There's new Instruct checkpoints as well!
OPEN SOURCE
-
OLMo 2 introduces reordered norm and QK-norm innovations
By
–
Yo! @allen_ai just dropped OLMo 2 Tech report, some interesting things I found Architecture
> Reordered norm: Normalizing outputs of attention and feedforward layers within transformer blocks instead of inputs
> QK-norm: Normalizing key and query projections with RMSNorm -
LoRA Tutorial and Additional Tips from Readers
By
–
Glad you found it useful, and thanks for the kind words! I think you probably already read that, but there is a LoRA from-scratch tutorial in the last Appendix. There was a reader how also had some useful additional tips regarding LoRA:
-
Volvo’s Three-Point Belt Patent: Precedent for Open AI Innovation
By
–
In 1959, Volvo set a powerful precedent by making the three-point seat belt patent-free, prioritizing lives over profits. As we step into 2025, we need the same boldness in AI innovation—where groundbreaking advancements are shared openly to benefit humanity. Let’s ensure AI
-
LLM Fine-Tuning Notebooks with OpenAI and Hugging Face
By
–
check out some of my latest notebooks containing llm fine-tuning using OAI and HF fine-tuners https://
github.com/avikumart/LLM-
GenAI-Transformers-Notebooks/tree/main/TMLC_LLM_projects
… -
LangChain maintains high issue ratio despite large star count
By
–
fun fact, langchain has a super high ratio as well, despite a very high star count
-
LangGraph: Production-Ready Agent Framework Low Level Control
By
–
Because langgraph is a production ready agent framework… it’s very low level, highly controllable, no hidden prompts Helps you get to production without getting in the way, so high usage But a bit of a steeper learning curve (we’re working on that!) so less GitHub stars
-
Open Source Advocacy Impacts Policymakers Through Nick Clegg
By
–
Thank you for your work championing open source to policymakers @nickclegg — this has made a real difference!
-
Qwen2.5 7B vs Ministral 8B: Detailed Model Testing and Comparison
By
–
right now qwen2.5 7b or ministral 8b. I’m running a more detailed test right now actually
-
Model Context Protocol Development Roadmap Published with Key Features
By
–
We just published a near-term development roadmap for the model context protocol (MCP) Some highlights:
– Remote support (and auth!)
– Reference implementations
– Better package management
– Agent support