@hao
Member since January 2026
Published Content
-
Tier 1 Deep Dive: Core LLM Techniques
================================================================================ TIER 1 DEEP DIVE: CORE LLM TECHNIQUES ================================================================================ Prerequisites: Transformer fundamentals, SFT (assumed known from nanoGPT +...
-
Modern LLM Technology Knowledge Framework
================================================================================ MODERN LLM TECHNOLOGY KNOWLEDGE FRAMEWORK ================================================================================ Background: Built GPT-2 124M (nanoGPT), fine-tuned with Dolly 15k using...
-
Claude Code proxy: use lowercase https_proxy (and include http://)
While setting up Claude Code behind a proxy, I found that the uppercase env var `HTTPS_PROXY` wasn’t picked up, but the lowercase `https_proxy` worked. One more gotcha: the proxy value needs to include the scheme (e.g., `http://`). Example: ```sh export https_proxy=http://127.0.0.1:7890 ```
-
Claude Code proxy env var: https_proxy vs HTTPS_PROXY
I was installing Claude Code with a proxy. I set the upper-case `HTTPS_PROXY` but it didn't work; the lower-case one worked. Also, it has to start with `http://`, e.g.: ```sh export https_proxy=http://127.0.0.1:7890 ```
-
Transformer Forward Pass: End-to-End Walkthrough
# Transformer Forward Pass: End-to-End Walkthrough *A detailed visual guide to how input tokens become output predictions* --- ## Overview ``` ┌────────────────────────────────────────────────────────────┐ │ HIGH-LEVEL FLOW │ │ ...
-
Attention Mechanisms Deep Dive (Part 2: Patterns, Implementation & Modern Optimizations)
# Attention Mechanisms in Large Language Models: A Deep Dive (Part 2) *Continuation: Patterns, Implementation, and Modern Optimizations* --- ## Part 6: Attention Patterns and What They Learn ### Induction Heads and In-Context Learning One of the most important discoveries in mechanistic...
-
Attention Mechanisms Deep Dive (Part 1: Foundations to Positional Encoding)
# Attention Mechanisms in Large Language Models: A Deep Dive *From foundational concepts to deep understanding of how attention powers modern AI* --- ## Table of Contents 1. [Foundations: What Problem Does Attention Solve?](#part-1-foundations-what-problem-does-attention-solve) 2. [The...
-
Ilya Sutskever, U of T honorary degree recipient, June 6, 2025
• Reflections on Educational Journey and Gratitude - Ilya Sutskever expresses gratitude for receiving an honorary degree from the University of Toronto, where he studied 20 years ago. - He shares the impact of studying under Geoff Hinton and the significant AI research conducted at the...
-
Ilya Sutskever, U of T honorary degree recipient, June 6, 2025
• Reflections on Educational Journey and Gratitude - Ilya Sutskever expresses gratitude for receiving an honorary degree from the University of Toronto, where he studied 20 years ago. - He shares the impact of studying under Geoff Hinton and the significant AI research conducted at the...
-
Claude 4 + Claude Code + Strands Agents in Action | AWS Show & Tell
• Introduction and Overview - Hosts Anil, Adam, and Scott introduce the AWS Show & Tell with excitement about discussing Claude Code and Claude 4. - Guest speakers Johnny Cook and Eric Burns from Anthropic join to talk about the latest advancements in AI models and tools. • Claude 4 and Claude...