{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.996,
  "eval_steps": 500,
  "global_step": 249,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 599.9553680419922,
      "epoch": 0.004,
      "grad_norm": 0.24663478136062622,
      "kl": 0.0,
      "learning_rate": 8.000000000000001e-07,
      "loss": -0.0302,
      "reward": 0.1160714365541935,
      "reward_std": 0.20195358991622925,
      "rewards/accuracy_reward": 0.1160714365541935,
      "step": 1
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 644.7827648586697,
      "epoch": 0.04,
      "grad_norm": 0.2229214757680893,
      "kl": 0.00017437669965955947,
      "learning_rate": 8.000000000000001e-06,
      "loss": -0.0164,
      "reward": 0.13591270407454836,
      "reward_std": 0.1599799825085534,
      "rewards/accuracy_reward": 0.13591270407454836,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 693.7643196105957,
      "epoch": 0.08,
      "grad_norm": 0.21021594107151031,
      "kl": 0.005759429931640625,
      "learning_rate": 1.6000000000000003e-05,
      "loss": -0.0059,
      "reward": 0.22232143711298705,
      "reward_std": 0.23315932899713515,
      "rewards/accuracy_reward": 0.22232143711298705,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 821.8294998168946,
      "epoch": 0.12,
      "grad_norm": 0.2534916400909424,
      "kl": 0.02135162353515625,
      "learning_rate": 1.997807645233842e-05,
      "loss": -0.001,
      "reward": 0.33482144214212894,
      "reward_std": 0.34336078241467477,
      "rewards/accuracy_reward": 0.33482144214212894,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 877.2330703735352,
      "epoch": 0.16,
      "grad_norm": 0.2218112200498581,
      "kl": 0.0388702392578125,
      "learning_rate": 1.9803328406604252e-05,
      "loss": 0.004,
      "reward": 0.36428573317825796,
      "reward_std": 0.34514649510383605,
      "rewards/accuracy_reward": 0.36428573317825796,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 610.845565032959,
      "epoch": 0.2,
      "grad_norm": 0.3174823522567749,
      "kl": 0.064056396484375,
      "learning_rate": 1.9457233587073177e-05,
      "loss": 0.0214,
      "reward": 0.4857143074274063,
      "reward_std": 0.36878507286310197,
      "rewards/accuracy_reward": 0.4857143074274063,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 555.7643127441406,
      "epoch": 0.24,
      "grad_norm": 0.34336915612220764,
      "kl": 0.214013671875,
      "learning_rate": 1.8946528335730344e-05,
      "loss": 0.0412,
      "reward": 0.4758928783237934,
      "reward_std": 0.382133661583066,
      "rewards/accuracy_reward": 0.4758928783237934,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 738.0321739196777,
      "epoch": 0.28,
      "grad_norm": 0.3574662208557129,
      "kl": 0.13843994140625,
      "learning_rate": 1.8281152949374527e-05,
      "loss": 0.0249,
      "reward": 0.4142857350409031,
      "reward_std": 0.37416205164045097,
      "rewards/accuracy_reward": 0.4142857350409031,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.2402069091797,
      "epoch": 0.32,
      "grad_norm": 0.23834314942359924,
      "kl": 0.12186279296875,
      "learning_rate": 1.7474058203047863e-05,
      "loss": 0.0057,
      "reward": 0.4830357365310192,
      "reward_std": 0.3151938086375594,
      "rewards/accuracy_reward": 0.4830357365310192,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 633.9777038574218,
      "epoch": 0.36,
      "grad_norm": 0.3009245991706848,
      "kl": 0.10828857421875,
      "learning_rate": 1.6540953277930925e-05,
      "loss": 0.0113,
      "reward": 0.5000000216066838,
      "reward_std": 0.3544993594288826,
      "rewards/accuracy_reward": 0.5000000216066838,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 609.4643142700195,
      "epoch": 0.4,
      "grad_norm": 0.2935497462749481,
      "kl": 0.12471923828125,
      "learning_rate": 1.55e-05,
      "loss": 0.0182,
      "reward": 0.5151785992085933,
      "reward_std": 0.3775656022131443,
      "rewards/accuracy_reward": 0.5151785992085933,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 521.8821655273438,
      "epoch": 0.44,
      "grad_norm": 0.31697750091552734,
      "kl": 0.14866943359375,
      "learning_rate": 1.437145934074321e-05,
      "loss": 0.0108,
      "reward": 0.5276785962283611,
      "reward_std": 0.3804761566221714,
      "rewards/accuracy_reward": 0.5276785962283611,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 525.0000259399415,
      "epoch": 0.48,
      "grad_norm": 0.38678815960884094,
      "kl": 0.13438720703125,
      "learning_rate": 1.317729706039701e-05,
      "loss": 0.0101,
      "reward": 0.5142857365310192,
      "reward_std": 0.3935286600142717,
      "rewards/accuracy_reward": 0.5142857365310192,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 613.3500282287598,
      "epoch": 0.52,
      "grad_norm": 0.30911684036254883,
