Draft: SOTA+ TTT + RoPE50K + EMA + Curriculum (pending H100 run)#223
Draft
0xjaishy wants to merge 7 commits intoopenai:mainfrom
Draft
Draft: SOTA+ TTT + RoPE50K + EMA + Curriculum (pending H100 run)#2230xjaishy wants to merge 7 commits intoopenai:mainfrom
0xjaishy wants to merge 7 commits intoopenai:mainfrom
Conversation
added 4 commits
March 21, 2026 00:33
… validation script - records/track_10min_16mb/2026-03-20_AllInOne_SmearGate_Int6QAT_SlidingWindow/ - scripts/validate_submission.py (CPU checks, no CUDA) - docs/WITHOUT_GRANT.md, docs/GRANT_APPLICATION.md Made-with: Cursor
Made-with: Cursor
…nimal) Made-with: Cursor
Rebuild from the proven openai#1 submission (PR openai#198, 1.1326 BPB) and stack four untried improvements: - RoPE base 50K (smoother position interpolation at seq2048) - LAWA-EMA replacing periodic SWA (continuous exponential moving average) - Context-length curriculum (seq1024 early for 60% more steps, seq2048 late) - Full-model SGD test-time training (1 epoch, lr=3e-4, on val data) Architecture: 11L 512d MLP3x SmearGate BigramHash OrthoInit WD=0.04 Artifact: ~15.7MB (int6+zstd-22), 26.8M params, FA3 with SDPA fallback Pending 8xH100 run. Target: sub-1.13 BPB. Made-with: Cursor
0xjaishy
changed the title
added 3 commits
March 21, 2026 01:53
Single map of GitHub vs Mac workspace; scripts are not part of the CUDA submission artifact but back up local workflow. Made-with: Cursor
…ission - Document one clone only (parameter-golf-fork); data/.venv stay local gitignored - README: sample_fineweb_tokens, Mac submission notes, prep checklist - HANDOFF: remove duplicate Desktop workspace; point to this repo only Made-with: Cursor
…DOFF validate cmd Made-with: Cursor
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.
SOTA+ submission: PR #198 base + 4 untried improvements
Target: sub-1.13 BPB (pending 8xH100 run)
Base: PR #198 Stack (current #1 at 1.1326 BPB)
New techniques (none tried on the #198 stack before)
Architecture
torchrun --standalone --nproc_per_node=8 train_gpt.pyExpected outcome
Status