approached the problem in the form of open conversation. One of which was The Commit to C.A.R.E (Care About Racial Equity) Now initiative that was launched as a response to the consistent injustices, stereotypes, and negative attitudes about Black men in America. The initiative partnered with Dove Men+Care and the NBPA (National Basketball Player&rsquo;s Association) to create a portfolio focusing on &ldquo;celebrating the contributions of Black men in communities across the country and telling their stories as a part of our commitment to change the way Black men are seen and treated in our society.&rdquo; This initiative brought to light the perspectives and stories from CEOs, comedians, NBA players, authors, and actors alike to capture them in a series of videos and photo shoots. I was fortunate enough to participate in this collaboration and got to share my own experience on what it is like to be a Black man in America. These conversations moved the needle forward in the larger conversation about racial inequity and emphasized the need for tangible benchmarks for measuring this change.&nbsp;
&nbsp;
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://lh5.googleusercontent.com/MC-c6NkZ9xHbl5KiZg1PIjgUigQuZ3nlGGzPuxy1SvnmZ8T_EZtqqwpS6cRDQIDXOe6RFzqm4jupcz2sx0DBYtTKDBM-fiymx9WAqC2StoQbeYGfZFGUmDBdiuH1YoqRq4jXkqSXPAUr" width="235" height="235">
<figcaption>Here I am posing for the The Commit to C.A.R.E Now portfolio in 2020. This initiative partnered with Dove Men+Care and the National Basketball Players Association.</figcaption>
</figure>
Participating in this program and listening to the other men who joined me in it, taught me what it means to persevere and implement change. I heard their struggles, their hopes, and their own thoughts of what could be done to improve our culture&rsquo;s place in this world. Their views reaffirmed my belief: that when there is representation in any environment, avenues for change become achievable. At a granular level, having a diverse group of leaders can create networking opportunities for all communities through continuous involvement and influence. What were once one time donations, turns into institutionalized funding. Higher quality education becomes accessible rather than the faraway dream it used to be. These are examples of how equipping Black communities with the right tools can snowball into something that my community continually strives for: generational wealth in the form of capital, knowledge, and opportunity. Access to these tools allows for all generations to stand a chance against cronyism and social stratification.&nbsp;
While learning lessons from our current generation is powerful, I feel that it&rsquo;s important to also spotlight and commemorate the feats that Black Americans have accomplished in the past. Despite the history of racism and oppression, the Black community has paved the way for people like me. This community has been fighting for equality and fair opportunities so that Black people today may experience life in a way that they could not. They have demonstrated constant efforts to leave the world a better place than when they entered it and these very efforts have percolated from leaders like Martin Luther King Jr. down into the communities he served. As a result, these communities place a great deal of importance on sharing the impact that these historical endeavors have had on their lives by passing their history down to the next generation. So when the National Museum of African American History and Culture opened in 2016, my grandparents jumped at the opportunity to donate an artifact to the collection. Donating this artifact meant the world to my grandmother in particular because behind it, lies the history of her father.&nbsp;
Percival Leon Burrows was born in Barbados, and later in life, would move to Buffalo, NY and then Harlem to join the Universal Negro Improvement Association and African Communities League (UNIA-ACL) alongside Marcus Garvey. He then received a Black Star Line certificate for his service and passed it down to my grandmother who donated this very certificate to the museum. I&rsquo;ll never forget how proud my grandmother stood while looking at a small piece of my family&rsquo;s Black History on display in the nation&rsquo;s capital.&nbsp;
&nbsp;
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://lh3.googleusercontent.com/Wlk_aR5NJZju3VJC7bnlsxxBPRXA7J1_Yw-EEctsfLTYUu65St43VRqht19euAUmP62SOC5QQ7p9tmtLlIi6_NMRc9SkF_CHmfvoxuhGF6LD09XNQZZz_DC8fe3PA8KKPoc8wTfAbrdC" width="246" height="207">
<figcaption>UNIA-ACL promoted the purchase of stock in this shipping company to encourage commercial trading among African people worldwide.</figcaption>
</figure>
<figure class="image"><img style="display: block; margin-left: auto; margin-right: auto;" src="https://lh6.googleusercontent.com/Ke6ICNeyeAbtQLjJgR1MPS4-YxNIMJ0r7df5V3ba_ww1p8CsITwsEo_AMvEgOWXY3MzRajudXX2UhpCFCEkopzkTlh-sWUsDr8BV9ihsK-9h5HuGBulhydnXxVu3N8lyte2pN4_lWvWC" width="188" height="252">
<figcaption>Leota &ldquo;Burrows&rdquo; &amp; Edgar Henry pose in front of the Voices of Protest and Resistance exhibit at the National Museum of African American History and Culture.</figcaption>
</figure>
I am thankful for everything the Black community has done for me and I am especially grateful to my family and colleagues who have shaped the way I approach my work and my life. For without these efforts, I would not be in the position I am today. I would not have had the tools to become a Carnegie Mellon graduate. I would not have had the skills to play college ball and try out for the NBA&rsquo;s official minor league basketball organization. I would not have had the connections to intern under a U.S Senator. And I most definitely would not be a Security Engineer at Scale AI. But the efforts made by Commit to C.A.R.E Now, UNIA-ACL, and my grandmother, exhibit the lengths our community goes to progress the state of all African Americans and demonstrates what can happen when our community has a chance to shine their own light on the world. I hope that I am able to pay it forward with half the capacity that my community has done for me.&nbsp;
And Black history doesn&rsquo;t just stop in February! If you live in D.C like me and you&rsquo;re interested in learning more about the African American saga, I recommend visiting the <a href="https://nmaahc.si.edu/">National Museum of African American History and Culture</a>. It&rsquo;s packed full with extraordinary exhibits and thousands of artifacts that have been donated by people all over the world. If you aren&rsquo;t in the area but want to learn more, check out the Commit to C.A.R.E Now video (14:25)&nbsp;I was featured in! Or, if you use Instagram, checkout @blackarchives.co or @brownhistory.&nbsp;
<a href="https://www.youtube.com/watch?v=I82ioHNSEss">Commit to C.A.R.E Now video </a>
But for now, &ldquo;Alexa, play Umi Says, by Mos Def.&rdquo;
&nbsp;

