fix(workshop): serialize LLMP histories at return-scale precision#14
Merged
Conversation
Reading an actual rendered prompt (per the new trace-audit rule) showed the default precision=2 quantizes daily log-return histories and return covariates to +/-0.01/0.00 — the model was forecasting from coarsened near-noise. All four workshop LLMP factories (sp500 + tsx, +/-cov) now serialize at precision=5. Prior LLMP smoke artifacts describe the quantized-input variant; a re-smoke follows this change. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Reading an actual rendered prompt (per the trace-audit rule) showed the default
precision=2quantizes daily log-return histories and return covariates to ±0.01/0.00 — the LLMP rungs were forecasting from coarsened near-noise. All four workshop LLMP factories (sp500 + tsx, ±cov) now serialize atprecision=5. Prior LLMP smoke artifacts describe the quantized-input variant; the forced re-smoke (with opus-4-8 in the matrix) follows this merge.675 passed, 7 skipped; pre-commit clean.
🤖 Generated with Claude Code