{
  "best_metric": 1.9192386865615845,
  "best_model_checkpoint": "experiments/checkpoint-500",
  "epoch": 100.0,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.8e-06,
      "loss": 1.7688,
      "step": 5
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-06,
      "loss": 1.7684,
      "step": 10
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.799999999999998e-06,
      "loss": 1.7564,
      "step": 15
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 1.7424,
      "step": 20
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.3799999999999998e-05,
      "loss": 1.727,
      "step": 25
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.68e-05,
      "loss": 1.7135,
      "step": 30
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.98e-05,
      "loss": 1.701,
      "step": 35
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.28e-05,
      "loss": 1.6797,
      "step": 40
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.5799999999999997e-05,
      "loss": 1.6547,
      "step": 45
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.88e-05,
      "loss": 1.6245,
      "step": 50
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.5684312582015991,
      "eval_runtime": 1.4166,
      "eval_samples_per_second": 7.059,
      "eval_steps_per_second": 1.412,
      "step": 50
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.1799999999999994e-05,
      "loss": 1.5841,
      "step": 55
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.48e-05,
      "loss": 1.5316,
      "step": 60
    },
    {
      "epoch": 13.0,
      "learning_rate": 3.78e-05,
      "loss": 1.4644,
      "step": 65
    },
    {
      "epoch": 14.0,
      "learning_rate": 4.08e-05,
      "loss": 1.3728,
      "step": 70
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.3799999999999994e-05,
      "loss": 1.2692,
      "step": 75
    },
    {
      "epoch": 16.0,
      "learning_rate": 4.56e-05,
      "loss": 1.1998,
      "step": 80
    },
    {
      "epoch": 17.0,
      "learning_rate": 4.8599999999999995e-05,
      "loss": 1.1159,
      "step": 85
    },
    {
      "epoch": 18.0,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 1.0442,
      "step": 90
    },
    {
      "epoch": 19.0,
      "learning_rate": 5.459999999999999e-05,
      "loss": 0.9944,
      "step": 95
    },
    {
      "epoch": 20.0,
      "learning_rate": 5.76e-05,
      "loss": 0.9518,
      "step": 100
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.9818406105041504,
      "eval_runtime": 1.3835,
      "eval_samples_per_second": 7.228,
      "eval_steps_per_second": 1.446,
      "step": 100
    },
    {
      "epoch": 21.0,
      "learning_rate": 6.0599999999999996e-05,
      "loss": 0.908,
      "step": 105
    },
    {
      "epoch": 22.0,
      "learning_rate": 6.359999999999999e-05,
      "loss": 0.8678,
      "step": 110
    },
    {
      "epoch": 23.0,
      "learning_rate": 6.659999999999999e-05,
      "loss": 0.8303,
      "step": 115
    },
    {
      "epoch": 24.0,
      "learning_rate": 6.96e-05,
      "loss": 0.7928,
      "step": 120
    },
    {
      "epoch": 25.0,
      "learning_rate": 7.259999999999999e-05,
      "loss": 0.7594,
      "step": 125
    },
    {
      "epoch": 26.0,
      "learning_rate": 7.56e-05,
      "loss": 0.73,
      "step": 130
    },
    {
      "epoch": 27.0,
      "learning_rate": 7.86e-05,
      "loss": 0.7034,
      "step": 135
    },
    {
      "epoch": 28.0,
      "learning_rate": 8.16e-05,
      "loss": 0.6777,
      "step": 140
    },
    {
      "epoch": 29.0,
      "learning_rate": 8.459999999999998e-05,
      "loss": 0.6493,
      "step": 145
    },
    {
      "epoch": 30.0,
      "learning_rate": 8.759999999999999e-05,
      "loss": 0.6249,
      "step": 150
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.8599645495414734,
      "eval_runtime": 1.4099,
      "eval_samples_per_second": 7.093,
      "eval_steps_per_second": 1.419,
      "step": 150
    },
    {
      "epoch": 31.0,
      "learning_rate": 9.059999999999999e-05,
      "loss": 0.6007,
      "step": 155
    },
    {
      "epoch": 32.0,
      "learning_rate": 9.36e-05,
      "loss": 0.5716,
      "step": 160
    },
    {
      "epoch": 33.0,
      "learning_rate": 9.659999999999999e-05,
      "loss": 0.5465,
      "step": 165
    },
    {
      "epoch": 34.0,
      "learning_rate": 9.96e-05,
      "loss": 0.5191,
      "step": 170
    },
    {
      "epoch": 35.0,
      "learning_rate": 0.0001026,
      "loss": 0.4947,
