Merged
Conversation
- make_figures.py generates all paper figures from results_baseline.tsv and results_enhanced.tsv (fig1 head-to-head, fig2 efficiency, fig3 cost, progress.png README teaser) - progress.png: new teaser showing baseline vs web-enhanced convergence curves with paper-citation annotations on the 5 kept improvements (replaces karpathy's original placeholder image) Run: uv run python make_figures.py Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Separate training cost (always local) from code agent cost (optional API) - Add per-experiment and 53-run cost estimates for Claude Haiku, Sonnet, GPT-4o - Remove inaccurate "no cloud billing, no API tokens" claim - Show that even with GPT-4o API, local training is still ~5x cheaper than full cloud H100 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Reverting our custom progress.png while we redesign the experiments to start both conditions from the same raw config for a proper A/B comparison. Will regenerate figures once new enhanced_v2 data is ready. make_figures.py is kept for when we re-run. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2imi9
added a commit
that referenced
this pull request
Apr 1, 2026
The original train.py selected FA3 repo by compute capability: - varunneal/flash-attention-3 for Hopper (SM 9.0) - kernels-community/flash-attn3 for all other GPUs This was lost during the V4 rewrite. Restored now. Note: neither repo supports Blackwell (SM 12.0) yet — falls through to FlexAttention. See #1, #4. Co-Authored-By: Claude Opus 4.6 <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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
No description provided.