submission: LeakyReLU2 + TrigramHashEmbedding (1.1448 bpb)#884
Open
BhatiaUday wants to merge 1 commit intoopenai:mainfrom
Open
submission: LeakyReLU2 + TrigramHashEmbedding (1.1448 bpb)#884BhatiaUday wants to merge 1 commit intoopenai:mainfrom
BhatiaUday wants to merge 1 commit intoopenai:mainfrom
Conversation
Novel hash-based trigram embedding on PR openai#414 stack with LeakyReLU(0.5)^2. 3-seed mean sliding window val_bpb: 1.14485, all artifacts under 16MB. Validated on 1xH100 NVL with proportional wallclock (4054s = 600s x 6.76). Track: non-record (1xH100 validation).
70c9ab9 to
16e1e8c
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
LeakyReLU² + TrigramHashEmbedding (non-record track)
val_bpb: 1.1448 (3-seed mean, sliding window stride 64) | ~15.6 MB
Summary
Novel hash-based TrigramHashEmbedding on PR #414 stack (11L EMA + GPTQ-lite) with LeakyReLU(0.5)² from PR #549. The trigram embedding extends BigramHashEmbedding (PR #198) from 2-grams to 3-grams using XOR prime hashing into 2048 buckets, capturing richer local context at the input layer.
Results (3 seeds, 1×H100 NVL, proportional wallclock)
Compute Note
Validated on 1×H100 NVL 96GB with proportional wallclock (4054s = 600s x 6.76) to match the 8xH100 training trajectory. The script uses
grad_accum = 8 // world_size(auto: 1 on 8-GPU, 8 on 1-GPU) for identical effective batch size. Defaults toMAX_WALLCLOCK_SECONDS=600on 8xH100.Key Changes from Base
Files
records/track_non_record_16mb/2026-03-26_LeakyReLU2_TrigramHash/train_gpt.pyrecords/track_non_record_16mb/2026-03-26_LeakyReLU2_TrigramHash/submission.jsonrecords/track_non_record_16mb/2026-03-26_LeakyReLU2_TrigramHash/requirements.txtrecords/track_non_record_16mb/2026-03-26_LeakyReLU2_TrigramHash/README.md