      "step": 175
    },
    {
      "epoch": 36.0,
      "learning_rate": 0.00010559999999999998,
      "loss": 0.4681,
      "step": 180
    },
    {
      "epoch": 37.0,
      "learning_rate": 0.00010859999999999998,
      "loss": 0.4417,
      "step": 185
    },
    {
      "epoch": 38.0,
      "learning_rate": 0.00011159999999999999,
      "loss": 0.4116,
      "step": 190
    },
    {
      "epoch": 39.0,
      "learning_rate": 0.0001146,
      "loss": 0.3804,
      "step": 195
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0001176,
      "loss": 0.3544,
      "step": 200
    },
    {
      "epoch": 40.0,
      "eval_loss": 1.1081293821334839,
      "eval_runtime": 1.3764,
      "eval_samples_per_second": 7.265,
      "eval_steps_per_second": 1.453,
      "step": 200
    },
    {
      "epoch": 41.0,
      "learning_rate": 0.00012059999999999999,
      "loss": 0.3248,
      "step": 205
    },
    {
      "epoch": 42.0,
      "learning_rate": 0.0001236,
      "loss": 0.2931,
      "step": 210
    },
    {
      "epoch": 43.0,
      "learning_rate": 0.0001266,
      "loss": 0.2677,
      "step": 215
    },
    {
      "epoch": 44.0,
      "learning_rate": 0.00012959999999999998,
      "loss": 0.2386,
      "step": 220
    },
    {
      "epoch": 45.0,
      "learning_rate": 0.0001326,
      "loss": 0.2142,
      "step": 225
    },
    {
      "epoch": 46.0,
      "learning_rate": 0.0001356,
      "loss": 0.1932,
      "step": 230
    },
    {
      "epoch": 47.0,
      "learning_rate": 0.0001386,
      "loss": 0.1709,
      "step": 235
    },
    {
      "epoch": 48.0,
      "learning_rate": 0.00014159999999999997,
      "loss": 0.1571,
      "step": 240
    },
    {
      "epoch": 49.0,
      "learning_rate": 0.0001446,
      "loss": 0.1417,
      "step": 245
    },
    {
      "epoch": 50.0,
      "learning_rate": 0.00014759999999999998,
      "loss": 0.1184,
      "step": 250
    },
    {
      "epoch": 50.0,
      "eval_loss": 1.5212451219558716,
      "eval_runtime": 1.3827,
      "eval_samples_per_second": 7.232,
      "eval_steps_per_second": 1.446,
      "step": 250
    },
    {
      "epoch": 51.0,
      "learning_rate": 0.00015059999999999997,
      "loss": 0.1096,
      "step": 255
    },
    {
      "epoch": 52.0,
      "learning_rate": 0.0001536,
      "loss": 0.1037,
      "step": 260
    },
    {
      "epoch": 53.0,
      "learning_rate": 0.00015659999999999998,
      "loss": 0.095,
      "step": 265
    },
    {
      "epoch": 54.0,
      "learning_rate": 0.0001596,
      "loss": 0.0865,
      "step": 270
    },
    {
      "epoch": 55.0,
      "learning_rate": 0.0001626,
      "loss": 0.0808,
      "step": 275
    },
    {
      "epoch": 56.0,
      "learning_rate": 0.0001656,
      "loss": 0.0794,
      "step": 280
    },
    {
      "epoch": 57.0,
      "learning_rate": 0.0001686,
      "loss": 0.075,
      "step": 285
    },
    {
      "epoch": 58.0,
      "learning_rate": 0.00017159999999999997,
      "loss": 0.0726,
      "step": 290
    },
    {
      "epoch": 59.0,
      "learning_rate": 0.00017459999999999996,
      "loss": 0.0696,
      "step": 295
    },
    {
      "epoch": 60.0,
      "learning_rate": 0.00017759999999999998,
      "loss": 0.0665,
      "step": 300
    },
    {
      "epoch": 60.0,
      "eval_loss": 1.7048699855804443,
      "eval_runtime": 1.3753,
      "eval_samples_per_second": 7.271,
      "eval_steps_per_second": 1.454,
      "step": 300
    },
    {
      "epoch": 61.0,
      "learning_rate": 0.00018059999999999997,
      "loss": 0.065,
      "step": 305
    },
    {
      "epoch": 62.0,
      "learning_rate": 0.0001836,
      "loss": 0.0623,
      "step": 310
    },
    {
      "epoch": 63.0,
      "learning_rate": 0.00018659999999999998,
      "loss": 0.0574,
      "step": 315
    },
    {
      "epoch": 64.0,
      "learning_rate": 0.00018959999999999997,
      "loss": 0.0577,
      "step": 320
    },
    {
      "epoch": 65.0,
      "learning_rate": 0.0001926,
      "loss": 0.0597,
      "step": 325
    },
    {
      "epoch": 66.0,
      "learning_rate": 0.00019559999999999998,
      "loss": 0.0546,
      "step": 330
    },
    {
      "epoch": 67.0,
      "learning_rate": 0.0001986,
      "loss": 0.0603,
      "step": 335
    },
    {
      "epoch": 68.0,
      "learning_rate": 0.0002016,
      "loss": 0.0555,
      "step": 340
    },
    {
      "epoch": 69.0,
      "learning_rate": 0.00020459999999999999,
      "loss": 0.0551,
      "step": 345
    },
