HLM5 — 1B frozen trunk for certified knowledge editing
HLM5 is a frozen decoder-only language model paired with a certified, auditable,
exact-key edit layer. It is the trunk for "Certified Knowledge Editing:
Closed-Form Reachability Guarantees for Additive Edits to a Frozen Language
Model" (Marius Dima, Qriton). A slot memory attaches to this frozen trunk; a
hidden-state query selects slots through a hard cosine gate and adds the value to
the pre-head representation. Because the gate and read see only the unperturbed
hidden state, the post-edit logits are affine in the edit strength, so
single-token reachability is one-dimensional linear feasibility — solved in closed
form as a certificate that accepts or refuses each edit before injection and
doses the accepted ones at a certified strength.
- Code, scripts, and result artifacts: https://github.com/qriton/hlm5
- Paper: in the repository (
paper/); arXiv link forthcoming.
Model description
- Architecture: decoder-only transformer, dim 1792, 24 layers, 14 heads,
context length 1024.
- Vocabulary: 65,536-token byte-pair tokenizer (
fineweb-65536-compat.json).
- Training data: FineWeb (~20B tokens).
- Validation PPL: 18.01 (baseline trunk).
- Parameters: ~1.045B.
This repository hosts hlm5_lm_baseline_fineweb_g3_final.pt (the frozen baseline
trunk, step 150,999), hlm5_lm_hybrid_fineweb_g3_final.pt (the matched
co-trained-memory arm, step 149,999), and the tokenizer.
Usage
Use the package loader from the GitHub repository. It loads checkpoints with
torch.load(..., weights_only=True), freezes the trunk, and pairs the checkpoint
with the bundled tokenizer:
1from hlm5.io import load_trunk, load_tokenizer
2
3model, ck = load_trunk("baseline")
4tok = load_tokenizer()
Avoid older snippets that call torch.load(..., weights_only=False) directly.
Results
Evaluated as the deployed certificate-governed pipeline (whitened degree-5 gate,
threshold 0.95) on a 17-fact set; every number traces to a JSON in the GitHub
repository under results/.
- Certificate-dosed efficacy 0.765 (13/17), rising to 1.000 (17/17) with
residual-synthesis rescue through the full memory path, at locality 1.000 (bit-identical logits on
neutral prompts) and paraphrase transfer 0.020 (exact-key by design). The
0.529 figure some earlier material led with is the global-boost dosing
ablation — a single strength that overshoots the certified interval — not the
method.
- Operating envelope: 78.4% of 1,200 single-token targets reachable at a
novel-entity key; 79.4% ± 2.2% across 60 keys.
- Gate: perfectly separable (exact-key 1.0, all paraphrase/typo/other-relation
0.0); realized own-slot weight 0.9993.
- No-tax: matched baseline-vs-hybrid validation-PPL delta 0.116% at 1B
(0.018% at 136M).
Implementation note: the current certificate engine uses float64 arithmetic and
exact slope signs (EPS = 0.0), with a regression test for near-zero negative
slopes. Some copied JSON artifacts from the initial bundle were generated under
the older slope-dead-zone convention, whether or not they include an eps field;
rerun the affected certificate/dosing scripts before publishing refreshed
quantitative artifacts.
Intended use
Research on certified, auditable knowledge editing: reachability certificates,
certificate-governed dosing, exact-key slot memories, and edit/forget/rollback
receipts on a frozen trunk. Suitable for reproducing the paper and building on the
certificate machinery.
Out of scope
This is a research base model, not an instruction-tuned or chat model. It is
not aligned, not safety-tuned, and not intended for deployment as an assistant or
for generating end-user-facing text. The edit layer is exact-key by design: it
does not generalize edits to paraphrases.
Citation
1@article{dima2026certified,
2 title = {Certified Knowledge Editing: Closed-Form Reachability Guarantees
3 for Additive Edits to a Frozen Language Model},
4 author = {Dima, Marius},
5 year = {2026},
6 doi = {10.5281/zenodo.21258599},
7 note = {arXiv link forthcoming}
8}
License
Business Source License 1.1 (BSL 1.1), © 2026 Qriton Technologies S.R.L.
Academic research, education, personal, and non-commercial evaluation use is
permitted; commercial use is not. Converts to Apache-2.0 on 2030-01-01. See the
LICENSE file in the GitHub repository; commercial inquiries:
license@qriton.com.
