Automated README summary
Automatically summarized from the README on 20 Oct 2026; not a run by jevbooks. Run instructions and author-reported outcomes still need verification.
What it is
A calibration study of TypeSafe Jev and open-weights Laya against ~150,000 human ratings on 4,554 short messages, plus stdlib-only Python scripts (metrics.py, recal.py, recal_curve.py, transfer.py) so others can run the same check on their own data.
How it uses Jev
Jev is asked the same yes/no question about every message and returns a probability. The study compares that number to the fraction of ~24 human raters who agreed, measuring calibration and ordering, then fits an isotonic correction on labelled examples. The scripts take model-score/label pairs and report whether the problem is scale or ordering.
Primitives:score
Technique worth stealing
Separate ordering from scale: rank with the raw score, and only trust the number after fitting a correction on your own labelled examples.
Try it
git clone the repo, then run python3 metrics.py results-compact.jsonl --arm jev-full --label jev (Python 3.10+, no installs).
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 projectnot found in the excerpt0.22
- Shows a System One patternnot found in the excerpt0.14
- Describes handling uncertaintynot found in the excerpt0.05
- Reports measured numbersFound in material0.82
- Is a runnable toolnot found in the excerpt0.27
- Top-10 learning shortlist signalnot found in the excerpt0.21
- Is a replica of the modelnot found in the excerpt0.08
- Solves a common problem (0–2)score on a 0–2 scale1.85 / 2
- About JevFound in material0.97
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 20 Oct 2026.