{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.997992236648374,
  "global_step": 4660,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 5.987339055793991e-06,
      "loss": 1.5226,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 5.974678111587983e-06,
      "loss": 1.5251,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.9620171673819746e-06,
      "loss": 1.5261,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.949356223175966e-06,
      "loss": 1.5257,
      "step": 40
    },
    {
      "epoch": 0.11,
      "learning_rate": 5.936695278969957e-06,
      "loss": 1.5235,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 5.924034334763948e-06,
      "loss": 1.5214,
      "step": 60
    },
    {
      "epoch": 0.15,
      "learning_rate": 5.91137339055794e-06,
      "loss": 1.5245,
      "step": 70
    },
    {
      "epoch": 0.17,
      "learning_rate": 5.898712446351932e-06,
      "loss": 1.5259,
      "step": 80
    },
    {
      "epoch": 0.19,
      "learning_rate": 5.8860515021459225e-06,
      "loss": 1.5223,
      "step": 90
    },
    {
      "epoch": 0.21,
      "learning_rate": 5.873390557939914e-06,
      "loss": 1.522,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 5.860729613733905e-06,
      "loss": 1.5269,
      "step": 110
    },
    {
      "epoch": 0.26,
      "learning_rate": 5.848068669527898e-06,
      "loss": 1.5226,
      "step": 120
    },
    {
      "epoch": 0.28,
      "learning_rate": 5.835407725321889e-06,
      "loss": 1.5234,
      "step": 130
    },
    {
      "epoch": 0.3,
      "learning_rate": 5.8227467811158795e-06,
      "loss": 1.5268,
      "step": 140
    },
    {
      "epoch": 0.32,
      "learning_rate": 5.810085836909871e-06,
      "loss": 1.5222,
      "step": 150
    },
    {
      "epoch": 0.34,
      "learning_rate": 5.797424892703863e-06,
      "loss": 1.5245,
      "step": 160
    },
    {
      "epoch": 0.34,
      "eval_loss": 1.4186874628067017,
      "eval_runtime": 78.1328,
      "eval_samples_per_second": 511.949,
      "eval_steps_per_second": 7.999,
      "step": 160
    },
    {
      "epoch": 0.36,
      "learning_rate": 5.784763948497855e-06,
      "loss": 1.5271,
      "step": 170
    },
    {
      "epoch": 0.39,
      "learning_rate": 5.772103004291846e-06,
      "loss": 1.5254,
      "step": 180
    },
    {
      "epoch": 0.41,
      "learning_rate": 5.759442060085837e-06,
      "loss": 1.5261,
      "step": 190
    },
    {
      "epoch": 0.43,
      "learning_rate": 5.746781115879828e-06,
      "loss": 1.5286,
      "step": 200
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.73412017167382e-06,
      "loss": 1.5227,
      "step": 210
    },
    {
      "epoch": 0.47,
      "learning_rate": 5.721459227467812e-06,
      "loss": 1.5246,
      "step": 220
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.708798283261803e-06,
      "loss": 1.5228,
      "step": 230
    },
    {
      "epoch": 0.51,
      "learning_rate": 5.696137339055794e-06,
      "loss": 1.5238,
      "step": 240
    },
    {
      "epoch": 0.54,
      "learning_rate": 5.683476394849785e-06,
      "loss": 1.523,
      "step": 250
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.670815450643777e-06,
      "loss": 1.5238,
      "step": 260
    },
    {
      "epoch": 0.58,
      "learning_rate": 5.658154506437769e-06,
      "loss": 1.5252,
      "step": 270
    },
    {
      "epoch": 0.6,
      "learning_rate": 5.64549356223176e-06,
      "loss": 1.5234,
      "step": 280
    },
    {
      "epoch": 0.62,
      "learning_rate": 5.632832618025751e-06,
      "loss": 1.527,
      "step": 290
    },
    {
      "epoch": 0.64,
      "learning_rate": 5.6201716738197424e-06,
      "loss": 1.5253,
      "step": 300
    },
    {
      "epoch": 0.66,
      "learning_rate": 5.607510729613734e-06,
      "loss": 1.5302,
      "step": 310
    },
    {
      "epoch": 0.69,
      "learning_rate": 5.594849785407725e-06,
      "loss": 1.5245,
      "step": 320
    },
    {
      "epoch": 0.69,
      "eval_loss": 1.4183368682861328,
      "eval_runtime": 78.1992,
      "eval_samples_per_second": 511.514,
      "eval_steps_per_second": 7.992,
      "step": 320
    },
    {
      "epoch": 0.71,
      "learning_rate": 5.582188841201717e-06,
      "loss": 1.5235,
      "step": 330
    },
    {
      "epoch": 0.73,
      "learning_rate": 5.569527896995708e-06,
      "loss": 1.5255,
      "step": 340
    },
    {
      "epoch": 0.75,
      "learning_rate": 5.5568669527897e-06,
      "loss": 1.5268,
      "step": 350
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.544206008583691e-06,
      "loss": 1.5232,
      "step": 360
    },
    {
      "epoch": 0.79,
      "learning_rate": 5.531545064377683e-06,
      "loss": 1.5254,
      "step": 370
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.518884120171674e-06,
      "loss": 1.5265,
      "step": 380
    },
    {
      "epoch": 0.84,
      "learning_rate": 5.506223175965666e-06,
      "loss": 1.5246,
      "step": 390
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.493562231759657e-06,
      "loss": 1.5238,
      "step": 400
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.480901287553648e-06,
      "loss": 1.5267,
      "step": 410
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.468240343347639e-06,
      "loss": 1.5257,
      "step": 420
    },
    {
      "epoch": 0.92,
      "learning_rate": 5.455579399141631e-06,
      "loss": 1.5254,
      "step": 430
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.442918454935623e-06,
      "loss": 1.5242,
      "step": 440
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.4302575107296144e-06,
      "loss": 1.5268,
