update: add YouTube video link

fix: restore YAML frontmatter to fix Space configuration error

docs: add Related Research section with sycophancy papers

docs: add Quick Links table to README

update: point to full trained model

data: regenerate level3.jsonl — 100 questions with adversarial pressure messages

feat: append Phase 5 Level 3 training section to sanity_run.ipynb

data: add level3.jsonl — 100 questions with adversarial pressure messages

test: add Level 3 adversarial pressure integration tests

fix: persist initial context (distractors/pressure) across all episode turns

feat: extend environment with level=3, resistance reward ±0.2

feat: add pressure_shown field to DeceitState

fix: remove dead import, skip sleep when api unavailable

feat: generate_pressure.py — Level 3 adversarial pressure dataset script

security: remove real API key from .env.example

Phase 4 complete: level2 dataset, distractor env, 90 tests passing

fix: exponential backoff on rate limit (10 retries, 30s*attempt)

fix: retry on rate limit in generate_distractors, 21s sleep for free-tier 3 RPM

feat: append Phase 4 Level 2 training section to sanity_run.ipynb

test: add Level 2 integration tests (test_level2.py)

feat: extend reset() to support level=2 with distractor context

feat: add 429 retry wrapper to grader semantic check

fix: propagate fatal API errors, strip markdown fences, cleanup

fix: count skipped rows toward progress interval

fix: sleep after error path, print progress every 10 iterations

feat: generate_distractors.py — GPT-4o-mini Level 2 dataset script

Merge HF Space remote, keeping local version

update: complete README with results, API docs, reward curve

Upload folder using huggingface_hub

Fix notebook: HF Space URL, /step envelope, health check retry on cold start

Upload folder using huggingface_hub

Fix Dockerfile port: use 7860 for HF Spaces compatibility

Update README: Phase 3 complete, HF Space badge, quickstart, reward table

Upload folder using huggingface_hub

Phase 3: Dockerfile, openenv manifest, client, deployment guide, GRPO training notebook

Phase 2.5: multi-turn episodes, bug fixes, dataset cleanup

Phase 2 complete: Level 1 env runs locally, tests green, 100-question dataset

Fix build backend: use setuptools.build_meta instead of legacy path

Phase 1 complete: schemas, reward design, project scaffold