Raw Inspect AI logs (.eval files) of every run made for runs. The intent is that no run can be left out afterwards. - Each file is at runs/ /.eval. - The manifest manifests/.json in the code repository records the sha256 of each file, the code version, the model files and the settings of the run, and the dataset commit at which the file was published and verified. - The files are published only after a scrub check: no key material, no home-directory paths, no private addresses or host names. Status: infrastructure in progress. There are no findings yet. The runs published so far are tests of the plumbing and measurements of the substrate, not results about monitoring. An.eval file is a zip…
Independent publisher
Dan W
dang-w
Models in Library0
Datasets in Library1
Models on Hugging Face—
Followers1
Compare the published evidence
This profile connects the 1 model and dataset record shown below to their upstream cards and sources. The library counts describe SAVRN's coverage, rather than this publisher's entire output.
Reported licenses: mit. Open a record to review its specific license and conditions.
Compare model specifications, source dates, and hardware estimates on the individual records. Publisher identity verification is separate from a model's quality, deployment readiness, or suitability for your work.