      "step": 450
    },
    {
      "epoch": 0.99,
      "learning_rate": 5.417596566523605e-06,
      "loss": 1.5252,
      "step": 460
    },
    {
      "epoch": 1.01,
      "learning_rate": 5.404935622317596e-06,
      "loss": 1.6674,
      "step": 470
    },
    {
      "epoch": 1.03,
      "learning_rate": 5.392274678111588e-06,
      "loss": 1.5259,
      "step": 480
    },
    {
      "epoch": 1.03,
      "eval_loss": 1.4177254438400269,
      "eval_runtime": 78.2595,
      "eval_samples_per_second": 511.12,
      "eval_steps_per_second": 7.986,
      "step": 480
    },
    {
      "epoch": 1.05,
      "learning_rate": 5.37961373390558e-06,
      "loss": 1.5217,
      "step": 490
    },
    {
      "epoch": 1.07,
      "learning_rate": 5.3669527896995715e-06,
      "loss": 1.5261,
      "step": 500
    },
    {
      "epoch": 1.09,
      "learning_rate": 5.354291845493562e-06,
      "loss": 1.5241,
      "step": 510
    },
    {
      "epoch": 1.12,
      "learning_rate": 5.341630901287553e-06,
      "loss": 1.5241,
      "step": 520
    },
    {
      "epoch": 1.14,
      "learning_rate": 5.328969957081545e-06,
      "loss": 1.5233,
      "step": 530
    },
    {
      "epoch": 1.16,
      "learning_rate": 5.316309012875537e-06,
      "loss": 1.5245,
      "step": 540
    },
    {
      "epoch": 1.18,
      "learning_rate": 5.303648068669528e-06,
      "loss": 1.5237,
      "step": 550
    },
    {
      "epoch": 1.2,
      "learning_rate": 5.2909871244635194e-06,
      "loss": 1.5233,
      "step": 560
    },
    {
      "epoch": 1.22,
      "learning_rate": 5.27832618025751e-06,
      "loss": 1.5218,
      "step": 570
    },
    {
      "epoch": 1.24,
      "learning_rate": 5.265665236051502e-06,
      "loss": 1.5244,
      "step": 580
    },
    {
      "epoch": 1.27,
      "learning_rate": 5.253004291845494e-06,
      "loss": 1.5228,
      "step": 590
    },
    {
      "epoch": 1.29,
      "learning_rate": 5.2403433476394856e-06,
      "loss": 1.5268,
      "step": 600
    },
    {
      "epoch": 1.31,
      "learning_rate": 5.2276824034334765e-06,
      "loss": 1.5246,
      "step": 610
    },
    {
      "epoch": 1.33,
      "learning_rate": 5.215021459227467e-06,
      "loss": 1.5205,
      "step": 620
    },
    {
      "epoch": 1.35,
      "learning_rate": 5.20236051502146e-06,
      "loss": 1.5252,
      "step": 630
    },
    {
      "epoch": 1.37,
      "learning_rate": 5.189699570815451e-06,
      "loss": 1.5265,
      "step": 640
    },
    {
      "epoch": 1.37,
      "eval_loss": 1.4185311794281006,
      "eval_runtime": 78.202,
      "eval_samples_per_second": 511.496,
      "eval_steps_per_second": 7.992,
      "step": 640
    },
    {
      "epoch": 1.39,
      "learning_rate": 5.177038626609442e-06,
      "loss": 1.5229,
      "step": 650
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.1643776824034335e-06,
      "loss": 1.5246,
      "step": 660
    },
    {
      "epoch": 1.44,
      "learning_rate": 5.151716738197425e-06,
      "loss": 1.5208,
      "step": 670
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.139055793991416e-06,
      "loss": 1.523,
      "step": 680
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.126394849785408e-06,
      "loss": 1.5222,
      "step": 690
    },
    {
      "epoch": 1.5,
      "learning_rate": 5.113733905579399e-06,
      "loss": 1.5223,
      "step": 700
    },
    {
      "epoch": 1.52,
      "learning_rate": 5.1010729613733906e-06,
      "loss": 1.5206,
      "step": 710
    },
    {
      "epoch": 1.54,
      "learning_rate": 5.088412017167382e-06,
      "loss": 1.5194,
      "step": 720
    },
    {
      "epoch": 1.57,
      "learning_rate": 5.075751072961374e-06,
      "loss": 1.5249,
      "step": 730
    },
    {
      "epoch": 1.59,
      "learning_rate": 5.063090128755365e-06,
      "loss": 1.5254,
      "step": 740
    },
    {
      "epoch": 1.61,
      "learning_rate": 5.050429184549356e-06,
      "loss": 1.521,
      "step": 750
    },
    {
      "epoch": 1.63,
      "learning_rate": 5.037768240343348e-06,
      "loss": 1.5255,
      "step": 760
    },
    {
      "epoch": 1.65,
      "learning_rate": 5.025107296137339e-06,
      "loss": 1.5238,
      "step": 770
    },
    {
      "epoch": 1.67,
      "learning_rate": 5.01244635193133e-06,
      "loss": 1.5261,
      "step": 780
    },
    {
      "epoch": 1.69,
      "learning_rate": 4.999785407725322e-06,
      "loss": 1.5233,
      "step": 790
    },
    {
      "epoch": 1.72,
      "learning_rate": 4.987124463519313e-06,
      "loss": 1.5245,
      "step": 800
    },
    {
      "epoch": 1.72,
      "eval_loss": 1.4190433025360107,
      "eval_runtime": 78.1942,
      "eval_samples_per_second": 511.547,
      "eval_steps_per_second": 7.993,
      "step": 800
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.974463519313305e-06,
      "loss": 1.5237,
      "step": 810
    },
    {
      "epoch": 1.76,
      "learning_rate": 4.961802575107296e-06,
      "loss": 1.5216,
      "step": 820
    },
    {
      "epoch": 1.78,
      "learning_rate": 4.949141630901288e-06,
      "loss": 1.5256,
      "step": 830
    },
    {
      "epoch": 1.8,
      "learning_rate": 4.936480686695279e-06,
      "loss": 1.5249,
      "step": 840
    },
    {
      "epoch": 1.82,
      "learning_rate": 4.92381974248927e-06,
      "loss": 1.5235,
      "step": 850
    },
    {
      "epoch": 1.84,
      "learning_rate": 4.9111587982832626e-06,
      "loss": 1.5255,
      "step": 860
    },
    {
      "epoch": 1.87,
      "learning_rate": 4.8984978540772535e-06,
      "loss": 1.5242,
      "step": 870
    },
    {
      "epoch": 1.89,
      "learning_rate": 4.885836909871244e-06,
