{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 10000,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "advantage": 2.050441083312035,
      "completion_length": 549.2318519592285,
      "epoch": 0.064,
      "grad_norm": 0.3377276062965393,
      "kl": 0.00016334056854248047,
      "learning_rate": 1.875e-06,
      "loss": 0.0,
      "reward": 1.4327381163835526,
      "reward_std": 0.2913255948573351,
      "rewards/accuracy_reward": 0.43273810148239134,
      "rewards/format_reward": 1.0,
      "step": 5
    },
    {
      "advantage": 1.8731657922267915,
      "completion_length": 535.2622138977051,
      "epoch": 0.128,
      "grad_norm": 0.2055761218070984,
      "kl": 0.001010751724243164,
      "learning_rate": 2.993961440992859e-06,
      "loss": 0.0,
      "reward": 1.500892886519432,
      "reward_std": 0.25598252601921556,
      "rewards/accuracy_reward": 0.5008928671479225,
      "rewards/format_reward": 1.0,
      "step": 10
    },
    {
      "advantage": 1.8132987678050996,
      "completion_length": 546.9449508666992,
      "epoch": 0.192,
      "grad_norm": 0.22043277323246002,
      "kl": 0.00526275634765625,
      "learning_rate": 2.9265847744427307e-06,
      "loss": 0.0002,
      "reward": 1.535119077563286,
      "reward_std": 0.23049755450338125,
      "rewards/accuracy_reward": 0.5351190570741892,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "advantage": 1.9496307998895646,
      "completion_length": 559.6901908874512,
      "epoch": 0.256,
      "grad_norm": 0.18047218024730682,
      "kl": 0.006356048583984375,
      "learning_rate": 2.7876731904027993e-06,
      "loss": 0.0003,
      "reward": 1.5279762238264083,
      "reward_std": 0.21949078943580388,
      "rewards/accuracy_reward": 0.5279762007296085,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "advantage": 1.9445811256766319,
      "completion_length": 561.3247123718262,
      "epoch": 0.32,
      "grad_norm": 0.17815344035625458,
      "kl": 0.00570068359375,
      "learning_rate": 2.584192295741087e-06,
      "loss": 0.0002,
      "reward": 1.5446428924798965,
      "reward_std": 0.23059090217575431,
      "rewards/accuracy_reward": 0.5446428753435612,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "advantage": 1.7600619524717331,
      "completion_length": 547.6345359802247,
      "epoch": 0.384,
      "grad_norm": 0.18666820228099823,
      "kl": 0.00573883056640625,
      "learning_rate": 2.3263454721781537e-06,
      "loss": 0.0002,
      "reward": 1.5431547909975052,
      "reward_std": 0.21657919492572547,
      "rewards/accuracy_reward": 0.5431547746062279,
      "rewards/format_reward": 1.0,
      "step": 30
    },
    {
      "advantage": 1.7993007361888886,
      "completion_length": 546.9157859802247,
      "epoch": 0.448,
      "grad_norm": 0.1700781136751175,
      "kl": 0.00593109130859375,
      "learning_rate": 2.027062236122014e-06,
      "loss": 0.0002,
      "reward": 1.5273809880018234,
      "reward_std": 0.2096662785857916,
      "rewards/accuracy_reward": 0.5273809637874365,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "advantage": 1.896444058418274,
      "completion_length": 552.3208389282227,
      "epoch": 0.512,
      "grad_norm": 0.1920849233865738,
      "kl": 0.005233001708984375,
      "learning_rate": 1.7013498987264833e-06,
      "loss": 0.0002,
      "reward": 1.5434524178504945,
      "reward_std": 0.22238524220883846,
      "rewards/accuracy_reward": 0.5434523917734623,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "advantage": 1.870384192466736,
      "completion_length": 558.927098083496,
      "epoch": 0.576,
      "grad_norm": 0.18718139827251434,
      "kl": 0.0053081512451171875,
      "learning_rate": 1.3655410366448499e-06,
      "loss": 0.0002,
      "reward": 1.5154762089252471,
      "reward_std": 0.21234574727714062,
      "rewards/accuracy_reward": 0.5154761992394924,
      "rewards/format_reward": 1.0,
      "step": 45
    },
    {
      "advantage": 1.9293582290410995,
      "completion_length": 554.5389984130859,
      "epoch": 0.64,
      "grad_norm": 0.2384500652551651,
      "kl": 0.0057861328125,
      "learning_rate": 1.036474508437579e-06,
      "loss": 0.0002,
      "reward": 1.5339285999536514,
      "reward_std": 0.2257133638486266,
      "rewards/accuracy_reward": 0.5339285805821419,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "advantage": 1.9427187949419022,
      "completion_length": 556.450008392334,
      "epoch": 0.704,
      "grad_norm": 0.19755816459655762,
      "kl": 0.005307769775390625,
      "learning_rate": 7.30651083891141e-07,
      "loss": 0.0002,
      "reward": 1.5452381312847137,
      "reward_std": 0.2197699649259448,
      "rewards/accuracy_reward": 0.545238108932972,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "advantage": 1.8897637754678727,
      "completion_length": 560.4640045166016,
      "epoch": 0.768,
      "grad_norm": 0.24272561073303223,
      "kl": 0.005287933349609375,
      "learning_rate": 4.63406026519703e-07,
      "loss": 0.0002,
      "reward": 1.525297647714615,
      "reward_std": 0.24526160284876825,
      "rewards/accuracy_reward": 0.5252976305782795,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "advantage": 1.9503421396017075,
      "completion_length": 557.6199478149414,
      "epoch": 0.832,
      "grad_norm": 0.18535475432872772,
      "kl": 0.005422210693359375,
      "learning_rate": 2.48140119418046e-07,
      "loss": 0.0002,
      "reward": 1.517559552192688,
      "reward_std": 0.21927062328904867,
      "rewards/accuracy_reward": 0.5175595372915268,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "advantage": 1.9777270182967186,
      "completion_length": 550.9339363098145,
      "epoch": 0.896,
      "grad_norm": 0.23573845624923706,
      "kl": 0.00623931884765625,
      "learning_rate": 9.564769404039419e-08,
      "loss": 0.0002,
      "reward": 1.5324405014514924,
      "reward_std": 0.22246364038437605,
      "rewards/accuracy_reward": 0.5324404895305633,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "advantage": 1.8454515248537064,
      "completion_length": 566.992268371582,
      "epoch": 0.96,
      "grad_norm": 0.18028907477855682,
      "kl": 0.00513458251953125,
      "learning_rate": 1.357535734809795e-08,
      "loss": 0.0002,
      "reward": 1.515476217865944,
      "reward_std": 0.22508260188624263,
      "rewards/accuracy_reward": 0.5154761977493763,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "advantage": 2.0446028262376785,
      "completion_length": 550.128978729248,
      "epoch": 0.9984,
      "kl": 0.005794525146484375,
      "reward": 1.536210338274638,
      "reward_std": 0.22843041829764843,
      "rewards/accuracy_reward": 0.5362103283405304,
      "rewards/format_reward": 1.0,
      "step": 78,
      "total_flos": 0.0,
      "train_loss": 0.0001987359993598567,
      "train_runtime": 7692.2558,
      "train_samples_per_second": 0.975,
      "train_steps_per_second": 0.01
    }
  ],
  "logging_steps": 5,
  "max_steps": 78,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}