#LangChain streamlines the development of applications, such as chatbots, Generative Question-Answering (GQA), and summarization. By “chaining” components from multiple modules, it allows for the creation of applications built around an #LLM
https://t.co/c2tw6KMCrv
LIMA: Less Is More for Alignment
LIMA, a 65B parameter LLaMa language model fine-tuned with the standard supervised loss on only 1,000 carefully curated prompts and responses, without any reinforcement learning or human preference modeling. LIMA demonstrates remarkably strong performance, learning to follow specific response formats from only a handful of examples in the training data, including complex queries that range from planning trip itineraries to speculating about alternate history. Moreover, the model tends to generalize well to unseen tasks that did not appear in the training data. In a controlled human study, responses from LIMA are either equivalent or strictly preferred to GPT-4 in 43% of cases; this statistic is as high as 58% when compared to Bard and 65% versus DaVinci003, which was trained with human feedback
paper page: https://t.co/MjiZ6mZn0x
@nntaleb It comes in really handy if you want the thoughts to be put together quickly given you do have some basic understanding of the topic. It takes few iterations to get a good output.
ChatGPT is a statistical representation of things found on the web, which will increasingly include ITS OWN output (directly and second hand).
You post something picked up from it & it will use it to reinforce its own knowledge.
Progressively a self-licking lollipop.
Enjoy
#AI
1. Create a Strong Password
2. Disallow Third Party Access
3. Avoid Unprotected WiFi
4. Don’t Link Your Twitter to Webmail
5. Keep Web Software Up-To-Date
6. Opt-In to Two-Factor Authentication
https://t.co/d24zenjN2c
@WallStreetSilv@WallStreetSilv Unsafe? Yes, very.
Human behavior is a complex equation shaped by various socioeconomic factors, regulations, and education - to name a few.
Sometimes cost of transportation may take precedence over safety.
🔥 Go through this list of 101 top data science interview questions & answers over the weekend: https://t.co/ahLJdIDYAC
#DataScience#Interviews#DSDojo
📢 On May 3rd, we are hosting a talk that will provide an overview of BevelCloud's decentralized cloud service. Our speaker, Timothy Chou, will discuss how AI applications are key to reducing healthcare inequity both locally and globally.
RSVP now: https://t.co/bUPcW2HXjz
🥁 🥁 Drum roll….Get ready for some exciting news!
We are thrilled to announce the winners of our free giveaway for the Future of Data and AI conference, held on March 1st-2nd, 2023. We received an overwhelming response, and we couldn't be more grateful to our community for making it such a huge success.
Firstly, congratulations to Rym Oualha, the winner of the first giveaway - a free seat in our award-winning Data Science Bootcamp! Rym is a Junior Machine Learning Engineer from Sousse, Tunisia, and we are excited to have her join our bootcamp.
Data Science Dojo's Future of Data and AI conference hosted over 10,000 attendees, 30+ speakers, and 20 panels & tutorials - safe to say, it was a virtual tech extravaganza. We would like to express our gratitude to the speakers for sharing their expertise and providing a variety of valuable tutorials to our community; @Z_Score, Albar Wahab, @nickdesb, @camimanera, @DataMovesHer, @SamPartee, @JackyGKoh, @svonava, Harmke Alkemade, @bobvanluijt, @htahir111, Muntazir Mehdi, Faisal Masood, @natalevinto, Jimmy Nguyen, @averydata, @NickSinghTech, @MuazmaZahid, Hamza Farooq, @BatoolArhamna, @frlazzeri, @timothychou, @frankzliu, Arif Iqbal, @adashofdata, @thinking_code, Corey Weisinger, Emilio Silvestri, @touseefliaqat, @hemanham, Sanjay Pant, and Ebad Ullah Khan,
Additionally, we want to thank our community partners for their contribution to the success of the conference; @zilliz_universe, @mlopscommunity, @opensv, @Redisinc, Algolab, @zenml_io, @kdnuggets, and @knime. We look forward to continuing our partnerships with all of you in the near future.
We express our gratitude to all who participated in the conference. Stay tuned for the second edition of the conference!
#futureofdataandai
@paulg@paulg Even getting traffic that is legitimately yours isnt possible. A competitor has been bidding on our brand keyword ‘data science dojo’ and getting placed above our home page. Landing page has /dataciencedojo
Our only option is to pay Google for the traffic.