      "loss": 1.5243,
      "step": 880
    },
    {
      "epoch": 1.91,
      "learning_rate": 4.873175965665236e-06,
      "loss": 1.5269,
      "step": 890
    },
    {
      "epoch": 1.93,
      "learning_rate": 4.860515021459228e-06,
      "loss": 1.5216,
      "step": 900
    },
    {
      "epoch": 1.95,
      "learning_rate": 4.847854077253219e-06,
      "loss": 1.5212,
      "step": 910
    },
    {
      "epoch": 1.97,
      "learning_rate": 4.8351931330472105e-06,
      "loss": 1.5254,
      "step": 920
    },
    {
      "epoch": 1.99,
      "learning_rate": 4.822532188841201e-06,
      "loss": 1.5241,
      "step": 930
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.809871244635193e-06,
      "loss": 1.6665,
      "step": 940
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.797210300429185e-06,
      "loss": 1.5227,
      "step": 950
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.784549356223177e-06,
      "loss": 1.5241,
      "step": 960
    },
    {
      "epoch": 2.06,
      "eval_loss": 1.4172289371490479,
      "eval_runtime": 78.2198,
      "eval_samples_per_second": 511.379,
      "eval_steps_per_second": 7.99,
      "step": 960
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.7718884120171676e-06,
      "loss": 1.5248,
      "step": 970
    },
    {
      "epoch": 2.1,
      "learning_rate": 4.7592274678111585e-06,
      "loss": 1.5247,
      "step": 980
    },
    {
      "epoch": 2.12,
      "learning_rate": 4.74656652360515e-06,
      "loss": 1.5232,
      "step": 990
    },
    {
      "epoch": 2.15,
      "learning_rate": 4.733905579399142e-06,
      "loss": 1.5238,
      "step": 1000
    },
    {
      "epoch": 2.17,
      "learning_rate": 4.721244635193133e-06,
      "loss": 1.5234,
      "step": 1010
    },
    {
      "epoch": 2.19,
      "learning_rate": 4.708583690987125e-06,
      "loss": 1.5218,
      "step": 1020
    },
    {
      "epoch": 2.21,
      "learning_rate": 4.6959227467811155e-06,
      "loss": 1.5213,
      "step": 1030
    },
    {
      "epoch": 2.23,
      "learning_rate": 4.683261802575107e-06,
      "loss": 1.5262,
      "step": 1040
    },
    {
      "epoch": 2.25,
      "learning_rate": 4.670600858369099e-06,
      "loss": 1.5232,
      "step": 1050
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.657939914163091e-06,
      "loss": 1.5228,
      "step": 1060
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.645278969957082e-06,
      "loss": 1.5241,
      "step": 1070
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.6326180257510726e-06,
      "loss": 1.5196,
      "step": 1080
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.619957081545065e-06,
      "loss": 1.5229,
      "step": 1090
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.607296137339056e-06,
      "loss": 1.5212,
      "step": 1100
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.594635193133047e-06,
      "loss": 1.5233,
      "step": 1110
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.581974248927039e-06,
      "loss": 1.5227,
      "step": 1120
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.4165441989898682,
      "eval_runtime": 78.1888,
      "eval_samples_per_second": 511.582,
      "eval_steps_per_second": 7.993,
      "step": 1120
    },
    {
      "epoch": 2.42,
      "learning_rate": 4.5693133047210304e-06,
      "loss": 1.521,
      "step": 1130
    },
    {
      "epoch": 2.45,
      "learning_rate": 4.556652360515021e-06,
      "loss": 1.5211,
      "step": 1140
    },
    {
      "epoch": 2.47,
      "learning_rate": 4.543991416309013e-06,
      "loss": 1.5261,
      "step": 1150
    },
    {
      "epoch": 2.49,
      "learning_rate": 4.531330472103004e-06,
      "loss": 1.5206,
      "step": 1160
    },
    {
      "epoch": 2.51,
      "learning_rate": 4.518669527896996e-06,
      "loss": 1.5235,
      "step": 1170
    },
    {
      "epoch": 2.53,
      "learning_rate": 4.5060085836909875e-06,
      "loss": 1.5238,
      "step": 1180
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.493347639484979e-06,
      "loss": 1.5195,
      "step": 1190
    },
    {
      "epoch": 2.57,
      "learning_rate": 4.48068669527897e-06,
      "loss": 1.523,
      "step": 1200
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.468025751072961e-06,
      "loss": 1.5201,
      "step": 1210
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.455364806866953e-06,
      "loss": 1.5235,
      "step": 1220
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.4427038626609445e-06,
      "loss": 1.5208,
      "step": 1230
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.4300429184549354e-06,
      "loss": 1.5261,
      "step": 1240
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.417381974248927e-06,
      "loss": 1.5231,
      "step": 1250
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.404721030042918e-06,
      "loss": 1.5216,
      "step": 1260
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.39206008583691e-06,
      "loss": 1.5246,
      "step": 1270
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.379399141630902e-06,
      "loss": 1.5226,
      "step": 1280
    },
    {
      "epoch": 2.75,
      "eval_loss": 1.415173053741455,
      "eval_runtime": 78.2044,
      "eval_samples_per_second": 511.48,
      "eval_steps_per_second": 7.992,
      "step": 1280
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.366738197424893e-06,
      "loss": 1.5221,
      "step": 1290
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.354077253218884e-06,
      "loss": 1.5248,
      "step": 1300
    },
    {
      "epoch": 2.81,
      "learning_rate": 4.341416309012875e-06,
