{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5004255319148936,
  "eval_steps": 50,
  "global_step": 588,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.00013539849850576911,
      "loss": 4.8195,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017615750792387034,
      "loss": 2.2427,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 2.0574,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001967741935483871,
      "loss": 2.0078,
      "step": 40
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001931899641577061,
      "loss": 1.9905,
      "step": 50
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.6544462424008755,
      "eval_loss": 1.5293824672698975,
      "eval_runtime": 32.8926,
      "eval_samples_per_second": 61.047,
      "eval_steps_per_second": 0.243,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001896057347670251,
      "loss": 1.9798,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00018602150537634407,
      "loss": 1.9546,
      "step": 70
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001824372759856631,
      "loss": 1.9234,
      "step": 80
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00017885304659498208,
      "loss": 1.929,
      "step": 90
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00017526881720430107,
      "loss": 1.907,
      "step": 100
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.5062341056107924,
      "eval_loss": 3.5476841926574707,
      "eval_runtime": 32.158,
      "eval_samples_per_second": 62.442,
      "eval_steps_per_second": 0.249,
      "step": 100
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001716845878136201,
      "loss": 1.8956,
      "step": 110
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00016810035842293908,
      "loss": 1.8982,
      "step": 120
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00016451612903225807,
      "loss": 1.8738,
      "step": 130
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00016093189964157706,
      "loss": 1.8788,
      "step": 140
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00015734767025089608,
      "loss": 1.8675,
      "step": 150
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.5123508896963466,
      "eval_loss": 3.6367032527923584,
      "eval_runtime": 33.3455,
      "eval_samples_per_second": 60.218,
      "eval_steps_per_second": 0.24,
      "step": 150
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00015376344086021504,
      "loss": 1.8457,
      "step": 160
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00015017921146953406,
      "loss": 1.8669,
      "step": 170
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00014659498207885305,
      "loss": 1.8484,
      "step": 180
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00014301075268817205,
      "loss": 1.8312,
      "step": 190
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00013942652329749104,
      "loss": 1.8359,
      "step": 200
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.514210995704377,
      "eval_loss": 3.6230080127716064,
      "eval_runtime": 33.3113,
      "eval_samples_per_second": 60.28,
      "eval_steps_per_second": 0.24,
      "step": 200
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00013584229390681005,
      "loss": 1.8261,
      "step": 210
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00013225806451612905,
      "loss": 1.8447,
      "step": 220
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012867383512544804,
      "loss": 1.836,
      "step": 230
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00012508960573476703,
      "loss": 1.8209,
      "step": 240
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00012150537634408603,
      "loss": 1.8103,
      "step": 250
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.5280159907778466,
      "eval_loss": 3.4809513092041016,
      "eval_runtime": 33.3217,
      "eval_samples_per_second": 60.261,
      "eval_steps_per_second": 0.24,
      "step": 250
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00011792114695340501,
      "loss": 1.8104,
      "step": 260
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00011433691756272403,
      "loss": 1.8002,
      "step": 270
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000110752688172043,
      "loss": 1.7979,
      "step": 280
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00010716845878136201,
      "loss": 1.7871,
      "step": 290
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000103584229390681,
      "loss": 1.7905,
      "step": 300
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.5359403052501626,
      "eval_loss": 3.469621419906616,
      "eval_runtime": 33.3288,
      "eval_samples_per_second": 60.248,
      "eval_steps_per_second": 0.24,
      "step": 300
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.7734,
      "step": 310
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.6415770609319e-05,
      "loss": 1.7913,
      "step": 320
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.2831541218638e-05,
      "loss": 1.76,
      "step": 330
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.924731182795699e-05,
      "loss": 1.7728,
      "step": 340
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.566308243727598e-05,
      "loss": 1.7578,
      "step": 350
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.5407767755955649,
      "eval_loss": 3.409736156463623,
      "eval_runtime": 32.2132,
      "eval_samples_per_second": 62.335,
      "eval_steps_per_second": 0.248,
      "step": 350
    },
    {
      "epoch": 0.31,
      "learning_rate": 8.207885304659499e-05,
      "loss": 1.7371,
      "step": 360
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.849462365591398e-05,
      "loss": 1.7593,
      "step": 370
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.491039426523297e-05,
      "loss": 1.7437,
      "step": 380
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.132616487455197e-05,
      "loss": 1.7414,
      "step": 390
    },
    {
      "epoch": 0.34,
      "learning_rate": 6.774193548387096e-05,
      "loss": 1.729,
      "step": 400
    },
    {
      "epoch": 0.34,
      "eval_accuracy": 0.5427147714128822,
      "eval_loss": 3.4043824672698975,
      "eval_runtime": 33.2117,
      "eval_samples_per_second": 60.461,
      "eval_steps_per_second": 0.241,
      "step": 400
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.415770609318996e-05,
      "loss": 1.7609,
      "step": 410
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.057347670250897e-05,
      "loss": 1.7368,
      "step": 420
    },
    {
      "epoch": 0.37,
      "learning_rate": 5.6989247311827965e-05,
      "loss": 1.7354,
      "step": 430
    },
    {
      "epoch": 0.37,
      "learning_rate": 5.340501792114696e-05,
      "loss": 1.7269,
      "step": 440
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.982078853046595e-05,
      "loss": 1.7308,
      "step": 450
    },
    {
      "epoch": 0.38,
      "eval_accuracy": 0.5484289625466852,
      "eval_loss": 3.3802289962768555,
      "eval_runtime": 32.0517,
      "eval_samples_per_second": 62.649,
      "eval_steps_per_second": 0.25,
      "step": 450
    },
    {
      "epoch": 0.39,
      "learning_rate": 4.6236559139784944e-05,
      "loss": 1.7006,
      "step": 460
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.265232974910394e-05,
      "loss": 1.7124,
      "step": 470
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.906810035842295e-05,
      "loss": 1.7225,
      "step": 480
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.548387096774194e-05,
      "loss": 1.7029,
      "step": 490
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.1899641577060935e-05,
      "loss": 1.6896,
      "step": 500
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.5529602022019449,
      "eval_loss": 3.345804214477539,
      "eval_runtime": 32.1251,
      "eval_samples_per_second": 62.506,
      "eval_steps_per_second": 0.249,
      "step": 500
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.831541218637993e-05,
      "loss": 1.7141,
      "step": 510
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.4731182795698928e-05,
      "loss": 1.6969,
      "step": 520
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.1146953405017922e-05,
      "loss": 1.7087,
      "step": 530
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7562724014336916e-05,
      "loss": 1.6779,
      "step": 540
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.3978494623655914e-05,
      "loss": 1.6721,
      "step": 550
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.5576710752298722,
      "eval_loss": 3.291583776473999,
      "eval_runtime": 32.0142,
      "eval_samples_per_second": 62.722,
      "eval_steps_per_second": 0.25,
      "step": 550
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.039426523297491e-05,
      "loss": 1.6997,
      "step": 560
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.810035842293908e-06,
      "loss": 1.6679,
      "step": 570
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.225806451612903e-06,
      "loss": 1.6949,
      "step": 580
    },
    {
      "epoch": 0.5,
      "step": 588,
      "total_flos": 6.465698107324629e+18,
      "train_loss": 1.8579450334821428,
      "train_runtime": 7071.4809,
      "train_samples_per_second": 21.266,
      "train_steps_per_second": 0.083
    }
  ],
  "logging_steps": 10,
  "max_steps": 588,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 6.465698107324629e+18,
  "trial_name": null,
  "trial_params": null
}