soup-cli 0.71.34__tar.gz → 0.71.36__tar.gz

{soup_cli-0.71.34 → soup_cli-0.71.36}/CHANGELOG.md

@@ -12,6 +12,130 @@ reproducing 70+ versions of notes.
 
 ## [Unreleased]
 
+## [0.71.36] - 2026-07-16
+
+**Data Moat II** — a semantic layer over your training data, plus two tools for
+what a fine-tune forgets and leaks.
+
+### Added
+
+- **`soup data dedup --semantic`** — near-duplicate removal over embedding
+  cosine instead of MinHash shingling. Catches *reworded* duplicates that
+  MinHash misses (measured: 0.88–0.91 cosine on rewordings MinHash scored as
+  distinct) while correctly keeping distinct-but-similar instructions. Zero new
+  dependencies: uses `transformers` from the `[train]` extra. **Read the
+  known-limitation below before lowering `--threshold`.**
+- **`soup data topics <data>`** — cluster a dataset and label each cluster with
+  c-TF-IDF terms, plus a coverage table (`82% code · 6% math`) and a warning for
+  thin topics. Labels are *emergent term clusters*, not a fixed taxonomy.
+- **`soup data canary insert|check`** — Secret-Sharer memorization probe. Insert
+  K high-entropy secrets, then check any model/adapter: each secret's loss is
+  ranked against never-inserted controls drawn from the same space. Exit 2 on
+  MAJOR so CI can gate. Measured on SmolLM2-135M: a memorized set lands at
+  percentile 0.0 (loss 1.7–2.5) against a clean model's 4.1–6.2.
+- **`soup train --replay old.jsonl --replay-ratio 0.1`** — continual-learning
+  rehearsal. Interleaves a seeded sample of an old dataset into training so a
+  new task does not erase the old one. `r` is the fraction of the **final**
+  mixed set (`n_replay = round(r/(1-r) · n_new)`), rows are interleaved rather
+  than appended, and an undersized pool reports the shortfall instead of
+  repeating rows. Mixed into `train` only — validation stays pure new-task.
+
+### Fixed
+
+- **The hardware-fit gate refused to train any local checkpoint.** A merged
+  model (`soup merge -o ./mymodel`) has no size marker in its name, so the size
+  guesser returned its 7B default, predicted ~16 GB of VRAM and refused. Local
+  checkpoints are now *measured* from their safetensors header (0.135B actual vs
+  7.0B guessed) — this had blocked `soup merge` → train-from-merged entirely.
+  Third instance of this class after the v0.71.32 (Whisper) and v0.71.33
+  (`M` suffix) fixes.
+- **`pip install 'soup-cli[extra]'` hints printed without the extra.** Rich ate
+  the bracket, so every "install the missing dependency" message across 17 sites
+  told users to run `pip install 'soup-cli'` — which succeeds and still leaves
+  the feature broken. Affected `[eval]`, `[data]`, `[serve]`, `[ui]`, `[tui]`,
+  `[compile]`, `[mcp]`, `[carbon]` and others, including Typer help text.
+- Replay rows bypassed the image/audio path-traversal validation that the
+  primary dataset receives.
+
+### Known limitations
+
+- **Semantic dedup is not a paraphrase detector.** Measured with
+  all-MiniLM-L6-v2, paraphrase cosines (0.49–0.76) **overlap** with
+  genuinely-distinct rows (0.54–0.76): "Add two numbers" vs "Multiply two
+  numbers" scores 0.759, *higher* than the true paraphrase "reverse a string" /
+  "invert the order of characters" at 0.491. **No threshold separates them**, so
+  lowering `--threshold` to chase paraphrases deletes real training rows. The
+  default (0.8) is deliberately conservative.
+- **Replay is validated at proof-of-mechanism scale.** On SmolLM2-135M + LoRA,
+  replay retained the old task 7% better than a no-replay control — the correct
+  direction — but forgetting without it was only +4%, i.e. mild. The effect size
+  at full fine-tuning or 7B+ is unproven on a 4 GB box.
+- **Canary exposure is the sampled-control approximation**, not full-space rank
+  enumeration. "No exposure" is not proof of no memorization.
+- **`data topics` / `dedup --semantic` require `[train]`** (torch) and download
+  an embedding model. Plain MinHash `dedup` stays on the light core.
+- Replay v1 is `sft`/`pretrain` only and is incompatible with
+  `packing`/`multipack`.
+
+## [0.71.35] - 2026-07-15
+
+### Added
+- **Compliance templates — `soup init --template hipaa|soc2|eu-ai-act|sr-11-7`.**
+  Four regulation-shaped starting configs. Soup's compliance controls are CLI
+  flags/commands rather than config keys, so each template is a valid training
+  config plus header comments naming the exact commands for that regime
+  (PHI scrubbing + air-gap for HIPAA, BOM/attest/sign for SOC 2, Annex XI +
+  energy tracking for the EU AI Act, repro-receipt + diagnose/ship for SR 11-7).
+  Templates default to a license-clean Apache-2.0 base.
+- **`soup card <registry-id> -o MODELCARD.md` — model-card autogen.** Turns a
+  Local Model Registry entry into a publishable, provenance-carrying HF model
+  card: base model, training config, eval scorecard, config/data hashes,
+  lineage (ancestors) and a table of every registered artifact. Adapter vs
+  full-model is inferred from registered artifacts, falling back to the training
+  config (LoRA rank, with Spectrum/LISA full-FT correctly treated as dense), so
+  the card sets the right `library_name` and never misreports the model type.
+- **`soup push --card <registry-id>`** — render that registry-driven card and
+  upload it as `README.md`, overriding the auto-generated one. A bad ref fails
+  fast before any network call; HF hub only.
+- **`soup ci init` — fine-tuning CI.** Writes `.github/workflows/soup-gate.yml`,
+  a PR gate chaining `soup data validate` → `soup expect` →
+  `soup ship --evidence` (exit 2 blocks the merge). Every interpolated path is
+  validated to stay under the repo root and shell-quoted; the branch and Python
+  version are regex-gated; the write is atomic, symlink-rejecting, and refuses
+  to clobber an existing workflow without `--force`.
+- **Compliance quickstart** — a new [docs/compliance.md](docs/compliance.md)
+  walkthrough: template → PII scrub → train with receipt/Annex XI/energy →
+  registry → BOM + attestation → scan/sign/verify → air-gap → model card → CI gate.
+
+### Fixed
+- **GGUF export now actually works on Windows** (validated end-to-end against a
+  locally-built llama.cpp: SmolLM2-135M → q4_0 / q4_k_m / q8_0 / f16 → `soup deploy
+  ollama` → live inference). Four real bugs, each of which independently broke the
+  path:
+  - **`soup export --format gguf` cloned llama.cpp into your current directory.**
+    `SOUP_DIR` is the bare name `.soup`, but the lookup used it relatively rather
+    than anchoring to `~` like the rest of the codebase — so the canonical
+    `~/.soup/llama.cpp` was never found and a fresh ~200 MB checkout was dropped
+    into whatever directory you ran from.
+  - **The first GGUF export downgraded your PyTorch and broke CUDA.** The auto-clone
+    ran `pip install -r <llama.cpp>/requirements.txt` into your interpreter, and
+    llama.cpp pins `torch~=2.2.1` against the CPU wheel index (observed:
+    torch 2.5.1+cu → 2.2.2+cpu, transformers 4.57 → 4.46). Soup now installs only
+    the convert script's extra dependencies, unpinned, and never touches torch.
+  - **A correctly-built llama.cpp was not found on Windows.** MSVC (like Xcode) is a
+    multi-config generator and emits `build/bin/Release/llama-quantize.exe`; only the
+    flat single-config layout was searched.
+  - **`soup deploy ollama` failed on a relative GGUF path** with "pull model manifest:
+    file does not exist" — Ollama resolves `FROM` against the Modelfile's directory,
+    and Soup writes the Modelfile to a temp dir. The Modelfile now emits an absolute path.
+- **Model-card injection hardening (affects the pre-existing `soup push` card too).**
+  The `## Training` section interpolated `base` / `task` / `scheduler` / `recipe`
+  unescaped. Since `SoupConfig.base` and `scheduler` have no charset validator, a
+  crafted-but-valid config could smuggle raw HTML — or a backtick breaking out of
+  the surrounding code span — into a card published to the Hub. All values now go
+  through the markdown escaper, which additionally neutralises backticks and
+  strips C0/ESC control bytes.
+
 ## [0.71.34] - 2026-07-15
 
