{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9943502824858758,
  "eval_steps": 500,
  "global_step": 22,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 276.02735233306885,
      "epoch": 0.04519774011299435,
      "grad_norm": 0.13002647486414434,
      "kl": 0.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.0303,
      "reward": 1.3704427555203438,
      "reward_std": 0.1837239097803831,
      "rewards/fc_accuracy_reward": 0.4309895895421505,
      "rewards/format_reward": 0.9394531473517418,
      "step": 1
    },
    {
      "completion_length": 292.13477420806885,
      "epoch": 0.0903954802259887,
      "grad_norm": 0.14619501253296655,
      "kl": 0.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0469,
      "reward": 1.3990885764360428,
      "reward_std": 0.21469345316290855,
      "rewards/fc_accuracy_reward": 0.44856772013008595,
      "rewards/format_reward": 0.9505208507180214,
      "step": 2
    },
    {
      "completion_length": 268.9563903808594,
      "epoch": 0.13559322033898305,
      "grad_norm": 0.13036409926543296,
      "kl": 0.0002338886260986328,
      "learning_rate": 2e-05,
      "loss": 0.0319,
      "reward": 1.3704427406191826,
      "reward_std": 0.18289842084050179,
      "rewards/fc_accuracy_reward": 0.41992188431322575,
      "rewards/format_reward": 0.9505208656191826,
      "step": 3
    },
    {
      "completion_length": 237.50716876983643,
      "epoch": 0.1807909604519774,
      "grad_norm": 0.12085621262181591,
      "kl": 0.00720977783203125,
      "learning_rate": 1.9863613034027224e-05,
      "loss": 0.0065,
      "reward": 1.427734412252903,
      "reward_std": 0.1744816219434142,
      "rewards/fc_accuracy_reward": 0.46614584885537624,
      "rewards/format_reward": 0.9615885503590107,
      "step": 4
    },
    {
      "completion_length": 198.7910213470459,
      "epoch": 0.22598870056497175,
      "grad_norm": 1.1449772190729142,
      "kl": 0.0814666748046875,
      "learning_rate": 1.9458172417006347e-05,
      "loss": -0.007,
      "reward": 1.2617187947034836,
      "reward_std": 0.17365613672882318,
      "rewards/fc_accuracy_reward": 0.3222656352445483,
      "rewards/format_reward": 0.939453125,
      "step": 5
    },
    {
      "completion_length": 231.09896564483643,
      "epoch": 0.2711864406779661,
      "grad_norm": 0.1913660515760745,
      "kl": 0.045501708984375,
      "learning_rate": 1.879473751206489e-05,
      "loss": 0.0077,
      "reward": 1.3893229588866234,
      "reward_std": 0.14095465373247862,
      "rewards/fc_accuracy_reward": 0.43489584513008595,
      "rewards/format_reward": 0.9544270932674408,
      "step": 6
    },
    {
      "completion_length": 247.44727039337158,
      "epoch": 0.3163841807909605,
      "grad_norm": 0.13567082283951828,
      "kl": 0.045623779296875,
      "learning_rate": 1.789140509396394e-05,
      "loss": 0.0065,
      "reward": 1.3470052555203438,
      "reward_std": 0.14320992957800627,
      "rewards/fc_accuracy_reward": 0.3938802182674408,
      "rewards/format_reward": 0.9531250074505806,
      "step": 7
    },
    {
      "completion_length": 269.38086795806885,
      "epoch": 0.3615819209039548,
      "grad_norm": 0.16971700336774798,
      "kl": 0.060699462890625,
      "learning_rate": 1.6772815716257414e-05,
      "loss": 0.0085,
      "reward": 1.3372396156191826,
      "reward_std": 0.1406525035854429,
      "rewards/fc_accuracy_reward": 0.39257813431322575,
      "rewards/format_reward": 0.9446614682674408,
      "step": 8
    },
    {
      "completion_length": 295.05665016174316,
      "epoch": 0.4067796610169492,
      "grad_norm": 1.3287845887377334,
      "kl": 0.070892333984375,
      "learning_rate": 1.5469481581224274e-05,
      "loss": 0.0192,
      "reward": 1.4036458730697632,
      "reward_std": 0.18207293096929789,
      "rewards/fc_accuracy_reward": 0.43750000931322575,
      "rewards/format_reward": 0.9661458507180214,
      "step": 9
    },
    {
      "completion_length": 307.4420690536499,
      "epoch": 0.4519774011299435,
      "grad_norm": 54738.02031388349,
      "kl": 394.0651397705078,
      "learning_rate": 1.4016954246529697e-05,
      "loss": 37.1931,
      "reward": 1.3352864980697632,
      "reward_std": 0.2137870010919869,
      "rewards/fc_accuracy_reward": 0.4199218861758709,
      "rewards/format_reward": 0.915364608168602,
      "step": 10
    },
    {
      "completion_length": 314.0898551940918,
      "epoch": 0.4971751412429379,
      "grad_norm": 2.107091829265173,
      "kl": 0.095855712890625,
      "learning_rate": 1.2454854871407993e-05,
      "loss": 0.0338,
      "reward": 1.3177083730697632,
      "reward_std": 0.2345268540084362,
      "rewards/fc_accuracy_reward": 0.4108072966337204,
      "rewards/format_reward": 0.9069010689854622,
      "step": 11
    },
    {
      "completion_length": 318.7447986602783,
      "epoch": 0.5423728813559322,
      "grad_norm": 0.20351280734446026,
