Lilian Weng | LinkedIn
linkedin.com · 1,309 words · saved by 1 readers
13,139 followers 13,139 followers From Lilian's school From Lilian's school
Wei Jing (William) Wong NVIDIA • 1K followers NVIDIA Nemotron-RL-Agentic-Terminal-Pivot, the open agentic RL dataset released alongside Nemotron 3.5 Lightning, is trending on Hugging Face this week, in the top 10 out of more than a million public datasets. It is the reinforcement learning data used to post-train Lightning's command-line agent capabilities: training samples for terminal-use agents, built to run in NeMo Gym, and ready for commercial use. Useful for anyone building agents that plan, run commands, and recover from errors. Lightning is a 30B MoE (3B active) built for…
related reading
- RLHF: Reinforcement Learning from Human Feedbackhuyenchip.com
- Composer2.pdfcursor.com
- Inkling: Our Open-Weights Model - Thinking Machines Labthinkingmachines.ai
- State of RL for reasoning LLMs | A. Weersaweers.de
- Andrej Karpathy — AGI is still a decade awaydwarkesh.com
- A Taxonomy of RL Environments for LLM Agentsleehanchung.github.io
- PostTrainBenchposttrainbench.com
- LLM Resourcesforrestbicker.com
- RL Environments and RL for Science: Data Foundries and Multi-Agent Architecturesnewsletter.semianalysis.com
- karl.pdfdatabricks.com
- Mind Labmacaron.im
- nemotron-3-super-120b-a12b Model by NVIDIA | NVIDIA NIMbuild.nvidia.com