mirror of
https://github.com/harvard-edge/cs249r_book.git
synced 2026-08-10 06:35:05 -05:00
Minor correction on page 12 #473
Closed
opened 2026-03-22 15:42:10 -05:00 by GiteaMirror
·
4 comments
No Branch/Tag Specified
dev
gh-pages
dependabot/npm_and_yarn/design-grammar/js-yaml-4.3.1
vol1/all-final3
dependabot/uv/mlperf-edu/pypdf-6.15.0
dependabot/uv/mlperf-edu/pymdown-extensions-11.0.1
dependabot/npm_and_yarn/socratiq/dompurify-3.4.13
dependabot/npm_and_yarn/socratiq/mermaid-11.16.1
dependabot/uv/mlperf-edu/cryptography-50.0.0
dependabot/uv/mlperf-edu/aiohttp-3.14.3
dependabot/npm_and_yarn/interviews/staffml-vault-worker/multi-9648a5b66e
dependabot/npm_and_yarn/socratiq/linkify-it-5.0.2
audit/mlperf-edu-readiness
feat/mlperf-edu-precondition
vol1/all-final
main
vol1/appendices-final
vol1/ch16-final
vol1/ch15-final
vol1/ch14-final
vol1/ch13-final
vol1/ch11-final
vol1/ch12-final
vol1/ch10-final
vol1/ch9-final
vol1/ch8-final
vol1/ch7-final
vol1/ch6-final
vol1/ch5-final
vol1/ch4-final
vol1/ch3-final
vol1/ch2-final
vol1/frontmater-final
kai/fixing-profile-setting-and-map
chore/staffml-ci-path
fix/callout-flow
vol1/ch10-pass4
vol1/ch9-pass4
vol1/ch8-pass4
vol1/ch7-pass4
vol1/ch6-pass4
vol1/ch5-pass4
vol1/apC-pass3
vol1/ch4-pass4
vol1/ch3-pass4
vol1/ch2-pass4
vol1/ch1-pass4
vol1/frontmatter
vol1/apE-pass3
vol1/apD-pass3
fmt-fix
vol1/ch14-pass3
kai/clarify-community-map-totals
vol1/ch13-pass3
vol1/ch12-pass3
vol1/ch11-pass3
vol1/ch10-pass3
vol1/ch7-pass3
vol1/ch9-pass3
vol1/ch8-pass3
vol1/ch6-pass3
vol1/ch5-pass3
vol1/ch4-pass3
vol1/ch3-pass3
vol1/ch2-pass3
vol1/ch1-pass3
vol1/ch6-pass2
vol1/ch5-pass2
vol1/ch4-pass2
vol1/ch3-pass2
vol1/ch2-pass2
fix/badge-fixes
chore/precommit-cleanup
cleanup/book-validate-paths
fix/staffml-trigger-on-workflow-edits
fix/staffml-reusable-concurrency
feat/container-preflight-urls
instructors-v0.1.1
vol1-pr1957-merged
vol2-v0.2.1
vol1-v0.7.1
vol1-v0.7.0+vol2-v0.2.0
slides-v0.1.0
vol2-v0.2.0
vol1-v0.7.0
tinytorch-v0.1.13
mlsysim-v0.2.0
vol2-v0.1.2
vol1-v0.6.2
tinytorch-v0.1.12
vol2-v0.1.1
vol1-v0.6.1
mlsysim-v0.1.3
vol2-v0.1.0
vol1-v0.6.0
mlsysim-v0.1.2
staffml-v0.1.1
tinytorch-v0.1.11
labs-v0.1.0
kits-v0.1.0
slides-latest
instructors-v0.1.0
staffml-v0.1.0
mlsysim-v0.1.1
mlsysim-v0.1.0
tinytorch-v0.1.10
build-verified-windows-v1
tinytorch-v0.1.9
tinytorch-v0.1.8
mit-submission-v1
tinytorch-v0.1.7
tinytorch-v0.1.6
tinytorch-v0.1.5
tinytorch-slides-v0.1.0
tinytorch-v0.1.3
tinytorch-v0.1.4
tinytorch-v0.1.2
tinytorch-v0.1.1
tinytorch-v0.1.0
book-v0.5.1
tinytorch-audio-v0.1.1
tinytorch-audio-assets-v0.1
book-v0.5.0
book-v0.4.2
book-v0.4.1
book-v0.4.0
book-v0.3.0
book-v0.2.0
book-v0.1.0
Labels
Clear labels
area: book
area: collabs
area: kits
area: labs
area: socratiq
area: staffml
area: tinytorch
area: tools
area: website
bug
dependencies
format: epub
format: pdf
javascript
link-health
link-rot
priority-high
pull-request
staffml
type: bug
type: citation
type: code
type: errata
type: improvement
type: new
type: question
vault-sli
Mirrored from GitHub Pull Request
Milestone
No items
No Milestone
Projects
Clear projects
No projects
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/cs249r_book#473
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @lttelezynski on GitHub (Dec 31, 2025).
Originally assigned to: @profvjreddi on GitHub.
On page 12 where you talk about GPT-3 training energy consumption, a comma is used instead of a dot. It states 1,287 MWh, but should be 1.287 MWh.
@profvjreddi commented on GitHub (Dec 31, 2025):
Thanks @lttelezynski - I will check it out.
@lttelezynski commented on GitHub (Dec 31, 2025):
@profvjreddi I checked multiple sources and made some rough calculations, and it looks like 1,287 MWh is in fact more likely, so I truly apologize for confusion. However, I think that it would require at least 10 GPUs, not just 1, to be equivalent to 3-7 hair dryers in power consumption, since hair dryers consume roughly 1000-2000 W.
@profvjreddi commented on GitHub (Jan 2, 2026):
Hi @lttelezynski, thank you so much for taking the time to look into this and share your thoughts!
Please don't apologize, as we're all here to learn from one another 🤗 , and I genuinely appreciate the careful analysis you did here. The fact that you went back to check multiple sources and recalculated things yourself shows exactly the kind of critical thinking we hope to encourage in the "classroom." Our classroom here is everyone :)
You're right that 1,287 MWh is the correct figure. And your point about the power consumption comparison is a good observation. I'll take a look at that to make sure the context is clear for readers. Thanks again for engaging so thoughtfully with the material.
Contributions like yours help make this a better resource for everyone! 🙏
@profvjreddi commented on GitHub (Jan 2, 2026):
Fixed in commit
1e88b9ca8- removed the incorrect hair dryer comparison. You were right that the math didn't work out (GPUs at 300-700W are actually less than a single hair dryer at ~1500W, not 3-7x). Thanks for catching this!