Commit History

  1. update: add YouTube video link

    Jayant-Kernel
    4520926
  2. fix: restore YAML frontmatter to fix Space configuration error

    Jayant-Kernel
    4f7ce24
  3. docs: add Related Research section with sycophancy papers

    Jayant-Kernel
    72715b2
  4. docs: add Quick Links table to README

    Jayant-Kernel
    d07843b

  1. update: point to full trained model

    Jayant-Kernel
    Unverified signature
    ed01808
  2. data: regenerate level3.jsonl — 100 questions with adversarial pressure messages

    Jayant-Kernel
    Unverified signature
    43ba213
  3. feat: append Phase 5 Level 3 training section to sanity_run.ipynb⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    c893cdf
  4. data: add level3.jsonl — 100 questions with adversarial pressure messages⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    83e9993
  5. test: add Level 3 adversarial pressure integration tests⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    e83d409
  6. fix: persist initial context (distractors/pressure) across all episode turns⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    725414c
  7. feat: extend environment with level=3, resistance reward ±0.2⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    27deeb6
  8. feat: add pressure_shown field to DeceitState

    Jayant-Kernel
    Unverified signature
    ed81b09
  9. fix: remove dead import, skip sleep when api unavailable

    Jayant-Kernel
    Unverified signature
    ea54c9d
  10. feat: generate_pressure.py — Level 3 adversarial pressure dataset script

    Jayant-Kernel
    Unverified signature
    7937a1a
  11. security: remove real API key from .env.example

    Jayant-Kernel
    Unverified signature
    15b2fa9
  12. Phase 4 complete: level2 dataset, distractor env, 90 tests passing

    Jayant-Kernel
    Unverified signature
    46011b7
  13. fix: exponential backoff on rate limit (10 retries, 30s*attempt)

    Jayant-Kernel
    Unverified signature
    8fd96d5
  14. fix: retry on rate limit in generate_distractors, 21s sleep for free-tier 3 RPM

    Jayant-Kernel
    Unverified signature
    26d82a2
  15. feat: append Phase 4 Level 2 training section to sanity_run.ipynb⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    f8a8477
  16. test: add Level 2 integration tests (test_level2.py)⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    3380d3c
  17. feat: extend reset() to support level=2 with distractor context⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    f2049f5
  18. feat: add 429 retry wrapper to grader semantic check⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    b44d7b0
  19. fix: propagate fatal API errors, strip markdown fences, cleanup⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    d5d723b
  20. fix: count skipped rows toward progress interval

    Jayant-Kernel
    Unverified signature
    ba97ba8
  21. fix: sleep after error path, print progress every 10 iterations

    Jayant-Kernel
    Unverified signature
    b7f42c1
  22. feat: generate_distractors.py — GPT-4o-mini Level 2 dataset script

    Jayant-Kernel
    Unverified signature
    1c6af55
  23. Merge HF Space remote, keeping local version⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    c0e1de3
  24. update: complete README with results, API docs, reward curve

    Jayant-Kernel
    Unverified signature
    0235c7b
  25. Upload folder using huggingface_hub

    Verified signature
    d7298b2
  26. Fix notebook: HF Space URL, /step envelope, health check retry on cold start⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    db475da
  27. Upload folder using huggingface_hub

    Verified signature
    b0addf2
  28. Fix Dockerfile port: use 7860 for HF Spaces compatibility⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    97384d7
  29. Update README: Phase 3 complete, HF Space badge, quickstart, reward table⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    44808d9
  30. Upload folder using huggingface_hub

    Verified signature
    61af0e3
  31. Phase 3: Dockerfile, openenv manifest, client, deployment guide, GRPO training notebook⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    f89afce

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

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    9737348
  2. Phase 2 complete: Level 1 env runs locally, tests green, 100-question dataset⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    f577d1f
  3. Fix build backend: use setuptools.build_meta instead of legacy path⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    db07765
  4. Phase 1 complete: schemas, reward design, project scaffold⁠

    Jayant-Kernel +1: Claude Sonnet 4.6
    Unverified signature
    139d3d1