{
  "fetched_at": "2026-09-09T07:12:31.903918+00:00",
  "method": "Run.scan_history() without key filtering or sampling",
  "sdk_version": "0.30.0",
  "runs": [
    {
      "id": "8ov14yyc",
      "url": "https://wandb.ai/hzj-lab/shopsimrl/runs/8ov14yyc",
      "round": 0,
      "primary": true,
      "state": "finished",
      "created_at": "2026-09-06T10:26:00Z",
      "history_rows": 100,
      "keys": [
        "_runtime",
        "_step",
        "_timestamp",
        "perf/actor_train_tflops",
        "perf/actor_train_time",
        "perf/actor_train_tok_per_s",
        "perf/data_preprocess_time",
        "perf/effective_tokens_per_gpu_per_sec",
        "perf/log_probs_tflops",
        "perf/log_probs_time",
        "perf/longest_effective_sample_tokens_per_sec",
        "perf/longest_sample_tokens_per_sec",
        "perf/rollout_time",
        "perf/sleep_time",
        "perf/step_time",
        "perf/tokens_per_gpu_per_sec",
        "perf/train_time",
        "perf/train_wait_time",
        "perf/update_weights_time",
        "perf/wait_time_ratio",
        "perf/wake_up_time",
        "rollout/advantages",
        "rollout/avg_cached_tokens_per_sample",
        "rollout/dynamic_filter/drop_shopsim_unscored_technical_group",
        "rollout/dynamic_filter/drop_shopsim_zero_std_group",
        "rollout/kl",
        "rollout/log_probs",
        "rollout/prefix_cache_hit_rate",
        "rollout/raw_reward",
        "rollout/repetition_frac",
        "rollout/response_len/max",
        "rollout/response_len/mean",
        "rollout/response_len/median",
        "rollout/response_len/min",
        "rollout/response_lengths",
        "rollout/returns",
        "rollout/rewards",
        "rollout/rollout_log_probs",
        "rollout/shopsim/all_wrong_group_rate",
        "rollout/shopsim/assisted_empty_group_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/inclusion_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/target_rho",
        "rollout/shopsim/chunk/chunk-27a7e5bb9344/inclusion_rate",
        "rollout/shopsim/chunk/chunk-27a7e5bb9344/target_rho",
        "rollout/shopsim/chunk/chunk-4d958fc1d69e/inclusion_rate",
        "rollout/shopsim/chunk/chunk-4d958fc1d69e/target_rho",
        "rollout/shopsim/chunk/chunk-a71a5634ee99/inclusion_rate",
        "rollout/shopsim/chunk/chunk-a71a5634ee99/target_rho",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/inclusion_rate",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/target_rho",
        "rollout/shopsim/chunk/chunk-b64b0567ce98/inclusion_rate",
        "rollout/shopsim/chunk/chunk-b64b0567ce98/target_rho",
        "rollout/shopsim/chunk/chunk-bbe6cb9408eb/inclusion_rate",
        "rollout/shopsim/chunk/chunk-bbe6cb9408eb/target_rho",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/inclusion_rate",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/target_rho",
        "rollout/shopsim/chunk/chunk-d3a9ac5af7ef/inclusion_rate",
        "rollout/shopsim/chunk/chunk-d3a9ac5af7ef/target_rho",
        "rollout/shopsim/chunk/chunk-f58e352cdad0/inclusion_rate",
        "rollout/shopsim/chunk/chunk-f58e352cdad0/target_rho",
        "rollout/shopsim/full_skill_retry_rate",
        "rollout/shopsim/full_skill_retry_success_rate",
        "rollout/shopsim/group_count",
        "rollout/shopsim/internalization_deficit_rate",
        "rollout/shopsim/pending_analysis_rate",
        "rollout/shopsim/q",
        "rollout/shopsim/reward_mean",
        "rollout/shopsim/rho_max",
        "rollout/shopsim/rho_mean",
        "rollout/shopsim/rho_min",
        "rollout/shopsim/runtime_error_group_rate",
        "rollout/shopsim/scored_rate",
        "rollout/shopsim/skill_free_group_rate",
        "rollout/shopsim/skills_per_assisted_group_mean",
        "rollout/shopsim/strict_reward_mean",
        "rollout/shopsim/success_rate",
        "rollout/shopsim/technical_group_drop_count",
        "rollout/shopsim/technical_group_drop_rate",
        "rollout/shopsim/train_group_count",
        "rollout/shopsim/train_trajectory_count",
        "rollout/shopsim/trajectory_count",
        "rollout/shopsim/zero_std_group_drop_count",
        "rollout/shopsim/zero_std_group_drop_rate",
        "rollout/step",
        "rollout/total_lengths",
        "rollout/truncated",
        "rollout/truncated_ratio",
