Kev-4B (Qwen3)
Previous generation (Qwen3). This checkpoint is kept as the fast option on Apple Silicon (its attention-only backbone runs the packed forward at full speed on MPS). For accuracy and calibration use Kev-4B (Qwen3.5): on the locked test it scores 0.832 vs 0.806 out of domain against this model on the same items. Weights:
jaredpalmer/kev-4b@qwen3.
Kev-4B is a decision model: one document (the state) and a set of typed questions in, a probability distribution per question out, in one forward pass. No text generation. It is a LoRA adapter (r=16) plus a pointer head on Qwen/Qwen3-4B-Base, serving TypeSafe’s public /v1/systemone contract.
The recommended Kev. The best 4B checkpoint under a frozen, checksummed protocol after ~40 controlled 4B trials, and the first Kev within seven points of Jev out of domain on the same items. Same recipe run at three seeds: transfer 0.773 / 0.790 / 0.770; this checkpoint is the seed selected on the development partition (never on the locked test).
- Hub:
jaredpalmer/kev-4b, revision tagqwen3(trialv7-rc3/01-trial-1); the repo’s main revision now holds the Qwen3.5 checkpoint - Code, suites, every trial with hashes and paired bootstraps: github.com/jaredpalmer/kev —
PLAN.md(full record at git tagresearch-archive-2026-09-24),runs/leaderboard.md
Results (same frozen items for every row)
| Kev-0.5B (prototype) | Kev-0.6B | Kev-4B | Jev | |
|---|---|---|---|---|
| in-distribution accuracy (decision-v4 dev, 1,200 q) | 0.712 | 0.801 | 0.854 | 0.845 |
| out-of-domain accuracy (transfer-v4 dev, 560 q) | 0.561 | 0.620 | 0.790 | 0.857 |
| out-of-domain Brier | 0.50 | 0.536 | 0.328 | 0.211 |
| confident errors out of domain (p ≥ 0.9 and wrong) | – | 10.8% | 8.2% | 3.7% |
| held-out policy structures, both siblings correct | – | 0.08 | 0.73 | 0.86 |
| option-order flip rate | 0.21 | 0.07 | 0.06 | 0.00 |
Per-source out-of-domain accuracy (Kev-4B / Jev): QNLI 0.89 / 0.93, SciQ 0.99 / 0.99, TweetEval-offensive 0.75 / 0.81, PAWS 0.72 / 0.79, MMLU 0.65 / 0.90, Emotion 0.66 / 0.59, deadline (3-level date arithmetic) 0.53 / 0.93, (A and B) or not C 0.97 / 0.97, if A then not B else C 0.88 / 0.78.
Seeds: three seeds on decision-v7: transfer 0.773 / 0.790 / 0.770, held-out rule pairs 0.62 / 0.73 / 0.67 (Jev 0.86); this checkpoint is seed 1, selected on development transfer accuracy. Trained on decision-v7 (10k public records + 896 policy records over nine template families incl. four ordinal Score threshold families + 1,680 records from 60 random rule structures with negation anywhere); development/test items are byte-identical to v4, so every number here is comparable with earlier checkpoints.
Locked test, read once (runs/locked/kev-4b-v7-preview-ungated/): in-distribution 0.856 (Brier 0.211), out-of-domain 0.806 (Brier 0.294, confident errors 6.6%, held-out pairs 0.66). This partition will not be read again for this checkpoint.
What we learned building it
- Capacity dominates out of domain. With public examples and synthetic budget held equal, 0.6B → 4B is +14–19 pp; 4B → 8B is +1–7 pp.
- Fine-tuning erodes base capability, and the learning rate controls it. The 4B base, zero-shot with a letter readout, scores 0.688 on the same MMLU items and 0.787 on PAWS; the default recipe (lr 2e-4) trained down to 0.60–0.66 / 0.56–0.71. Lowering lr to 5e-5 recovers most of it and is the single largest recipe improvement we found; fewer LoRA target modules and smaller ranks help less.
- More public training data raises in-distribution accuracy and lowers transfer at 4B (10k vs 3.4k records: −3 pp). Knowledge MCQ sources (ARC, OpenBookQA, CommonsenseQA) raise in-distribution accuracy to 0.86 without moving transfer.
- Programmatic contrastive policy pairs teach the trained rule structures (both-correct 0.85–1.0) but transfer to unseen structures only partially (0.5–0.6 at 4B, 0.03–0.11 at 0.6B).
Known limits
- Held-out policy reasoning (unseen rule compositions, date arithmetic with grace periods) is far from Jev.
- Product-shaped questions with no training analogue are not guaranteed: on the TypeSafe docs example (“two charges on my card” → Is there a billing problem?) this checkpoint answers 0.48 (Kev-8B 0.95, Kev-0.6B 0.97) while picking the return reason correctly (wrong size 0.53; Kev-8B 0.84; Kev-0.6B prefers “none of the above” 0.58). Measure on your own inputs.
- Out-of-domain probabilities are usable but not calibrated (raw ECE 0.096); temperature fitted in-domain does not transfer.
- 4B fp32 needs ~16 GB; on a 32 GB Mac use
KEV_DTYPE=bf16. Latency on an H100 is ~45 ms per packed request; on an M5 several hundred ms.
Training
Frozen suite evals/v4/decision-v4: 10,000 public records (1,000 per source, ten sources) plus two programmatic policy arms of 448 records, two epochs, LoRA r=16 on attention and MLP projections, pointer head from scratch, cross-entropy on the option distribution, lr 5e-5 (OneCycle), effective batch 8, bf16 autocast with fp32 master weights, gradient checkpointing, one H100 (~40 min). Augmentation: option permutation, none-of-the-above insertion, distractors, none minimal pairs on 25% of Choice records. No Jev outputs were used for training.
Evaluation protocol
Development partitions select models; the locked test partition is read at most once per candidate. Every number carries suite hash, code hashes, and git commit in result.json. See PLAN.md at git tag research-archive-2026-09-24 (“Evidence and corrections”) for the corrections we made to our own earlier claims.
Use
uv run --extra serve python -m kev.serve --run jaredpalmer/kev-4b --port 8008 # KEV_DTYPE=bf16 on a 32 GB MacAny TypeSafe-compatible client works: TypeSafeClient(api_key="local", base_url="http://127.0.0.1:8008", model="kev-latest").
License
Apache-2.0 for the adapter and head; Qwen3 base is Apache-2.0; datasets carry their own licenses.