 ### Added

{soup_cli-0.71.34 → soup_cli-0.71.36}/CONTRIBUTING.md

@@ -117,10 +117,10 @@ src/soup_cli/
   data/traces/        - Trace-to-Preference harvester (v0.26.0)
   data/collators.py   - CrossDocCollator for sample packing (v0.33.0)
   utils/              - GPU, errors, MoE, GaLore, QAT, Unsloth, vLLM, SGLang, Liger, FlashAttn, FSDP, Ring Attention, long-context, quality, curriculum, freeze, dataset-registry, mlx, peft_builder, paths, topology, launcher, mii, pipeline, cut_ce, fp8, gradient_ckpt, kernel_picker, cross_doc_attn, activation_offload, hf, spec_pairing, structured_output, metrics, tracing, auto_quant, lr_finder, grad_accum, mixed_precision, warmup, spike_recovery, convergence, v028_features, multipack_sampler, multipack, neat_packing, jinja_analyzer, quant_menu, relora, peft_patches, peft_wiring, dpo_variants, optimizer_zoo, lr_groups, loftq_init, block_expansion, tts, classifier, distill, bitnet, ebft_gdpo, moe_quant, reasoning_effort, gguf_quant, kv_cache, advanced_precision, save_formats, deploy_measure, advise, advise_history, adapter_diff, adapter_merge, blame, adapter_branch, unlearning, unlearning_eval, knowledge_edit, edit_governor, edit_diff, ra_dit, steering, citation_faithful, grace_codebook, ingest_sources, prune_prompt, active_sampler, ab_test, drift_alarm, tunability, terraform_plan, env_lock, hardware_fit, completions, license_advisor, behavior_battery, capability_suite, checklist_dsl, irt, sae_diff, sleeper_probe, interference, probe_pack, cmaes_merge, vector_bank, mole_routing, adapter_pr, soup_lock, adapter_bisect, prompt_compile, prompt_distill, compile_tools, apple_adapter, local_rl, build_dag, expectations, magpie, persona_hub, brain_rot, reward_hacking, uld, minillm, rl_checkpoint, iterative_dpo, echo_trap, mod, local_rl_scheduler, spectrum_scan, ship_verdict, reward_hack_control, data_doctor, data_lint, shrink, best_of_n, evolve
-  templates/          - 17 built-in soup.yaml templates (YAML + manifest.json) with load_template loader (v0.39.0, +bco v0.40.0)
+  templates/          - 21 built-in soup.yaml templates (YAML + manifest.json) with load_template loader (v0.39.0, +bco v0.40.0, +4 compliance v0.71.35)
   ui/                 - Web UI (FastAPI + HTML/JS SPA)
 
