leona-qompact-1.0

An experimental LoRA adapter for natural-language-to-OpenQASM 3 generation. This is the final step (1,000 of 1,000) of the clean-v12 run. Training is complete, but the model is not yet quality-validated. No improvement over the base model is claimed; validation and final testing are incomplete. Every milestone is kept under its own tag (clean-v12-step-<N>); the tag, not main, identifies a checkpoint.

Base model and loading

This repository contains adapter weights, not the full 27B model. Use Qwen/Qwen3.8-27B at revision 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0, then load this repository with peft.PeftModel.from_pretrained. The training environment uses Transformers 5.22.1, PEFT 0.21.2, and PyTorch 2.8.0. The base is loaded through Qwen3_5ForConditionalGeneration, in BF16. Use the included chat template with enable_thinking=False. The evaluation protocol uses greedy decoding and max_new_tokens=4096.

Training

LoRA rank 16, alpha 32, dropout 0; maximum sequence length 4,096; gradient accumulation 16; peak learning rate 5e-5 with 5% warmup. The clean training candidate contains 22,260 rows, of which 22,123 fit the sequence limit. Training exposure is 70% public-source-derived examples and 30% original generated examples; this is a sampling ratio, not a row-count ratio. Sources include QCircuitBench, Agent-Q, QASM-Eval, MQT Bench, QASMBench, Qly verified OpenQASM circuits, and quantum-circuits-21k. PQID is not included in this run. Source-derived examples underwent filtering and task-specific repair/checking; checker coverage is not universal. No training dataset or optimizer state is redistributed in this upload. Upstream datasets retain their own terms; the repository license does not relicense those datasets.

Evaluation and limitations

Checkpoint selection uses 2,479 validation examples. The final held-out test has 2,494 examples and is not used for checkpoint selection. Neither evaluation is complete for this checkpoint. Training loss is not an evaluation result. Hardware timing/pulse behavior is not universally certified by the available checkers. Generated circuits must be checked before use.

training_provenance.json records the source snapshot hashes and training contract.