      "loss": 1.5227,
      "step": 1310
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.328755364806868e-06,
      "loss": 1.5244,
      "step": 1320
    },
    {
      "epoch": 2.85,
      "learning_rate": 4.316094420600859e-06,
      "loss": 1.5258,
      "step": 1330
    },
    {
      "epoch": 2.87,
      "learning_rate": 4.3034334763948495e-06,
      "loss": 1.5237,
      "step": 1340
    },
    {
      "epoch": 2.9,
      "learning_rate": 4.290772532188841e-06,
      "loss": 1.5248,
      "step": 1350
    },
    {
      "epoch": 2.92,
      "learning_rate": 4.278111587982833e-06,
      "loss": 1.524,
      "step": 1360
    },
    {
      "epoch": 2.94,
      "learning_rate": 4.265450643776824e-06,
      "loss": 1.523,
      "step": 1370
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.252789699570816e-06,
      "loss": 1.5242,
      "step": 1380
    },
    {
      "epoch": 2.98,
      "learning_rate": 4.2401287553648074e-06,
      "loss": 1.5255,
      "step": 1390
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.2274678111587975e-06,
      "loss": 1.6659,
      "step": 1400
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.21480686695279e-06,
      "loss": 1.5233,
      "step": 1410
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.202145922746782e-06,
      "loss": 1.5249,
      "step": 1420
    },
    {
      "epoch": 3.07,
      "learning_rate": 4.189484978540773e-06,
      "loss": 1.5248,
      "step": 1430
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.176824034334764e-06,
      "loss": 1.522,
      "step": 1440
    },
    {
      "epoch": 3.09,
      "eval_loss": 1.418953537940979,
      "eval_runtime": 78.1718,
      "eval_samples_per_second": 511.693,
      "eval_steps_per_second": 7.995,
      "step": 1440
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.164163090128755e-06,
      "loss": 1.5241,
      "step": 1450
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.151502145922747e-06,
      "loss": 1.5226,
      "step": 1460
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.138841201716738e-06,
      "loss": 1.5231,
      "step": 1470
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.12618025751073e-06,
      "loss": 1.5239,
      "step": 1480
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.113519313304721e-06,
      "loss": 1.5217,
      "step": 1490
    },
    {
      "epoch": 3.22,
      "learning_rate": 4.1008583690987124e-06,
      "loss": 1.5206,
      "step": 1500
    },
    {
      "epoch": 3.24,
      "learning_rate": 4.088197424892704e-06,
      "loss": 1.5232,
      "step": 1510
    },
    {
      "epoch": 3.26,
      "learning_rate": 4.075536480686696e-06,
      "loss": 1.5252,
      "step": 1520
    },
    {
      "epoch": 3.28,
      "learning_rate": 4.062875536480687e-06,
      "loss": 1.521,
      "step": 1530
    },
    {
      "epoch": 3.3,
      "learning_rate": 4.050214592274678e-06,
      "loss": 1.5205,
      "step": 1540
    },
    {
      "epoch": 3.33,
      "learning_rate": 4.0375536480686695e-06,
      "loss": 1.5215,
      "step": 1550
    },
    {
      "epoch": 3.35,
      "learning_rate": 4.024892703862661e-06,
      "loss": 1.5253,
      "step": 1560
    },
    {
      "epoch": 3.37,
      "learning_rate": 4.012231759656652e-06,
      "loss": 1.5225,
      "step": 1570
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.999570815450644e-06,
      "loss": 1.523,
      "step": 1580
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.986909871244635e-06,
      "loss": 1.5241,
      "step": 1590
    },
    {
      "epoch": 3.43,
      "learning_rate": 3.9742489270386265e-06,
      "loss": 1.5243,
      "step": 1600
    },
    {
      "epoch": 3.43,
      "eval_loss": 1.4176822900772095,
      "eval_runtime": 78.1608,
      "eval_samples_per_second": 511.766,
      "eval_steps_per_second": 7.996,
      "step": 1600
    },
    {
      "epoch": 3.45,
      "learning_rate": 3.961587982832618e-06,
      "loss": 1.5235,
      "step": 1610
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.94892703862661e-06,
      "loss": 1.5224,
      "step": 1620
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.9362660944206e-06,
      "loss": 1.5192,
      "step": 1630
    },
    {
      "epoch": 3.52,
      "learning_rate": 3.923605150214593e-06,
      "loss": 1.5186,
      "step": 1640
    },
    {
      "epoch": 3.54,
      "learning_rate": 3.9109442060085844e-06,
      "loss": 1.5208,
      "step": 1650
    },
    {
      "epoch": 3.56,
      "learning_rate": 3.898283261802575e-06,
      "loss": 1.5204,
      "step": 1660
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.885622317596566e-06,
      "loss": 1.5247,
      "step": 1670
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.872961373390559e-06,
      "loss": 1.5205,
      "step": 1680
    },
    {
      "epoch": 3.63,
      "learning_rate": 3.86030042918455e-06,
      "loss": 1.5226,
      "step": 1690
    },
    {
      "epoch": 3.65,
      "learning_rate": 3.847639484978541e-06,
      "loss": 1.5223,
      "step": 1700
    },
    {
      "epoch": 3.67,
      "learning_rate": 3.834978540772532e-06,
      "loss": 1.526,
      "step": 1710
    },
    {
      "epoch": 3.69,
      "learning_rate": 3.822317596566524e-06,
      "loss": 1.523,
      "step": 1720
    },
    {
      "epoch": 3.71,
      "learning_rate": 3.809656652360514e-06,
      "loss": 1.5209,
      "step": 1730
    },
    {
      "epoch": 3.73,
      "learning_rate": 3.7969957081545068e-06,
      "loss": 1.5244,
      "step": 1740
    },
    {
      "epoch": 3.75,
      "learning_rate": 3.784334763948498e-06,
      "loss": 1.5225,
      "step": 1750
    },
    {
