Downloads · 30 days
288
27% of all-time downloads
Joakimpalm-Zen/Muse-Glimmer-30B-Surgical-Q4_K-GGUF
Muse-Glimmer-30B-Surgical-Q4_K-GGUF is a text generation model from Joakimpalm-Zen. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
Research artifact, not a general-purpose release. - What was changed: 4.75% of the decoder removed, the damage healed by local distillation against the frozen parent, then quantised to Q4K by Runner. - Why it exists:…
Downloads · 30 days
288
27% of all-time downloads
All-time downloads
1.1K
Public
Repo size
15.7 GB
Likes
0
Public
Click a slice to open those files.
.gguf15.7 GB · 100%
From the Hugging Face model README
<!-- END xyntetik-status --> <!-- BEGIN xyntetik-actions -->Research artifact, not a general-purpose release.
<!-- BEGIN xyntetik-validated -->
- What was changed: 4.75% of the decoder removed, the damage healed by local distillation against the frozen parent, then quantised to Q4_K by Runner.
- Why it exists: one of the two artifacts that came out of a nine-run preregistered surgery study on this model.
- Parent model: meta-models/Muse-Glimmer-30B
- Useful for normal inference? it passes the fidelity bar (97.74% / 0.04810 on a held-out split) but the benchmark annex found a paired benchmark deficit; read the annex before using it as a stand-in for the parent.
- Experiment: Model Surgery and Scale Recovery
- Evidence: the surgery study (dataset), SURGERY.json and SHA256SUMS in this repo
- Runner compatibility: any Runner release; standard tensor shapes, so other runtimes load it too.
<!-- END xyntetik-validated -->
- Validation status (Runner support tiers): each file's row in the Runner roster says which tier it is in, what is re-checked on a release and the date and version of the last run.
Muse-Glimmer-30B-surgical-4p75-Q4_K.gguf: roster row
Want this for another model? The tools behind these files are open: requantization, expert pruning and sublayer removal in Xyntetik Runner, and the method for this file is on this card or in its evidence dataset.
Request a model · Report your result · Community results · All files
<p align="center"><a href="https://buy.stripe.com/9B69AUddpdx9auHgP27N600"><img src="https://xyntetik.com/assets/support-button.svg" alt="Support this work" height="44"></a></p> <p align="center">If this file is useful to you, a contribution funds the hardware time behind the next one.</p> <!-- END xyntetik-actions --> <!-- BEGIN runner-training -->xyntetik-runner — the engine that trains, merges, and quantizes the GGUF it servesThis GGUF is built to be served with xyntetik-runner, a single-binary, dependency-free C11 inference engine (CPU / CUDA / Metal, OpenAI-compatible server). For this artifact the relevant differentiator is the training side:
Runner trains, merges, and quantizes GGUFs deterministically: the same data and seed reproduce the same adapter file byte for byte, and every written artifact carries a provenance record.
One binary covers the loop: --train (AdamW LoRA on the served GGUF, deterministic by default, gated in CI by adapter sha256), --lora to serve the adapter beside the frozen base, --merge-lora to fold it in with per-tensor requantization and a merge.json provenance record, and --score for teacher-forced held-out evals. Adapters interoperate both ways: a Runner-trained adapter scores identically (1.000 on its held-out eval) served by stock llama.cpp, and community F16 adapters load straight back — measured, not assumed.
Measured — the finding every fine-tuner should know before merging (held-out eval where the fine-tune scores 1.00 and the base model 0.69; verified in two independent runtimes):
| merge output type | eval after merge | fine-tune survives? |
|---|---|---|
| F16 | 1.00 | yes |
| Q8_0 | 1.00 | yes |
| Q4_0 | 0.69 — the base model's score | no: erased |
A 4-bit merge silently deletes the adaptation while perplexity still looks fine. Merge to Q8_0 or F16, or keep serving base + adapter.
This artifact is that toolchain eating its own cooking. The surgery was applied by byte-level substitution into the parent GGUF, and the Q4_K file was written by Runner's own quantizer. The healed weights survive Q8_0 at a cost of +0.00006 KLD, and the Q4_K cost you see in the fidelity table was PREDICTED to 0.6% before the file was built, from the measured independence of quantization noise and surgery error. If you fine-tune on top of this model, the merge-erasure table above applies to you.
# the same binary that wrote this artifact:
runner -m model.gguf --train data.jsonl --train-out adapter.gguf
# same data + seed -> byte-identical adapter, build provenance in adapter.gguf.train.json
runner -m model.gguf --lora adapter.gguf --merge-lora merged --quant q8_0
# provenance lands in merged.gguf.merge.json; do not merge to 4-bit (table above)
The exact boundary of the determinism claim, including what is deliberately not promised: docs/determinism-scope.md.
<!-- END runner-training -->This file is 14.61 GB, so it wants a machine with roughly 17 GB of usable memory or more.
