Remove preparation status and unnecessary repository references

README.md

@@ -23,9 +23,9 @@ datasets:
 
 # OpenSML-JSONLite-175M-Instruct
 
-**Author:** William Zebrowski · **Model:** OpenSML-JSONLite-175M-Instruct · **Status:** private repository preparation / experimental extraction candidate
+**Author:** William Zebrowski · **Model:** OpenSML-JSONLite-175M-Instruct
 
-[Technical report](TECHNICAL_REPORT.md) · [Results](JSONLITE_RESULTS.json) · [Checkpoint provenance](checkpoint_provenance.json) · [Native inference verification](INFERENCE_VERIFICATION.json) · [Source repository](https://github.com/williamzebrowskI/OpenSML-JSONLite-175M)
+[Technical report](TECHNICAL_REPORT.md) · [Results](JSONLITE_RESULTS.json) · [Checkpoint provenance](checkpoint_provenance.json) · [Native inference verification](INFERENCE_VERIFICATION.json)
 
 **A 175.6M-parameter English field extractor with a schema-guided JSON adapter.**
 
@@ -35,14 +35,14 @@ The weights were trained from scratch through OpenSML Stage A pretraining, then
 
 ## Availability and inference
 
-This private model repository contains the model card, configuration, detailed technical report, tokenizer/inference source bundle, verified extraction results and checkpoint identity. **The selected weights have not yet been uploaded to this repository.** They remain preserved in the owner's research workspace. A weight artifact and model-weight license are separate release items.
+This repository contains the model card, configuration, detailed technical report, tokenizer/inference source bundle, verified extraction results and checkpoint identity. **The selected weights have not yet been uploaded to this repository.** They remain preserved in the owner's research workspace. A weight artifact and model-weight license are separate release items.
 
-The source GitHub repository is also private during preparation. This repository is not deployed by a hosted inference provider. It is not a Transformers AutoModel or mlx-lm auto-loading export.
+This repository is not deployed by a hosted inference provider. It is not a Transformers AutoModel or mlx-lm auto-loading export.
 
-The included `native_bundle.zip` contains the source repository's standalone inference entry point, frozen tokenizer, required model code, field adapter, examples, evaluation script, provenance and source notices. It contains no weight or optimizer binaries. Supply the matching local BF16 weight file to run it:
+The included `native_bundle.zip` contains the standalone inference entry point, frozen tokenizer, required model code, field adapter, examples, evaluation script, provenance and source notices. It contains no weight or optimizer binaries. Supply the matching local BF16 weight file to run it:
 
 ```bash
-# Authenticate first while this repository is private.
+# Authenticate if access requires it.
 hf auth login
 hf download wzebrowski/OpenSML-JSONLite-175M --local-dir ./OpenSML-JSONLite-175M
 cd OpenSML-JSONLite-175M
@@ -241,7 +241,7 @@ General dialogue, coding, multilingual quality, calibrated confidence, prompt-in
 
 ## Verification and provenance
 
-The selected BF16 weight file has **351,242,342 bytes** and the SHA-256 above. Original metadata and sanitized published metadata have separate identities in `checkpoint_provenance.json`. The source repository's snapshot verifier checks shipped code, tokenizer, docs and evidence; its native loader separately verifies the supplied weights.
+The selected BF16 weight file has **351,242,342 bytes** and the SHA-256 above. Original metadata and sanitized published metadata have separate identities in `checkpoint_provenance.json`. The included snapshot verifier checks shipped code, tokenizer, docs and evidence; its native loader separately verifies the supplied weights.
 
 The model card, metrics and technical report describe one fixed checkpoint. They do not include subsequent training. Clean-runtime 40-case parity is documented separately from the historical full-panel results. No automatic model updates, model promotion or hosted deployment is enabled.
 

SHA256SUMS

@@ -2,9 +2,9 @@
 82b9908c9fd21fd8c9d10ad809e7c48446485ee98a56bfd9d64bf9fb353761c2  INFERENCE_VERIFICATION.json
 d29cf5952a82db444a6df56446806b1f4e2ccbee1b4aa6b93cb7eeeabcab8536  JSONLITE_RESULTS.json
 c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4  LICENSE
-0f6af13be03ee3e6896e76737f395e323e9690ad92ce41a1361cfb287544f838  README.md
+12ca825efe5784d85a45ed2e98b7d12e7deb9b914e8d095cbc341ce92e0a0692  README.md
 64519f25aded3e1fa74b0b9eafdf35df4490e7bb550027338a9bf1a25f9203bb  SOURCE_NOTICE.md
-4f8c90fad555e380c0e99b61582fe6f3bcb9e6f43a1f17d253769384471f8374  TECHNICAL_REPORT.md
+6b7b7e931c7a4e119411ef93dd2535927ed2adb647273471ef5995110df144f8  TECHNICAL_REPORT.md
 423f0d1e0e2690505c24e6d31308db728fba60657c5166e177bb269d6a7198d0  THIRD_PARTY_NOTICES.md
 6f8de0b591c8a07885c8adcab359b8ca1cdc81be5f90638cc0ce286f3899c6c5  checkpoint_provenance.json
 fcc46a0f69fc4afb8d6e25361a013e7656ac6758ae5a49cc6b629f95bf728c1a  complete-request-exact.svg

TECHNICAL_REPORT.md

@@ -1,6 +1,6 @@
 # OpenSML-JSONLite-175M: SFT-093/update 160
 
-Hugging Face companion edition. Archive-relative links in this report resolve inside `native_bundle.zip`; the rendered source edition is in the private GitHub repository.
+Hugging Face companion edition. Archive-relative links in this report resolve inside `native_bundle.zip`.
 
 ## Technical report, training provenance, evaluation and public release specification
 
@@ -1056,6 +1056,6 @@ The saved full-test runner requires the pinned workspace, source panel, dependen
 
 ## Appendix F. Dedicated repository packaging verification
 
-The public source repository is [OpenSML-JSONLite-175M](https://github.com/williamzebrowskI/OpenSML-JSONLite-175M). The portable root `inference.py` reads a relative config/tokenizer and an explicitly supplied matching weight file. It has no dependence on workspace model-registration receipts or private path metadata. Its greedy fieldwise prompt/parser/model path reproduces all 40 saved playground values and stop reasons with stock MLX 0.32.3, NumPy 2.4.2, Tokenizers 0.22.2 and jsonschema 4.26.0 in a clean temporary environment. This verification has the same 29/40 reference accuracy as the original service. It checks inference packaging; it does not change weights, retrain the model, establish full-suite cross-runtime parity or make the test a fresh blind benchmark.
+The included `native_bundle.zip` contains the standalone inference source. The portable root `inference.py` reads a relative config/tokenizer and an explicitly supplied matching weight file. It has no dependence on workspace model-registration receipts or private path metadata. Its greedy fieldwise prompt/parser/model path reproduces all 40 saved playground values and stop reasons with stock MLX 0.32.3, NumPy 2.4.2, Tokenizers 0.22.2 and jsonschema 4.26.0 in a clean temporary environment. This verification has the same 29/40 reference accuracy as the original service. It checks inference packaging; it does not change weights, retrain the model, establish full-suite cross-runtime parity or make the test a fresh blind benchmark.
 
 The source snapshot includes no weight or optimizer binaries, raw training pools, packed arrays or private cluster/network receipts. Public checkpoint JSON has redacted workspace paths and separate published hashes. Original metadata/weight identities remain in `../provenance/LINEAGE.json`. Source-file inventory and the integrity manifest describe the actual package.