0.8128 BPB: Classical Compression Eval + N-gram Backoff on PR #549 Base#786
Open
shinegami-2002 wants to merge 2 commits intoopenai:mainfrom
Open
0.8128 BPB: Classical Compression Eval + N-gram Backoff on PR #549 Base#786shinegami-2002 wants to merge 2 commits intoopenai:mainfrom
shinegami-2002 wants to merge 2 commits intoopenai:mainfrom
Conversation
Novel approach bringing cmix/PAQ techniques (n-gram backoff, match model, APM error correction, logistic mixing) as eval-time augmentation on top of the PR openai#549 neural model stack. Initial proof of concept on 1xH100 shows compression pipeline working. Pending full 8xH100 run. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Full training run on 8xH100 SXM (7135 steps, 600s wallclock). Base model: 1.1218 BPB (sliding window). With n-gram backoff (orders 2-7) + entropy-adaptive alpha: 0.8128 BPB. Artifact: 15.88 MB. Eval time: 383s. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
shinegami-2002
changed the title
10 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Approach
Eval-time augmentation inspired by classical data compression (cmix/PAQ). Multi-order n-gram backoff with entropy-adaptive alpha, vectorized numpy implementation. All backward-looking, zero artifact cost.
Key Numbers
Test plan
Credits
Built on PR #549 base. N-gram technique inspired by PR #727. Classical compression research from cmix/PAQ.
🤖 Generated with Claude Code