# Apple Silicon Mac
curl -L https://github.com/Joakimpalm-Zen/xyntetik-runner/releases/latest/download/runner-macos-arm64.tar.gz | tar xz
curl -L -o Muse-Glimmer-30B-surgical-4p75-Q4_K.gguf \
https://huggingface.co/Joakimpalm-Zen/Muse-Glimmer-30B-Surgical-Q4_K-GGUF/resolve/main/Muse-Glimmer-30B-surgical-4p75-Q4_K.gguf
./runner-macos-arm64 -m Muse-Glimmer-30B-surgical-4p75-Q4_K.gguf -i
Linux and Windows builds are on the same release page. Any GGUF runtime with muse-glimmer support should load it: every tensor is an ordinary GGUF type in the standard shape; 9 FFN tensors happen to be exact zeros, which is the surgery.
A 14.61 GB GGUF of Meta's Muse-Glimmer-30B with the FFN sublayers of layers 4, 9 and 48 structurally removed (weights exactly zero, standard tensor shape), the damage repaired by short local distillation against the frozen parent, and the result quantized to Q4_K.
It passes a strict fidelity bar against the unmodified BF16 parent on a held-out split: margin-qualified top-1 agreement 97.74% (bar: >=97%) and mean KLD 0.04810 (bar: <=0.05).
| artifact | size | mean KLD | margin-qualified top-1 | bar |
|---|---|---|---|---|
| this file (surgery + heal, Q4_K) | 14.61 GB | 0.04810 | 97.74% | PASS |
| same weights, Q8_0 | 27.58 GB | 0.03076 | 98.65% | PASS |
| same weights, BF16 | 51.90 GB | 0.03056 | 98.64% | PASS |
| parent at Q4_K (quant-alone control) | 14.61 GB | 0.01930 | 99.12% | PASS |
Per-domain behavior: the healed model sits within a few percent of the parent in every measured domain (11-domain perplexity spread 0.046 vs 0.373 for un-healed surgery). Healing distils toward the parent: this is a smaller model that behaves like the original, not a different model that happens to score well.
The fidelity table above is a real, strict measurement. This section reports what it does not buy. One preregistered measurement night translated the house bar into paired public-benchmark deltas versus the frozen parent: lm-evaluation-harness 0.4.9, stock task configs, both models scored through the identical loglikelihood path on identical items, per-item pairing, exact binomial confidence intervals on the discordant pairs, margins fixed before sampling.
| config | benchmark | preregistered margin | net paired delta | 95% CI | verdict |
|---|---|---|---|---|---|
| these weights at BF16 (pre-quant) | HellaSwag acc_norm | ±0.5 pts | −0.916 | [−1.230, −0.583] | FAIL |
The flip counts say the deficit is directional, not noise: on the 2.89% of items where the two models disagree, the surgical model loses roughly 2:1 (191 parent-right-variant-wrong against 99 the reverse).
Stated plainly:
The parent-fidelity table remains this card's primary claim: these weights track the parent's token distribution closely, and that is what they are for. The benchmark row exists so nobody has to take distribution fidelity as a proxy for task equivalence. We measured that it is not one.
Base model: Muse-Glimmer-30B by Meta, Apache-2.0. This derivative keeps the license and changes 18 tensors (9 zeroed, 9 replaced with healed weights) as described.
A window-level audit of the healing corpus found that 283 of the eval split's 43,855 windows (0.65%, one domain) had leaked into the training pool through near-duplicate source documents. This affected every healed artifact's published number, so all of them were re-scored on the decontaminated subset.
This artifact passes on clean data: KLD 0.04903 (bar 0.05), margin-qualified top-1 97.70% (bar 97%). The published full-split numbers reproduce from the file on disk to five decimals.
Stated plainly, because thin margins should be said and not discovered: this artifact's clean-data headroom under the KLD bar is 0.00097, which is 1.37 times the measurement procedure's noise floor (plus or minus 0.00071), and its original build gate was cleared by 0.00014. It is a genuine, exactly reproducible pass with thin margins on every axis we have examined. If your use case needs comfortable headroom rather than a demonstrated boundary case, use the Q8_0 sibling, whose headroom is roughly 15 times the noise floor.
A control decomposition isolated the leak's effect: healed models score better on positions whose content appeared in healing data, by +0.0012 to +0.0014 KLD across three independent artifacts. Small, real, and measured; the naive before-and-after delta would have hidden it behind an opposing domain-mix shift. Full detail will accompany the method write-up's next revision.
sha256(Muse-Glimmer-30B-surgical-4p75-Q4_K.gguf) =
103aa117dfce88974bd78e03ee6ca6601c720f01a1ec24ee1cfd0e5fe2f77ce6
SURGERY.json in this repo records the byte-substitution provenance:
the parent revision, which tensors were zeroed and which were replaced
with healed weights.
This repository is published by Xyntetik, the Zenova AB program for running capable language models on hardware you own and proving what they did. Its released engine is Xyntetik Runner: a free, single-binary GGUF engine in C for CPU, CUDA and Metal that serves, scores and trains through the quantized file it deploys and records every run as a receipt. The measurements behind the program's claims are collected at xyntetik.com/evidence. The program's organization on Hugging Face is Xyntetik.