Ask me your LLM prompting/fine-tuning questions. Have an hour, will answer as many as I can.
LLMS
-
Mistral 7B Improvements Enable Better Fine-Tuning Capabilities
By
–
This is huge news. Mistral 7B was already the best model in its size class, and these improvements are a huge step up. I’ll be re-training many of my current fine-tunes over this model ASAP.
-
Google GIVT: Generative Infinite Vocabulary Transformers Breakthrough
By
–
https://
drdebashisdutta.com/introducing-gi
vt-google-researchs-breakthrough-in-real-valued-vector-sequences-with-generative-infinite-vocabulary-transformers/
… -
LangChain Chatchat: Open Source Local Q&A Application
By
–
LangChain Chatchat A local Q&A application. The goal is to build a KBQA(Knowledge based Q&A) solution that is friendly to Chinese scenarios and open source models and can run both offline and online. 25k stars on Github! https://
github.com/chatchat-space
/Langchain-Chatchat/blob/master/README_en.md
… -
Advanced RAG: Generation and Evaluation Techniques
By
–
Advanced RAG Series: Generation and Evaluation In the fifth part of this series, we look at techniques for: Generation (CRAG, Self-RAG, RRR) Evaluation (RAGAs, Langsmith, DeepEval) Another awesome blog by @divyanshu_van https://
div.beehiiv.com/p/advanced-rag
-series-generation-evaluation
… -
Better LLM Reward Model Performance Without Numerical Scales
By
–
If you’re using out of the box LLMs as a reward model/sample rater, here’s a trick to get way better performance: Don’t ask the LLM to rate examples on a numerical scale (i.e. 1-5). The model will almost always choose 1 or 5. Instead, use words as rating options (“very bad”,
-
Axolotl and Together Compute for Model Fine-tuning
By
–
I use Axolotl, but services like @togethercompute might be easier. Last I checked, they don’t support only training on outputs, which is important. May have changed since!
-
Fine-tuning LLMs: Progressive Data Selection Strategy
By
–
High alpha technique for fine-tuning LLMs: First, fine-tune your model on all your data Then, use AI to pick a smaller set of really good examples, and tune on them Finally, hand-pick and tune on a few hundred incredible examples >> Huge performance gains <<
-
Building RAG Systems for Powerful Data Access
By
–
How to Make a RAG System to Gain Powerful Access to Your Data This article is a great introduction to RAG, and walks through a few important steps of a RAG pipeline: Retrieve Data
Pre-process Data
Implement RAG
Test https://
towardsdatascience.com/how-to-make-a-
rag-system-to-gain-powerful-access-to-your-data-caf4bb9186ea
…
