| {"task_id": "e1_launch_readiness", "policy_kind": "model", "steps": 3, "total_reward": 1.39, "terminal_reward": 0.91, "reward_trace": [0.29000000000000004, 0.19, 0.9099999999999999], "verifier_pass_rate": 1.0, "passed_checks": ["qa_report_present", "final_rec_valid", "no_missed_milestones"], "failed_checks": [], "milestones_total": 2, "milestones_complete": 2, "milestones_missed": 0, "invalid_action_count": 0, "env_error_count": 0, "errors": [], "final_swd_version": 4, "success": true, "episode_index": 0, "model_stage": "base", "policy": "hf", "model": "Qwen/Qwen2.5-7B-Instruct", "adapter": ""} |
| {"task_id": "m1_budget_reallocation", "policy_kind": "model", "steps": 5, "total_reward": 1.766667, "terminal_reward": 0.806667, "reward_trace": [0.09000000000000001, 0.29000000000000004, 0.19, 0.39, 0.8066666666666666], "verifier_pass_rate": 0.666667, "passed_checks": ["required_agents_consulted", "conflict_logged", "conflict_resolved", "phased_plan"], "failed_checks": ["budget_constraint_acknowledged", "reasoning_documented"], "milestones_total": 3, "milestones_complete": 3, "milestones_missed": 0, "invalid_action_count": 0, "env_error_count": 0, "errors": [], "final_swd_version": 6, "success": false, "episode_index": 0, "model_stage": "base", "policy": "hf", "model": "Qwen/Qwen2.5-7B-Instruct", "adapter": ""} |
| {"task_id": "h1_acquisition_defence", "policy_kind": "model", "steps": 9, "total_reward": 2.6025, "terminal_reward": 0.7825, "reward_trace": [0.09000000000000001, 0.09000000000000001, 0.29000000000000004, 0.19, 0.39, 0.29000000000000004, 0.19, 0.29000000000000004, 0.7825], "verifier_pass_rate": 0.75, "passed_checks": ["all_agents_consulted", "conflict_explicitly_resolved", "resolution_has_type", "counter_offer_present", "deadline_present", "retention_addressed", "no_single_agent_copied", "all_phases_reached", "swd_version_rich"], "failed_checks": ["multi_conflict_logged", "rich_reasoning_log", "timeline_constraint_acknowledged"], "milestones_total": 5, "milestones_complete": 3, "milestones_missed": 0, "invalid_action_count": 0, "env_error_count": 0, "errors": [], "final_swd_version": 10, "success": false, "episode_index": 0, "model_stage": "base", "policy": "hf", "model": "Qwen/Qwen2.5-7B-Instruct", "adapter": ""} |
|
|
