---
base_model: Qwen/Qwen3.5-2B
library_name: peft
license: apache-2.0
pipeline_tag: text-generation
tags:
- lora
- transformers
- decision-model
- opencodifier
---

# OpenCodifier E1-E — Qwen3.5-2B verdict-slot decision LoRA

A LoRA adapter that turns [Qwen/Qwen3.5-2B](https://huggingface.co/Qwen/Qwen3.5-2B)
into a Jev-style **decision model**: probabilities are read from the
log-probs of verdict-slot label tokens at the answer position — nothing is
generated, nothing is parsed. It is the decision rung of the
[OpenCodifier](https://github.com/aliasfoxkde/OpenCodifier) ladder, and the
first own-trained arm to pass both release-gate legs.

## Results

All numbers from the OpenCodifier evaluation records
([BENCHMARKS.md](https://github.com/aliasfoxkde/OpenCodifier/blob/main/docs/BENCHMARKS.md),
TRAINING.md §9.9.1). Public 231-item JevBench split, deterministic
readout, fp16, single V100.

| metric | value |
|---|---|
| JevBench-231 accuracy | **0.6840** (publish bar 0.6494 — clear) |
| Internal holdout suite | **0.8083** (gate ≥ 0.75 — GATE_PASS; seed-replicated 0.8000) |
| Raw board ECE | 0.207 |
| Holdout-fitted temperature (T=3.362, analysis-only on board) | ECE 0.207 → 0.127 |
| Auto-decidable @5% risk | 0.100 (23/231) |
| Brier | 0.504 |
| p50 latency | 0.58 s |

Per-family shape: routing / tool-selection / ordinal families at 1.00;
relational/compositional families are the weakness (multi_hop 0.389,
judge_hard 0.412, long_policy 0.368). The board is lost on compositional
families, not on routing — the candidate lever is relational-heavy
soft-label data, not parameters.

## Training

- **Base**: Qwen3.5-2B, LoRA verdict-slot readout
- **Corpus**: merged-v3 (173,452 records; composition and join rates in
  TRAINING.md). The corpus itself is **not** released with this adapter.
- **Schedule**: 11,009 steps, single V100 (T5500), fp16
- **Calibration**: temperature T=3.362 fitted on a labelled holdout
  (holdout ECE 0.179 → 0.117); artifact ships in the OpenCodifier repo
  at `calibration/e1e-qwen3.5-2b-holdout.json`

## Intended use

Typed multiple-choice decisions (routing, tool selection, ordinal
grading) where calibrated token-level probabilities — not free text —
are the product. Pairs with the OpenCodifier engine's confidence gate:
low-confidence reads abstain or escalate to a stronger rung.

## Out of scope

Free-form generation; the base model's general chat ability is not
preserved or claimed. Not evaluated for safety of generated content —
the adapter is a readout head, not a chat model.

## How to use

```python
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-2B", torch_dtype="float16")
model = PeftModel.from_pretrained(base, "TaskWizerAI/opencodifier-e1e-2b")
```

Scores come from label-token log-probs at the answer position; see the
OpenCodifier repo for the readout harness and the Jev adapter wire shape.

## License

Apache-2.0, matching the OpenCodifier project.

## Citation

```bibtex
@software{opencodifier_e1e_2026,
  title  = {OpenCodifier E1-E: a gate-passing 2B decision-model LoRA},
  author = {Kinney, Michael},
  year   = {2026},
  url    = {https://github.com/aliasfoxkde/OpenCodifier}
}
```