{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8547008547008547,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio": 0.0,
      "completion_length": 503.30001945495604,
      "epoch": 0.042735042735042736,
      "grad_norm": 0.14690014719963074,
      "kl": 0.0005318611507391324,
      "learning_rate": 9.871794871794872e-05,
      "loss": -0.0402,
      "num_tokens": 33288.0,
      "reward": -1.1772359251976012,
      "reward_std": 2.312183880805969,
      "rewards/reward_function": -1.1772359311580658,
      "step": 10
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 509.85001373291016,
      "epoch": 0.08547008547008547,
      "grad_norm": 0.16410136222839355,
      "kl": 0.001959062390960753,
      "learning_rate": 9.72934472934473e-05,
      "loss": -0.008,
      "num_tokens": 67317.0,
      "reward": -1.455158567428589,
      "reward_std": 2.39455783367157,
      "rewards/reward_function": -1.4551584720611572,
      "step": 20
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 382.58334197998045,
      "epoch": 0.1282051282051282,
      "grad_norm": 0.5200854539871216,
      "kl": 0.003280931804329157,
      "learning_rate": 9.586894586894587e-05,
      "loss": 0.0011,
      "num_tokens": 93788.0,
      "reward": 2.4295308887958527,
      "reward_std": 1.948470675945282,
      "rewards/reward_function": 2.4295308709144594,
      "step": 30
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.25001068115233,
      "epoch": 0.17094017094017094,
      "grad_norm": 0.2215566486120224,
      "kl": 0.005862738820724189,
      "learning_rate": 9.444444444444444e-05,
      "loss": -0.0111,
      "num_tokens": 125045.0,
      "reward": 1.2611542105674745,
      "reward_std": 2.462652099132538,
      "rewards/reward_function": 1.2611541628837586,
      "step": 40
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 518.6166809082031,
      "epoch": 0.21367521367521367,
      "grad_norm": 0.15176545083522797,
      "kl": 0.008088351227343082,
      "learning_rate": 9.301994301994303e-05,
      "loss": -0.0085,
      "num_tokens": 159318.0,
      "reward": -1.0398478865623475,
      "reward_std": 1.9632926762104035,
      "rewards/reward_function": -1.039847767353058,
      "step": 50
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 436.50001125335694,
      "epoch": 0.2564102564102564,
      "grad_norm": 0.23773637413978577,
      "kl": 0.007962367637082935,
      "learning_rate": 9.15954415954416e-05,
      "loss": 0.0155,
      "num_tokens": 188622.0,
      "reward": -0.9993367567658424,
      "reward_std": 2.5938119292259216,
      "rewards/reward_function": -0.9993367329239845,
      "step": 60
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 465.0166717529297,
      "epoch": 0.29914529914529914,
      "grad_norm": 0.22686824202537537,
      "kl": 0.01036710049957037,
      "learning_rate": 9.017094017094018e-05,
      "loss": 0.0232,
      "num_tokens": 219649.0,
      "reward": -1.0854085847735404,
      "reward_std": 2.3031186699867248,
      "rewards/reward_function": -1.0854085013270378,
      "step": 70
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 594.3666900634765,
      "epoch": 0.3418803418803419,
      "grad_norm": 0.2330074906349182,
      "kl": 0.011198346642777324,
      "learning_rate": 8.874643874643875e-05,
      "loss": 0.0085,
      "num_tokens": 258521.0,
      "reward": 0.44582319259643555,
      "reward_std": 2.180183058977127,
      "rewards/reward_function": 0.44582313299179077,
      "step": 80
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 363.2166831970215,
      "epoch": 0.38461538461538464,
      "grad_norm": 1.2950036525726318,
      "kl": 0.01624974999576807,
      "learning_rate": 8.732193732193732e-05,
      "loss": -0.0279,
      "num_tokens": 283896.0,
      "reward": -0.22942656874656678,
      "reward_std": 2.698866534233093,
      "rewards/reward_function": -0.2294266164302826,
      "step": 90
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 366.3166763305664,
      "epoch": 0.42735042735042733,
      "grad_norm": 0.44916248321533203,
      "kl": 0.0371546683833003,
      "learning_rate": 8.58974358974359e-05,
      "loss": -0.034,
      "num_tokens": 309127.0,
      "reward": 1.030603051185608,
      "reward_std": 2.1768004059791566,
      "rewards/reward_function": 1.03060302734375,
      "step": 100