-tests/                - Test suite (312 files, 15906 tests)
+tests/                - Test suite (314 files, 16283 tests)
 examples/             - Real-world config examples and datasets
 ```
 
@@ -150,7 +150,7 @@ pytest tests/test_data.py::test_detect_alpaca_format -v
 pytest tests/ --cov=soup_cli --cov-report=html
 ```
 
-### Test Files (307 files)
+### Test Files (313 files)
 
 > A representative sample of the suite below. Run `pytest tests/ -v` for the complete list.
 
@@ -388,7 +388,7 @@ If adding a new training algorithm:
 
 1. Create `trainer/your_trainer.py` wrapping the appropriate TRL trainer
 2. Add config fields to `config/schema.py` (Pydantic v2)
-3. Add template to `config/schema.py` (see existing 15 templates)
+3. Add a template YAML under `src/soup_cli/templates/` + a `manifest.json` entry (see the existing 21 templates)
 4. Update `commands/train.py` to route to your trainer
 5. Add 30+ tests in `tests/test_your_trainer.py`
 6. Update `README.md` + the relevant page under `docs/`

{soup_cli-0.71.34 → soup_cli-0.71.36}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: soup-cli
-Version: 0.71.34
+Version: 0.71.36
 Summary: Fine-tune and post-train LLMs in one command. No SSH, no config hell.
 Project-URL: Homepage, https://github.com/MakazhanAlpamys/Soup
 Project-URL: Repository, https://github.com/MakazhanAlpamys/Soup
@@ -193,24 +193,36 @@ infrastructure instead of improving models. Soup fixes that.
 
 ## What's New
 
-**v0.71.34 — adapter algebra + LISA.** Compose fine-tunes like vectors, and train big-model quality on a small-model memory budget.
-
-- **`soup adapters arithmetic "coder + 0.5*math - toxic"`.** Task-vector algebra over LoRA
-  adapters (arXiv:2212.04089) — **add, scale, and NEGATE** trained behaviours into one merged
-  adapter. The math is done right: a LoRA's effective delta `ΔW = B·A` scales *linearly* with
-  each coefficient (subtracting an adapter actually removes its behaviour), not quadratically.
-  Same-base + backdoor-scan gated; mixed ranks refused with a clear message.
-- **LISA — layerwise importance sampling (`training.lisa_enabled`).** Full-fine-tuning quality
-  at LoRA-like memory (arXiv:2403.17919): every N steps LISA re-activates a small random set of
-  decoder layers (embeddings + head always on) and freezes the rest. Live on a 4 GB GPU.
-- **Verified on real models (RTX 3050).** LISA trains SmolLM2-135M end-to-end; `2·a` produces an
-  adapter whose ΔW is *exactly* 2× the original's — the merged adapter loads and serves.
+**v0.71.36 — Data Moat II.** A semantic layer over your training data, plus two tools for what a fine-tune forgets and what it leaks.
+
+- **`soup data dedup --semantic`** — dedup by meaning, not by shared tokens. Catches
+  *reworded* duplicates MinHash's shingling scores as distinct. Zero new dependencies.
+- **`soup data topics <data>`** — cluster your data and get a coverage table
+  (`82% code · 6% math · 1% safety`) with a warning for thin topics, so you can see what
+  you are actually training on.
+- **`soup data canary insert|check`** — insert unique secrets, then prove whether a model
+  memorized them: each secret's loss is ranked against never-inserted controls. Exit 2 on a
+  leak, so CI can gate. On SmolLM2-135M a memorized set lands at percentile 0.0 against a
+  clean model's 4%–93% spread.
+- **`soup train --replay old.jsonl`** — continual-learning rehearsal: interleave your old
+  data so the new task doesn't erase it.
+- **Honest result:** semantic dedup is **not** a paraphrase detector. Measured, paraphrase
+  cosines (0.49–0.76) *overlap* with genuinely-distinct rows (0.54–0.76) — "Add two numbers"
+  vs "Multiply two numbers" scores **higher** than a real paraphrase — so no threshold
+  separates them. The default is deliberately conservative; see the CHANGELOG.
+- **Two blocking bugs fixed:** the hardware-fit gate **refused to train any model you merged
+  yourself** (a local checkpoint's name has no size marker, so it guessed 7B and predicted
+  16 GB), and every `pip install 'soup-cli[extra]'` hint printed **without the extra** —
+  17 sites where the suggested command succeeds and leaves the feature still broken.
 
 ```bash
-soup adapters arithmetic "coder - toxic" \
-  --adapter coder=./coder-lora --adapter toxic=./toxic-lora -o ./cleaned
+soup data topics train.jsonl                    # what am I actually training on?
+soup data dedup train.jsonl --semantic -o clean.jsonl
+
+soup data canary insert train.jsonl -o canaried.jsonl --manifest secrets.json
+soup data canary check --manifest secrets.json --base ./my-model  # exit 2 = leak
 
-soup train --config sft.yaml   # with training.lisa_enabled: true, lisa_num_layers: 2
+soup train --config sft.yaml --replay old_task.jsonl --replay-ratio 0.1
 ```
 
