Repository navigation
Document stopped Sarvam confirmation and freeze corrected evidence cases - #59
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The fresh Sarvam confirmation was stopped after 32 of 240 episodes when source review found that one question family conflated a withdrawn historical request with the currently wanted replacement. All 111 calls, original grades, backend states and the explicit interruption reconciliation are preserved; this is an incomplete comparison and no candidate is promoted. A separate v2 generator corrects only the 12 affected questions and hash bindings, retains all cases and exactly reproduces the frozen, unexecuted corrected bank.
The runner reuses the ordinary native actor once per case/model and compares the two fixed reporters on the same resulting transcript. Reports separate actual factual answers from intended task completion and receipt-owned status. README, the Indic roadmap, English judge-bridge pointer and a new Sarvam research index now qualify unsupported universal-language/novelty claims and identify the remaining conditional-evidence leads without claiming a memory-product need. Earlier roadmap text is archived.
Validation: 170 focused tests; all 543 archived files hash-verified from a clean Git export; exact offline request/receipt/SQLite/meter replay and separate raw-parser/accounting reproduction; corrected v2 source/manifest and regenerated inputs verified. No retries, unknown new usage or second paid phase. Estimated new usage INR10.225170; native cumulative retained INR257.233050 under the standing INR6000 total ceiling. Assistant language review only, not independent adjudication. Independent code, documentation, correction and evidence commits are retained.