docs(ml): frontier-context note on the model ledger - #147
Merged
Conversation
Same-instrument numbers from the protocol ledger: GLM-5.2 2.51 DER plain completion vs GLM-5.3-Flash 8.57 / GLM-5.3 9.98 raw on SadeedDiac-25 (1,200 paragraphs, windowed zero-skip evaluator); attribution: wrong vowels, not convention. Dedicated distilled models framed as the reproducible, protocol-pinned path. Links rababa/docs/RESULTS.md for the disclosed rows.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds a frontier-context note under the model ledger on /ml:
Same-instrument numbers from the rababa protocol ledger: GLM-5.2 2.51 DER plain completion vs GLM-5.3-Flash 8.57 / GLM-5.3 9.98 raw DER on SadeedDiac-25 (same 1,200 paragraphs, windowed zero-skip evaluator); the regression is wrong vowels, not writing convention. Frames dedicated distilled models as the reproducible, protocol-pinned path; links the disclosed rows in rababa/docs/RESULTS.md.
glm-4.7-flash row lands separately (79/1,200 paragraphs still 429-blocked at the provider); this note covers the three rows already ledgered with bootstrap CIs and will be amended with the fourth when it lands.
Checks: site test 34/34 (whitespace guard), astro check clean, full unit suite 255/255.