Repository navigation
docs: deepen the README note demo - #100
Merged
Merged
Conversation
Rebuild both demo animations from a gpt-6.1-sol note on the Transformer paper, checked against the original PDF: full Table 2, ten Table 3 rows, three-line attention formula, positional and learning-rate formulas, the model configuration table, and the complexity table. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
The README demo now shows a gpt-6.1-sol note on Attention Is All You Need instead of a gpt-6.1-terra one. The middle of the note, where the depth is, now carries:
Two new in-page tags were added for the extra stops: "The easy-to-miss details, explained" and "Surprising results included".
Evidence
evals/papers/arxiv-transformer.pdf(Table 2, the ten Table 3 rows, optimizer and schedule, base/big configs, dropout, checkpoint averaging, Table 1). Encoded at 1440px q72 with-kmin 2 -kmax 3, the same settings that fixed the ghosting in docs: add the animated note demo to the README #99.Merge Danger
Door: two-way
Blast Radius: README
Only the two demo images change. They are larger than before (13.5 / 14.6 MB vs ~10 MB), so the README loads a little more slowly.
🤖 Generated with Claude Code