Spaces:
Paused
Paused
Commit History
-
fix: T4-safe 4bit compute, eval/audit resilience, async inference, plot_rewards CSV/table
raj92154c5725 -
Use SVG reward curve for HF Space Git (reject binary PNG)
raj921HF pre-receive hook blocks PNG in Space repo. Plot to docs/reward_curve.svg; plot_rewards saves SVG when output path ends in .svg. README + proof script updated.4a624ee -
Add DriftShield proof artifacts and generator script
raj921- reward_curve.png from illustrative CSV (docs/driftshield_proof_reward_log.csv) - eval_compare.md: naive vs scripted strong on all D1 tasks - before_after_prompt_injection.md: prompt-injection narrative - scripts/generate_proof_artifacts.py to regenerate without GPU - README links; eval_compare/plot_rewards wording; ignore .venv-proofaabbe72 -
Rebrand: drop 'Support Ops' user-facing text; add DriftShield* aliases; tighten README
raj921Audited every reference. Two layers: User-facing text (renamed everywhere): - 'Support Ops Env' / 'SupportOps Control Tower' / 'first-line support' / 'SaaS support triage' -> DriftShield language - Files touched: README.md, docs/BENCHMARK_BRIEF.md, train.py, train_gemma4.py, eval_compare.py, audit.py, plot_rewards.py, inference.py (SYS_PROMPT + BENCHMARK constant), server/support_ops_environment.py (env metadata name + description), __init__.py (module docstring). - docs/BENCHMARK_BRIEF.md rewritten end to end (was stale and still talked about 3 support-triage tasks). Python identifiers (kept as legacy slugs to avoid breaking the live HF Space URL https://huggingface.co/spaces/raj23211/support-ops-env and every notebook import): - Package still importable as 'support_ops_env'. - Classes still named SupportOpsEnv / SupportOpsAction / etc. - Added DriftShield-branded aliases in __init__.py: DriftShieldEnv = SupportOpsEnv DriftShieldAction…88930f5
-
feat: GRPO training stack (Qwen3 + Unsloth opt-in, curriculum, eval, audit)
raj921Adds the full RL training path for the Support Ops Env: * train.py: Qwen GRPO trainer with --difficulty curriculum, --use-unsloth opt-in path (unsloth/Qwen3.5-4B, bf16 LoRA), --load-in-4bit for T4, --use-vllm opt-in for A100. Default model Qwen/Qwen3-4B-Instruct-2507. * train_gemma4.py: experimental Gemma 4 path via TRL environment_factory. * tasks.py + __init__.py: curriculum helpers (easy/medium/hard/expert/all). * eval_compare.py: baseline-vs-trained evaluator -> JSON + Markdown table. * audit.py: reward-hack audit -> trajectory dumps + flag counts. * plot_rewards.py: 4-panel reward curves (total / components / penalty / rolling pass-rate). Reads v1 or v2 CSV schema. * pyproject.toml: [train] + [gemma] + [unsloth] optional-deps. * support_ops_colab.ipynb + support_ops_colab_gemma4.ipynb: full Colab notebooks with curriculum + Unsloth toggle + 4-bit fallback. * README.md: judge-facing rewrite aligned with the hackathon self-serve guide (tl;dr table, curriculum, reward desi…ffa6d1d