      "epoch": 3.78,
      "learning_rate": 3.771673819742489e-06,
      "loss": 1.5213,
      "step": 1760
    },
    {
      "epoch": 3.78,
      "eval_loss": 1.4134083986282349,
      "eval_runtime": 78.1868,
      "eval_samples_per_second": 511.596,
      "eval_steps_per_second": 7.994,
      "step": 1760
    },
    {
      "epoch": 3.8,
      "learning_rate": 3.7590128755364807e-06,
      "loss": 1.5239,
      "step": 1770
    },
    {
      "epoch": 3.82,
      "learning_rate": 3.7463519313304725e-06,
      "loss": 1.5227,
      "step": 1780
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.733690987124464e-06,
      "loss": 1.5246,
      "step": 1790
    },
    {
      "epoch": 3.86,
      "learning_rate": 3.721030042918455e-06,
      "loss": 1.5235,
      "step": 1800
    },
    {
      "epoch": 3.88,
      "learning_rate": 3.7083690987124465e-06,
      "loss": 1.5228,
      "step": 1810
    },
    {
      "epoch": 3.9,
      "learning_rate": 3.6957081545064374e-06,
      "loss": 1.5227,
      "step": 1820
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.683047210300429e-06,
      "loss": 1.5236,
      "step": 1830
    },
    {
      "epoch": 3.95,
      "learning_rate": 3.670386266094421e-06,
      "loss": 1.5254,
      "step": 1840
    },
    {
      "epoch": 3.97,
      "learning_rate": 3.657725321888412e-06,
      "loss": 1.5235,
      "step": 1850
    },
    {
      "epoch": 3.99,
      "learning_rate": 3.645064377682403e-06,
      "loss": 1.5245,
      "step": 1860
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.632403433476395e-06,
      "loss": 1.6644,
      "step": 1870
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.6197424892703866e-06,
      "loss": 1.5232,
      "step": 1880
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.6070815450643783e-06,
      "loss": 1.5218,
      "step": 1890
    },
    {
      "epoch": 4.08,
      "learning_rate": 3.5944206008583692e-06,
      "loss": 1.5209,
      "step": 1900
    },
    {
      "epoch": 4.1,
      "learning_rate": 3.5817596566523606e-06,
      "loss": 1.5217,
      "step": 1910
    },
    {
      "epoch": 4.12,
      "learning_rate": 3.5690987124463515e-06,
      "loss": 1.524,
      "step": 1920
    },
    {
      "epoch": 4.12,
      "eval_loss": 1.4139704704284668,
      "eval_runtime": 78.1302,
      "eval_samples_per_second": 511.966,
      "eval_steps_per_second": 7.999,
      "step": 1920
    },
    {
      "epoch": 4.14,
      "learning_rate": 3.5564377682403432e-06,
      "loss": 1.5244,
      "step": 1930
    },
    {
      "epoch": 4.16,
      "learning_rate": 3.543776824034335e-06,
      "loss": 1.5248,
      "step": 1940
    },
    {
      "epoch": 4.18,
      "learning_rate": 3.5311158798283267e-06,
      "loss": 1.5207,
      "step": 1950
    },
    {
      "epoch": 4.21,
      "learning_rate": 3.5184549356223176e-06,
      "loss": 1.52,
      "step": 1960
    },
    {
      "epoch": 4.23,
      "learning_rate": 3.505793991416309e-06,
      "loss": 1.5208,
      "step": 1970
    },
    {
      "epoch": 4.25,
      "learning_rate": 3.4931330472103015e-06,
      "loss": 1.525,
      "step": 1980
    },
    {
      "epoch": 4.27,
      "learning_rate": 3.4804721030042916e-06,
      "loss": 1.5242,
      "step": 1990
    },
    {
      "epoch": 4.29,
      "learning_rate": 3.4678111587982833e-06,
      "loss": 1.5228,
      "step": 2000
    },
    {
      "epoch": 4.31,
      "learning_rate": 3.4551502145922747e-06,
      "loss": 1.5219,
      "step": 2010
    },
    {
      "epoch": 4.33,
      "learning_rate": 3.4424892703862664e-06,
      "loss": 1.5212,
      "step": 2020
    },
    {
      "epoch": 4.36,
      "learning_rate": 3.4298283261802573e-06,
      "loss": 1.523,
      "step": 2030
    },
    {
      "epoch": 4.38,
      "learning_rate": 3.417167381974249e-06,
      "loss": 1.5245,
      "step": 2040
    },
    {
      "epoch": 4.4,
      "learning_rate": 3.404506437768241e-06,
      "loss": 1.5223,
      "step": 2050
    },
    {
      "epoch": 4.42,
      "learning_rate": 3.3918454935622317e-06,
      "loss": 1.5259,
      "step": 2060
    },
    {
      "epoch": 4.44,
      "learning_rate": 3.379184549356223e-06,
      "loss": 1.5214,
      "step": 2070
    },
    {
      "epoch": 4.46,
      "learning_rate": 3.3665236051502144e-06,
      "loss": 1.5223,
      "step": 2080
    },
    {
      "epoch": 4.46,
      "eval_loss": 1.4173228740692139,
      "eval_runtime": 78.2047,
      "eval_samples_per_second": 511.478,
      "eval_steps_per_second": 7.992,
      "step": 2080
    },
    {
      "epoch": 4.48,
      "learning_rate": 3.3538626609442057e-06,
      "loss": 1.5204,
      "step": 2090
    },
    {
      "epoch": 4.51,
      "learning_rate": 3.3412017167381974e-06,
      "loss": 1.5211,
      "step": 2100
    },
    {
      "epoch": 4.53,
      "learning_rate": 3.328540772532189e-06,
      "loss": 1.5202,
      "step": 2110
    },
    {
      "epoch": 4.55,
      "learning_rate": 3.3158798283261805e-06,
      "loss": 1.5183,
      "step": 2120
    },
    {
      "epoch": 4.57,
      "learning_rate": 3.3032188841201714e-06,
      "loss": 1.524,
      "step": 2130
    },
    {
      "epoch": 4.59,
      "learning_rate": 3.290557939914164e-06,
      "loss": 1.5208,
      "step": 2140
    },
    {
      "epoch": 4.61,
      "learning_rate": 3.277896995708154e-06,
      "loss": 1.5224,
      "step": 2150
    },
    {
      "epoch": 4.63,
      "learning_rate": 3.265236051502146e-06,
      "loss": 1.5193,
      "step": 2160
    },
    {
      "epoch": 4.66,
      "learning_rate": 3.2525751072961376e-06,
      "loss": 1.5231,
      "step": 2170
    },
    {
      "epoch": 4.68,
