LLMs-from-scratch
reor
LLMs-from-scratch | reor | |
---|---|---|
11 | 7 | |
18,902 | 6,379 | |
- | 34.6% | |
9.6 | 9.9 | |
7 days ago | 4 days ago | |
Jupyter Notebook | TypeScript | |
GNU General Public License v3.0 or later | GNU Affero General Public License v3.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
LLMs-from-scratch
- Evaluating LLMs locally, on a laptop, with Llama 3 and Ollama
-
Ask HN: What are some books/resources where we can learn by building
By happenchance today I learned that Manning recently started working on publishing a X From Scratch series, which currently includes:
* Container Orchestrator: https://www.manning.com/books/build-an-orchestrator-in-go-fr...
* LLM : https://www.manning.com/books/build-a-large-language-model-f...
* Frontend Framework: https://www.manning.com/books/build-a-frontend-web-framework...
- Finetuning an LLM-Based Spam Classifier with LoRA from Scratch
- Finetune a GPT Model for Spam Detection on Your Laptop in Just 5 Minutes
- Insights from Finetuning LLMs for Classification Tasks
-
Ask HN: Textbook Regarding LLMs
https://www.manning.com/books/build-a-large-language-model-f...
- Comparing 5 ways to implement Multihead Attention in PyTorch
- FLaNK Stack 29 Jan 2024
-
Implementing a ChatGPT-like LLM from scratch, step by step
The attention mechanism we implement in this book* is specific to LLMs in terms of the text inputs, but it's fundamentally the same attention mechanism that is used in vision transformers. The only difference is that in LLMs, you turn text into tokens, and convert these tokens into vector embeddings that go into an LLM. In vision transformers, instead of regarding images as tokens, you use an image patch as a token and turn those into vector embeddings (a bit hard to explain without visuals here). In both text or vision context, it's the same attention mechanism, and it both cases it receives vector embeddings.
(*Chapter 3, already submitted last week and should be online in the MEAP soon, in the meantime the code along with the notes is also available here: https://github.com/rasbt/LLMs-from-scratch/blob/main/ch03/01...)
reor
-
Ask HN: Has Anyone Trained a personal LLM using their personal notes?
[4] https://github.com/reorproject/reor
- FLaNK Stack Weekly 19 Feb 2024
-
Show HN: Reor – An AI note-taking app that runs models locally
Seems promising, but I couldn't get it to work at all. Seems like I'm not the only one having issues: https://github.com/reorproject/reor/issues
- FLaNK Stack 29 Jan 2024
-
Went down the rabbit hole of 100% local RAG, it works but are there better options?
I am working on local RAG that encapsulates all the embedding & llm logic inside of an electron app. It is meant to be kind of a writing/personal knowledge management app powered by local models: https://github.com/reorproject/reor
What are some alternatives?
s4 - Structured state space sequence models
pong-wars
obsidian-releases - Community plugins list, theme list, and releases of Obsidian.
Deep_Object_Pose - Deep Object Pose Estimation (DOPE) – ROS inference (CoRL 2018)
get-the-news-rss-atom-feed-summary - Get a summary of the most recent news from an RSS or Atom feed using Amazon Bedrock.
java-snapshot-testing - Facebook style snapshot testing for JAVA Tests
llm-classifier - Classify data instantly using an LLM
Weaviate - Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of a cloud-native database.
llm_steer - Steer LLM outputs towards a certain topic/subject and enhance response capabilities using activation engineering by adding steering vectors
hellodata-be - The Open-Source Enterprise Data Platform in a single Portal
prql - PRQL is a modern language for transforming data — a simple, powerful, pipelined SQL replacement
recon-ng - Open Source Intelligence gathering tool aimed at reducing the time spent harvesting information from open sources.