docs(benchmark): correct the #1782 record's byte-identity claim - #1942
Merged
Merged
Conversation
The 2026-09-11 record states that greedy text is byte-identical to classic decode on every after arm including block 16. Issue #1935 and PR #1939 established that this never held: the DFlash burst prefilled the prompt through `forward_speculative`, which drives full-attention layers one query position at a time instead of the single batched causal attention classic decode runs, so the burst entered its first round from different KV and gated-delta state at every width. `attend_per_position` predates the measurement, so the property had not held when the record was written rather than regressing afterwards. The record is corrected in place with a banner rather than edited, matching how the superseded #1820 record is handled, because #1797's issue body already cites this table and a reader arriving at the file directly needs the correction there. The banner also withdraws the attribution of block 16's divergence to the ignored chain-parity flag, which PR #1939 showed changes acceptance and burst time without changing the completion, and flags that every acceptance-dependent number needs re-checking against the fixed prefill. Refs #1935
inureyes
added
status:review
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The 2026-09-11 record for #1782 claims greedy text is byte-identical to classic decode on every after arm including block 16. Issue #1935 and PR #1939 established the claim never held: the DFlash burst prefilled the prompt through
forward_speculative, driving full-attention layers one query position at a time rather than the single batched causal attention classic decode runs, so the burst started every round from different KV and gated-delta state.attend_per_positionlanded 2026-05-24, before the measurement, so this was never a regression.Corrected in place with a banner rather than by editing the text, matching the superseded #1820 record. #1797's issue body already cites this table, so a reader arriving at the file needs the correction there rather than only in the newer record.
The banner also withdraws the record's attribution of block 16's divergence to the ignored chain-parity flag, and marks every acceptance-dependent number as needing re-checking, since the prefill fix changes the state each round starts from.
Refs #1935