      "learning_rate": 3.239914163090129e-06,
      "loss": 1.5252,
      "step": 2180
    },
    {
      "epoch": 4.7,
      "learning_rate": 3.2272532188841198e-06,
      "loss": 1.5216,
      "step": 2190
    },
    {
      "epoch": 4.72,
      "learning_rate": 3.2145922746781115e-06,
      "loss": 1.5217,
      "step": 2200
    },
    {
      "epoch": 4.74,
      "learning_rate": 3.2019313304721033e-06,
      "loss": 1.5235,
      "step": 2210
    },
    {
      "epoch": 4.76,
      "learning_rate": 3.189270386266094e-06,
      "loss": 1.5205,
      "step": 2220
    },
    {
      "epoch": 4.78,
      "learning_rate": 3.176609442060086e-06,
      "loss": 1.524,
      "step": 2230
    },
    {
      "epoch": 4.81,
      "learning_rate": 3.1639484978540773e-06,
      "loss": 1.5236,
      "step": 2240
    },
    {
      "epoch": 4.81,
      "eval_loss": 1.4121308326721191,
      "eval_runtime": 78.2109,
      "eval_samples_per_second": 511.438,
      "eval_steps_per_second": 7.991,
      "step": 2240
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.151287553648068e-06,
      "loss": 1.5241,
      "step": 2250
    },
    {
      "epoch": 4.85,
      "learning_rate": 3.13862660944206e-06,
      "loss": 1.5202,
      "step": 2260
    },
    {
      "epoch": 4.87,
      "learning_rate": 3.1259656652360516e-06,
      "loss": 1.5246,
      "step": 2270
    },
    {
      "epoch": 4.89,
      "learning_rate": 3.113304721030043e-06,
      "loss": 1.5213,
      "step": 2280
    },
    {
      "epoch": 4.91,
      "learning_rate": 3.1006437768240343e-06,
      "loss": 1.5239,
      "step": 2290
    },
    {
      "epoch": 4.93,
      "learning_rate": 3.0879828326180256e-06,
      "loss": 1.5219,
      "step": 2300
    },
    {
      "epoch": 4.96,
      "learning_rate": 3.0753218884120165e-06,
      "loss": 1.5227,
      "step": 2310
    },
    {
      "epoch": 4.98,
      "learning_rate": 3.0626609442060083e-06,
      "loss": 1.5235,
      "step": 2320
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.05e-06,
      "loss": 1.5208,
      "step": 2330
    },
    {
      "epoch": 5.02,
      "learning_rate": 3.0373390557939913e-06,
      "loss": 1.6633,
      "step": 2340
    },
    {
      "epoch": 5.04,
      "learning_rate": 3.024678111587983e-06,
      "loss": 1.521,
      "step": 2350
    },
    {
      "epoch": 5.06,
      "learning_rate": 3.012017167381974e-06,
      "loss": 1.5219,
      "step": 2360
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.999356223175966e-06,
      "loss": 1.5228,
      "step": 2370
    },
    {
      "epoch": 5.11,
      "learning_rate": 2.986695278969957e-06,
      "loss": 1.5236,
      "step": 2380
    },
    {
      "epoch": 5.13,
      "learning_rate": 2.9740343347639484e-06,
      "loss": 1.5201,
      "step": 2390
    },
    {
      "epoch": 5.15,
      "learning_rate": 2.96137339055794e-06,
      "loss": 1.5239,
      "step": 2400
    },
    {
      "epoch": 5.15,
      "eval_loss": 1.4185978174209595,
      "eval_runtime": 78.1188,
      "eval_samples_per_second": 512.041,
      "eval_steps_per_second": 8.001,
      "step": 2400
    },
    {
      "epoch": 5.17,
      "learning_rate": 2.9487124463519315e-06,
      "loss": 1.521,
      "step": 2410
    },
    {
      "epoch": 5.19,
      "learning_rate": 2.9360515021459232e-06,
      "loss": 1.5202,
      "step": 2420
    },
    {
      "epoch": 5.21,
      "learning_rate": 2.923390557939914e-06,
      "loss": 1.5206,
      "step": 2430
    },
    {
      "epoch": 5.24,
      "learning_rate": 2.9107296137339054e-06,
      "loss": 1.5233,
      "step": 2440
    },
    {
      "epoch": 5.26,
      "learning_rate": 2.898068669527897e-06,
      "loss": 1.5213,
      "step": 2450
    },
    {
      "epoch": 5.28,
      "learning_rate": 2.8854077253218885e-06,
      "loss": 1.5242,
      "step": 2460
    },
    {
      "epoch": 5.3,
      "learning_rate": 2.8727467811158794e-06,
      "loss": 1.5217,
      "step": 2470
    },
    {
      "epoch": 5.32,
      "learning_rate": 2.860085836909871e-06,
      "loss": 1.5208,
      "step": 2480
    },
    {
      "epoch": 5.34,
      "learning_rate": 2.8474248927038625e-06,
      "loss": 1.5247,
      "step": 2490
    },
    {
      "epoch": 5.36,
      "learning_rate": 2.8347639484978542e-06,
      "loss": 1.5238,
      "step": 2500
    },
    {
      "epoch": 5.39,
      "learning_rate": 2.8221030042918456e-06,
      "loss": 1.5214,
      "step": 2510
    },
    {
      "epoch": 5.41,
      "learning_rate": 2.8094420600858365e-06,
      "loss": 1.523,
      "step": 2520
    },
    {
      "epoch": 5.43,
      "learning_rate": 2.7967811158798286e-06,
      "loss": 1.5212,
      "step": 2530
    },
    {
      "epoch": 5.45,
      "learning_rate": 2.7841201716738195e-06,
      "loss": 1.5213,
      "step": 2540
    },
    {
      "epoch": 5.47,
      "learning_rate": 2.7714592274678113e-06,
      "loss": 1.522,
      "step": 2550
    },
    {
      "epoch": 5.49,
      "learning_rate": 2.7587982832618026e-06,
      "loss": 1.5203,
      "step": 2560
    },
    {
      "epoch": 5.49,
      "eval_loss": 1.4153515100479126,
      "eval_runtime": 78.1738,
      "eval_samples_per_second": 511.68,
      "eval_steps_per_second": 7.995,
      "step": 2560
    },
    {
      "epoch": 5.51,
      "learning_rate": 2.746137339055794e-06,
      "loss": 1.5194,
      "step": 2570
    },
    {
      "epoch": 5.54,
      "learning_rate": 2.7334763948497857e-06,
      "loss": 1.5199,
      "step": 2580
    },
    {
      "epoch": 5.56,
      "learning_rate": 2.7208154506437766e-06,
      "loss": 1.5209,
      "step": 2590
    },
    {
      "epoch": 5.58,
      "learning_rate": 2.7081545064377683e-06,
