Automated README summary
Automatically summarized from the README on 6 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
Measures whether Jev's probabilities stay honest on a rule-generated support-ticket task it cannot have seen, plus three public benchmarks. Publishes raw responses, ECE, temperature refit, and per-type miscalibration sign. For researchers evaluating Jev's epistemic honesty.
How it uses Jev
Jev answers Choice, Score, and Noul questions on support-ticket states and benchmark items. Its returned probabilities are compared to labels to compute accuracy, ECE, Brier, NLL, and a refit temperature. Results show overconfidence on Choice/Score and underconfidence on Noul for the synthetic task.
Primitives:choicescorenoul
Technique worth stealing
Refit temperature with endpoint floor at half a grid step to avoid artefacts from exact zeros.
Try it
Run scripts/endpoint_refit.py --dump results/jev_synth.jsonl
Model judgments from README excerpts
Project signals use repository metadata, an existing summary and the first 700 README characters, not the full project. “Not found” means not detected in that material; it does not establish absence. These are the original judgments with recorded models and dates, not source audits or run results.
Models & evaluation dates
- Inclusion gate
jev-1.13.0- Project signals
jev-1.13.0- Category
jev-1.13.0- Pattern tags
jev-1.13.0
Evidence
These signals use repository metadata, an existing summary and at most 700 cleaned README characters. ≥ 0.60 means reported in that excerpt; ≤ 0.40 means not found there. The middle band is inconclusive. None is a source-code audit or a test result. The runnable-tool question excludes research experiments; the recommendation question asks for a learning shortlist, not a quality verdict.
- Jev is core to the projectFound in material0.77
- Shows a System One patternnot found in the excerpt0.30
- Describes handling uncertaintyFound in material0.88
- Reports measured numbersFound in material0.79
- Is a runnable toolnot found in the excerpt0.06
- Top-10 learning shortlist signalnot found in the excerpt0.36
- Is a replica of the modelnot found in the excerpt0.03
- Solves a common problem (0–2)score on a 0–2 scale1.46 / 2
- About JevFound in material0.99
no-judgment band (0.40–0.60)
no-judgment band (0.40–0.60)
The Jev model and date for each judgment stage are listed above. Card written by DeepSeek V4.1 Flash from the README on 6 Oct 2026.