{
 "agi_status": "NOT ACHIEVED",
 "arm_accuracy": {
  "compute_matched": 0.6536,
  "frozen_e123": 0.0893,
  "full": 0.9786,
  "no_abstraction": 0.4321,
  "no_causal_memory": 0.3607,
  "no_learning": 0.1429,
  "random_policy": 0.1143,
  "retrieval_only": 0.2143
 },
 "base_model_arm": {
  "accuracy": 0.0,
  "elapsed_s": 26.21,
  "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
  "n_items": 20,
  "status": "RUN"
 },
 "elapsed_s": 43.4,
 "full_accuracy_wilson95": [
  0.954,
  0.9901
 ],
 "g05_passed": true,
 "g06_passed": true,
 "hypothesis_evaluations": {
  "compute_matched": 11686,
  "full": 17481,
  "no_causal_memory": 1916
 },
 "n_sealed_instances": 280,
 "n_task_families": 56,
 "paradigm_accuracy_full": {
  "changing_environment": 1.0,
  "conflicting_observations": 0.8571,
  "delayed_effect": 1.0,
  "hidden_precondition": 1.0,
  "partial_observability": 1.0,
  "resource_constraint": 0.9714,
  "unseen_combination": 1.0,
  "unseen_rule": 1.0
 },
 "schema": "cain.e124.capability_acquisition.v2",
 "stats": {
  "full_vs_compute_matched": {
   "a_correct": 274,
   "a_only_wins": 92,
   "b_correct": 183,
   "b_only_losses": 1,
   "gain": 0.325,
   "n": 280,
   "p_value": 0.0
  },
  "full_vs_frozen_e123": {
   "a_correct": 274,
   "a_only_wins": 249,
   "b_correct": 25,
   "b_only_losses": 0,
   "gain": 0.8893,
   "n": 280,
   "p_value": 0.0
  },
  "full_vs_no_abstraction": {
   "a_correct": 274,
   "a_only_wins": 153,
   "b_correct": 121,
   "b_only_losses": 0,
   "gain": 0.5464,
   "n": 280,
   "p_value": 0.0
  },
  "holm": {
   "vs_compute_matched": 0.0,
   "vs_frozen_e123": 0.0,
   "vs_no_abstraction": 0.0
  }
 },
 "verified_families": 56
}