| { |
| "version": "0.4.0", |
| "source_commit": "49359eae9aa197d5b92631c16efb74085180a27c", |
| "source_archive_sha256": "7586bc75e2f797b27429f6cf8e0c44bdfd2589c2437397bc5717ab517fff5a59", |
| "exact_frozen_checkpoint_archive_sha256": "e2719b814e762936047afa79c9336c754a1c7e400903ac1ab7b090addc486448", |
| "official_harness_original_checkpoint_predictions": 2000, |
| "official_harness_original_checkpoint_full_replies_exact": 2000, |
| "extracted_release_official_predictions_exact": 2000, |
| "extracted_release_full_replies_exact": 2000, |
| "portable_source_checksums_verified": 1042, |
| "new_core_unit_tests_passed": 15, |
| "authored_confirmation": { |
| "positive_total": 60, |
| "candidate_correct": 60, |
| "baseline_correct": 12, |
| "focused_boundaries_candidate": 16, |
| "known_family_maintainer_ai_assisted": true |
| }, |
| "inherited_regression": { |
| "passed": 671, |
| "remaining_old_failures": 4, |
| "old_failures_fixed": 25 |
| }, |
| "reproductions_are_not_additional_unique_capability_cases": true, |
| "resource_bundle_version": "0.3.7", |
| "weights_protected_solvers_and_CA_kernels_unchanged": true, |
| "metadata_only_changes": [ |
| "cael_unified/__init__.py", |
| "cael_unified/self_knowledge.json", |
| "bundle_config.json" |
| ], |
| "official_metrics": { |
| "baseline": { |
| "name": "cael_040", |
| "kwargs": { |
| "fallback": "majority", |
| "default_symbol": 0 |
| } |
| }, |
| "benchmark": "cellarc_100k", |
| "splits": [ |
| { |
| "split": "test_interpolation", |
| "episodes": 1000, |
| "solved": 79, |
| "episode_accuracy": 0.079, |
| "token_total": 11099, |
| "token_correct": 6361, |
| "token_accuracy": 0.5731146950175692 |
| }, |
| { |
| "split": "test_extrapolation", |
| "episodes": 1000, |
| "solved": 5, |
| "episode_accuracy": 0.005, |
| "token_total": 10918, |
| "token_correct": 3815, |
| "token_accuracy": 0.34942297124015387 |
| } |
| ], |
| "overall": { |
| "episodes": 2000, |
| "solved": 84, |
| "episode_accuracy": 0.042, |
| "token_total": 22017, |
| "token_correct": 10176, |
| "token_accuracy": 0.4621883090339283 |
| }, |
| "max_episodes": null |
| }, |
| "official_rank_claim": false, |
| "general_chat_or_Qwen_superiority_claim": false, |
| "live_worker_or_tunnel_changed": false |
| } |
|
|
