{
  "bundle_doctor": {
    "artifact_hashes": {
      "files": 14,
      "problems": [],
      "status": "ok"
    },
    "config_profile": {
      "model_handoff_problems": [],
      "profile": "single-gpu-online-distillation-v1",
      "roots": [
        "actor_rollout_ref",
        "custom_reward_function",
        "data",
        "trainer"
      ],
      "status": "ok"
    },
    "distributed_execution_status": "not tested",
    "installed_verl": {
      "direct_url": {
        "url": "https://github.com/verl-project/verl.git",
        "vcs_info": {
          "commit_id": "7aed6b230776f963fa09509c10d9c3a767d1102c",
          "requested_revision": "7aed6b230776f963fa09509c10d9c3a767d1102c",
          "vcs": "git"
        }
      },
      "expected_commit": "7aed6b230776f963fa09509c10d9c3a767d1102c",
      "status": "ok",
      "version": "0.8.0.dev0"
    },
    "local_smoke_status": "passed",
    "model_adapter_loadability": {
      "base_model": "hf-internal-testing/tiny-random-gpt2",
      "detail": "1 tensor header(s)",
      "load_scope": "PEFT config plus safetensors structure; base weights not loaded",
      "peft_config_load": "loaded LoraConfig",
      "peft_type": "LORA",
      "status": "ok"
    },
    "parquet_schema": {
      "schemas": {
        "train": [
          "data_source",
          "prompt",
          "ability",
          "reward_model",
          "extra_info"
        ],
        "val": [
          "data_source",
          "prompt",
          "ability",
          "reward_model",
          "extra_info"
        ]
      },
      "status": "ok"
    },
    "privacy": {
      "problems": [],
      "status": "ok"
    },
    "reward_scaffold_importability": {
      "detail": "side-effect-free import; scaffold intentionally not executed",
      "status": "ok"
    },
    "target_verl": {
      "commit": "7aed6b230776f963fa09509c10d9c3a767d1102c",
      "mismatches": {},
      "profile": "single-gpu-online-distillation-v1",
      "status": "ok",
      "tag": "v0.8.0"
    },
    "tokenizer_identity": {
      "files": [
        "tokenizer_config.json"
      ],
      "status": "ok",
      "structural_digest": "5a652912e64f5edd8c1b7c2a6f0d2e4d77d6522c546bf66b6bc38e906d05be4b"
    },
    "unsupported_semantics": [
      "optimizer state",
      "distributed RNG state",
      "FSDP native checkpoint",
      "Megatron native checkpoint",
      "Ray runtime state",
      "teacher cache as PPO reference cache",
      "PPO advantage or clipping semantics",
      "GRPO group semantics"
    ],
    "verdict": "ok"
  },
  "distributed_execution_status": "not tested",
  "model_or_adapter_load": "loaded LoraConfig",
  "official_config": {
    "exported_override_roots": [
      "actor_rollout_ref",
      "custom_reward_function",
      "data",
      "trainer"
    ],
    "missing_export_fields": [],
    "missing_fields": [],
    "parse_status": "passed with OmegaConf",
    "path": "verl/trainer/config/_generated_ppo_trainer.yaml",
    "required_export_fields": [
      "actor_rollout_ref.model.lora_rank",
      "actor_rollout_ref.model.lora_alpha",
      "actor_rollout_ref.model.target_modules",
      "actor_rollout_ref.model.lora_adapter_path",
      "custom_reward_function.path",
      "custom_reward_function.name"
    ],
    "required_fields": [
      "data.train_files",
      "data.val_files",
      "data.prompt_key",
      "data.max_prompt_length",
      "data.max_response_length",
      "data.seed",
      "actor_rollout_ref.model.path",
      "actor_rollout_ref.model.enable_gradient_checkpointing",
      "actor_rollout_ref.actor.optim.lr",
      "trainer.save_freq",
      "trainer.test_freq",
      "trainer.project_name",
      "trainer.experiment_name",
      "trainer.total_epochs"
    ],
    "structured_merge_status": "passed"
  },
  "parquet_load": "train and val passed",
  "privacy_status": "ok",
  "profile": "single-gpu-online-distillation-v1",
  "reward_scaffold_import": "passed; fail-closed scorer not executed",
  "schema_version": 1,
  "target_verl": {
    "commit": "7aed6b230776f963fa09509c10d9c3a767d1102c",
    "observed_package_version": "0.8.0.dev0",
    "observed_vcs_commit": "7aed6b230776f963fa09509c10d9c3a767d1102c",
    "tag": "v0.8.0"
  },
  "tiny_cpu_dry_run": {
    "reason": "full PPO execution requires the excluded distributed inference stack",
    "status": "artifact-only"
  },
  "verdict": "ok"
}