RomanNum3ral
/
soup
mirror of https://github.com/razor-ai/soup.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
soup/soup_cli/commands
History
Shivam 4e95d4c71f
feat(echo-trap): add tokenizer-aware repetition scoring (#242) 
Closes #241.

Adds opt-in tokenizer-aware n-gram path for the v0.70.0 Part F echo-trap detector. The existing whitespace `score_echo_signal` is unchanged; callers opt in via the new `score_trajectory_repetition_tokenized` / `score_echo_signal_tokenized` helpers or the `--echo-trap-tokenizer-aware` train flag.

Acceptance criterion from #241 verified: synthetic case where decoded strings differ by punctuation but token-id sequence repeats — whitespace path returns OK, tokenizer-aware path returns TRAP.
 		2026-05-25 19:46:51 +05:00
..
__init__.py Initial project setup: CLI skeleton + config + trainer + data pipeline 2026-02-20 16:14:56 +05:00
_eval_v0550.py feat(eval): soup eval design — derive evals from data (v0.55.0) 2026-05-15 12:44:14 +05:00
_eval_v0610.py feat(unlearn/edit): v0.61.0 — Unlearning & Knowledge Edit (NPO/SimNPO/RMU + ROME/MEMIT/AlphaEdit) 2026-05-19 15:16:15 +05:00
_eval_v0650.py feat(v0.65.0): Eval Depth — judge calibration + behaviour battery + capability suite + CheckList DSL + IRT subset 2026-05-21 15:35:47 +05:00
ab.py feat(v0.63.0): Production Trace Ecosystem — soup ingest + prune-prompt + active-sample + ab + drift-alarm 2026-05-20 22:02:54 +05:00
active_sample.py feat(v0.63.0): Production Trace Ecosystem — soup ingest + prune-prompt + active-sample + ab + drift-alarm 2026-05-20 22:02:54 +05:00
adapters.py feat(v0.67.0): Adapter Lifecycle Finish — CMA-ES merge + VeRA bank + MoLE + PRs + soup.lock + bisect 2026-05-24 15:58:30 +05:00
advise.py feat(advise): soup advise — pre-flight decision (v0.54.0) 2026-05-14 21:48:19 +05:00
agent.py feat(agent): Agent Forge — OpenAPI/MCP/GraphQL spec → tool-calling SFT dataset (v0.46.0 Part B) 2026-05-11 17:12:13 +05:00
airgap.py feat(security): v0.60.0 — Supply Chain Security wedge (adapter scan/sign/verify, strict-safetensors, namespace-pin, license-matrix, airgap-bundle) 2026-05-19 13:05:14 +05:00
apple_adapter.py feat(v0.68.0): Anti-trend Insurance — compile (DSPy/GEPA) + distill-prompt + compile-tools + apple-adapter + local-rl 2026-05-24 20:24:39 +05:00
attest.py feat(governance): v0.59.0 — CycloneDX/SPDX BOM + in-toto/SLSA-3 attest + Annex XI/XII + audit-log + repro-receipt + energy schema 2026-05-18 21:05:12 +05:00
audit_log.py feat(governance): v0.59.0 — CycloneDX/SPDX BOM + in-toto/SLSA-3 attest + Annex XI/XII + audit-log + repro-receipt + energy schema 2026-05-18 21:05:12 +05:00
autopilot.py fix(autopilot): Windows py3.9 path traversal false-positive 2026-04-13 13:31:43 +05:00
bench.py feat(v0.53.9): Live Dashboard + UX + Bench + Standalone CLIs 2026-05-14 04:28:19 +05:00
bom.py feat(governance): v0.59.0 — CycloneDX/SPDX BOM + in-toto/SLSA-3 attest + Annex XI/XII + audit-log + repro-receipt + energy schema 2026-05-18 21:05:12 +05:00
build.py feat(v0.69.0): Data Engineering Pro — soup build + expect + gen-magpie + persona-mix + brain-rot 2026-05-25 16:00:15 +05:00
can.py fix(v0.33.0): missed checklist items (CONTRIBUTING + Unicode arrow) 2026-04-27 20:11:22 +05:00
chat.py feat(v0.53.10): Quick wins + packaging + UX wiring 2026-05-14 13:58:20 +05:00