      "kl": 0.081268310546875,
      "learning_rate": 1.0825793454723325e-05,
      "loss": 0.0176,
      "reward": 1.2675781697034836,
      "reward_std": 0.3210527803748846,
      "rewards/fc_accuracy_reward": 0.397135429084301,
      "rewards/format_reward": 0.8704427294433117,
      "step": 12
    },
    {
      "completion_length": 311.57813262939453,
      "epoch": 0.5875706214689266,
      "grad_norm": 0.1737011976596144,
      "kl": 0.07958984375,
      "learning_rate": 9.174206545276678e-06,
      "loss": 0.0088,
      "reward": 1.3684896156191826,
      "reward_std": 0.23807168938219547,
      "rewards/fc_accuracy_reward": 0.4479166828095913,
      "rewards/format_reward": 0.9205729365348816,
      "step": 13
    },
    {
      "completion_length": 301.56706619262695,
      "epoch": 0.632768361581921,
      "grad_norm": 18.70037949885207,
      "kl": 0.586456298828125,
      "learning_rate": 7.545145128592009e-06,
      "loss": 0.0599,
      "reward": 1.3203125298023224,
      "reward_std": 0.20432353112846613,
      "rewards/fc_accuracy_reward": 0.40820313803851604,
      "rewards/format_reward": 0.9121094010770321,
      "step": 14
    },
    {
      "completion_length": 289.6764450073242,
      "epoch": 0.6779661016949152,
      "grad_norm": 0.15577324164046125,
      "kl": 0.077789306640625,
      "learning_rate": 5.983045753470308e-06,
      "loss": 0.011,
      "reward": 1.3496094048023224,
      "reward_std": 0.21356581384316087,
      "rewards/fc_accuracy_reward": 0.4186198003590107,
      "rewards/format_reward": 0.9309896044433117,
      "step": 15
    },
    {
      "completion_length": 286.3886785507202,
      "epoch": 0.7231638418079096,
      "grad_norm": 5.326707764035434,
      "kl": 0.0885009765625,
      "learning_rate": 4.530518418775734e-06,
      "loss": 0.0102,
      "reward": 1.3658854514360428,
      "reward_std": 0.23490996472537518,
      "rewards/fc_accuracy_reward": 0.44401043094694614,
      "rewards/format_reward": 0.9218750260770321,
      "step": 16
    },
    {
      "completion_length": 272.7597780227661,
      "epoch": 0.768361581920904,
      "grad_norm": 0.33239438768575263,
      "kl": 0.075653076171875,
      "learning_rate": 3.2271842837425917e-06,
      "loss": 0.0083,
      "reward": 1.3222656697034836,
      "reward_std": 0.23243350349366665,
      "rewards/fc_accuracy_reward": 0.4173177182674408,
      "rewards/format_reward": 0.9049479365348816,
      "step": 17
    },
    {
      "completion_length": 267.5846462249756,
      "epoch": 0.8135593220338984,
      "grad_norm": 0.16069383676764726,
      "kl": 0.07232666015625,
      "learning_rate": 2.1085949060360654e-06,
      "loss": 0.0114,
      "reward": 1.3854166939854622,
      "reward_std": 0.242501275613904,
      "rewards/fc_accuracy_reward": 0.47070313803851604,
      "rewards/format_reward": 0.9147135652601719,
      "step": 18
    },
    {
      "completion_length": 266.1367254257202,
      "epoch": 0.8587570621468926,
      "grad_norm": 0.14326034602692705,
      "kl": 0.0689239501953125,
      "learning_rate": 1.2052624879351105e-06,
      "loss": 0.0075,
      "reward": 1.3854166939854622,
      "reward_std": 0.20935742277652025,
      "rewards/fc_accuracy_reward": 0.44531251676380634,
      "rewards/format_reward": 0.940104179084301,
      "step": 19
    },
    {
      "completion_length": 260.7949285507202,
      "epoch": 0.903954802259887,
      "grad_norm": 0.14485988944724762,
      "kl": 0.06939697265625,
      "learning_rate": 5.418275829936537e-07,
      "loss": 0.0098,
      "reward": 1.395182341337204,
      "reward_std": 0.2115317303687334,
      "rewards/fc_accuracy_reward": 0.47330730222165585,
      "rewards/format_reward": 0.9218750260770321,
      "step": 20
    },
    {
      "completion_length": 262.3704528808594,
      "epoch": 0.9491525423728814,
      "grad_norm": 0.19459843380674965,
      "kl": 0.0714874267578125,
      "learning_rate": 1.3638696597277678e-07,
      "loss": 0.0072,
      "reward": 1.3489583656191826,
      "reward_std": 0.20432352973148227,
      "rewards/fc_accuracy_reward": 0.4348958469927311,
      "rewards/format_reward": 0.9140625186264515,
      "step": 21
    },
    {
      "completion_length": 263.5794382095337,
      "epoch": 0.9943502824858758,
      "grad_norm": 0.16933108100699076,
      "kl": 0.067138671875,
      "learning_rate": 0.0,
      "loss": 0.0169,
      "reward": 1.354817733168602,
      "reward_std": 0.242501275613904,
      "rewards/fc_accuracy_reward": 0.4348958469927311,
      "rewards/format_reward": 0.9199219010770321,
      "step": 22
    },
    {
      "epoch": 0.9943502824858758,
      "step": 22,
      "total_flos": 0.0,
      "train_loss": 1.7066374281421304,
      "train_runtime": 5498.7014,
      "train_samples_per_second": 2.055,
      "train_steps_per_second": 0.004
    }
  ],
  "logging_steps": 1,
  "max_steps": 22,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}