{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9893390191897654,
  "eval_steps": 500,
  "global_step": 58,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 395.3694381713867,
      "epoch": 0.017057569296375266,
      "grad_norm": 0.24042215943336487,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.048,
      "reward": 1.1350447088479996,
      "reward_std": 0.5315618552267551,
      "rewards/sum_reward": 1.1350447088479996,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 400.4135227203369,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.23158641159534454,
      "kl": 0.0001577436923980713,
      "learning_rate": 2.5e-06,
      "loss": 0.0631,
      "reward": 1.006835974752903,
      "reward_std": 0.5563086168840528,
      "rewards/sum_reward": 1.006835974752903,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 384.19622192382815,
      "epoch": 0.17057569296375266,
      "grad_norm": 161.1280059814453,
      "kl": 4.255555725097656,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.1142,
      "reward": 1.0516741514205932,
      "reward_std": 0.5486123532056808,
      "rewards/sum_reward": 1.0516741514205932,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 370.5531402587891,
      "epoch": 0.255863539445629,
      "grad_norm": 0.3267284035682678,
      "kl": 0.02659912109375,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0595,
      "reward": 1.1354911133646965,
      "reward_std": 0.5183316260576248,
      "rewards/sum_reward": 1.1354911133646965,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 355.5297027587891,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.5259718894958496,
      "kl": 0.028741455078125,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0472,
      "reward": 1.1906250610947609,
      "reward_std": 0.5139306262135506,
      "rewards/sum_reward": 1.1906250610947609,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 372.9276954650879,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.9172742366790771,
      "kl": 0.0341339111328125,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0427,
      "reward": 1.1500000417232514,
      "reward_std": 0.5094686262309551,
      "rewards/sum_reward": 1.1500000417232514,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 379.5109550476074,
      "epoch": 0.511727078891258,
      "grad_norm": 0.38280367851257324,
      "kl": 0.038134765625,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0477,
      "reward": 1.1505580887198448,
      "reward_std": 0.4839449178427458,
      "rewards/sum_reward": 1.1505580887198448,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 366.76407928466796,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.34553492069244385,
      "kl": 0.04346923828125,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 0.0465,
      "reward": 1.1466518342494965,
      "reward_std": 0.49975207149982454,
      "rewards/sum_reward": 1.1466518342494965,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 356.1212215423584,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.8228252530097961,
      "kl": 0.0513427734375,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0707,
      "reward": 1.1147321864962578,
      "reward_std": 0.5065454341471195,
      "rewards/sum_reward": 1.1147321864962578,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 384.7980072021484,
      "epoch": 0.767590618336887,
      "grad_norm": 1.0524908304214478,
      "kl": 0.070587158203125,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.116,
      "reward": 1.1046875521540642,
      "reward_std": 0.5838725090026855,
      "rewards/sum_reward": 1.1046875521540642,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 399.55358734130857,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.8683111667633057,
      "kl": 0.08956298828125,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.1583,
      "reward": 1.021428617835045,
      "reward_std": 0.6321207888424396,
      "rewards/sum_reward": 1.021428617835045,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 401.3645248413086,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.8170827627182007,
      "kl": 0.095947265625,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.1683,
      "reward": 1.0492187902331351,
      "reward_std": 0.6428232848644256,
      "rewards/sum_reward": 1.0492187902331351,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 401.34786860148114,
      "epoch": 0.9893390191897654,
      "kl": 0.09938557942708333,
      "reward": 1.052455407877763,
      "reward_std": 0.6334689930081367,
      "rewards/sum_reward": 1.052455407877763,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.08800288617354014,
      "train_runtime": 13838.5765,
      "train_samples_per_second": 0.542,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 5,
  "max_steps": 58,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}