Docs for calibration, the LLM tools, the ask-back and a month of cost - #26
Merged
Merged
Conversation
…ost and the Bayesian sensor
The ask-back section described the probability split. It now describes the name comparison. It also restores the "What it costs" heading that the section had replaced. The diagnostics section named the debug view as the place for the agent detail. The note now shows in the Assist dialog and as an intent-progress event in the pipeline debug view. The refusal row and the budget sensor now say what the budget actually refuses: a command whose estimate does not fit, with tokens left.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Docs for the 1.16 changes in #20 to #25. Merge this after those PRs, because the pages describe what they add.
New pages:
calibrate.md: jev.calibrate, its fields, its response, and how many occasions to collect before you trust it.llm-tools.md: the jev__noul and jev__choice tools, why the option is off by default, and what Jev sees.bayesian.md: Jev and the Bayesian binary sensor compared, when to use each, and a Jev sensor as a Bayesian observation.Changed pages:
cost.md: a month worked out for three setups, from the measured 339 tokens for one entity, 65.8 per more entity, and 1,329 to 1,371 per voice command, at $0.042 per million. Only the question preview still checks what was already spent.conversation.md: the ask-back when two devices fit the name, and the Assist debug view.limitations.md: why there is no local model. The llama.cpp/v1/decisionendpoint returns only the probability of the chosen value, which covers a noul and not a choice or a score.actions.md: a pointer to jev.calibrate.mkdocs.yml: nav, llms.txt sections and the site description.Numbers checked against the code: the 30 second floor for trigger refreshes (
request_refresh_debouncercooldown), the 250 entity target cap, the 0.2 ask-back floor, and the 5 minute expiry (CONVERSATION_TIMEOUTin Home Assistant).Gate on this branch:
EXIT 0. ruff clean, mypy clean, 407 passed and 12 skipped, 97% coverage, config_flow 100%, mkdocs --strict built, hassfest 0 invalid.