{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.83629191321499,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 6.2827061767850974e-06,
      "loss": 3.8041,
      "step": 10
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.721930086076435e-06,
      "loss": 3.5706,
      "step": 20
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.1424336578396131e-05,
      "loss": 3.2293,
      "step": 30
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.2565412353570195e-05,
      "loss": 2.8198,
      "step": 40
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.3424920790724471e-05,
      "loss": 2.6382,
      "step": 50
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.4114674634318977e-05,
      "loss": 2.5852,
      "step": 60
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.4690794260273606e-05,
      "loss": 2.4718,
      "step": 70
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5185478617142983e-05,
      "loss": 2.3908,
      "step": 80
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.561892416495773e-05,
      "loss": 2.3321,
      "step": 90
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.600463626286153e-05,
      "loss": 2.3622,
      "step": 100
    },
    {
      "epoch": 0.79,
      "eval_loss": 2.0142996311187744,
      "eval_runtime": 13.3285,
      "eval_samples_per_second": 92.958,
      "eval_steps_per_second": 18.607,
      "step": 100
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.6352093070986755e-05,
      "loss": 2.2679,
      "step": 110
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.666820294156779e-05,
      "loss": 2.2493,
      "step": 120
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.6958156901812732e-05,
      "loss": 2.3766,
      "step": 130
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.7225954557575116e-05,
      "loss": 2.1449,
      "step": 140
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.747474292314672e-05,
      "loss": 2.1136,
      "step": 150
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.770704275518123e-05,
      "loss": 2.107,
      "step": 160
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.7924904433402933e-05,
      "loss": 2.1469,
      "step": 170
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.8130018169564944e-05,
      "loss": 2.1229,
      "step": 180
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.832379371735486e-05,
      "loss": 2.0893,
      "step": 190
    },
    {
      "epoch": 1.58,
      "learning_rate": 1.8507419182561513e-05,
      "loss": 2.0836,
      "step": 200
    },
    {
      "epoch": 1.58,
      "eval_loss": 1.813197135925293,
      "eval_runtime": 13.3546,
      "eval_samples_per_second": 92.777,
      "eval_steps_per_second": 18.57,
      "step": 200
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.868190518296623e-05,
      "loss": 2.0307,
      "step": 210
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.8848118530355293e-05,
      "loss": 2.0025,
      "step": 220
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.9006808283301558e-05,
      "loss": 2.0116,
      "step": 230
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.915862615484805e-05,
      "loss": 2.0355,
      "step": 240
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.9304142682139013e-05,
      "loss": 2.0056,
      "step": 250
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.945753054066002e-05,
      "loss": 2.1508,
      "step": 260
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.959138057324679e-05,
      "loss": 1.9203,
      "step": 270
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.9720308518695147e-05,
      "loss": 1.9358,
      "step": 280
    },
    {
      "epoch": 2.3,
      "learning_rate": 1.9844663575115566e-05,
      "loss": 1.877,
      "step": 290
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.9964759049286476e-05,
      "loss": 1.9239,
      "step": 300
    },
    {
      "epoch": 2.38,
      "eval_loss": 1.726140022277832,
      "eval_runtime": 13.3534,
      "eval_samples_per_second": 92.785,
      "eval_steps_per_second": 18.572,
      "step": 300
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.994594594594595e-05,
      "loss": 1.9449,
      "step": 310
    },
    {
      "epoch": 2.54,
      "learning_rate": 1.9855855855855857e-05,
      "loss": 1.895,
      "step": 320
    },
    {
      "epoch": 2.62,
