hypneum-lab/micro-kiki-stack-spice

micro-kiki stack-spice — SPICE-simulation LoRA for Qwen3.5-35B-A3B

One standard LoRA adapter (adapters.safetensors, 2.05 GB) trained with mlx_lm on Qwen/Qwen3.5-35B-A3B for SPICE / ngspice / LTspice netlist and simulation assistance. (The previous card called it a "merged adapter"; the shipped adapter_config.json shows a plain LoRA run.) Base model: Qwen/Qwen3.5-35B-A3B (not confirmed by a config file; stated by the authors). The shipped adapter_config.json records only the local path models/qwen3.5-35b-a3b.

Files and format

Hub revision 0528f72688d55de3f4c6ca82c3e40daa6604c6fc (last modified 2026-10-25T09:09:45Z), 4 files, 2 051 649 378 bytes (2.05 GB): adapters.safetensors, adapter_config.json, README.md, .gitattributes.

Recipe (from adapter_config.json)

Parameter	Value
Base model (declared)	Qwen/Qwen3.5-35B-A3B (local models/qwen3.5-35b-a3b, BF16)
Framework	mlx_lm.lora, fine_tune_type: lora
LoRA rank / alpha / scale	16 / 32 / 2.0 (= alpha/rank; unlike V4 which used scale 20)
Dropout	0.0
Target layers	16
Iterations	11 542
Batch size	1 (grad_accumulation_steps 1)
Learning rate	5e-5, no schedule, optimizer adam
Max sequence length	2048
Gradient checkpointing	true
Seed	0
Data	outputs/stacks/stack-spice (micro-kiki spice domain; example counts not recorded; contents and licence of this data are not documented)
Adapter size	2 051 644 214 bytes (16 layers × ~128 MB)

Training host & cost

Mac Studio M3 Ultra 512 GB, MLX BF16. Peak memory, wall-clock, energy: not recorded (no log shipped). Host powered off since 2026-10.

Evaluation

Not recorded for this adapter; none found for it in the source repository. Related evidence: the project README reports that the base model is already strong on SPICE/electronics and that only 3 of 10 probed domains improved over base (README.md, section "Stacks vs base"); in the June 2026 iact-bench run on the sibling Qwen3.6 ailiance adapters, no adapter beat the base except html-css, and a SPICE adapter was removed from production routing for degrading numeric (RC) calculations. (statement of the 2026-10-25 card; no source file in the repository, unverified on 2026-10-31) Validate on your own netlists before use.

Limitations

Research artefact; may produce syntactically plausible but electrically wrong netlists; not for safety-critical design sign-off. No safety tuning is documented; generated netlists must be checked in a simulator. The source repository follows an errata policy under which a number is reported only if a committed artefact reproduces it (docs/ERRATA.md). Research artefact of the pre-registered dreamOfkiki programme (parent registration OSF 10.17605/OSF.IO/Q6JYN).

Citation

micro-kiki concept DOI 10.5281/zenodo.19666759.

Provenance

https://github.com/hypneum-lab/micro-kiki · https://github.com/ailiance/ailiance-mac-tuner · base https://huggingface.co/Qwen/Qwen3.5-35B-A3B

• Source repository: https://github.com/hypneum-lab/micro-kiki (Apache-2.0 LICENSE), checked at commit a3968d9e1f0b6a6b8cc4cb028d4fef3ced24770c.
• Zenodo DOI: 10.5281/zenodo.19666759, concept DOI of the source repository; no per-artefact DOI.
• Parent pre-registration: OSF 10.17605/OSF.IO/Q6JYN (CITATION.cff).
• Hub revision described by the file listing: 0528f72688d55de3f4c6ca82c3e40daa6604c6fc.
• Card revision date: 2026-10-31.

EU AI Act transparency

LoRA adapter weights only, base unchanged; provider Clément Saillant / L'Electron Rare for Hypneum Lab; limited risk, Article 50; contact clement@saillant.cc. Outputs may be inaccurate, biased or fabricated.

AI assistance (transparency)

This repository is developed by a one-person lab with AI assistants used as tools, never as authors:

• Claude Code (Anthropic) assists with code, documentation, operations and literature search in interactive sessions driven by the PI.
• Codex CLI (OpenAI, gpt-6.1-sol / gpt-6-astra) runs independent review passes before a release, a pre-registration or a submission; the briefs, reports and verdict tables are published next to the reviewed material whenever a pass is run.

Every scientific claim, verdict and pre-registration decision is taken by the PI; results are reproduced by the test suite and CI, and review reports are evidence, not endorsements. No model is listed as a co-author of a publication or as a co-author of a commit.