Chat-Your-Data Update Our starter repo for creating a chatbot over your data has got an update! Big thanks to @bllchmbrs for updating it (new chains, new LLMs) The space moves fast, is great to have community members like Bill keeping us up-to-date!
SOFTWARE
-
Curated Transformers Library Reaches v1.0 with PyTorch Implementations
By
–
Curated Transformers is our new library of PyTorch LLM and other transformer implementations — now at v1.0! The library features well-factored implementations of common architectures, including Llama 2, Falcon, BERT and many more. Architectures are implemented from common https://
x.com/spacy_io/statu
/spacy_io/status/1687103018628722688
… -
DataRobot Extends Data Safely with LLMs and Chatapps
By
–
Generative AI is getting a lot of media attention, but there is also a lot of vaporware. We're doing things differently. Join Chief Customer Officer Jay Schuren to hear how DataRobot makes it easy for you to safely extend your data with #LLMs, build and prototype #chatapps,
-
Browser extension focused, multi-platform expansion planned
By
–
Not quite yet, as we're focused on the browser extension, but we do plan to bring this to more platforms over time!
-
GitHub Copilot Code Referencing Helps Developers Make Informed Decisions
By
–
This is awesome from @github
, helping developers make more informed decisions about the Copilot code they use. >> Introducing code referencing for GitHub Copilot https://
github.blog/2023-08-03-int
roducing-code-referencing-for-github-copilot/
… -
Replicate Integration with Vercel AI SDK for Llama 2 Chatbot
By
–
Replicate is now supported by the @vercel AI SDK. `npm ai replicate` Here's how to build your own Llama 2 Chatbot, with streaming: https://
sdk.vercel.ai/docs/guides/pr
oviders/replicate
… -
File Upload Beta: Ask Questions Directly to Your Documents
By
–
👩🔬 Now in Beta: File Upload
— Perplexity (@perplexity_ai) 3 août 2023
Try it now: https://t.co/pICSQrcaRY
Start asking your questions directly to your documents, code, or research papers. Dive into information faster and get insights quicker. It's in beta, and we'd love your feedback. pic.twitter.com/GNbQU0D0nRNow in Beta: File Upload Try it now: https://
pplx.ai/file-upload-be
ta
… Start asking your questions directly to your documents, code, or research papers. Dive into information faster and get insights quicker. It's in beta, and we'd love your feedback. -
Optimized LLM Training Framework Reduces Host Memory Overhead
By
–
If you use an optimized #LLM training framework like https://
pbase.ai/3DHqnE5, you can get the host memory overhead back down to a more reasonable 7 * 4 = 28 GiB of host memory even when training on multiple GPUs. -
Loading Model Weights into GPU Memory with Ray Object Store
By
–
How did we do it? By loading the model weights into memory once before training begins and inserting them as numpy arrays into the #Ray object store, we can then zero-copy read the weights directly from shared memory into each GPU worker process.
-
Loading Pretrained Checkpoints: Multi-GPU Memory Challenge
By
–
Before you even get to multi-GPU training with model parallel frameworks like #Deepspeed, you need to load the pretrained checkpoint into memory. To make matters worse for machines with multiple GPUs, you need to load the checkpoint into host memory once for each GPU in your job!