Awesome, thanks a lot for this direct head-to-head comparison! Actually, after reading the R1 paper, wasn't the pure motivation behind GRPO computational (/memory) efficiency?
@rasbt
-
Adding Official OpenAI Transformer Weights to Notebook
By
–
Ah nice! I thought they only had the transformer-library-formatted weights on the hub! In that case, I can actually add that to the notebook directly. Cool stuff! (Btw the reason why I opted for the OpenAI Tf ones was that they were the original/official ones)
-
Lower-precision formats and distilled models enabling local long-context LLMs
By
–
Me neither :). I think with new lower-precision formats, and better distilled models, we'll maybe increasingly adopt long-context LLMs run locally.
-
Long-context needle-in-the-haystack improvements driven by cost optimization
By
–
I think that's exactly what's happening but for cost reasons rather than accuracy reasons. I think long-context needle-in-the-haystack issues have improved a lot last year since all major LLMs now have a dedicated long-context finetuning stage in pre/post-training.
-
Loading weights from Hugging Face model hub as alternative
By
–
Sorry about the hassle loading the weights directly. But I actually have some bonus material on that to load them from the HF model hub as an alternative (should be referenced in the book itself):
-
Pre-training remains superior to expensive knowledge injection methods
By
–
Interesting! But knowledge is still best instilled via pre-training. "Relatively" cheap still remains very expensive and cumbersome, even for 7B models, compared to just adding some more embeddings to a database.
-
2025: The Year of LLM Specialization and Multimodal Models
By
–
Remember when we were arguing about the term "foundation models"? It feels like ages ago! With those base models culminating in DeepSeek v3 in Dec 2024, 2025 will likely be the year of LLM specialization! 1. Multimodal LLMs: It was my big prediction for 2024. Most products
-
Best ArXiv Papers Read Throughout 2024
By
–
Ha, yeah, it's basically the "best of" of the arxiv slice I read in ~365 days in 2024 😛
-
Noteworthy AI Research Papers of 2024 Compiled Mega-Post
By
–
But before I get to the reasoning model space… if you are looking to do some focused offline reading this weekend, I just re-compiled my take on the "noteworthy AI research papers of 2024" into one PDF-export-friendly 47-page mega-post with TOC and all:
-
Fast models replacing personal notes and reasoning for writing assistance
By
–
Fast/cheap models are replacing my personal notes… Even things like "bash one-liner to do X". And yes. Reasoning models are my go to when I am writing something: "Does {text} make sense, can you find any flaws?"
Btw I don't like the term "agents" either, but hey there's also