{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 84.0,
      "completions/max_terminated_length": 84.0,
      "completions/mean_length": 52.15178680419922,
      "completions/mean_terminated_length": 52.15178680419922,
      "completions/min_length": 30.0,
      "completions/min_terminated_length": 30.0,
      "entropy": 0.5559625327587128,
      "epoch": 0.25,
      "frac_reward_zero_std": 0.0714285746216774,
      "grad_norm": 0.3681248426437378,
      "learning_rate": 0.0,
      "loss": -0.0119,
      "num_tokens": 217107.0,
      "reward": 0.3636607229709625,
      "reward_std": 0.5160845518112183,
      "rewards/env_alfworld_reward/mean": 0.3636607229709625,
      "rewards/env_alfworld_reward/std": 0.516084611415863,
      "sampling/importance_sampling_ratio/max": 2.9794723987579346,
      "sampling/importance_sampling_ratio/mean": 0.9032093286514282,
      "sampling/importance_sampling_ratio/min": 0.0,
      "sampling/sampling_logp_difference/max": 1.4227659702301025,
      "sampling/sampling_logp_difference/mean": 0.031906239688396454,
      "step": 1,
      "step_time": 1695.3378897999974
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 87.0,
      "completions/max_terminated_length": 87.0,
      "completions/mean_length": 50.145835876464844,
      "completions/mean_terminated_length": 50.145835876464844,
      "completions/min_length": 31.0,
      "completions/min_terminated_length": 31.0,
      "entropy": 0.5740033835172653,
      "epoch": 0.5,
      "frac_reward_zero_std": 0.0714285746216774,
      "grad_norm": 0.44471269845962524,
      "learning_rate": 2.84304e-07,
      "loss": -0.019,
      "num_tokens": 458068.0,
      "reward": 0.2202083021402359,
      "reward_std": 0.4740038514137268,
      "rewards/env_alfworld_reward/mean": 0.2202083021402359,
      "rewards/env_alfworld_reward/std": 0.4740039110183716,
      "sampling/importance_sampling_ratio/max": 2.9810025691986084,
      "sampling/importance_sampling_ratio/mean": 1.0009801387786865,
      "sampling/importance_sampling_ratio/min": 0.0,
      "sampling/sampling_logp_difference/max": 4.155694007873535,
      "sampling/sampling_logp_difference/mean": 0.032881855964660645,
      "step": 2,
      "step_time": 1649.8384637760028
    },
    {
      "clip_ratio/high_max": 0.007568976026959717,
      "clip_ratio/high_mean": 0.006711035966873169,
      "clip_ratio/low_mean": 0.003108485194388777,
      "clip_ratio/low_min": 0.0022171867603901774,
      "clip_ratio/region_mean": 0.009819521103054285,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 104.0,
      "completions/max_terminated_length": 104.0,
      "completions/mean_length": 50.770835876464844,
      "completions/mean_terminated_length": 50.770835876464844,
      "completions/min_length": 25.0,
      "completions/min_terminated_length": 25.0,
      "entropy": 0.6086474657058716,
      "epoch": 0.75,
      "frac_reward_zero_std": 0.095238097012043,
      "grad_norm": 0.35934266448020935,
      "learning_rate": 5.68608e-07,
      "loss": -0.0351,
      "num_tokens": 693527.0,
      "reward": 0.33044642210006714,
      "reward_std": 0.5141857266426086,
      "rewards/env_alfworld_reward/mean": 0.33044642210006714,
      "rewards/env_alfworld_reward/std": 0.5141857266426086,
      "sampling/importance_sampling_ratio/max": 2.783240795135498,
      "sampling/importance_sampling_ratio/mean": 0.9537404775619507,
      "sampling/importance_sampling_ratio/min": 0.0,
      "sampling/sampling_logp_difference/max": 1.1146756410598755,
      "sampling/sampling_logp_difference/mean": 0.03519035875797272,
      "step": 3,
      "step_time": 1672.7712960310018
    },
    {
      "clip_ratio/high_max": 0.007020744145847857,
      "clip_ratio/high_mean": 0.006485109799541533,
      "clip_ratio/low_mean": 0.004376257362309843,
      "clip_ratio/low_min": 0.003922754956874996,
      "clip_ratio/region_mean": 0.010861367220059037,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 95.0,
      "completions/max_terminated_length": 95.0,
      "completions/mean_length": 51.60714340209961,
      "completions/mean_terminated_length": 51.60714340209961,
      "completions/min_length": 28.0,
      "completions/min_terminated_length": 28.0,
      "entropy": 0.5597293674945831,
      "epoch": 1.0,
      "frac_reward_zero_std": 0.0535714291036129,
      "grad_norm": 0.409810870885849,
      "learning_rate": 8.529119999999999e-07,
      "loss": 0.0229,
      "num_tokens": 942371.0,
      "reward": 0.03812500089406967,
      "reward_std": 0.29200857877731323,
      "rewards/env_alfworld_reward/mean": 0.03812500089406967,
      "rewards/env_alfworld_reward/std": 0.2920086085796356,
      "sampling/importance_sampling_ratio/max": 2.8310158252716064,
      "sampling/importance_sampling_ratio/mean": 0.9631035327911377,
      "sampling/importance_sampling_ratio/min": 0.0,
      "sampling/sampling_logp_difference/max": 1.3009870052337646,
      "sampling/sampling_logp_difference/mean": 0.03364055976271629,
      "step": 4,
      "step_time": 1871.100103906003
    },
    {
      "epoch": 1.0,
      "eval_clip_ratio/high_max": 0.0,
      "eval_clip_ratio/high_mean": 0.0,
      "eval_clip_ratio/low_mean": 0.0,
      "eval_clip_ratio/low_min": 0.0,
      "eval_clip_ratio/region_mean": 0.0,
      "eval_completions/clipped_ratio": 0.0,
      "eval_completions/max_length": 68.96,
      "eval_completions/max_terminated_length": 68.96,
      "eval_completions/mean_length": 51.315,
      "eval_completions/mean_terminated_length": 51.315,
      "eval_completions/min_length": 37.8,
      "eval_completions/min_terminated_length": 37.8,
      "eval_entropy": 0.6129815554618836,
      "eval_frac_reward_zero_std": 0.09,
      "eval_loss": -0.010877630673348904,
      "eval_num_tokens": 942371.0,
      "eval_reward": 0.19167500026524067,
      "eval_reward_std": 0.38457470636814833,
      "eval_rewards/env_alfworld_reward/mean": 0.19167500026524067,
      "eval_rewards/env_alfworld_reward/std": 0.38457471802830695,
      "eval_runtime": 2099.7568,
      "eval_samples_per_second": 0.095,
      "eval_sampling/importance_sampling_ratio/max": 1.5984654092788697,
      "eval_sampling/importance_sampling_ratio/mean": 0.8831864225864411,
      "eval_sampling/importance_sampling_ratio/min": 0.3859718208014965,
      "eval_sampling/sampling_logp_difference/max": 0.6316441667079925,
      "eval_sampling/sampling_logp_difference/mean": 0.03440669998526573,
      "eval_steps_per_second": 0.012,
      "step": 4
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 8,
  "num_input_tokens_seen": 942371,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 42,
  "trial_name": null,
  "trial_params": null
}