      "learning_rate": 1.9765765765765768e-05,
      "loss": 1.864,
      "step": 330
    },
    {
      "epoch": 2.69,
      "learning_rate": 1.967567567567568e-05,
      "loss": 1.8928,
      "step": 340
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.9585585585585586e-05,
      "loss": 1.8801,
      "step": 350
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.9495495495495497e-05,
      "loss": 1.911,
      "step": 360
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.9405405405405408e-05,
      "loss": 1.8945,
      "step": 370
    },
    {
      "epoch": 3.02,
      "learning_rate": 1.930630630630631e-05,
      "loss": 1.9508,
      "step": 380
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.9216216216216216e-05,
      "loss": 1.7709,
      "step": 390
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.9126126126126127e-05,
      "loss": 1.7693,
      "step": 400
    },
    {
      "epoch": 3.17,
      "eval_loss": 1.673058032989502,
      "eval_runtime": 13.3468,
      "eval_samples_per_second": 92.832,
      "eval_steps_per_second": 18.581,
      "step": 400
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.9036036036036038e-05,
      "loss": 1.7773,
      "step": 410
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.8945945945945945e-05,
      "loss": 1.8198,
      "step": 420
    },
    {
      "epoch": 3.41,
      "learning_rate": 1.8855855855855856e-05,
      "loss": 1.7813,
      "step": 430
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.8765765765765767e-05,
      "loss": 1.7736,
      "step": 440
    },
    {
      "epoch": 3.57,
      "learning_rate": 1.8675675675675678e-05,
      "loss": 1.8544,
      "step": 450
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.8585585585585585e-05,
      "loss": 1.7672,
      "step": 460
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.8495495495495496e-05,
      "loss": 1.8223,
      "step": 470
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.8405405405405407e-05,
      "loss": 1.7832,
      "step": 480
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.8315315315315318e-05,
      "loss": 1.7937,
      "step": 490
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.822522522522523e-05,
      "loss": 1.8078,
      "step": 500
    },
    {
      "epoch": 3.96,
      "eval_loss": 1.6267061233520508,
      "eval_runtime": 13.3535,
      "eval_samples_per_second": 92.784,
      "eval_steps_per_second": 18.572,
      "step": 500
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.8126126126126127e-05,
      "loss": 1.8374,
      "step": 510
    },
    {
      "epoch": 4.13,
      "learning_rate": 1.8036036036036037e-05,
      "loss": 1.7228,
      "step": 520
    },
    {
      "epoch": 4.21,
      "learning_rate": 1.7945945945945948e-05,
      "loss": 1.6986,
      "step": 530
    },
    {
      "epoch": 4.28,
      "learning_rate": 1.7855855855855856e-05,
      "loss": 1.6858,
      "step": 540
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.7765765765765767e-05,
      "loss": 1.7169,
      "step": 550
    },
    {
      "epoch": 4.44,
      "learning_rate": 1.7675675675675677e-05,
      "loss": 1.7069,
      "step": 560
    },
    {
      "epoch": 4.52,
      "learning_rate": 1.7585585585585588e-05,
      "loss": 1.6725,
      "step": 570
    },
    {
      "epoch": 4.6,
      "learning_rate": 1.7495495495495496e-05,
      "loss": 1.7052,
      "step": 580
    },
    {
      "epoch": 4.68,
      "learning_rate": 1.7405405405405406e-05,
      "loss": 1.6597,
      "step": 590
    },
    {
      "epoch": 4.76,
      "learning_rate": 1.7315315315315317e-05,
      "loss": 1.7565,
      "step": 600
    },
    {
      "epoch": 4.76,
      "eval_loss": 1.5994915962219238,
      "eval_runtime": 13.2972,
      "eval_samples_per_second": 93.177,
      "eval_steps_per_second": 18.651,
      "step": 600
    },
    {
      "epoch": 4.84,
      "learning_rate": 1.7225225225225225e-05,
      "loss": 1.723,
      "step": 610
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.7135135135135135e-05,
      "loss": 1.6777,
      "step": 620
    },
    {
      "epoch": 4.99,
      "learning_rate": 1.7045045045045046e-05,
      "loss": 1.6943,
      "step": 630
    },
    {
      "epoch": 5.08,
      "learning_rate": 1.6954954954954957e-05,
      "loss": 1.7274,
      "step": 640
    },
    {
      "epoch": 5.16,
      "learning_rate": 1.6864864864864868e-05,