 <details>
@@ -310,6 +322,7 @@ The full feature reference lives in [`docs/`](docs/). Start here:
 | [Evaluation & probes](docs/evaluation.md) | Eval design/gate, eval-gated training, benchmarks, NLG metrics, calibration, Elo arena, diagnose, post-train X-ray probes, A/B, drift, tunability, `soup advise` |
 | [Serving & export](docs/serving-and-export.md) | OpenAI-compatible server, batch inference, benchmarking, merge/export, Anthropic Messages endpoint, speculative decoding (train + measure your own draft), deploy autopilot, Web UI, Agent Forge |
 | [Adapters, registry & governance](docs/adapters-and-governance.md) | Adapter lifecycle/management, model registry, Soup Cans, the data flywheel (`soup loop`), knowledge editing, steering, supply-chain controls (scan/sign/BOM/attest/audit/airgap) |
+| [Compliance & governance quickstart](docs/compliance.md) | HIPAA/SOC2/EU-AI-Act/SR-11-7 `init` templates, provenance (BOM/attest/repro-receipt), audit log, air-gap, model-card autogen (`soup card`), CI gate (`soup ci init`) |
 | [Backends, platform & ops](docs/backends-and-ops.md) | MLX/Unsloth backends, alternative hubs, HF Hub integration, autopilot, experiment tracking, plan/apply, env lockfiles, hardware-fit, completions, plugins, utility commands |
 | [Command reference](docs/commands.md) | The full `soup` command list |
 | [Supported models & extras](docs/models.md) | Recommended model families, the VRAM size guide, the pip extras matrix |

{soup_cli-0.71.34 → soup_cli-0.71.36}/README.md

@@ -49,24 +49,36 @@ infrastructure instead of improving models. Soup fixes that.
 
 ## What's New
 
-**v0.71.34 — adapter algebra + LISA.** Compose fine-tunes like vectors, and train big-model quality on a small-model memory budget.
-
-- **`soup adapters arithmetic "coder + 0.5*math - toxic"`.** Task-vector algebra over LoRA
-  adapters (arXiv:2212.04089) — **add, scale, and NEGATE** trained behaviours into one merged
-  adapter. The math is done right: a LoRA's effective delta `ΔW = B·A` scales *linearly* with
-  each coefficient (subtracting an adapter actually removes its behaviour), not quadratically.
-  Same-base + backdoor-scan gated; mixed ranks refused with a clear message.
-- **LISA — layerwise importance sampling (`training.lisa_enabled`).** Full-fine-tuning quality
-  at LoRA-like memory (arXiv:2403.17919): every N steps LISA re-activates a small random set of
-  decoder layers (embeddings + head always on) and freezes the rest. Live on a 4 GB GPU.
-- **Verified on real models (RTX 3050).** LISA trains SmolLM2-135M end-to-end; `2·a` produces an
-  adapter whose ΔW is *exactly* 2× the original's — the merged adapter loads and serves.
+**v0.71.36 — Data Moat II.** A semantic layer over your training data, plus two tools for what a fine-tune forgets and what it leaks.
+
+- **`soup data dedup --semantic`** — dedup by meaning, not by shared tokens. Catches
+  *reworded* duplicates MinHash's shingling scores as distinct. Zero new dependencies.
+- **`soup data topics <data>`** — cluster your data and get a coverage table
+  (`82% code · 6% math · 1% safety`) with a warning for thin topics, so you can see what
+  you are actually training on.
+- **`soup data canary insert|check`** — insert unique secrets, then prove whether a model
+  memorized them: each secret's loss is ranked against never-inserted controls. Exit 2 on a
+  leak, so CI can gate. On SmolLM2-135M a memorized set lands at percentile 0.0 against a
+  clean model's 4%–93% spread.
+- **`soup train --replay old.jsonl`** — continual-learning rehearsal: interleave your old
+  data so the new task doesn't erase it.
+- **Honest result:** semantic dedup is **not** a paraphrase detector. Measured, paraphrase
+  cosines (0.49–0.76) *overlap* with genuinely-distinct rows (0.54–0.76) — "Add two numbers"
+  vs "Multiply two numbers" scores **higher** than a real paraphrase — so no threshold
+  separates them. The default is deliberately conservative; see the CHANGELOG.
+- **Two blocking bugs fixed:** the hardware-fit gate **refused to train any model you merged
+  yourself** (a local checkpoint's name has no size marker, so it guessed 7B and predicted
+  16 GB), and every `pip install 'soup-cli[extra]'` hint printed **without the extra** —
+  17 sites where the suggested command succeeds and leaves the feature still broken.
 
 ```bash
-soup adapters arithmetic "coder - toxic" \
-  --adapter coder=./coder-lora --adapter toxic=./toxic-lora -o ./cleaned
+soup data topics train.jsonl                    # what am I actually training on?
+soup data dedup train.jsonl --semantic -o clean.jsonl
 
-soup train --config sft.yaml   # with training.lisa_enabled: true, lisa_num_layers: 2
+soup data canary insert train.jsonl -o canaried.jsonl --manifest secrets.json
+soup data canary check --manifest secrets.json --base ./my-model  # exit 2 = leak
+
+soup train --config sft.yaml --replay old_task.jsonl --replay-ratio 0.1
 ```
 
