{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3589375448671931,
  "eval_steps": 50,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 138.55,
      "epoch": 0.03589375448671931,
      "grad_norm": 0.0010730335488915443,
      "kl": 0.0005896038687205874,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.0,
      "reward": 0.99,
      "reward_std": 0.014142135381698609,
      "rewards/reward_fn": 0.99,
      "step": 25
    },
    {
      "completion_length": 131.66,
      "epoch": 0.07178750897343862,
      "grad_norm": 0.0010653886711224914,
      "kl": 0.0006372866354649886,
      "learning_rate": 2.4500000000000003e-06,
      "loss": 0.0,
      "reward": 0.965,
      "reward_std": 0.04949747383594513,
      "rewards/reward_fn": 0.965,
      "step": 50
    },
    {
      "epoch": 0.07178750897343862,
      "eval_completion_length": 137.90506329113924,
      "eval_kl": 0.0006547578797628514,
      "eval_loss": 2.6065379643114284e-05,
      "eval_reward": 0.995253164556962,
      "eval_reward_std": 0.0067130389470088335,
      "eval_rewards/reward_fn": 0.995253164556962,
      "eval_runtime": 1868.146,
      "eval_samples_per_second": 0.168,
      "eval_steps_per_second": 0.021,
      "step": 50
    },
    {
      "completion_length": 135.84,
      "epoch": 0.10768126346015794,
      "grad_norm": 0.0006992756389081478,
      "kl": 0.00064181099034613,
      "learning_rate": 3.7e-06,
      "loss": 0.0,
      "reward": 0.98,
      "reward_std": 0.028284270763397217,
      "rewards/reward_fn": 0.98,
      "step": 75
    },
    {
      "completion_length": 132.96,
      "epoch": 0.14357501794687724,
      "grad_norm": 0.0013170982711017132,
      "kl": 0.0006926136535184924,
      "learning_rate": 4.95e-06,
      "loss": 0.0,
      "reward": 0.975,
      "reward_std": 0.03535533845424652,
      "rewards/reward_fn": 0.975,
      "step": 100
    },
    {
      "epoch": 0.14357501794687724,
      "eval_completion_length": 134.5253164556962,
      "eval_kl": 0.0006582768799006185,
      "eval_loss": 2.6447118216310628e-05,
      "eval_reward": 0.9746835443037974,
      "eval_reward_std": 0.03580287438404711,
      "eval_rewards/reward_fn": 0.9746835443037974,
      "eval_runtime": 1865.6091,
      "eval_samples_per_second": 0.168,
      "eval_steps_per_second": 0.021,
      "step": 100
    },
    {
      "completion_length": 134.35,
      "epoch": 0.17946877243359655,
      "grad_norm": 0.000720625976100564,
      "kl": 0.0006820108561078086,
      "learning_rate": 4.907120743034056e-06,
      "loss": 0.0,
      "reward": 0.99,
      "reward_std": 0.014142135381698609,
      "rewards/reward_fn": 0.99,
      "step": 125
    },
    {
      "completion_length": 132.83,
      "epoch": 0.21536252692031588,
      "grad_norm": 0.0009368288447149098,
      "kl": 0.0006625658516713884,
      "learning_rate": 4.8103715170278645e-06,
      "loss": 0.0,
      "reward": 0.98,
      "reward_std": 0.028284270763397217,
      "rewards/reward_fn": 0.98,
      "step": 150
    },
    {
      "epoch": 0.21536252692031588,
      "eval_completion_length": 136.07753164556962,
      "eval_kl": 0.0006587035147932983,
      "eval_loss": 2.6281333703082055e-05,
      "eval_reward": 0.9889240506329114,
      "eval_reward_std": 0.01566375754302061,
      "eval_rewards/reward_fn": 0.9889240506329114,
      "eval_runtime": 1842.521,
      "eval_samples_per_second": 0.17,
      "eval_steps_per_second": 0.022,
      "step": 150
    },
    {
      "completion_length": 137.885,
      "epoch": 0.25125628140703515,
      "grad_norm": 0.0010509429266676307,
      "kl": 0.0006368485454004257,
      "learning_rate": 4.713622291021672e-06,
      "loss": 0.0,
      "reward": 0.975,
      "reward_std": 0.03535533845424652,
      "rewards/reward_fn": 0.975,
      "step": 175
    },
    {
      "completion_length": 133.24,
      "epoch": 0.2871500358937545,
      "grad_norm": 0.0007394957356154919,
      "kl": 0.0006816260272171349,
      "learning_rate": 4.61687306501548e-06,
      "loss": 0.0,
      "reward": 0.995,
      "reward_std": 0.007071067690849304,
      "rewards/reward_fn": 0.995,
      "step": 200
    },
    {
      "epoch": 0.2871500358937545,
      "eval_completion_length": 141.43037974683546,
      "eval_kl": 0.0006899324428478751,
      "eval_loss": 2.767087789834477e-05,
      "eval_reward": 0.9857594936708861,
      "eval_reward_std": 0.01566375754302061,
      "eval_rewards/reward_fn": 0.9857594936708861,
      "eval_runtime": 1829.9816,
      "eval_samples_per_second": 0.171,
      "eval_steps_per_second": 0.022,
      "step": 200
    },
    {
      "completion_length": 138.26,
      "epoch": 0.3230437903804738,
      "grad_norm": 0.0025397809222340584,
      "kl": 0.0007233391245245002,
      "learning_rate": 4.5201238390092885e-06,
      "loss": 0.0,
      "reward": 0.98,
      "reward_std": 0.028284270763397217,
      "rewards/reward_fn": 0.98,
      "step": 225
    },
    {
      "completion_length": 136.39,
      "epoch": 0.3589375448671931,
      "grad_norm": 0.001854633097536862,
      "kl": 0.0006911460669653025,
      "learning_rate": 4.423374613003097e-06,
      "loss": 0.0,
      "reward": 0.995,
      "reward_std": 0.007071067690849304,
      "rewards/reward_fn": 0.995,
      "step": 250
    },
    {
      "epoch": 0.3589375448671931,
      "eval_completion_length": 142.51740506329114,
      "eval_kl": 0.000728936185117198,
      "eval_loss": 2.9135468139429577e-05,
      "eval_reward": 0.9873417721518988,
      "eval_reward_std": 0.017901437192023556,
      "eval_rewards/reward_fn": 0.9873417721518988,
      "eval_runtime": 1854.9938,
      "eval_samples_per_second": 0.169,
      "eval_steps_per_second": 0.022,
      "step": 250
    }
  ],
  "logging_steps": 25,
  "max_steps": 1392,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}