Frequently Asked Questions
Q: What is Njx’Log?
A: A public notebook on the engineering side of modern LLM training: harness and rollout systems, model architecture, mid-training and post-training, evals and benchmarks, reasoning RL and agentic RL, RSI, data synthesis — the details that usually hide inside framework defaults.
Q: What kind of posts will appear here?
A: Mostly field notes and mechanism teardowns: what a knob actually does, what breaks when you turn it, and what the metric said. Longer posts often come with an interactive report edition where the charts are explorable.
Q: What are the interactive reports?
A: The Reports section turns longer posts into standalone interactive pages — formulas become sliders, curves become probes. Each report is linked from its matching blog post.
Q: What will you not write here?
A: I try not to write generic AI commentary. If a post has no code, no experiment trace, no eval detail, or no honest uncertainty, it probably belongs somewhere else.
Q: Can I cite, translate, or share a post?
A: Yes. Keep a link to the original post, and feel free to tell me where it traveled. I like seeing notes find another desk.
Q: Are you open to collaboration or conversation?
A: Yes, especially around RL infrastructure (rollout and serving), post-training recipes, eval design, agentic systems, and strange little tools that make research easier to use.
Q: How can I reach you?
A: The cleanest paths are GitHub, Hugging Face, or X.