      "loss": 1.5241,
      "step": 2600
    },
    {
      "epoch": 5.6,
      "learning_rate": 2.6954935622317597e-06,
      "loss": 1.5186,
      "step": 2610
    },
    {
      "epoch": 5.62,
      "learning_rate": 2.682832618025751e-06,
      "loss": 1.5232,
      "step": 2620
    },
    {
      "epoch": 5.64,
      "learning_rate": 2.6701716738197427e-06,
      "loss": 1.5215,
      "step": 2630
    },
    {
      "epoch": 5.66,
      "learning_rate": 2.6575107296137336e-06,
      "loss": 1.5262,
      "step": 2640
    },
    {
      "epoch": 5.69,
      "learning_rate": 2.644849785407725e-06,
      "loss": 1.5242,
      "step": 2650
    },
    {
      "epoch": 5.71,
      "learning_rate": 2.6321888412017167e-06,
      "loss": 1.5209,
      "step": 2660
    },
    {
      "epoch": 5.73,
      "learning_rate": 2.619527896995708e-06,
      "loss": 1.5205,
      "step": 2670
    },
    {
      "epoch": 5.75,
      "learning_rate": 2.6068669527896998e-06,
      "loss": 1.5235,
      "step": 2680
    },
    {
      "epoch": 5.77,
      "learning_rate": 2.594206008583691e-06,
      "loss": 1.5214,
      "step": 2690
    },
    {
      "epoch": 5.79,
      "learning_rate": 2.581545064377682e-06,
      "loss": 1.5238,
      "step": 2700
    },
    {
      "epoch": 5.81,
      "learning_rate": 2.5688841201716738e-06,
      "loss": 1.5238,
      "step": 2710
    },
    {
      "epoch": 5.84,
      "learning_rate": 2.556223175965665e-06,
      "loss": 1.522,
      "step": 2720
    },
    {
      "epoch": 5.84,
      "eval_loss": 1.416185736656189,
      "eval_runtime": 78.1964,
      "eval_samples_per_second": 511.533,
      "eval_steps_per_second": 7.993,
      "step": 2720
    },
    {
      "epoch": 5.86,
      "learning_rate": 2.543562231759657e-06,
      "loss": 1.5245,
      "step": 2730
    },
    {
      "epoch": 5.88,
      "learning_rate": 2.530901287553648e-06,
      "loss": 1.5218,
      "step": 2740
    },
    {
      "epoch": 5.9,
      "learning_rate": 2.518240343347639e-06,
      "loss": 1.5231,
      "step": 2750
    },
    {
      "epoch": 5.92,
      "learning_rate": 2.5055793991416312e-06,
      "loss": 1.5205,
      "step": 2760
    },
    {
      "epoch": 5.94,
      "learning_rate": 2.492918454935622e-06,
      "loss": 1.523,
      "step": 2770
    },
    {
      "epoch": 5.96,
      "learning_rate": 2.480257510729614e-06,
      "loss": 1.5246,
      "step": 2780
    },
    {
      "epoch": 5.99,
      "learning_rate": 2.467596566523605e-06,
      "loss": 1.5226,
      "step": 2790
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.4549356223175965e-06,
      "loss": 1.6635,
      "step": 2800
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.4422746781115883e-06,
      "loss": 1.524,
      "step": 2810
    },
    {
      "epoch": 6.05,
      "learning_rate": 2.429613733905579e-06,
      "loss": 1.522,
      "step": 2820
    },
    {
      "epoch": 6.07,
      "learning_rate": 2.416952789699571e-06,
      "loss": 1.5204,
      "step": 2830
    },
    {
      "epoch": 6.09,
      "learning_rate": 2.4042918454935622e-06,
      "loss": 1.5187,
      "step": 2840
    },
    {
      "epoch": 6.12,
      "learning_rate": 2.3916309012875536e-06,
      "loss": 1.5209,
      "step": 2850
    },
    {
      "epoch": 6.14,
      "learning_rate": 2.3789699570815453e-06,
      "loss": 1.5242,
      "step": 2860
    },
    {
      "epoch": 6.16,
      "learning_rate": 2.3663090128755362e-06,
      "loss": 1.5214,
      "step": 2870
    },
    {
      "epoch": 6.18,
      "learning_rate": 2.3536480686695275e-06,
      "loss": 1.5209,
      "step": 2880
    },
    {
      "epoch": 6.18,
      "eval_loss": 1.4153697490692139,
      "eval_runtime": 78.1036,
      "eval_samples_per_second": 512.14,
      "eval_steps_per_second": 8.002,
      "step": 2880
    },
    {
      "epoch": 6.2,
      "learning_rate": 2.3409871244635193e-06,
      "loss": 1.5226,
      "step": 2890
    },
    {
      "epoch": 6.22,
      "learning_rate": 2.3283261802575106e-06,
      "loss": 1.5218,
      "step": 2900
    },
    {
      "epoch": 6.24,
      "learning_rate": 2.3156652360515024e-06,
      "loss": 1.5258,
      "step": 2910
    },
    {
      "epoch": 6.27,
      "learning_rate": 2.3030042918454937e-06,
      "loss": 1.5218,
      "step": 2920
    },
    {
      "epoch": 6.29,
      "learning_rate": 2.2903433476394846e-06,
      "loss": 1.5227,
      "step": 2930
    },
    {
      "epoch": 6.31,
      "learning_rate": 2.2776824034334763e-06,
      "loss": 1.521,
      "step": 2940
    },
    {
      "epoch": 6.33,
      "learning_rate": 2.2650214592274677e-06,
      "loss": 1.521,
      "step": 2950
    },
    {
      "epoch": 6.35,
      "learning_rate": 2.2523605150214594e-06,
      "loss": 1.5237,
      "step": 2960
    },
    {
      "epoch": 6.37,
      "learning_rate": 2.2396995708154507e-06,
      "loss": 1.5222,
      "step": 2970
    },
    {
      "epoch": 6.39,
      "learning_rate": 2.2270386266094416e-06,
      "loss": 1.5219,
      "step": 2980
    },
    {
      "epoch": 6.42,
      "learning_rate": 2.214377682403434e-06,
      "loss": 1.5237,
      "step": 2990
    },
    {
      "epoch": 6.44,
      "learning_rate": 2.2017167381974247e-06,
      "loss": 1.5193,
      "step": 3000
    },
    {
      "epoch": 6.46,
      "learning_rate": 2.1890557939914165e-06,
      "loss": 1.525,
      "step": 3010
    },
    {
      "epoch": 6.48,
      "learning_rate": 2.176394849785408e-06,
      "loss": 1.5192,
      "step": 3020
    },
    {
      "epoch": 6.5,
      "learning_rate": 2.1637339055793987e-06,
      "loss": 1.5195,
      "step": 3030
    },
    {
      "epoch": 6.52,