      "loss": 1.7062,
      "step": 650
    },
    {
      "epoch": 5.24,
      "learning_rate": 1.6774774774774775e-05,
      "loss": 1.6078,
      "step": 660
    },
    {
      "epoch": 5.32,
      "learning_rate": 1.6684684684684686e-05,
      "loss": 1.6484,
      "step": 670
    },
    {
      "epoch": 5.39,
      "learning_rate": 1.6594594594594597e-05,
      "loss": 1.6097,
      "step": 680
    },
    {
      "epoch": 5.47,
      "learning_rate": 1.6504504504504508e-05,
      "loss": 1.6255,
      "step": 690
    },
    {
      "epoch": 5.55,
      "learning_rate": 1.641441441441442e-05,
      "loss": 1.6188,
      "step": 700
    },
    {
      "epoch": 5.55,
      "eval_loss": 1.5767285823822021,
      "eval_runtime": 13.2952,
      "eval_samples_per_second": 93.191,
      "eval_steps_per_second": 18.653,
      "step": 700
    },
    {
      "epoch": 5.63,
      "learning_rate": 1.6324324324324326e-05,
      "loss": 1.68,
      "step": 710
    },
    {
      "epoch": 5.71,
      "learning_rate": 1.6234234234234237e-05,
      "loss": 1.618,
      "step": 720
    },
    {
      "epoch": 5.79,
      "learning_rate": 1.6144144144144144e-05,
      "loss": 1.5747,
      "step": 730
    },
    {
      "epoch": 5.87,
      "learning_rate": 1.6054054054054055e-05,
      "loss": 1.6617,
      "step": 740
    },
    {
      "epoch": 5.95,
      "learning_rate": 1.5963963963963966e-05,
      "loss": 1.6466,
      "step": 750
    },
    {
      "epoch": 6.03,
      "learning_rate": 1.5864864864864867e-05,
      "loss": 1.7621,
      "step": 760
    },
    {
      "epoch": 6.11,
      "learning_rate": 1.5774774774774778e-05,
      "loss": 1.5679,
      "step": 770
    },
    {
      "epoch": 6.19,
      "learning_rate": 1.5684684684684685e-05,
      "loss": 1.6356,
      "step": 780
    },
    {
      "epoch": 6.27,
      "learning_rate": 1.5594594594594596e-05,
      "loss": 1.5499,
      "step": 790
    },
    {
      "epoch": 6.35,
      "learning_rate": 1.5504504504504504e-05,
      "loss": 1.5783,
      "step": 800
    },
    {
      "epoch": 6.35,
      "eval_loss": 1.5619192123413086,
      "eval_runtime": 13.2986,
      "eval_samples_per_second": 93.168,
      "eval_steps_per_second": 18.649,
      "step": 800
    },
    {
      "epoch": 6.43,
      "learning_rate": 1.5414414414414414e-05,
      "loss": 1.5848,
      "step": 810
    },
    {
      "epoch": 6.5,
      "learning_rate": 1.5324324324324325e-05,
      "loss": 1.5019,
      "step": 820
    },
    {
      "epoch": 6.58,
      "learning_rate": 1.5234234234234236e-05,
      "loss": 1.6008,
      "step": 830
    },
    {
      "epoch": 6.66,
      "learning_rate": 1.5144144144144147e-05,
      "loss": 1.5757,
      "step": 840
    },
    {
      "epoch": 6.74,
      "learning_rate": 1.5054054054054054e-05,
      "loss": 1.5855,
      "step": 850
    },
    {
      "epoch": 6.82,
      "learning_rate": 1.4963963963963965e-05,
      "loss": 1.629,
      "step": 860
    },
    {
      "epoch": 6.9,
      "learning_rate": 1.4873873873873874e-05,
      "loss": 1.5948,
      "step": 870
    },
    {
      "epoch": 6.98,
      "learning_rate": 1.4783783783783785e-05,
      "loss": 1.5814,
      "step": 880
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.4684684684684686e-05,
      "loss": 1.7112,
      "step": 890
    },
    {
      "epoch": 7.14,
      "learning_rate": 1.4594594594594596e-05,
      "loss": 1.4981,
      "step": 900
    },
    {
      "epoch": 7.14,
      "eval_loss": 1.546679139137268,
      "eval_runtime": 13.2814,
      "eval_samples_per_second": 93.288,
      "eval_steps_per_second": 18.673,
      "step": 900
    },
    {
      "epoch": 7.22,
      "learning_rate": 1.4504504504504506e-05,
      "loss": 1.5341,
      "step": 910
    },
    {
      "epoch": 7.3,
      "learning_rate": 1.4414414414414416e-05,
      "loss": 1.5301,
      "step": 920
    },
    {
      "epoch": 7.38,
      "learning_rate": 1.4324324324324326e-05,
      "loss": 1.5253,
      "step": 930
    },
    {
      "epoch": 7.46,
      "learning_rate": 1.4234234234234234e-05,
      "loss": 1.5697,
      "step": 940
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.4144144144144145e-05,
      "loss": 1.5482,
      "step": 950
    },
    {
      "epoch": 7.62,
      "learning_rate": 1.4054054054054055e-05,
      "loss": 1.4849,
      "step": 960
    },
    {
      "epoch": 7.69,