        "rollout/zero_std/count_0.0",
        "rollout/zero_std/count_0.5",
        "rollout/zero_std/count_0.7",
        "rollout/zero_std/count_1.0",
        "train/entropy_loss",
        "train/global_batch_size",
        "train/grad_norm",
        "train/loss",
        "train/lr-pg_0",
        "train/lr-pg_1",
        "train/pg_clipfrac",
        "train/pg_loss",
        "train/ppo_kl",
        "train/step",
        "train/train_rollout_logprob_abs_diff"
      ],
      "sha256": "f38c31eac9be4fa4286680dcc96241d3f3936233351301f45da25420885eead6"
    },
    {
      "id": "chjizc8u",
      "url": "https://wandb.ai/hzj-lab/shopsimrl/runs/chjizc8u",
      "round": 1,
      "primary": true,
      "state": "finished",
      "created_at": "2026-09-06T15:04:16Z",
      "history_rows": 100,
      "keys": [
        "_runtime",
        "_step",
        "_timestamp",
        "perf/actor_train_tflops",
        "perf/actor_train_time",
        "perf/actor_train_tok_per_s",
        "perf/data_preprocess_time",
        "perf/effective_tokens_per_gpu_per_sec",
        "perf/log_probs_tflops",
        "perf/log_probs_time",
        "perf/longest_effective_sample_tokens_per_sec",
        "perf/longest_sample_tokens_per_sec",
        "perf/rollout_time",
        "perf/sleep_time",
        "perf/step_time",
        "perf/tokens_per_gpu_per_sec",
        "perf/train_time",
        "perf/train_wait_time",
        "perf/update_weights_time",
        "perf/wait_time_ratio",
        "perf/wake_up_time",
        "rollout/advantages",
        "rollout/avg_cached_tokens_per_sample",
        "rollout/dynamic_filter/drop_shopsim_zero_std_group",
        "rollout/kl",
        "rollout/log_probs",
        "rollout/prefix_cache_hit_rate",
        "rollout/raw_reward",
        "rollout/repetition_frac",
        "rollout/response_len/max",
        "rollout/response_len/mean",
        "rollout/response_len/median",
        "rollout/response_len/min",
        "rollout/response_lengths",
        "rollout/returns",
        "rollout/rewards",
        "rollout/rollout_log_probs",
        "rollout/shopsim/all_wrong_group_rate",
        "rollout/shopsim/assisted_empty_group_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/inclusion_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/target_rho",
        "rollout/shopsim/chunk/chunk-27a7e5bb9344/inclusion_rate",
        "rollout/shopsim/chunk/chunk-27a7e5bb9344/target_rho",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/inclusion_rate",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/target_rho",
        "rollout/shopsim/chunk/chunk-bbe6cb9408eb/inclusion_rate",
        "rollout/shopsim/chunk/chunk-bbe6cb9408eb/target_rho",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/inclusion_rate",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/target_rho",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/inclusion_rate",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/target_rho",
        "rollout/shopsim/chunk/online-bbef274e1764d953/inclusion_rate",
        "rollout/shopsim/chunk/online-bbef274e1764d953/target_rho",
        "rollout/shopsim/full_skill_retry_rate",
        "rollout/shopsim/full_skill_retry_success_rate",
        "rollout/shopsim/group_count",
        "rollout/shopsim/internalization_deficit_rate",
        "rollout/shopsim/pending_analysis_rate",
        "rollout/shopsim/q",
        "rollout/shopsim/reward_mean",
        "rollout/shopsim/rho_max",
        "rollout/shopsim/rho_mean",
        "rollout/shopsim/rho_min",
        "rollout/shopsim/runtime_error_group_rate",
        "rollout/shopsim/scored_rate",
        "rollout/shopsim/skill_free_group_rate",
        "rollout/shopsim/skills_per_assisted_group_mean",
        "rollout/shopsim/strict_reward_mean",
        "rollout/shopsim/success_rate",
        "rollout/shopsim/train_group_count",
        "rollout/shopsim/train_trajectory_count",
        "rollout/shopsim/trajectory_count",
        "rollout/shopsim/zero_std_group_drop_count",
        "rollout/shopsim/zero_std_group_drop_rate",
        "rollout/step",
        "rollout/total_lengths",
        "rollout/truncated",
        "rollout/truncated_ratio",
        "rollout/zero_std/count_0.0",
        "rollout/zero_std/count_0.4",
        "rollout/zero_std/count_0.5",
