M1 workflow rule specialist — reference seed 17
Author: Megern Qaisse (megern). Source and all seeds: M1 Memory Research.
An original local LoRA adapter over Qwen3 0.6B 4-bit, trained on an Apple M1 with 16 GiB unified memory. This contains adapter weights only. Base revision: 73e3e38d981303bc594367cd910ea6eb48349da8.
The task is workflow-policy priority classification from explicit permissions, trigger and untrusted-checkout flags. Prompts state the rule. No real incident logs, hosted training or teacher API was used. These synthetic tasks can also be solved directly by rule-based software.
Recorded evaluation
Final seed-17 checkpoint on the frozen 96-case synthetic test: strict accuracy 96/96 (100.00%), macro-F1 1.0000, valid JSON 96/96. All three training seeds and a common base baseline must be read together in the source study; this adapter was not selected for its test score.
The unadapted base scored 55.21% on the same test after removing only a complete outer JSON fence. All three workflow adapters scored 96/96 on this synthetic test; this does not establish production review accuracy.
Local use
Install the pinned study dependencies and acquire the pinned base checkpoint locally. Then use the study evaluator with --model /absolute/local/base --adapter /absolute/local/this-adapter --data /absolute/local/test.jsonl --output /absolute/local/fresh-report.json.
Generation uses the base chat template with enable_thinking=False, greedy decoding and a 48-token output cap. The adapter expects the study's system instruction, a stated classification rule, and JSON evidence. It is not a general conversational model.
Training and limitations
384 training and 48 validation examples, 400 iterations, batch one, last 16 layers, query/value LoRA rank eight, learning rate 0.0002, prompt masking and gradient checkpointing. No test-based checkpoint selection. See provenance.json for hashes, measured allocation and timing. MLX allocation is different from total process RAM.
This is an experimental synthetic classifier. Scores do not demonstrate general Arabic proficiency, production SOC accuracy, adversarial robustness or security guarantees. No peer review or novel training algorithm is claimed. Base weights remain under their Apache-2.0 terms; this adapter is released under Apache-2.0. The original study tools and synthetic data use MIT.
Quantized