Scale Celebrates 2023 Black History Month

Seth Henry shares his experiences as a Black man to celebrate Black History Month

Visions of Color

URL redirections for Website. Disclaimer: takes 5~ min for each added or edited.

url-redirects

Use relative path. use `/blog/foo` not `https://scale.com/blog/foo`

blog-filter-type

homepage-quotes

logos

homepage-content

quotes

blog-category-model

announcement-banner

homepage-resources

custom-fonts

homepage-event

author

Event model for new marketing events page

event

Optional second image intended to show on the grid view, should be a more squared image closer to 1:1 aspect ratio than the one used in the Cover field.

guides

announcement-news

homepage-research

legal

slugs must be all lowercase. ie: ‘my-slug’

eval

datastream

summit-2023-speakers

customers

Display order on the page /customers. Higher values show first

summit-2023-agenda

blog-model

Option to list or not list this post in the Blog index. Enabled = not list in the index, Disabled = list in the index

summit-2023-partners

enterprise-use-cases-categories

enterprise-use-cases-categories-items

partners

data-samples

enterprise-use-cases

Use only one image or video option. Add another item to add another source.

collapsible-sections

research

basecamp-speakers

Example: https://www.twitter.com/username

Example: https://www.linekin.com/in/username

leaderboards

Text for a short description like: Deprecated (as of January 2025)

Keyword for search. Add multiple keyword with comma. Example: hle, hleto, vista, enigma

Blog

Scale's Solution for Aerial Object Tracking

A letter from our CEO

Scale + Anthropic: Full-Stack Generative AI for Enterprises

Scale’s DISA IL-4 Provisional Authorization Increases Accessibility to AI for the Department of Defense (DoD)

Women’s History Month at Scale

Scale AI + Accenture: Enhancing Business Value with Human Ingenuity

Create And Deploy LLM Apps in Less Than 10 Minutes

Introducing: Scale’s Large Language Model Toolkit

ChatGPT API: Cheaper, Faster, Wordier

Visions of Color

The future of your industry starts here

Blog

Scale's Solution for Aerial Object Tracking

A letter from our CEO

Scale + Anthropic: Full-Stack Generative AI for Enterprises

Scale’s DISA IL-4 Provisional Authorization Increases Accessibility to AI for the Department of Defense (DoD)

Women’s History Month at Scale

Scale AI + Accenture: Enhancing Business Value with Human Ingenuity

Create And Deploy LLM Apps in Less Than 10 Minutes

Introducing: Scale’s Large Language Model Toolkit

ChatGPT API: Cheaper, Faster, Wordier

Visions of Color

The future of your industry starts here