Writing log70 posts277 tagslatest October 8, 2026
Open-source.
Latest posts
What Crosses the Bridge Between Two Games? Twelve Passthrough Mods Compared
This post opens the code of twelve passthrough mods that merge two running games into one and sorts them by what crosses the bridge: pixels, meshes, logic, or nothing at all.
Inside the Bridge That Runs Minecraft in Elden Ring: How Do Two Games Share One Frame?
This post reads minecraft-crossover-bridge and walks through what its shared-memory protocol, depth-tested compositing, terrain raycasts and hidden stand-in character each solve.
Archify dissected: what changes when the agent writes an IR instead of drawing the diagram?
Archify has the agent write a typed JSON IR and lets a deterministic compiler turn it into a self-contained HTML artifact. I read the repository end to end and built four diagrams with it, watching what the validator refuses, why deliver, visual-check, and human review are kept apart, and what changes when this blog’s own mermaid is ported over.
Llama 3 (2024) Paper Notes
Paper notes on The Llama 3 Herd of Models: 15T tokens, a 405B flagship, 128K context, and a design philosophy of 'managing complexity' (dense over MoE, SFT → rejection sampling → DPO instead of PPO) — the paper where an open-weight model first reaches GPT-4-class performance. It completes Llama 2's skeleton through scale, data, and simplicity.
Llama 2 (2023) Paper Notes
Paper notes on Llama 2 covering its commercial license, a base model scaled to 2T tokens with 4k context and GQA, and the Llama 2-Chat alignment pipeline (SFT → two reward models → iterative RLHF with rejection sampling + PPO → GAtt). It fuses LLaMA's open base lineage with InstructGPT's RLHF at production quality.
LLaMA (2023) Paper Notes
Paper notes on LLaMA covering its core strategy — train smaller models for longer on publicly available data, optimized for inference — the result where a 13B model beats 175B GPT-3, the RMSNorm/SwiGLU/RoPE architecture changes, the 2048-A100 training scale, and how it kicked off the open LLM ecosystem.
Paperclip Architecture Analysis - AI Agent Virtual Company Orchestration
An architectural analysis of Paperclip, an open-source control plane that operates AI agents as a single virtual company under an org chart, budget, and governance structure.