      "learning_rate": 1.3963963963963964e-05,
      "loss": 1.536,
      "step": 970
    },
    {
      "epoch": 7.77,
      "learning_rate": 1.3873873873873875e-05,
      "loss": 1.5267,
      "step": 980
    },
    {
      "epoch": 7.85,
      "learning_rate": 1.3783783783783784e-05,
      "loss": 1.5003,
      "step": 990
    },
    {
      "epoch": 7.93,
      "learning_rate": 1.3693693693693695e-05,
      "loss": 1.5296,
      "step": 1000
    },
    {
      "epoch": 7.93,
      "eval_loss": 1.5358564853668213,
      "eval_runtime": 13.2903,
      "eval_samples_per_second": 93.226,
      "eval_steps_per_second": 18.66,
      "step": 1000
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.3594594594594597e-05,
      "loss": 1.6728,
      "step": 1010
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.3504504504504506e-05,
      "loss": 1.4944,
      "step": 1020
    },
    {
      "epoch": 8.17,
      "learning_rate": 1.3414414414414417e-05,
      "loss": 1.5016,
      "step": 1030
    },
    {
      "epoch": 8.25,
      "learning_rate": 1.3324324324324324e-05,
      "loss": 1.5407,
      "step": 1040
    },
    {
      "epoch": 8.33,
      "learning_rate": 1.3234234234234235e-05,
      "loss": 1.5129,
      "step": 1050
    },
    {
      "epoch": 8.41,
      "learning_rate": 1.3144144144144144e-05,
      "loss": 1.4774,
      "step": 1060
    },
    {
      "epoch": 8.49,
      "learning_rate": 1.3054054054054055e-05,
      "loss": 1.5181,
      "step": 1070
    },
    {
      "epoch": 8.57,
      "learning_rate": 1.2963963963963966e-05,
      "loss": 1.4897,
      "step": 1080
    },
    {
      "epoch": 8.65,
      "learning_rate": 1.2873873873873875e-05,
      "loss": 1.4918,
      "step": 1090
    },
    {
      "epoch": 8.73,
      "learning_rate": 1.2783783783783785e-05,
      "loss": 1.4734,
      "step": 1100
    },
    {
      "epoch": 8.73,
      "eval_loss": 1.5299837589263916,
      "eval_runtime": 13.3097,
      "eval_samples_per_second": 93.09,
      "eval_steps_per_second": 18.633,
      "step": 1100
    },
    {
      "epoch": 8.8,
      "learning_rate": 1.2693693693693695e-05,
      "loss": 1.4425,
      "step": 1110
    },
    {
      "epoch": 8.88,
      "learning_rate": 1.2603603603603605e-05,
      "loss": 1.5102,
      "step": 1120
    },
    {
      "epoch": 8.96,
      "learning_rate": 1.2513513513513516e-05,
      "loss": 1.4614,
      "step": 1130
    },
    {
      "epoch": 9.05,
      "learning_rate": 1.2423423423423424e-05,
      "loss": 1.5378,
      "step": 1140
    },
    {
      "epoch": 9.13,
      "learning_rate": 1.2333333333333334e-05,
      "loss": 1.5182,
      "step": 1150
    },
    {
      "epoch": 9.21,
      "learning_rate": 1.2243243243243244e-05,
      "loss": 1.4606,
      "step": 1160
    },
    {
      "epoch": 9.28,
      "learning_rate": 1.2153153153153154e-05,
      "loss": 1.4675,
      "step": 1170
    },
    {
      "epoch": 9.36,
      "learning_rate": 1.2063063063063063e-05,
      "loss": 1.5261,
      "step": 1180
    },
    {
      "epoch": 9.44,
      "learning_rate": 1.1972972972972974e-05,
      "loss": 1.4602,
      "step": 1190
    },
    {
      "epoch": 9.52,
      "learning_rate": 1.1882882882882885e-05,
      "loss": 1.4415,
      "step": 1200
    },
    {
      "epoch": 9.52,
      "eval_loss": 1.521682620048523,
      "eval_runtime": 13.2939,
      "eval_samples_per_second": 93.2,
      "eval_steps_per_second": 18.655,
      "step": 1200
    },
    {
      "epoch": 9.6,
      "learning_rate": 1.1792792792792792e-05,
      "loss": 1.4601,
      "step": 1210
    },
    {
      "epoch": 9.68,
      "learning_rate": 1.1702702702702703e-05,
      "loss": 1.4676,
      "step": 1220
    },
    {
      "epoch": 9.76,
      "learning_rate": 1.1612612612612612e-05,
      "loss": 1.4213,
      "step": 1230
    },
    {
      "epoch": 9.84,
      "learning_rate": 1.1522522522522523e-05,
      "loss": 1.4096,
      "step": 1240
    },
    {
      "epoch": 9.92,
      "learning_rate": 1.1432432432432434e-05,
      "loss": 1.4475,
      "step": 1250
    },
    {
      "epoch": 9.99,
      "learning_rate": 1.1342342342342343e-05,
      "loss": 1.4477,
      "step": 1260
    },
    {
      "epoch": 10.08,
      "learning_rate": 1.1243243243243245e-05,
      "loss": 1.4979,
      "step": 1270
    },
    {
      "epoch": 10.16,
      "learning_rate": 1.1153153153153154e-05,
