A 2-bit quantization-aware byte-level telemetry student (S: 4 layers, d_model 192, 4 heads), distilled from the float student plawanrath/static-student-tel-mlx-S-w3, from the static-student project (Weights as Constants: compiling distilled task models into native executables as read-only data). Code, RUNBOOK and results: https://github.com/plawanrath/static-student.
Input: the raw bytes of one telemetry payload (a metric line in one of many ad-hoc formats). Output: per-byte span pointers for the metric name, value and unit, a unit class, a value score, and a defer logit. The student is meant to run only on payloads a frozen regex parser rejects, and to answer only when its confidence clears a threshold certified on the integer kernel that ships; everything else is deferred. It is a research artifact for one synthetic task family, not a general log parser.
| curriculum | data/curriculum/telemetry-mlx-s0-x12k (written by Mistral Small 3.2 24B from the spec line, expanded with drift rewrites) |
| init / distillation teacher | models/tel-mlx-S-w3 / models/tel-mlx-S-fp |
| rendered training rows | 500,000 |
| epochs | 2 |
| seed | 0 |
| parameters | 1,936,330 |
| packed weight bytes (trunk / total at 2 bit) | 580,608 / 685,258 |
| dev struct-exact after quantization, before fine-tuning | 75.87% |
| dev struct-exact, final (step sizes snapped to float16, as shipped) | 99.43% |
| wall time (Apple silicon, MPS) | 31 min |
The weights were compiled into a C kernel with the weights as a constant array; the deferral threshold below was certified (Learn-then-Test) on that exact integer kernel, on a locked calibration pool, and travels in the binary.
| risk bound α / confidence δ / min coverage | 0.01 / 0.05 / 0.5 |
| calibration pool n | 20,000 |
| threshold τ | -1.790019 |
| certified coverage | 79.77% (15,953 accepted, 131 wrong) |
| empirical risk on accepted | 0.82% |
| thresholds tested | 96 |
| pool sha256 | e06a4a53395c4486… |
model.safetensors, config.json — framework-neutral weights and constructor arguments.student.pt — the original PyTorch checkpoint the RUNBOOK commands expect (torch.load, dict with config, state, bits).train_report.json — the training log above.git clone https://github.com/plawanrath/static-student && cd static-student && bash scripts/env/setup.sh
from static_student.student.release import load_released
model, cfg = load_released("plawanrath/static-student-tel-mlx-S-w2")
To compile it into a certified binary (needs a C compiler, make, RE2):
.venv/bin/python scripts/release/download_hf.py tel-mlx-S-w2
.venv/bin/python -m static_student.build --spec specs/telemetry.spec --model models/tel-mlx-S-w2 --out build/tel-mlx-S-w2
If you use this checkpoint, please cite the repository:
@software{rath2026staticstudent,
author = {Rath, Plawan Kumar},
title = {static-student: Weights as Constants --- compiling distilled task models into executables as read-only data},
year = {2026},
doi = {10.5281/zenodo.23202887},
url = {https://github.com/plawanrath/static-student},
license = {Apache-2.0},
version = {0.1.0}
}
CITATION.cff in the repository carries the same entry (GitHub's "Cite this repository" button).
Plawan Kumar Rath (@plawanrath). This work was conducted in the author's personal capacity. The views expressed here are those of the author and do not necessarily reflect the views of Meta.
student.pt 8 MBmodel.safetensors 8 MBplawanrath/static-student-tel-mlx-S-w3