 <details>
@@ -166,6 +178,7 @@ The full feature reference lives in [`docs/`](docs/). Start here:
 | [Evaluation & probes](docs/evaluation.md) | Eval design/gate, eval-gated training, benchmarks, NLG metrics, calibration, Elo arena, diagnose, post-train X-ray probes, A/B, drift, tunability, `soup advise` |
 | [Serving & export](docs/serving-and-export.md) | OpenAI-compatible server, batch inference, benchmarking, merge/export, Anthropic Messages endpoint, speculative decoding (train + measure your own draft), deploy autopilot, Web UI, Agent Forge |
 | [Adapters, registry & governance](docs/adapters-and-governance.md) | Adapter lifecycle/management, model registry, Soup Cans, the data flywheel (`soup loop`), knowledge editing, steering, supply-chain controls (scan/sign/BOM/attest/audit/airgap) |
+| [Compliance & governance quickstart](docs/compliance.md) | HIPAA/SOC2/EU-AI-Act/SR-11-7 `init` templates, provenance (BOM/attest/repro-receipt), audit log, air-gap, model-card autogen (`soup card`), CI gate (`soup ci init`) |
 | [Backends, platform & ops](docs/backends-and-ops.md) | MLX/Unsloth backends, alternative hubs, HF Hub integration, autopilot, experiment tracking, plan/apply, env lockfiles, hardware-fit, completions, plugins, utility commands |
 | [Command reference](docs/commands.md) | The full `soup` command list |
 | [Supported models & extras](docs/models.md) | Recommended model families, the VRAM size guide, the pip extras matrix |

{soup_cli-0.71.34 → soup_cli-0.71.36}/docs/README.md

@@ -14,6 +14,7 @@ feature reference — every `soup` capability, grouped by area.
 | [Evaluation & probes](evaluation.md) | Eval design/gate, eval-gated training, benchmarks, NLG metrics, calibration, Elo arena, diagnose, `soup ship` verdict, post-train X-ray probes, A/B, drift, tunability, `soup advise` |
 | [Serving & export](serving-and-export.md) | OpenAI-compatible server, batch inference, benchmarking, merge/export, Anthropic Messages endpoint, speculative decoding (train + measure your own draft), deploy autopilot, Web UI, Agent Forge |
 | [Adapters, registry & governance](adapters-and-governance.md) | Adapter lifecycle/management, model registry, Soup Cans, the data flywheel (`soup loop`), knowledge editing, steering, supply-chain controls |
+| [Compliance & governance quickstart](compliance.md) | HIPAA/SOC2/EU-AI-Act/SR-11-7 `init` templates, provenance (BOM/attest/repro-receipt), audit log, air-gap, model-card autogen (`soup card`), CI gate (`soup ci init`) |
 | [Backends, platform & ops](backends-and-ops.md) | MLX/Unsloth backends, Modal cloud GPU training, alternative hubs, HF Hub integration, autopilot, experiment tracking, plan/apply, env lockfiles, hardware-fit, completions, plugins, utility commands |
 | [Command reference](commands.md) | The full `soup` command list |
 | [Supported models & extras](models.md) | Recommended model families, the VRAM size guide, the pip extras matrix |

