{
  "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": 0.22084756195545197,
      "kl": 0.0,
      "learning_rate": 5e-07,
      "loss": 0.0679,
      "reward": 1.090959869325161,
      "reward_std": 0.5451452471315861,
      "rewards/sum_reward": 1.090959869325161,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 399.8663663864136,
      "epoch": 0.08528784648187633,
      "grad_norm": 0.22432541847229004,
      "kl": 0.00016561150550842285,
      "learning_rate": 2.5e-06,
      "loss": 0.0816,
      "reward": 0.9741908963769674,
      "reward_std": 0.5656327093020082,
      "rewards/sum_reward": 0.9741908963769674,
      "step": 5
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 374.9484527587891,
      "epoch": 0.17057569296375266,
      "grad_norm": 26829.416015625,
      "kl": 11.960232734680176,
      "learning_rate": 2.956412726139078e-06,
      "loss": 0.917,
      "reward": 1.0515625387430192,
      "reward_std": 0.5815188780426979,
      "rewards/sum_reward": 1.0515625387430192,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.54956970214846,
      "epoch": 0.255863539445629,
      "grad_norm": 3.2569870948791504,
      "kl": 0.46447906494140623,
      "learning_rate": 2.7836719084521715e-06,
      "loss": 0.0741,
      "reward": 1.1245536178350448,
      "reward_std": 0.5835043609142303,
      "rewards/sum_reward": 1.1245536178350448,
      "step": 15
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 325.2422065734863,
      "epoch": 0.3411513859275053,
      "grad_norm": 0.45262619853019714,
      "kl": 0.105426025390625,
      "learning_rate": 2.4946839873611927e-06,
      "loss": 0.0516,
      "reward": 1.141741119325161,
      "reward_std": 0.5650397419929505,
      "rewards/sum_reward": 1.141741119325161,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 344.40380973815917,
      "epoch": 0.42643923240938164,
      "grad_norm": 0.4256417751312256,
      "kl": 0.048638916015625,
      "learning_rate": 2.1156192081791355e-06,
      "loss": 0.0594,
      "reward": 1.1424107611179353,
      "reward_std": 0.5505612261593342,
      "rewards/sum_reward": 1.1424107611179353,
      "step": 25
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 350.17479248046874,
      "epoch": 0.511727078891258,
      "grad_norm": 0.41853660345077515,
      "kl": 0.0392486572265625,
      "learning_rate": 1.6808050203829845e-06,
      "loss": 0.0428,
      "reward": 1.1254464760422707,
      "reward_std": 0.5392456278204918,
      "rewards/sum_reward": 1.1254464760422707,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 345.260062789917,
      "epoch": 0.5970149253731343,
      "grad_norm": 0.4747031629085541,
      "kl": 23.548565673828126,
      "learning_rate": 1.2296174432791415e-06,
      "loss": 1.1464,
      "reward": 1.119977732002735,
      "reward_std": 0.5517967157065868,
      "rewards/sum_reward": 1.119977732002735,
      "step": 35
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 330.96072883605956,
      "epoch": 0.6823027718550106,
      "grad_norm": 0.37849438190460205,
      "kl": 0.043377685546875,
      "learning_rate": 8.029152419343472e-07,
      "loss": 0.0458,
      "reward": 1.2135045230388641,
      "reward_std": 0.4967778742313385,
      "rewards/sum_reward": 1.2135045230388641,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 339.3346115112305,
      "epoch": 0.767590618336887,
      "grad_norm": 0.26795679330825806,
      "kl": 0.036431884765625,
      "learning_rate": 4.3933982822017883e-07,
      "loss": 0.0405,
      "reward": 1.2203125447034835,
      "reward_std": 0.5353297784924507,
      "rewards/sum_reward": 1.2203125447034835,
      "step": 45
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 334.0805953979492,
      "epoch": 0.8528784648187633,
      "grad_norm": 0.3841504156589508,
      "kl": 0.044720458984375,
      "learning_rate": 1.718159615201853e-07,
      "loss": 0.0349,
      "reward": 1.1978795155882835,
      "reward_std": 0.530747976154089,
      "rewards/sum_reward": 1.1978795155882835,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 334.7330490112305,
      "epoch": 0.9381663113006397,
      "grad_norm": 0.4773261547088623,
      "kl": 0.0414764404296875,
      "learning_rate": 2.4570139579284723e-08,
      "loss": 0.0476,
      "reward": 1.2361607775092125,
      "reward_std": 0.5607758149504661,
      "rewards/sum_reward": 1.2361607775092125,
      "step": 55
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 320.73202896118164,
      "epoch": 0.9893390191897654,
      "kl": 0.039357503255208336,
      "reward": 1.219494101901849,
      "reward_std": 0.5131760289271673,
      "rewards/sum_reward": 1.219494101901849,
      "step": 58,
      "total_flos": 0.0,
      "train_loss": 0.22030245705411353,
      "train_runtime": 13222.5433,
      "train_samples_per_second": 0.567,
      "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
}