compile_cmd.py feat(v0.68.0): Anti-trend Insurance — compile (DSPy/GEPA) + distill-prompt + compile-tools + apple-adapter + local-rl 2026-05-24 20:24:39 +05:00
compile_tools.py feat(v0.68.0): Anti-trend Insurance — compile (DSPy/GEPA) + distill-prompt + compile-tools + apple-adapter + local-rl 2026-05-24 20:24:39 +05:00
completions.py feat(v0.64.0): Pre-flight & Tooling — tunability, plan/apply, env, hardware-fit, completions, license-advisor 2026-05-20 23:45:04 +05:00
cost.py refactor(cost): polish soup cost from PR #42 2026-04-22 23:13:42 +05:00
data.py feat(v0.69.0): Data Engineering Pro — soup build + expect + gen-magpie + persona-mix + brain-rot 2026-05-25 16:00:15 +05:00
data_forge.py feat(v0.53.7): Data Forge + Pipeline live (wave 1) — 11 items 2026-05-13 19:28:11 +05:00
data_mix.py feat(adaptive): v0.53.5 — Adaptive Training (BETA → stable) 2026-05-13 15:55:28 +05:00
data_score.py feat(v0.53.7): Data Forge + Pipeline live (wave 1) — 11 items 2026-05-13 19:28:11 +05:00
delinearize_llama4.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
deploy.py feat(v0.53.8): Remote data + Hubs + Trackers (wave 2) — 6 features 2026-05-13 23:55:23 +05:00
diagnose.py feat(diagnose): soup diagnose — post-training model report card (v0.56.0) 2026-05-15 15:03:00 +05:00
diff.py feat(security): trust_remote_code opt-in across non-SFT trainers + 5 commands (v0.40.4 Part A) 2026-05-09 13:20:13 +05:00
distill_prompt.py feat(v0.68.0): Anti-trend Insurance — compile (DSPy/GEPA) + distill-prompt + compile-tools + apple-adapter + local-rl 2026-05-24 20:24:39 +05:00
doctor.py refactor(doctor): harden NCCL benchmark from PR #178 2026-05-17 18:38:34 +05:00
drift_alarm.py feat(v0.63.0): Production Trace Ecosystem — soup ingest + prune-prompt + active-sample + ab + drift-alarm 2026-05-20 22:02:54 +05:00
edit.py feat(unlearn/edit): v0.61.0 — Unlearning & Knowledge Edit (NPO/SimNPO/RMU + ROME/MEMIT/AlphaEdit) 2026-05-19 15:16:15 +05:00
env.py feat(v0.64.0): Pre-flight & Tooling — tunability, plan/apply, env, hardware-fit, completions, license-advisor 2026-05-20 23:45:04 +05:00
eval.py feat(v0.65.0): Eval Depth — judge calibration + behaviour battery + capability suite + CheckList DSL + IRT subset 2026-05-21 15:35:47 +05:00
expect.py feat(v0.69.0): Data Engineering Pro — soup build + expect + gen-magpie + persona-mix + brain-rot 2026-05-25 16:00:15 +05:00
export.py feat(v0.53.10): Quick wins + packaging + UX wiring 2026-05-14 13:58:20 +05:00
fetch.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
generate.py feat(security): trust_remote_code opt-in across non-SFT trainers + 5 commands (v0.40.4 Part A) 2026-05-09 13:20:13 +05:00
history.py fix(v0.40.1): QA Hardening — UTF-8 bootstrap, schema strictness, multi-objective preference runtime, CLI UX 2026-05-08 12:03:16 +05:00
infer.py feat(v0.53.10): Quick wins + packaging + UX wiring 2026-05-14 13:58:20 +05:00
ingest.py feat(v0.63.0): Production Trace Ecosystem — soup ingest + prune-prompt + active-sample + ab + drift-alarm 2026-05-20 22:02:54 +05:00
init.py fix(v0.40.1): QA Hardening — UTF-8 bootstrap, schema strictness, multi-objective preference runtime, CLI UX 2026-05-08 12:03:16 +05:00
iterative_dpo.py feat(v0.70.0): Loop Hardening — reward-hacking + ULD + MiniLLM + RL ckpt + iterative DPO + echo-trap 2026-05-25 17:17:03 +05:00