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 627.7833419799805,
      "epoch": 0.4700854700854701,
      "grad_norm": 0.24189692735671997,
      "kl": 0.02031996361911297,
      "learning_rate": 8.447293447293447e-05,
      "loss": 0.0057,
      "num_tokens": 350124.0,
      "reward": 1.7054275810718535,
      "reward_std": 2.761507248878479,
      "rewards/reward_function": 1.7054275453090668,
      "step": 110
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 466.7833419799805,
      "epoch": 0.5128205128205128,
      "grad_norm": 0.3966105282306671,
      "kl": 0.03020444568246603,
      "learning_rate": 8.304843304843305e-05,
      "loss": 0.0158,
      "num_tokens": 381323.0,
      "reward": 0.4135805606842041,
      "reward_std": 2.793270480632782,
      "rewards/reward_function": 0.41358056366443635,
      "step": 120
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 418.7166778564453,
      "epoch": 0.5555555555555556,
      "grad_norm": 0.8130244016647339,
      "kl": 0.050436797365546224,
      "learning_rate": 8.162393162393163e-05,
      "loss": 0.0109,
      "num_tokens": 409632.0,
      "reward": 3.6511381447315214,
      "reward_std": 2.057768177986145,
      "rewards/reward_function": 3.651138073205948,
      "step": 130
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 401.18334617614744,
      "epoch": 0.5982905982905983,
      "grad_norm": 0.237253338098526,
      "kl": 0.08577278926968575,
      "learning_rate": 8.01994301994302e-05,
      "loss": 0.005,
      "num_tokens": 436613.0,
      "reward": 3.9179525792598726,
      "reward_std": 2.5252403259277343,
      "rewards/reward_function": 3.9179525285959245,
      "step": 140
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 534.9666793823242,
      "epoch": 0.6410256410256411,
      "grad_norm": 0.23871329426765442,
      "kl": 0.0823534980416298,
      "learning_rate": 7.877492877492878e-05,
      "loss": 0.0435,
      "num_tokens": 471891.0,
      "reward": 4.100382626056671,
      "reward_std": 2.5033695101737976,
      "rewards/reward_function": 4.100382459163666,
      "step": 150
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 332.50001373291013,
      "epoch": 0.6837606837606838,
      "grad_norm": 0.8124951720237732,
      "kl": 0.15444251857697963,
      "learning_rate": 7.735042735042735e-05,
      "loss": 0.0189,
      "num_tokens": 494961.0,
      "reward": 5.122303450107575,
      "reward_std": 2.0907008171081545,
      "rewards/reward_function": 5.122303307056427,
      "step": 160
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 405.25001373291013,
      "epoch": 0.7264957264957265,
      "grad_norm": 0.7909404635429382,
      "kl": 0.13212636522948742,
      "learning_rate": 7.592592592592593e-05,
      "loss": -0.0367,
      "num_tokens": 522258.0,
      "reward": 5.544608736038208,
      "reward_std": 1.7265446126461028,
      "rewards/reward_function": 5.544608497619629,
      "step": 170
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 595.9833587646484,
      "epoch": 0.7692307692307693,
      "grad_norm": 0.15963077545166016,
      "kl": 0.09406365267932415,
      "learning_rate": 7.450142450142451e-05,
      "loss": -0.0366,
      "num_tokens": 561197.0,
      "reward": 7.151281929016113,
      "reward_std": 2.8219340562820436,
      "rewards/reward_function": 7.15128173828125,
      "step": 180
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 566.9666809082031,
      "epoch": 0.811965811965812,
      "grad_norm": 0.18807636201381683,
      "kl": 0.06498432569205762,
      "learning_rate": 7.307692307692307e-05,
      "loss": -0.0543,
      "num_tokens": 598575.0,
      "reward": 1.7874940395355225,
      "reward_std": 2.3280829310417177,
      "rewards/reward_function": 1.7874938309192658,
      "step": 190
    },
    {
      "clip_ratio": 0.0,
      "completion_length": 831.9500183105469,
      "epoch": 0.8547008547008547,
      "grad_norm": 0.19820700585842133,
      "kl": 0.03369698449969292,
      "learning_rate": 7.165242165242165e-05,
      "loss": -0.0132,
      "num_tokens": 651618.0,
      "reward": 2.0692047476768494,
      "reward_std": 2.6171678781509398,
      "rewards/reward_function": 2.06920468211174,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 702,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "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
}