{soup_cli-0.71.34 → soup_cli-0.71.36}/docs/commands.md

@@ -8,8 +8,10 @@
 
 ```
 soup init [--template chat|code|...|audio]       Create config
+soup init --template hipaa|soc2|eu-ai-act|sr-11-7  Compliance-shaped starting config + the commands for that regime (v0.71.35)
 soup autopilot --model <id> --data d.jsonl --goal <g>  Zero-configsoup train --config soup.yaml                 Start training
 soup train --config soup.yaml --tensorboard   Train with TensorBoard logging
+soup train --config soup.yaml --replay old.jsonl --replay-ratio 0.1  Continual-learning rehearsal: interleave old data so the new task doesn't erase it (sft/pretrain)
 soup train --config soup.yaml --fsdp full_shard  Train with FSDP2
 soup train --config soup.yaml --deepspeed zero++  DeepSpeed ZeRO++ (quantized comms)
 soup train --config soup.yaml --gpus auto|N      Multi-GPU launch hint
@@ -76,6 +78,10 @@ soup data lint <path>                         Preference-data linter: length bia
 soup data convert <path> --to chatml          Convert between formats
 soup data merge data1.jsonl data2.jsonl       Combine datasets
 soup data dedup <path> --threshold 0.8        Remove duplicates (MinHash)
+soup data dedup <path> --semantic             Dedup by embedding cosine — catches rewordings MinHash misses ([train])
+soup data topics <path> [--clusters N|auto]   Cluster + c-TF-IDF labels + coverage table + thin-topic warnings ([train])
+soup data canary insert <path> -o <out> --manifest <m>  Insert K secrets to later prove memorization (manifest = SECRET)
+soup data canary check --manifest <m> --base <model>    Rank each secret's loss vs never-inserted controls; exit 2 = leak
 soup data stats <path>                        Extended statistics
 soup data generate --prompt "..." --count 100 Generate synthetic data
 soup data generate ... --provider ollama      Use local Ollama instance
@@ -159,6 +165,9 @@ soup ship --evidence ev.json [--output v.json]  Decide offline from pre-computed
 soup ship ... --task-mode judge_score --judge-model ollama://llama3.1  Leg-1 via LLM-as-a-judge
 soup ship ... --task-mode pairwise --judge-model ollama://llama3.1  Leg-1 via swap-debiased judge win-rate (base=0.5) (v0.71.31)
 soup ship ... --general-suite mmlu,gsm8k --baseline base.json  lm-eval leg-2 + recorded base scores
+soup card <registry-id> -o MODELCARD.md       HF model card from a registry entry: training config, evals, hashes, lineage, artifacts (v0.71.35)
+soup push --model ./out --repo you/m --card <registry-id>  Upload that registry-driven card as the README (HF only) (v0.71.35)
+soup ci init [--data d.jsonl --suite s.yaml --evidence ev.json] [--branch main --python 3.11] [--force]  Write .github/workflows/soup-gate.yml: data validate -> expect -> ship gate on every PR (v0.71.35)
 soup mcp serve                                MCP server over stdio (drive Soup from Claude Code / Cursor / Cline; requires [mcp] extra) (v0.71.28)
 soup mcp serve --allow-mutating               Also expose plan-only train_start / export tools (never execute) (v0.71.28)
 soup shrink --model <id|path> --drop-ratio 0.25 --calib c.jsonl -o shrunk  Depth-prune least-important layer block + SHIP/DON'T-SHIP ppl verdict (exit 0/2/1) (v0.71.29)

soup_cli-0.71.36/docs/compliance.md

@@ -0,0 +1,136 @@
+# Compliance & governance quickstart
+
+[← Back to the docs index](README.md)
+
+Soup ships an end-to-end compliance workflow: start from a regulation-shaped
+config, train with provenance capture, sign + attest the artifact, keep an audit
+trail, and move it across an air gap — then publish a documented model card and
+gate future changes in CI.
+
+- **Pre-wired configs:** `soup init --template hipaa|soc2|eu-ai-act|sr-11-7`
+- **Provenance:** `soup train --repro-receipt` · `soup bom emit` · `soup attest emit`
+- **Integrity:** `soup adapters sign` / `verify` · `soup adapters scan`
+- **Audit trail:** the HIPAA/SOC2 audit log is on by default (`soup audit-log tail`)
+- **Air gap:** `soup airgap-bundle`
+- **Publish:** `soup card <registry-id>` → `MODELCARD.md` (or `soup push --card`)
+- **CI gate:** `soup ci init` → a PR workflow that runs validate → expect → ship
+
+---
+
+## 1. Start from a compliance template
+
+Each template is a normal training config plus header comments listing the exact
+compliance commands to run around it. Pick the regime you operate under:
+
+```bash
+soup init --template hipaa      # Protected Health Information
+soup init --template soc2       # SOC 2 Trust Services Criteria
+soup init --template eu-ai-act  # EU AI Act Annex XI/XII
+soup init --template sr-11-7    # SR 11-7 Model Risk Management
+```
+
+The compliance controls are Soup **CLI flags/commands, not config keys** — the
+template header documents which ones apply. The steps below are the common path.
+
+## 2. Clean the data before training
+
+```bash
+soup data pii ./data/train.jsonl            # flag emails / phones / SSNs / MRNs
+soup data decontaminate ./data/train.jsonl  # drop public-benchmark overlap
