intro-to-large-language-models SKILL.md skillA
intro-to-large-language-models from michalstrnadel/source-to-skill is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add michalstrnadel/source-to-skill/intro-to-large-language-models@git:20261004.d77ddfaPins git:20261004.d77ddfa in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- michalstrnadel/source-to-skill
- Path
examples/intro-to-large-language-models/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261004.d77ddfa, committed Oct 4, 2026
- Size
- 8,092 bytes
- Repository stars
- 7
- License
- MIT
Andrej Karpathy's 1-hour "Intro to Large Language Models" talk (Nov 2023) distilled - what an LLM is (two files - weights plus run code), how it is trained (pretraining as lossy compression of the internet, fine-tuning into an assistant, RLHF), scaling laws, tool use and multimodality, System 1/2 thinking, self-improvement, the "LLM OS" analogy, and LLM security (jailbreaks, prompt injection, data poisoning). Load when explaining how LLMs work to a non-specialist, reasoning about training…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Intro to Large Language Models (Andrej Karpathy) Source: [1hr Talk] Intro to Large Language Models — Andrej Karpathy, YouTube, uploaded 2023-11-23, 59:48. https://www.youtube.com/watch?v=zjkBMFhNj_g A re-recording of a 30-minute "busy person's intro" talk. Three acts: what LLMs are and how they are made, where they are going, and how they get attacked. **Date-sensitive:** numbers, leaderboards and "LLMs can't do X yet" claims describe the state of the field in November 2023. ## Core ideas 1. **An LLM is two files.** A parameters file (Llama 2 70B: 70B params × 2 bytes in float16 = 140 GB) plus a small program that runs them (~500 lines of dependency-free C). Runs offline on a MacBook. The architecture is fully understood; all the "magic" is in the parameters. → [01](segments/01-intro-and-llm-inference.md) 2. **Pretraining = lossy compression of the internet.** Llama 2 70B: ~10 TB of crawled text, ~6,000 GPUs, ~12 days, ~$2M → 140 GB of weights (~100× compression, lossy — a "gestalt", not a copy). Frontier models were 10×+ beyond this, costing tens to hundreds of millions of dollars. → [02](segments/02-llm-training.md) …
Read the whole file at its exact version.
How to install
mdr add michalstrnadel/source-to-skill/intro-to-large-language-models@sha256:755a3d93bb9ccd2eThe line at the top of this page pins the label git:20261004.d77ddfa, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_uwqr5wyls77q5ghr)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 4, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (8092 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
michalstrnadel/source-to-skill · 7 stars · license MIT · pushed 2026-10-05 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_uwqr5wyls77q5ghr GET https://markdownregistry.com/api/v1/resolve?ref=michalstrnadel/source-to-skill/intro-to-large-language-models GET https://markdownregistry.com/api/v1/blob/755a3d93bb9ccd2e1a8456c09843ee80dcafcd5638b542693e3403042039509e
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.