"Machine Learning Engineering Open Book" by Stas Bekman This open book has a lot of information on the Engineering aspects of building DL / ML models specifically LLM and Multi-modal models. This open book is continuously being updated. A pdf version can also be downloaded.
@sudalairajkumar
-
Allen AI releases OLMo open language model with technical documentation
By
–
Technical Report: https://
allenai.org/olmo/olmo-pape
r.pdf
…
Blog: https://
blog.allenai.org/olmo-open-lang
uage-model-87ccfc95f580
…
Github: https://
github.com/allenai/OLMo
Models: -
AI2 Launches OLMo Open Language Model Framework
By
–
@allen_ai launched "OLMo: Open Language Model". AI2 opens its framework that provide access to data, training code, models, and evaluation code. The training data Dolma is also released, a dataset of 3 trillion tokens from a diverse mix of web content, academic
-
AI LLM Retrieval Evaluation with LlamaIndex
By
–
Link to blog: https://
srk.ai/blog/004-ai-ll
m-retrieval-eval-llamaindex
… -
RAG Encoder and Reranker Evaluation with Open-Source Models
By
–
Inspired by @ravithejads
, wrote a simple blog post on "RAG – Encoder and Reranker evaluation" using @llama_index on a custom QA dataset. Used open-source embeddings and re-rankers for the evaluation
"JinaAI-base" embedding with "bge-reranker-large" gave the best result among -
Building Enterprise RAG Systems: Key Components and Failure Points
By
–
A very good blog post by @nlpguy_ on "Building Enterprise RAG Systems" that covers Common failure points of RAG systems Guardrails on inputs and outputs Retrieval part including query rewriter, encoder, chunker, reranker. Storage part including indexer, data
-
New Book on Kaggle Notebooks Covers GenAI and Deepfakes
By
–
Got my copy of "Developing #Kaggle Notebooks" by @PredaGabi Glad to see that a section of the book covers #GenAI, Deep fake as well. Excited to read the same this week!
-
Tensoic Releases Kan-LLaMA 7B Kannada-Optimized Language Model
By
–
@tensoic team has released Kan-LLaMA [ಕನ್-LLama] — a 7B Llama-2 LoRA PreTrained and FineTuned specifically for "Kannada" tokens. Key highlights: ✦ Llama-2's existing vocabulary has been expanded from 32K tokens to a comprehensive 48k (49420) tokens, enhancing its efficiency
-
Tamil-Llama: New LLaMA-Based Language Model for Tamil
By
–
Abhinand Balachandran has unveiled "Tamil-Llama," an LLM based on the LLaMA model, with a specific focus on the Tamil language. Key highlights:
✦ This model is built upon the LLaMA model, incorporating additional Tamil tokens and leveraging the LoRA methodology for efficient