      "loss": 1.4436,
      "step": 1280
    },
    {
      "epoch": 10.24,
      "learning_rate": 1.1063063063063065e-05,
      "loss": 1.4913,
      "step": 1290
    },
    {
      "epoch": 10.32,
      "learning_rate": 1.0972972972972974e-05,
      "loss": 1.4513,
      "step": 1300
    },
    {
      "epoch": 10.32,
      "eval_loss": 1.5171560049057007,
      "eval_runtime": 13.2984,
      "eval_samples_per_second": 93.169,
      "eval_steps_per_second": 18.649,
      "step": 1300
    },
    {
      "epoch": 10.39,
      "learning_rate": 1.0882882882882884e-05,
      "loss": 1.4649,
      "step": 1310
    },
    {
      "epoch": 10.47,
      "learning_rate": 1.0792792792792795e-05,
      "loss": 1.4126,
      "step": 1320
    },
    {
      "epoch": 10.55,
      "learning_rate": 1.0702702702702703e-05,
      "loss": 1.3916,
      "step": 1330
    },
    {
      "epoch": 10.63,
      "learning_rate": 1.0612612612612613e-05,
      "loss": 1.393,
      "step": 1340
    },
    {
      "epoch": 10.71,
      "learning_rate": 1.0522522522522523e-05,
      "loss": 1.3972,
      "step": 1350
    },
    {
      "epoch": 10.79,
      "learning_rate": 1.0432432432432433e-05,
      "loss": 1.4867,
      "step": 1360
    },
    {
      "epoch": 10.87,
      "learning_rate": 1.0342342342342344e-05,
      "loss": 1.4109,
      "step": 1370
    },
    {
      "epoch": 10.95,
      "learning_rate": 1.0252252252252253e-05,
      "loss": 1.4215,
      "step": 1380
    },
    {
      "epoch": 11.03,
      "learning_rate": 1.0153153153153155e-05,
      "loss": 1.5288,
      "step": 1390
    },
    {
      "epoch": 11.11,
      "learning_rate": 1.0063063063063064e-05,
      "loss": 1.3782,
      "step": 1400
    },
    {
      "epoch": 11.11,
      "eval_loss": 1.5077377557754517,
      "eval_runtime": 13.2758,
      "eval_samples_per_second": 93.328,
      "eval_steps_per_second": 18.681,
      "step": 1400
    },
    {
      "epoch": 11.19,
      "learning_rate": 9.972972972972975e-06,
      "loss": 1.4292,
      "step": 1410
    },
    {
      "epoch": 11.27,
      "learning_rate": 9.882882882882884e-06,
      "loss": 1.4404,
      "step": 1420
    },
    {
      "epoch": 11.35,
      "learning_rate": 9.792792792792793e-06,
      "loss": 1.4192,
      "step": 1430
    },
    {
      "epoch": 11.43,
      "learning_rate": 9.702702702702704e-06,
      "loss": 1.324,
      "step": 1440
    },
    {
      "epoch": 11.5,
      "learning_rate": 9.612612612612613e-06,
      "loss": 1.4342,
      "step": 1450
    },
    {
      "epoch": 11.58,
      "learning_rate": 9.522522522522524e-06,
      "loss": 1.4233,
      "step": 1460
    },
    {
      "epoch": 11.66,
      "learning_rate": 9.432432432432433e-06,
      "loss": 1.3475,
      "step": 1470
    },
    {
      "epoch": 11.74,
      "learning_rate": 9.342342342342344e-06,
      "loss": 1.4178,
      "step": 1480
    },
    {
      "epoch": 11.82,
      "learning_rate": 9.252252252252253e-06,
      "loss": 1.3842,
      "step": 1490
    },
    {
      "epoch": 11.9,
      "learning_rate": 9.162162162162162e-06,
      "loss": 1.4103,
      "step": 1500
    },
    {
      "epoch": 11.9,
      "eval_loss": 1.507853388786316,
      "eval_runtime": 13.2911,
      "eval_samples_per_second": 93.22,
      "eval_steps_per_second": 18.659,
      "step": 1500
    },
    {
      "epoch": 11.98,
      "learning_rate": 9.072072072072073e-06,
      "loss": 1.3918,
      "step": 1510
    },
    {
      "epoch": 12.06,
      "learning_rate": 8.972972972972974e-06,
      "loss": 1.4841,
      "step": 1520
    },
    {
      "epoch": 12.14,
      "learning_rate": 8.882882882882883e-06,
      "loss": 1.3866,
      "step": 1530
    },
    {
      "epoch": 12.22,
      "learning_rate": 8.792792792792794e-06,
      "loss": 1.3713,
      "step": 1540
    },
    {
      "epoch": 12.3,
      "learning_rate": 8.702702702702703e-06,
      "loss": 1.3384,
      "step": 1550
    },
    {
      "epoch": 12.38,
      "learning_rate": 8.612612612612612e-06,
      "loss": 1.4079,
      "step": 1560
    },
    {
      "epoch": 12.46,
      "learning_rate": 8.522522522522523e-06,
      "loss": 1.3715,
      "step": 1570
    },
    {
      "epoch": 12.54,
      "learning_rate": 8.432432432432434e-06,
      "loss": 1.3541,
      "step": 1580
    },
    {
      "epoch": 12.62,
      "learning_rate": 8.342342342342343e-06,
      "loss": 1.3764,