      "kl": 0.12950439453125,
      "learning_rate": 1.1940756169408882e-05,
      "loss": 0.006,
      "reward": 0.46875002309679986,
      "reward_std": 0.3613659642636776,
      "rewards/accuracy_reward": 0.46875002309679986,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 636.9446754455566,
      "epoch": 0.56,
      "grad_norm": 0.28472012281417847,
      "kl": 0.12186279296875,
      "learning_rate": 1.0685904529677496e-05,
      "loss": 0.0075,
      "reward": 0.449107164144516,
      "reward_std": 0.36206651106476784,
      "rewards/accuracy_reward": 0.449107164144516,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 864.2687881469726,
      "epoch": 0.6,
      "grad_norm": 0.2347540706396103,
      "kl": 0.09737548828125,
      "learning_rate": 9.437166400997629e-06,
      "loss": 0.0228,
      "reward": 0.4017857346683741,
      "reward_std": 0.37069899663329126,
      "rewards/accuracy_reward": 0.4017857346683741,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 839.4205764770508,
      "epoch": 0.64,
      "grad_norm": 0.2780732214450836,
      "kl": 0.11231689453125,
      "learning_rate": 8.218847050625476e-06,
      "loss": 0.0131,
      "reward": 0.43303573224693537,
      "reward_std": 0.3525656070560217,
      "rewards/accuracy_reward": 0.43303573224693537,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 762.250927734375,
      "epoch": 0.68,
      "grad_norm": 0.24334457516670227,
      "kl": 0.11146240234375,
      "learning_rate": 7.054659678898304e-06,
      "loss": -0.0166,
      "reward": 0.4669643059372902,
      "reward_std": 0.4058806154876947,
      "rewards/accuracy_reward": 0.4669643059372902,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 653.9178886413574,
      "epoch": 0.72,
      "grad_norm": 0.293698787689209,
      "kl": 0.10758056640625,
      "learning_rate": 5.96726386876328e-06,
      "loss": 0.0285,
      "reward": 0.5089285977184772,
      "reward_std": 0.3675716958940029,
      "rewards/accuracy_reward": 0.5089285977184772,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 607.0893157958984,
      "epoch": 0.76,
      "grad_norm": 0.3261861801147461,
      "kl": 0.10914306640625,
      "learning_rate": 4.977824542770279e-06,
      "loss": -0.0012,
      "reward": 0.5026785939931869,
      "reward_std": 0.387619011849165,
      "rewards/accuracy_reward": 0.5026785939931869,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 641.3634246826172,
      "epoch": 0.8,
      "grad_norm": 0.34083282947540283,
      "kl": 0.11761474609375,
      "learning_rate": 4.1056000119292e-06,
      "loss": 0.0296,
      "reward": 0.5035714529454708,
      "reward_std": 0.3847282975912094,
      "rewards/accuracy_reward": 0.5035714529454708,
      "step": 200
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 684.0607452392578,
      "epoch": 0.84,
      "grad_norm": 0.35424429178237915,
      "kl": 0.13974609375,
      "learning_rate": 3.367567134592167e-06,
      "loss": 0.0385,
      "reward": 0.5000000253319741,
      "reward_std": 0.3738659627735615,
      "rewards/accuracy_reward": 0.5000000253319741,
      "step": 210
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 743.4545013427735,
      "epoch": 0.88,
      "grad_norm": 0.5738213658332825,
      "kl": 0.18140869140625,
      "learning_rate": 2.778090881216592e-06,
      "loss": -0.0031,
      "reward": 0.48482145369052887,
      "reward_std": 0.36493739411234855,
      "rewards/accuracy_reward": 0.48482145369052887,
      "step": 220
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 736.5125328063965,
      "epoch": 0.92,
      "grad_norm": 0.34698280692100525,
      "kl": 0.1845947265625,
      "learning_rate": 2.34864473655513e-06,
      "loss": 0.0267,
      "reward": 0.5008928813040257,
      "reward_std": 0.3658943545073271,
      "rewards/accuracy_reward": 0.5008928813040257,
      "step": 230
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 654.1107444763184,
      "epoch": 0.96,
      "grad_norm": 0.27743178606033325,
      "kl": 0.1630615234375,
      "learning_rate": 2.087587381325867e-06,
      "loss": 0.0044,
      "reward": 0.5339285939931869,
      "reward_std": 0.35324631109833715,
      "rewards/accuracy_reward": 0.5339285939931869,
      "step": 240
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 649.8173192342123,
      "epoch": 0.996,
      "kl": 0.14461263020833334,
      "reward": 0.5496032014489174,
      "reward_std": 0.36488261736101574,
      "rewards/accuracy_reward": 0.5496032014489174,
      "step": 249,
      "total_flos": 0.0,
      "train_loss": 0.012884760174496347,
      "train_runtime": 16946.151,
      "train_samples_per_second": 0.413,
      "train_steps_per_second": 0.015
    }
  ],
  "logging_steps": 10,
  "max_steps": 250,
  "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": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}