        "rollout/zero_std/count_1.0",
        "train/entropy_loss",
        "train/global_batch_size",
        "train/grad_norm",
        "train/loss",
        "train/lr-pg_0",
        "train/lr-pg_1",
        "train/pg_clipfrac",
        "train/pg_loss",
        "train/ppo_kl",
        "train/step",
        "train/train_rollout_logprob_abs_diff"
      ],
      "sha256": "7052dc3a60e8d4925f737c879b8bd6deb0f6ad18f2a4aec1cd3760cd2adc2e86"
    },
    {
      "id": "7vhbv5j0",
      "url": "https://wandb.ai/hzj-lab/shopsimrl/runs/7vhbv5j0",
      "round": 2,
      "primary": true,
      "state": "finished",
      "created_at": "2026-09-07T01:49:09Z",
      "history_rows": 100,
      "keys": [
        "_runtime",
        "_step",
        "_timestamp",
        "perf/actor_train_tflops",
        "perf/actor_train_time",
        "perf/actor_train_tok_per_s",
        "perf/data_preprocess_time",
        "perf/effective_tokens_per_gpu_per_sec",
        "perf/log_probs_tflops",
        "perf/log_probs_time",
        "perf/longest_effective_sample_tokens_per_sec",
        "perf/longest_sample_tokens_per_sec",
        "perf/rollout_time",
        "perf/sleep_time",
        "perf/step_time",
        "perf/tokens_per_gpu_per_sec",
        "perf/train_time",
        "perf/train_wait_time",
        "perf/update_weights_time",
        "perf/wait_time_ratio",
        "perf/wake_up_time",
        "rollout/advantages",
        "rollout/avg_cached_tokens_per_sample",
        "rollout/dynamic_filter/drop_shopsim_zero_std_group",
        "rollout/kl",
        "rollout/log_probs",
        "rollout/prefix_cache_hit_rate",
        "rollout/raw_reward",
        "rollout/repetition_frac",
        "rollout/response_len/max",
        "rollout/response_len/mean",
        "rollout/response_len/median",
        "rollout/response_len/min",
        "rollout/response_lengths",
        "rollout/returns",
        "rollout/rewards",
        "rollout/rollout_log_probs",
        "rollout/shopsim/all_wrong_group_rate",
        "rollout/shopsim/assisted_empty_group_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/inclusion_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/target_rho",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/inclusion_rate",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/target_rho",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/inclusion_rate",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/target_rho",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/inclusion_rate",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/target_rho",
        "rollout/shopsim/chunk/online-bbef274e1764d953/inclusion_rate",
        "rollout/shopsim/chunk/online-bbef274e1764d953/target_rho",
        "rollout/shopsim/full_skill_retry_rate",
        "rollout/shopsim/full_skill_retry_success_rate",
        "rollout/shopsim/group_count",
        "rollout/shopsim/internalization_deficit_rate",
        "rollout/shopsim/pending_analysis_rate",
        "rollout/shopsim/q",
        "rollout/shopsim/reward_mean",
        "rollout/shopsim/rho_max",
        "rollout/shopsim/rho_mean",
        "rollout/shopsim/rho_min",
        "rollout/shopsim/runtime_error_group_rate",
        "rollout/shopsim/scored_rate",
        "rollout/shopsim/skill_free_group_rate",
        "rollout/shopsim/skills_per_assisted_group_mean",
        "rollout/shopsim/strict_reward_mean",
        "rollout/shopsim/success_rate",
        "rollout/shopsim/train_group_count",
        "rollout/shopsim/train_trajectory_count",
        "rollout/shopsim/trajectory_count",
        "rollout/shopsim/zero_std_group_drop_count",
        "rollout/shopsim/zero_std_group_drop_rate",
        "rollout/step",
        "rollout/total_lengths",
        "rollout/truncated",
        "rollout/truncated_ratio",
        "rollout/zero_std/count_0.0",
        "rollout/zero_std/count_0.5",
        "rollout/zero_std/count_1.0",
        "train/entropy_loss",
        "train/global_batch_size",
        "train/grad_norm",
        "train/loss",
        "train/lr-pg_0",
        "train/lr-pg_1",
        "train/pg_clipfrac",
        "train/pg_loss",
        "train/ppo_kl",
        "train/step",
        "train/train_rollout_logprob_abs_diff"
      ],
      "sha256": "4012c89e093afc5ecce65a340946a4ff48574a3de71714ac1ddad0ccd300892c"