      "step": 1590
    },
    {
      "epoch": 12.69,
      "learning_rate": 8.252252252252254e-06,
      "loss": 1.3907,
      "step": 1600
    },
    {
      "epoch": 12.69,
      "eval_loss": 1.5033966302871704,
      "eval_runtime": 13.2864,
      "eval_samples_per_second": 93.254,
      "eval_steps_per_second": 18.666,
      "step": 1600
    },
    {
      "epoch": 12.77,
      "learning_rate": 8.162162162162163e-06,
      "loss": 1.3626,
      "step": 1610
    },
    {
      "epoch": 12.85,
      "learning_rate": 8.072072072072072e-06,
      "loss": 1.4094,
      "step": 1620
    },
    {
      "epoch": 12.93,
      "learning_rate": 7.981981981981983e-06,
      "loss": 1.3579,
      "step": 1630
    },
    {
      "epoch": 13.02,
      "learning_rate": 7.891891891891894e-06,
      "loss": 1.4517,
      "step": 1640
    },
    {
      "epoch": 13.09,
      "learning_rate": 7.801801801801803e-06,
      "loss": 1.301,
      "step": 1650
    },
    {
      "epoch": 13.17,
      "learning_rate": 7.711711711711712e-06,
      "loss": 1.3513,
      "step": 1660
    },
    {
      "epoch": 13.25,
      "learning_rate": 7.621621621621622e-06,
      "loss": 1.3487,
      "step": 1670
    },
    {
      "epoch": 13.33,
      "learning_rate": 7.531531531531532e-06,
      "loss": 1.3894,
      "step": 1680
    },
    {
      "epoch": 13.41,
      "learning_rate": 7.441441441441442e-06,
      "loss": 1.3619,
      "step": 1690
    },
    {
      "epoch": 13.49,
      "learning_rate": 7.3513513513513525e-06,
      "loss": 1.3663,
      "step": 1700
    },
    {
      "epoch": 13.49,
      "eval_loss": 1.501574993133545,
      "eval_runtime": 13.3092,
      "eval_samples_per_second": 93.094,
      "eval_steps_per_second": 18.634,
      "step": 1700
    },
    {
      "epoch": 13.57,
      "learning_rate": 7.2612612612612625e-06,
      "loss": 1.3268,
      "step": 1710
    },
    {
      "epoch": 13.65,
      "learning_rate": 7.1711711711711716e-06,
      "loss": 1.3703,
      "step": 1720
    },
    {
      "epoch": 13.73,
      "learning_rate": 7.0810810810810815e-06,
      "loss": 1.4246,
      "step": 1730
    },
    {
      "epoch": 13.8,
      "learning_rate": 6.9909909909909915e-06,
      "loss": 1.3642,
      "step": 1740
    },
    {
      "epoch": 13.88,
      "learning_rate": 6.900900900900901e-06,
      "loss": 1.3467,
      "step": 1750
    },
    {
      "epoch": 13.96,
      "learning_rate": 6.810810810810811e-06,
      "loss": 1.3802,
      "step": 1760
    },
    {
      "epoch": 14.05,
      "learning_rate": 6.711711711711713e-06,
      "loss": 1.4328,
      "step": 1770
    },
    {
      "epoch": 14.13,
      "learning_rate": 6.621621621621622e-06,
      "loss": 1.3316,
      "step": 1780
    },
    {
      "epoch": 14.21,
      "learning_rate": 6.531531531531532e-06,
      "loss": 1.3634,
      "step": 1790
    },
    {
      "epoch": 14.28,
      "learning_rate": 6.441441441441442e-06,
      "loss": 1.3565,
      "step": 1800
    },
    {
      "epoch": 14.28,
      "eval_loss": 1.4980181455612183,
      "eval_runtime": 13.2875,
      "eval_samples_per_second": 93.246,
      "eval_steps_per_second": 18.664,
      "step": 1800
    },
    {
      "epoch": 14.36,
      "learning_rate": 6.351351351351351e-06,
      "loss": 1.3217,
      "step": 1810
    },
    {
      "epoch": 14.44,
      "learning_rate": 6.261261261261262e-06,
      "loss": 1.33,
      "step": 1820
    },
    {
      "epoch": 14.52,
      "learning_rate": 6.171171171171172e-06,
      "loss": 1.3513,
      "step": 1830
    },
    {
      "epoch": 14.6,
      "learning_rate": 6.081081081081082e-06,
      "loss": 1.3649,
      "step": 1840
    },
    {
      "epoch": 14.68,
      "learning_rate": 5.990990990990992e-06,
      "loss": 1.3462,
      "step": 1850
    },
    {
      "epoch": 14.76,
      "learning_rate": 5.900900900900901e-06,
      "loss": 1.3454,
      "step": 1860
    },
    {
      "epoch": 14.84,
      "learning_rate": 5.810810810810811e-06,
      "loss": 1.3316,
      "step": 1870
    },
    {
      "epoch": 14.92,
      "learning_rate": 5.720720720720722e-06,
      "loss": 1.3347,
      "step": 1880
    },
    {
      "epoch": 14.99,
      "learning_rate": 5.6306306306306316e-06,
      "loss": 1.3039,
      "step": 1890
    },
    {
      "epoch": 15.08,
      "learning_rate": 5.531531531531532e-06,
      "loss": 1.4057,
      "step": 1900
    },
    {