+```
+
+## 3. Train with a reproducibility receipt (+ Annex XI / energy for the EU)
+
+```bash
+# SR 11-7 / SOC 2 / HIPAA: capture seeds, kernels, GPU, OS
+soup train --config soup.yaml --repro-receipt receipt.json
+
+# EU AI Act: auto-generate the Annex XI/XII documentation + measure energy
+soup train --config soup.yaml \
+    --annex-xi annex_xi.md \
+    --track-energy --energy-country DEU --energy-out energy.json
+```
+
+The audit log records every command automatically:
+
+```bash
+soup audit-log tail          # review the trail
+soup audit-log rotate        # force a rotation pass
+```
+
+## 4. Register the run, then emit BOM + attestation
+
+```bash
+soup registry push --run-id <run-id> --name my-model --tag v1
+
+soup bom emit --name my-model --base-sha <hex> --config-sha <hex> \
+    --energy energy.json --format both       # CycloneDX + SPDX
+soup attest emit --stage train --subject my-model --sha <hex> \
+    --sign ed25519 --key key.pem             # in-toto + SLSA-3
+```
+
+## 5. Sign, scan, and verify the artifact
+
+```bash
+soup adapters scan ./output                              # weight-space backdoor scan
+soup adapters sign ./output --backend ed25519 --generate-key key.pem
+soup adapters verify ./output --strict --public-key key.pub.pem
+```
+
+## 6. Air-gap transfer (optional)
+
+```bash
+soup airgap-bundle --model ./output --output my-model.tar --repro-receipt receipt.json
+```
+
+## 7. Generate a documented model card
+
+Turn the registry entry into a provenance-rich `MODELCARD.md` — base model,
+training config, eval scorecard, config/data hashes, lineage, and every
+registered artifact:
+
+```bash
+soup card my-model:v1 -o MODELCARD.md
+# or, when uploading to the Hub, override the auto-generated card:
+soup push --model ./output --repo you/my-model --card my-model:v1
+```
+
+## 8. Gate future changes in CI
+
+Write a GitHub Actions workflow that blocks a PR unless the data validates, the
+expectations suite passes, and the SHIP verdict is green:
+
+```bash
+soup ci init --data data/train.jsonl --suite expectations.yaml --evidence ship_evidence.json
+# writes .github/workflows/soup-gate.yml
+```
+
+The generated job runs, in order:
+
+```
+soup data validate <data>       # dataset format compliance
+soup expect <data> <suite>      # PII / token-length / refusal / judge expectations
+soup ship --evidence <ev.json>  # SHIP / DON'T-SHIP (exit 2 blocks the merge)
+```
+
+A minimal `expectations.yaml` for the second step:
+
+```yaml
+expectations:
+  - name: expect_no_pii
+  - name: expect_token_length_between
+    min_tokens: 1
+    max_tokens: 512
+```
+
+Supported names: `expect_no_pii`, `expect_token_length_between`,
+`expect_no_refusal_pattern`, `expect_chosen_preferred_over_rejected_by_judge`.
+
+Every path is shell-quoted and validated to stay under the repo root, so the
+rendered workflow is injection-safe. Edit the paths to match your repo.
+
+---
+
+See also: [Adapters, registry & governance](adapters-and-governance.md) for the
+full supply-chain command set, and [Evaluation & probes](evaluation.md) for the
+`soup ship` verdict engine.

{soup_cli-0.71.34 → soup_cli-0.71.36}/docs/data.md

@@ -22,6 +22,9 @@
 - [Synthetic Data Forge](#synthetic-data-forge)
 - [Data Quality Scorecard](#data-quality-scorecard)
 - [Remote Datasets (S3 / GCS / Azure / OCI)](#remote-datasets-s3--gcs--azure--oci)
+- [Semantic dedup (`soup data dedup --semantic`)](#semantic-dedup-soup-data-dedup---semantic)
+- [Topic map (`soup data topics`)](#topic-map-soup-data-topics)
+- [Canaries (`soup data canary insertcheck`)](#canaries-soup-data-canary-insertcheck)
 - [Data Recipe DAG](#data-recipe-dag)
 - [Data Mixing Optimizer (BETA)](#data-mixing-optimizer-beta)
 - [AOT Tokenization with `soup data preprocess`](#aot-tokenization-with-soup-data-preprocess)
@@ -29,6 +32,125 @@
 
 ---
 
+## Semantic dedup (`soup data dedup --semantic`)
+
+`soup data dedup` removes near-duplicates with MinHash by default — fast, no
+torch, but **lexical**: it compares shared token shingles, so two rows that say
+the same thing in different words look unrelated to it.
+
+`--semantic` compares embedding cosine instead:
+
+```bash
+soup data dedup train.jsonl --semantic -o clean.jsonl
+soup data dedup train.jsonl --semantic --threshold 0.85 --field text -o clean.jsonl
+soup data dedup train.jsonl --semantic --embed-model sentence-transformers/all-mpnet-base-v2
+```
+
+Requires the `[train]` extra (it reuses `transformers`; there is no new
+dependency) and downloads a small embedding model on first use. Plain MinHash
+`dedup` stays on the light core.
+
+**What it buys you.** Measured against MinHash on the same rows
+(all-MiniLM-L6-v2):
+
+| pair | cosine | MinHash | `--semantic` |
+|---|---|---|---|
+| exact duplicate | 1.000 | caught | caught |
+| "sorts **a list** of integers" / "sorts **an array** of integers" | 0.908 | **missed** | caught |
+| "which sorts a list of **ints**" (reworded) | 0.880 | **missed** | caught |
