PR #287 base + Overtone init + 12L option#5
Closed
machdragon wants to merge 1 commit intomainfrom
Closed
Conversation
PR openai#287 (SOTA val_bpb=1.1271) with Overtone embedding init added. Overtone reshapes tok_emb singular values to power-law decay for better int6 quantization. No KURE, R2, tanh, or TTT — clean minimal delta over proven SOTA. Modal launcher with CLI flags for EMA decay, num_layers, and XSA sweep. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
val_bpb=1.1271, 11L XSA+EMA) as base, adds Overtone embedding init (+5 lines, zero training cost)tok_embsingular value spectrum to power-law decay (S_i ∝ i^{-0.5}) for better int6 quantization fidelityscripts/modal_train_xsa_overtone.py) with CLI flags for--ema-decay,--num-layers,--xsa-last-nsweepsRun plan
Test plan
python3 -c "import ast; ast.parse(...)"— syntax OK_xsa_efficient,use_xsa)torch.linalg.svd,target_S)EMA_ENABLED,EMA_DECAY)🤖 Generated with Claude Code