      "epoch": 15.08,
      "eval_loss": 1.4985554218292236,
      "eval_runtime": 13.2747,
      "eval_samples_per_second": 93.335,
      "eval_steps_per_second": 18.682,
      "step": 1900
    },
    {
      "epoch": 15.16,
      "learning_rate": 5.441441441441442e-06,
      "loss": 1.3082,
      "step": 1910
    },
    {
      "epoch": 15.24,
      "learning_rate": 5.351351351351351e-06,
      "loss": 1.3589,
      "step": 1920
    },
    {
      "epoch": 15.32,
      "learning_rate": 5.261261261261261e-06,
      "loss": 1.3235,
      "step": 1930
    },
    {
      "epoch": 15.39,
      "learning_rate": 5.171171171171172e-06,
      "loss": 1.3153,
      "step": 1940
    },
    {
      "epoch": 15.47,
      "learning_rate": 5.081081081081082e-06,
      "loss": 1.3345,
      "step": 1950
    },
    {
      "epoch": 15.55,
      "learning_rate": 4.990990990990991e-06,
      "loss": 1.3824,
      "step": 1960
    },
    {
      "epoch": 15.63,
      "learning_rate": 4.900900900900901e-06,
      "loss": 1.291,
      "step": 1970
    },
    {
      "epoch": 15.71,
      "learning_rate": 4.810810810810811e-06,
      "loss": 1.3106,
      "step": 1980
    },
    {
      "epoch": 15.79,
      "learning_rate": 4.720720720720721e-06,
      "loss": 1.3559,
      "step": 1990
    },
    {
      "epoch": 15.87,
      "learning_rate": 4.630630630630631e-06,
      "loss": 1.3406,
      "step": 2000
    },
    {
      "epoch": 15.87,
      "eval_loss": 1.4952143430709839,
      "eval_runtime": 13.2782,
      "eval_samples_per_second": 93.31,
      "eval_steps_per_second": 18.677,
      "step": 2000
    },
    {
      "epoch": 15.95,
      "learning_rate": 4.540540540540541e-06,
      "loss": 1.328,
      "step": 2010
    },
    {
      "epoch": 16.03,
      "learning_rate": 4.441441441441442e-06,
      "loss": 1.4218,
      "step": 2020
    },
    {
      "epoch": 16.11,
      "learning_rate": 4.351351351351352e-06,
      "loss": 1.2962,
      "step": 2030
    },
    {
      "epoch": 16.19,
      "learning_rate": 4.2612612612612615e-06,
      "loss": 1.3122,
      "step": 2040
    },
    {
      "epoch": 16.27,
      "learning_rate": 4.1711711711711715e-06,
      "loss": 1.3641,
      "step": 2050
    },
    {
      "epoch": 16.35,
      "learning_rate": 4.0810810810810815e-06,
      "loss": 1.3058,
      "step": 2060
    },
    {
      "epoch": 16.43,
      "learning_rate": 3.990990990990991e-06,
      "loss": 1.2986,
      "step": 2070
    },
    {
      "epoch": 16.5,
      "learning_rate": 3.900900900900901e-06,
      "loss": 1.2902,
      "step": 2080
    },
    {
      "epoch": 16.58,
      "learning_rate": 3.810810810810811e-06,
      "loss": 1.3725,
      "step": 2090
    },
    {
      "epoch": 16.66,
      "learning_rate": 3.720720720720721e-06,
      "loss": 1.3031,
      "step": 2100
    },
    {
      "epoch": 16.66,
      "eval_loss": 1.495803713798523,
      "eval_runtime": 13.2846,
      "eval_samples_per_second": 93.266,
      "eval_steps_per_second": 18.668,
      "step": 2100
    },
    {
      "epoch": 16.74,
      "learning_rate": 3.6306306306306312e-06,
      "loss": 1.3091,
      "step": 2110
    },
    {
      "epoch": 16.82,
      "learning_rate": 3.5405405405405408e-06,
      "loss": 1.3003,
      "step": 2120
    },
    {
      "epoch": 16.9,
      "learning_rate": 3.4504504504504503e-06,
      "loss": 1.2694,
      "step": 2130
    },
    {
      "epoch": 16.98,
      "learning_rate": 3.3603603603603607e-06,
      "loss": 1.3349,
      "step": 2140
    },
    {
      "epoch": 17.06,
      "learning_rate": 3.2702702702702706e-06,
      "loss": 1.3553,
      "step": 2150
    },
    {
      "epoch": 17.14,
      "learning_rate": 3.1801801801801806e-06,
      "loss": 1.2844,
      "step": 2160
    },
    {
      "epoch": 17.22,
      "learning_rate": 3.0900900900900905e-06,
      "loss": 1.2815,
      "step": 2170
    },
    {
      "epoch": 17.3,
      "learning_rate": 3e-06,
      "loss": 1.2756,
      "step": 2180
    },
    {
      "epoch": 17.38,
      "learning_rate": 2.9099099099099105e-06,
      "loss": 1.33,
      "step": 2190
    },
    {
      "epoch": 17.46,
      "learning_rate": 2.81981981981982e-06,
      "loss": 1.31,
      "step": 2200
    },
    {
      "epoch": 17.46,
      "eval_loss": 1.4959148168563843,
      "eval_runtime": 13.2926,
      "eval_samples_per_second": 93.21,
      "eval_steps_per_second": 18.657,