license_advisor.py feat(v0.64.0): Pre-flight & Tooling — tunability, plan/apply, env, hardware-fit, completions, license-advisor 2026-05-20 23:45:04 +05:00
llama.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
local_rl.py feat(v0.68.0): Anti-trend Insurance — compile (DSPy/GEPA) + distill-prompt + compile-tools + apple-adapter + local-rl 2026-05-24 20:24:39 +05:00
lock.py feat(v0.67.0): Adapter Lifecycle Finish — CMA-ES merge + VeRA bank + MoLE + PRs + soup.lock + bisect 2026-05-24 15:58:30 +05:00
loop.py feat(loop): soup loop CLI-first data flywheel capstone (v0.58.0) 2026-05-15 22:52:37 +05:00
merge.py feat(v0.53.10): Quick wins + packaging + UX wiring 2026-05-14 13:58:20 +05:00
merge_sharded_fsdp_weights.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
migrate.py fix(v0.40.1): QA Hardening — UTF-8 bootstrap, schema strictness, multi-objective preference runtime, CLI UX 2026-05-08 12:03:16 +05:00
monitor.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
plan.py feat(v0.64.0): Pre-flight & Tooling — tunability, plan/apply, env, hardware-fit, completions, license-advisor 2026-05-20 23:45:04 +05:00
plugins.py feat(v0.45.0): Plugin System & Ecosystem Wins — 5 Parts, +169 tests 2026-05-10 21:21:04 +05:00
probe.py feat(v0.66.0): Post-train X-rays — SAE diff + live blame + sleeper probe + interference matrix + probe pack 2026-05-22 14:08:12 +05:00
profile.py fix: v0.24.1 — Windows Unicode fix, AWQ/GPTQ output path traversal 2026-04-03 23:41:44 +05:00
prune_prompt.py feat(v0.63.0): Production Trace Ecosystem — soup ingest + prune-prompt + active-sample + ab + drift-alarm 2026-05-20 22:02:54 +05:00
push.py feat(v0.53.10): Quick wins + packaging + UX wiring 2026-05-14 13:58:20 +05:00
quantize.py feat(v0.44.0): Live Dashboard & UX - 21 features, +192 tests 2026-05-10 19:20:18 +05:00
quickstart.py feat(v0.40.2): Quick polish + v0.40.1 carry-overs (#36, #50, #51 + 7 papercuts) 2026-05-08 13:20:21 +05:00
recipes.py fix(v0.40.1): QA Hardening — UTF-8 bootstrap, schema strictness, multi-objective preference runtime, CLI UX 2026-05-08 12:03:16 +05:00
registry.py feat(registry): add Local Model Registry / Provenance Vault (v0.26.0 Part A) 2026-04-20 20:07:54 +05:00
runs.py fix(v0.48.0): lstat symlink check on original path, not realpath 2026-05-11 19:25:22 +05:00
serve.py feat(v0.62.0): RAG & Activation Steering — RAFT + RA-DIT + soup steer + citation-faithful + GRACE codebook 2026-05-20 20:20:30 +05:00
steer.py feat(v0.62.0): RAG & Activation Steering — RAFT + RA-DIT + soup steer + citation-faithful + GRACE codebook 2026-05-20 20:20:30 +05:00
sweep.py feat(preference): v0.40.0 — Preference Variety (4 Parts: BCO + dispatcher + DPO variants + multi-objective) 2026-05-01 22:26:30 +05:00
tokenizer.py feat(v0.53.9): Live Dashboard + UX + Bench + Standalone CLIs 2026-05-14 04:28:19 +05:00
train.py feat(echo-trap): add tokenizer-aware repetition scoring (#242) 2026-05-25 19:46:51 +05:00
tui.py feat(observability): v0.34.0 — Observability & Dev UX (7 Parts) 2026-04-28 13:14:49 +05:00
tunability.py feat(v0.64.0): Pre-flight & Tooling — tunability, plan/apply, env, hardware-fit, completions, license-advisor 2026-05-20 23:45:04 +05:00
ui.py feat(v0.53.9): Live Dashboard + UX + Bench + Standalone CLIs 2026-05-14 04:28:19 +05:00
why.py feat(observability): v0.34.0 — Observability & Dev UX (7 Parts) 2026-04-28 13:14:49 +05:00