    {
      "epoch": 70.0,
      "learning_rate": 0.00020759999999999998,
      "loss": 0.0529,
      "step": 350
    },
    {
      "epoch": 70.0,
      "eval_loss": 1.901442527770996,
      "eval_runtime": 1.3785,
      "eval_samples_per_second": 7.254,
      "eval_steps_per_second": 1.451,
      "step": 350
    },
    {
      "epoch": 71.0,
      "learning_rate": 0.00021059999999999997,
      "loss": 0.051,
      "step": 355
    },
    {
      "epoch": 72.0,
      "learning_rate": 0.00021359999999999996,
      "loss": 0.05,
      "step": 360
    },
    {
      "epoch": 73.0,
      "learning_rate": 0.00021659999999999998,
      "loss": 0.0493,
      "step": 365
    },
    {
      "epoch": 74.0,
      "learning_rate": 0.00021959999999999997,
      "loss": 0.0465,
      "step": 370
    },
    {
      "epoch": 75.0,
      "learning_rate": 0.0002226,
      "loss": 0.0504,
      "step": 375
    },
    {
      "epoch": 76.0,
      "learning_rate": 0.00022559999999999998,
      "loss": 0.0491,
      "step": 380
    },
    {
      "epoch": 77.0,
      "learning_rate": 0.00022859999999999997,
      "loss": 0.0485,
      "step": 385
    },
    {
      "epoch": 78.0,
      "learning_rate": 0.0002316,
      "loss": 0.0451,
      "step": 390
    },
    {
      "epoch": 79.0,
      "learning_rate": 0.00023459999999999998,
      "loss": 0.0478,
      "step": 395
    },
    {
      "epoch": 80.0,
      "learning_rate": 0.0002376,
      "loss": 0.0435,
      "step": 400
    },
    {
      "epoch": 80.0,
      "eval_loss": 1.9840269088745117,
      "eval_runtime": 1.3787,
      "eval_samples_per_second": 7.253,
      "eval_steps_per_second": 1.451,
      "step": 400
    },
    {
      "epoch": 81.0,
      "learning_rate": 0.0002406,
      "loss": 0.0429,
      "step": 405
    },
    {
      "epoch": 82.0,
      "learning_rate": 0.00024359999999999999,
      "loss": 0.0464,
      "step": 410
    },
    {
      "epoch": 83.0,
      "learning_rate": 0.0002466,
      "loss": 0.0458,
      "step": 415
    },
    {
      "epoch": 84.0,
      "learning_rate": 0.00024959999999999994,
      "loss": 0.0441,
      "step": 420
    },
    {
      "epoch": 85.0,
      "learning_rate": 0.00025259999999999996,
      "loss": 0.0421,
      "step": 425
    },
    {
      "epoch": 86.0,
      "learning_rate": 0.0002556,
      "loss": 0.0433,
      "step": 430
    },
    {
      "epoch": 87.0,
      "learning_rate": 0.0002586,
      "loss": 0.0444,
      "step": 435
    },
    {
      "epoch": 88.0,
      "learning_rate": 0.00026159999999999996,
      "loss": 0.0472,
      "step": 440
    },
    {
      "epoch": 89.0,
      "learning_rate": 0.0002646,
      "loss": 0.0442,
      "step": 445
    },
    {
      "epoch": 90.0,
      "learning_rate": 0.0002676,
      "loss": 0.0431,
      "step": 450
    },
    {
      "epoch": 90.0,
      "eval_loss": 1.839582085609436,
      "eval_runtime": 1.3848,
      "eval_samples_per_second": 7.221,
      "eval_steps_per_second": 1.444,
      "step": 450
    },
    {
      "epoch": 91.0,
      "learning_rate": 0.00027059999999999996,
      "loss": 0.0431,
      "step": 455
    },
    {
      "epoch": 92.0,
      "learning_rate": 0.0002736,
      "loss": 0.044,
      "step": 460
    },
    {
      "epoch": 93.0,
      "learning_rate": 0.0002766,
      "loss": 0.0429,
      "step": 465
    },
    {
      "epoch": 94.0,
      "learning_rate": 0.00027959999999999997,
      "loss": 0.042,
      "step": 470
    },
    {
      "epoch": 95.0,
      "learning_rate": 0.0002826,
      "loss": 0.0415,
      "step": 475
    },
    {
      "epoch": 96.0,
      "learning_rate": 0.00028559999999999995,
      "loss": 0.0433,
      "step": 480
    },
    {
      "epoch": 97.0,
      "learning_rate": 0.00028859999999999997,
      "loss": 0.0405,
      "step": 485
    },
    {
      "epoch": 98.0,
      "learning_rate": 0.0002916,
      "loss": 0.0416,
      "step": 490
    },
    {
      "epoch": 99.0,
      "learning_rate": 0.00029459999999999995,
      "loss": 0.0378,
      "step": 495
    },
    {
      "epoch": 100.0,
      "learning_rate": 0.00029759999999999997,
      "loss": 0.0379,
      "step": 500
    },
    {
      "epoch": 100.0,
      "eval_loss": 1.9192386865615845,
      "eval_runtime": 1.3756,
      "eval_samples_per_second": 7.269,
      "eval_steps_per_second": 1.454,
      "step": 500
    }
  ],
  "max_steps": 10000,
  "num_train_epochs": 2000,
  "total_flos": 8.0956959522816e+16,
  "trial_name": null,
  "trial_params": null
}