Thank you Angela Yang for including us in this piece. Llama 3 is so exciting for AI and it's great to see the focus on enterprises having more options to maintain ownership of their own data, as well as the output of their work. Read the full story: https://
buff.ly/4bcneel
OPEN SOURCE
-
Llama 3 Empowers Enterprises with Data Ownership Control
By
–
-
Running Whisper-cpp on macOS via Homebrew and Hugging Face
By
–
Thanks to @slhck I also now have a recipe for running CLI whisper-cpp on macOS via Homebrew and a model file downloaded from Hugging Face – detailed notes on that here (including converting mp3 to wav using ffmpeg):
-
LangChain Upstage Integration: Layout Analysis and LLM Chat
By
–
Super excited to have the @langchain @upstageai integration package including Layout Analysis, Text Embedding, #solarllm Chat, and Groundedness Check! Simply get your key at https://
console.upstage.ai, and LangChain will take care of the rest. Enjoy langchain_upstage! -
Upstage releases LangChain integration package with Solar models
By
–
Upstage integration package @upstageai has released their `langchain-upstage` integration package, including their Solar chat model, embedding model, a task-specific groundedness check model, and a layout analysis document loader. Check them out here:
-
Sinequa Facilitates Access to NASA Open-Source Data
By
–
[#Article] #Opensource: How Sinequa Facilitates Access to NASA Data https://actuia.com/actualite/open-source-comment-sinequa-facilite-lacces-aux-donnees-de-la-nasa/
… #AI #artificialintelligence -
Free Drupal 9 Quick Code Reference Guide for Developers
By
–
If you're a Drupal developer and haven't bookmarked this (free) quick code reference guide yet, you may want to… https://
selwynpolit.github.io/d9book/ #drupal #drupaldevs #programming -
Extending LLaMA 3 Context Window with PoSE Method
By
–
Starting on using PoSE to extend LLaMA 3's context window, with @winglian
's help. This one was really fun to work on. Eval loss started insanely high, let's see how far down we can get it. -
Extending L3 70B context length with Unsloth on single A100
By
–
I almost always use (and LOVE) Axolotl… but wanted to try Unsloth to try to extend L3 70B's ctxlen w/ just one A100.
-
LLaMA 3-Instruct Dataset Format Structure Guide
By
–
Specifically, I have a dataset in the LLaMA 3-Instruct format:
“`
system {{ system_prompt }}user {{ user_message_1 }}assistant {{ -
SFTTrainer Response Template Training Logic Issue
By
–
Is it just me, or does the @huggingface SFTTrainer's approach to training only on completions make no sense? What if your response template shows up multiple times in your example? Am I misunderstanding this?