+| "Add two numbers" / "Multiply two numbers" | 0.759 | kept | kept (correct) |
+
+So `--semantic` catches **rewordings** MinHash's shingling scores as distinct.
+
+### It is not a paraphrase detector — and why the default is 0.8
+
+Heavier paraphrases are **not** reliably separable. Measured, paraphrase cosines
+(0.49–0.76) *overlap* with genuinely-distinct rows (0.54–0.76):
+
+- "reverse a string" / "invert the order of characters" — a **true paraphrase** — scores **0.491**
+- "Add two numbers" / "Multiply two numbers" — **two rows you must keep** — scores **0.759**
+
+A real paraphrase can score *lower* than two rows that must both survive, so **no
+threshold cleanly separates them**. Lowering `--threshold` to chase paraphrase
+recall deletes real training rows — silent data loss, which is worse than keeping
+a duplicate. The 0.8 default is deliberately conservative. Raise or lower it only
+against your own data, and check what got dropped.
+
+`--threshold` means Jaccard for MinHash and cosine for `--semantic`. They are
+different scales; a value tuned for one is not meaningful for the other.
+
+## Topic map (`soup data topics`)
+
+See what you are actually training on:
+
+```bash
+soup data topics train.jsonl                       # 'auto' picks the cluster count
+soup data topics train.jsonl --clusters 8 -o topics.json
+```
+
+Embeds every row, clusters with k-means, and labels each cluster with c-TF-IDF
+terms — terms frequent in *that* cluster and rare elsewhere, so filler words like
+"the" never become a label. Prints a coverage table plus a warning for any topic
+under 2% of the data:
+
+```
+        Topic map — 4200 rows, 6 clusters
+┏━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━┳━━━━━━━━━━┓
+┃ Topic                    ┃ Rows ┃ Coverage ┃
+┡━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━╇━━━━━━━━━━┩
+│ function / python / code │ 3444 │    82.0% │
+│ theorem / proof / math   │  252 │     6.0% │
+│ refuse / harmful / safe  │   42 │     1.0% │
+└──────────────────────────┴──────┴──────────┘
+topic 'refuse / harmful / safe' is thin: 1.0% of rows (42/4200)
+```
+
+Labels are **emergent term clusters**, not a classification against a fixed
+taxonomy: "82% code" means 82% of rows landed in a cluster whose top terms look
+like code. Requires `[train]`.
+
+## Canaries (`soup data canary insert|check`)
+
+Prove whether a model memorized your data — for leak detection and provenance.
+
+```bash
+# 1. insert K unique secrets (keep the manifest OUT of your repo)
+soup data canary insert train.jsonl -o canaried.jsonl --count 16 --manifest secrets.json
+
+# 2. train on canaried.jsonl as usual, then:
+soup data canary check --manifest secrets.json --base ./my-model --adapter ./lora
+```
+
+`check` measures the model's loss on each inserted secret and ranks it against
+never-inserted **controls** drawn from the same secret space and sharing the same
+carrier prompt — so a low loss means the *secret* is unusually likely, not the
+prompt. Exit **2** on MAJOR, so CI can gate on a leak.
+
+This is loss-vs-controls (Carlini et al., *The Secret Sharer*), not "ask the model
+and see if it says the secret": a model can memorize a canary and still not emit
+it under greedy decoding, so "nothing came back" would be false reassurance.
+
+The verdict asks whether **more** canaries look memorized than chance explains
+(binomial tail, α=0.05) rather than whether any single one dipped low — with 16
+canaries, an "any single one" rule fires on a **clean** model about 15% of the
+time, which would make a CI gate useless.
+
+Measured on SmolLM2-135M:
+
+| model | loss | percentiles | verdict |
+|---|---|---|---|
+| trained on the canaries | 1.7–2.5 | all 0.0% | **MAJOR** (exit 2) |
+| never saw them | 4.1–6.2 | 1.6%–93% | OK (exit 0) |
+
+**The manifest is the sensitive artifact**, not the dataset: anyone holding it can
+reproduce the secrets. It is written `0600` on POSIX and must not be committed
+alongside the data it protects. `check --output` embeds the same secrets.
+
+Exposure is a **sampled-control approximation**, not full-space rank enumeration —
+"no exposure" is not proof of no memorization.
+
+---
+
 ## Data Engineering Pro
 
 The v0.69.0 release ships 5 surfaces that turn dataset prep from "throw a JSONL at the trainer" into a first-class engineering workflow.

{soup_cli-0.71.34 → soup_cli-0.71.36}/docs/peft-and-efficiency.md

@@ -248,7 +248,7 @@ configs where expert FFNs need lower rank than attention. Caps: 256 keys × valu
 `lora_dropout` strip) auto-fire when the model name and architecture match. Both are
 gated and silent on unrelated models.
 
-**Template registry** — the 16 built-in templates now live as
+**Template registry** — the 21 built-in templates now live as
 `src/soup_cli/templates/*.yaml` with a `manifest.json` index. `soup init --template <name>`
 reads the YAML; the inline copies in `schema.py` stay as a back-compat fallback,
 deprecated in favour of the YAML registry.