    },
    {
      "id": "y4xk6byf",
      "url": "https://wandb.ai/hzj-lab/shopsimrl/runs/y4xk6byf",
      "round": 3,
      "primary": true,
      "state": "finished",
      "created_at": "2026-09-07T07:15:56Z",
      "history_rows": 100,
      "keys": [
        "_runtime",
        "_step",
        "_timestamp",
        "perf/actor_train_tflops",
        "perf/actor_train_time",
        "perf/actor_train_tok_per_s",
        "perf/data_preprocess_time",
        "perf/effective_tokens_per_gpu_per_sec",
        "perf/log_probs_tflops",
        "perf/log_probs_time",
        "perf/longest_effective_sample_tokens_per_sec",
        "perf/longest_sample_tokens_per_sec",
        "perf/rollout_time",
        "perf/sleep_time",
        "perf/step_time",
        "perf/tokens_per_gpu_per_sec",
        "perf/train_time",
        "perf/train_wait_time",
        "perf/update_weights_time",
        "perf/wait_time_ratio",
        "perf/wake_up_time",
        "rollout/advantages",
        "rollout/avg_cached_tokens_per_sample",
        "rollout/dynamic_filter/drop_shopsim_zero_std_group",
        "rollout/kl",
        "rollout/log_probs",
        "rollout/prefix_cache_hit_rate",
        "rollout/raw_reward",
        "rollout/repetition_frac",
        "rollout/response_len/max",
        "rollout/response_len/mean",
        "rollout/response_len/median",
        "rollout/response_len/min",
        "rollout/response_lengths",
        "rollout/returns",
        "rollout/rewards",
        "rollout/rollout_log_probs",
        "rollout/shopsim/all_wrong_group_rate",
        "rollout/shopsim/assisted_empty_group_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/inclusion_rate",
        "rollout/shopsim/chunk/chunk-1f8d27322e9f/target_rho",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/inclusion_rate",
        "rollout/shopsim/chunk/chunk-a9e98e128a35/target_rho",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/inclusion_rate",
        "rollout/shopsim/chunk/chunk-c83ed5ae8b70/target_rho",
        "rollout/shopsim/chunk/online-214dfd1028dc4157/inclusion_rate",
        "rollout/shopsim/chunk/online-214dfd1028dc4157/target_rho",
        "rollout/shopsim/chunk/online-3297dc4c0f4f6109/inclusion_rate",
        "rollout/shopsim/chunk/online-3297dc4c0f4f6109/target_rho",
        "rollout/shopsim/chunk/online-3d2cdd6a2aac9717/inclusion_rate",
        "rollout/shopsim/chunk/online-3d2cdd6a2aac9717/target_rho",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/inclusion_rate",
        "rollout/shopsim/chunk/online-8e3e179b9270814a/target_rho",
        "rollout/shopsim/chunk/online-ea5dd903b14ec0fc/inclusion_rate",
        "rollout/shopsim/chunk/online-ea5dd903b14ec0fc/target_rho",
        "rollout/shopsim/full_skill_retry_rate",
        "rollout/shopsim/full_skill_retry_success_rate",
        "rollout/shopsim/group_count",
        "rollout/shopsim/internalization_deficit_rate",
        "rollout/shopsim/pending_analysis_rate",
        "rollout/shopsim/q",
        "rollout/shopsim/reward_mean",
        "rollout/shopsim/rho_max",
        "rollout/shopsim/rho_mean",
        "rollout/shopsim/rho_min",
        "rollout/shopsim/runtime_error_group_rate",
        "rollout/shopsim/scored_rate",
        "rollout/shopsim/skill_free_group_rate",
        "rollout/shopsim/skills_per_assisted_group_mean",
        "rollout/shopsim/strict_reward_mean",
        "rollout/shopsim/success_rate",
        "rollout/shopsim/train_group_count",
        "rollout/shopsim/train_trajectory_count",
        "rollout/shopsim/trajectory_count",
        "rollout/shopsim/zero_std_group_drop_count",
        "rollout/shopsim/zero_std_group_drop_rate",
        "rollout/step",
        "rollout/total_lengths",
        "rollout/truncated",
        "rollout/truncated_ratio",
        "rollout/zero_std/count_0.0",
        "rollout/zero_std/count_0.5",
        "rollout/zero_std/count_1.0",
        "train/entropy_loss",
        "train/global_batch_size",
        "train/grad_norm",
        "train/loss",
        "train/lr-pg_0",
        "train/lr-pg_1",
        "train/pg_clipfrac",
        "train/pg_loss",
        "train/ppo_kl",
        "train/step",
        "train/train_rollout_logprob_abs_diff"
      ],
      "sha256": "f48ecf5e2ba5fc0a77f2b845c60c8a8d9178237c21ff3dd742cbc18762309a56"
    }
  ]
}