      "step": 2200
    },
    {
      "epoch": 17.54,
      "learning_rate": 2.72972972972973e-06,
      "loss": 1.3306,
      "step": 2210
    },
    {
      "epoch": 17.62,
      "learning_rate": 2.63963963963964e-06,
      "loss": 1.331,
      "step": 2220
    },
    {
      "epoch": 17.69,
      "learning_rate": 2.54954954954955e-06,
      "loss": 1.3347,
      "step": 2230
    },
    {
      "epoch": 17.77,
      "learning_rate": 2.45945945945946e-06,
      "loss": 1.3291,
      "step": 2240
    },
    {
      "epoch": 17.85,
      "learning_rate": 2.3693693693693693e-06,
      "loss": 1.2774,
      "step": 2250
    },
    {
      "epoch": 17.93,
      "learning_rate": 2.2792792792792793e-06,
      "loss": 1.3102,
      "step": 2260
    },
    {
      "epoch": 18.02,
      "learning_rate": 2.1801801801801804e-06,
      "loss": 1.4393,
      "step": 2270
    },
    {
      "epoch": 18.09,
      "learning_rate": 2.0900900900900904e-06,
      "loss": 1.2779,
      "step": 2280
    },
    {
      "epoch": 18.17,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.2708,
      "step": 2290
    },
    {
      "epoch": 18.25,
      "learning_rate": 1.90990990990991e-06,
      "loss": 1.3565,
      "step": 2300
    },
    {
      "epoch": 18.25,
      "eval_loss": 1.494253396987915,
      "eval_runtime": 13.2974,
      "eval_samples_per_second": 93.176,
      "eval_steps_per_second": 18.65,
      "step": 2300
    },
    {
      "epoch": 18.33,
      "learning_rate": 1.81981981981982e-06,
      "loss": 1.2744,
      "step": 2310
    },
    {
      "epoch": 18.41,
      "learning_rate": 1.72972972972973e-06,
      "loss": 1.2697,
      "step": 2320
    },
    {
      "epoch": 18.49,
      "learning_rate": 1.6396396396396397e-06,
      "loss": 1.2921,
      "step": 2330
    },
    {
      "epoch": 18.57,
      "learning_rate": 1.5495495495495497e-06,
      "loss": 1.3096,
      "step": 2340
    },
    {
      "epoch": 18.65,
      "learning_rate": 1.4594594594594596e-06,
      "loss": 1.2971,
      "step": 2350
    },
    {
      "epoch": 18.73,
      "learning_rate": 1.3693693693693694e-06,
      "loss": 1.3132,
      "step": 2360
    },
    {
      "epoch": 18.8,
      "learning_rate": 1.2792792792792793e-06,
      "loss": 1.3367,
      "step": 2370
    },
    {
      "epoch": 18.88,
      "learning_rate": 1.1891891891891893e-06,
      "loss": 1.2648,
      "step": 2380
    },
    {
      "epoch": 18.96,
      "learning_rate": 1.0990990990990993e-06,
      "loss": 1.3025,
      "step": 2390
    },
    {
      "epoch": 19.05,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.3732,
      "step": 2400
    },
    {
      "epoch": 19.05,
      "eval_loss": 1.4953521490097046,
      "eval_runtime": 13.2927,
      "eval_samples_per_second": 93.209,
      "eval_steps_per_second": 18.657,
      "step": 2400
    },
    {
      "epoch": 19.13,
      "learning_rate": 9.0990990990991e-07,
      "loss": 1.2938,
      "step": 2410
    },
    {
      "epoch": 19.21,
      "learning_rate": 8.198198198198199e-07,
      "loss": 1.3374,
      "step": 2420
    },
    {
      "epoch": 19.28,
      "learning_rate": 7.297297297297298e-07,
      "loss": 1.297,
      "step": 2430
    },
    {
      "epoch": 19.36,
      "learning_rate": 6.396396396396397e-07,
      "loss": 1.2748,
      "step": 2440
    },
    {
      "epoch": 19.44,
      "learning_rate": 5.495495495495496e-07,
      "loss": 1.3259,
      "step": 2450
    },
    {
      "epoch": 19.52,
      "learning_rate": 4.5945945945945953e-07,
      "loss": 1.3099,
      "step": 2460
    },
    {
      "epoch": 19.6,
      "learning_rate": 3.693693693693694e-07,
      "loss": 1.3277,
      "step": 2470
    },
    {
      "epoch": 19.68,
      "learning_rate": 2.792792792792793e-07,
      "loss": 1.2708,
      "step": 2480
    },
    {
      "epoch": 19.76,
      "learning_rate": 1.8918918918918921e-07,
      "loss": 1.2546,
      "step": 2490
    },
    {
      "epoch": 19.84,
      "learning_rate": 9.90990990990991e-08,
      "loss": 1.2705,
      "step": 2500
    },
    {
      "epoch": 19.84,
      "eval_loss": 1.494584560394287,
      "eval_runtime": 13.2864,
      "eval_samples_per_second": 93.253,
      "eval_steps_per_second": 18.666,
      "step": 2500
    }
  ],
  "max_steps": 2520,
  "num_train_epochs": 20,
  "total_flos": 5.993865079244718e+17,
  "trial_name": null,
  "trial_params": null
}