---
title: "Discussion on \"Stored depth vs executed depth: how to read a looped model card\" | Hashnode"
description: "Discussion on \"Stored depth vs executed depth: how to read a looped model card\". A 36-layer model can run 96 layers. Here's what that changes about the numbers you compare."
url: "https://hashnode.com/posts/stored-depth-vs-executed-depth-how-to-read-a-looped-model-card/6afa2a7ef63d30664c3f80bf"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Stored depth vs executed depth: how to read a looped model card](https://piszczek.hashnode.dev/stored-depth-vs-executed-depth-how-to-read-a-looped-model-card?utm_source=hashnode&utm_medium=feed)

A 36-layer model can run 96 layers. Here's what that changes about the numbers you compare.

1d ago · 4 min read · piszczek.hashnode.dev  
Read the article

-   [#ai](https://hashnode.com/tag/ai)
-   [#machine-learning](https://hashnode.com/tag/machine-learning)
-   [#llm](https://hashnode.com/tag/llm)
-   [#deep-learning](https://hashnode.com/tag/deep-learning)

## Written by

[Michał Piszczek](https://hashnode.com/@michalpiszczek)

@michalpiszczek · Kraków, Poland

  
[View profile](https://hashnode.com/@michalpiszczek)

CEO / CTO / Founder. AI infrastructure & economics. Author of the Joule Wars concept — piszczek.pl

CTO of Archdesk; founder and former CEO of Robotero (acquired), Rejsomat.pl, Inclify and Lextron.ai. Ex white-hat hacker. I write about AI infrastructure &amp; economics — author of the Joule Wars concept. Home: https://piszczek.pl

## Responses

Markdown supported · ⌘↵ to post

No responses yet.