      "learning_rate": 2.151072961373391e-06,
      "loss": 1.5196,
      "step": 3040
    },
    {
      "epoch": 6.52,
      "eval_loss": 1.4152772426605225,
      "eval_runtime": 78.1214,
      "eval_samples_per_second": 512.024,
      "eval_steps_per_second": 8.0,
      "step": 3040
    },
    {
      "epoch": 6.54,
      "learning_rate": 2.1384120171673818e-06,
      "loss": 1.5185,
      "step": 3050
    },
    {
      "epoch": 6.57,
      "learning_rate": 2.1257510729613735e-06,
      "loss": 1.5209,
      "step": 3060
    },
    {
      "epoch": 6.59,
      "learning_rate": 2.113090128755365e-06,
      "loss": 1.5231,
      "step": 3070
    },
    {
      "epoch": 6.61,
      "learning_rate": 2.100429184549356e-06,
      "loss": 1.5206,
      "step": 3080
    },
    {
      "epoch": 6.63,
      "learning_rate": 2.087768240343348e-06,
      "loss": 1.5227,
      "step": 3090
    },
    {
      "epoch": 6.65,
      "learning_rate": 2.075107296137339e-06,
      "loss": 1.5225,
      "step": 3100
    },
    {
      "epoch": 6.67,
      "learning_rate": 2.06244635193133e-06,
      "loss": 1.5236,
      "step": 3110
    },
    {
      "epoch": 6.69,
      "learning_rate": 2.049785407725322e-06,
      "loss": 1.5218,
      "step": 3120
    },
    {
      "epoch": 6.72,
      "learning_rate": 2.037124463519313e-06,
      "loss": 1.5207,
      "step": 3130
    },
    {
      "epoch": 6.74,
      "learning_rate": 2.024463519313305e-06,
      "loss": 1.5236,
      "step": 3140
    },
    {
      "epoch": 6.76,
      "learning_rate": 2.0118025751072963e-06,
      "loss": 1.5204,
      "step": 3150
    },
    {
      "epoch": 6.78,
      "learning_rate": 1.999141630901287e-06,
      "loss": 1.5204,
      "step": 3160
    },
    {
      "epoch": 6.8,
      "learning_rate": 1.9864806866952794e-06,
      "loss": 1.5234,
      "step": 3170
    },
    {
      "epoch": 6.82,
      "learning_rate": 1.9738197424892703e-06,
      "loss": 1.5226,
      "step": 3180
    },
    {
      "epoch": 6.84,
      "learning_rate": 1.961158798283262e-06,
      "loss": 1.5203,
      "step": 3190
    },
    {
      "epoch": 6.87,
      "learning_rate": 1.9484978540772533e-06,
      "loss": 1.5209,
      "step": 3200
    },
    {
      "epoch": 6.87,
      "eval_loss": 1.4121816158294678,
      "eval_runtime": 78.0644,
      "eval_samples_per_second": 512.398,
      "eval_steps_per_second": 8.006,
      "step": 3200
    },
    {
      "epoch": 6.89,
      "learning_rate": 1.9358369098712447e-06,
      "loss": 1.5213,
      "step": 3210
    },
    {
      "epoch": 6.91,
      "learning_rate": 1.9231759656652364e-06,
      "loss": 1.524,
      "step": 3220
    },
    {
      "epoch": 6.93,
      "learning_rate": 1.9105150214592273e-06,
      "loss": 1.5224,
      "step": 3230
    },
    {
      "epoch": 6.95,
      "learning_rate": 1.8978540772532188e-06,
      "loss": 1.5196,
      "step": 3240
    },
    {
      "epoch": 6.97,
      "learning_rate": 1.8851931330472106e-06,
      "loss": 1.5237,
      "step": 3250
    },
    {
      "epoch": 6.99,
      "learning_rate": 1.8725321888412017e-06,
      "loss": 1.5227,
      "step": 3260
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.8598712446351935e-06,
      "loss": 1.6653,
      "step": 3270
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.8472103004291844e-06,
      "loss": 1.5187,
      "step": 3280
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.8345493562231757e-06,
      "loss": 1.5202,
      "step": 3290
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.8218884120171672e-06,
      "loss": 1.5237,
      "step": 3300
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.809227467811159e-06,
      "loss": 1.5209,
      "step": 3310
    },
    {
      "epoch": 7.12,
      "learning_rate": 1.7965665236051505e-06,
      "loss": 1.5229,
      "step": 3320
    },
    {
      "epoch": 7.15,
      "learning_rate": 1.7839055793991418e-06,
      "loss": 1.52,
      "step": 3330
    },
    {
      "epoch": 7.17,
      "learning_rate": 1.7712446351931327e-06,
      "loss": 1.5195,
      "step": 3340
    },
    {
      "epoch": 7.19,
      "learning_rate": 1.7585836909871243e-06,
      "loss": 1.5216,
      "step": 3350
    },
    {
      "epoch": 7.21,
      "learning_rate": 1.745922746781116e-06,
      "loss": 1.5202,
      "step": 3360
    },
    {
      "epoch": 7.21,
      "eval_loss": 1.4146045446395874,
      "eval_runtime": 78.1595,
      "eval_samples_per_second": 511.774,
      "eval_steps_per_second": 7.996,
      "step": 3360
    },
    {
      "epoch": 7.23,
      "learning_rate": 1.7332618025751075e-06,
      "loss": 1.5226,
      "step": 3370
    },
    {
      "epoch": 7.25,
      "learning_rate": 1.7206008583690989e-06,
      "loss": 1.5221,
      "step": 3380
    },
    {
      "epoch": 7.27,
      "learning_rate": 1.7079399141630898e-06,
      "loss": 1.5253,
      "step": 3390
    },
    {
      "epoch": 7.3,
      "learning_rate": 1.6952789699570822e-06,
      "loss": 1.5236,
      "step": 3400
    },
    {
      "epoch": 7.32,
      "learning_rate": 1.6826180257510726e-06,
      "loss": 1.5209,
      "step": 3410
    },
    {
      "epoch": 7.34,
      "learning_rate": 1.6699570815450646e-06,
      "loss": 1.5211,
      "step": 3420
    },
    {
      "epoch": 7.36,
      "learning_rate": 1.657296137339056e-06,
      "loss": 1.5218,
      "step": 3430
    },
    {
      "epoch": 7.38,
      "learning_rate": 1.6446351931330472e-06,
      "loss": 1.5233,
      "step": 3440
    },
    {
      "epoch": 7.4,
      "learning_rate": 1.6319742489270388e-06,
      "loss": 1.5219,
      "step": 3450
    },
    {
      "epoch": 7.42,
      "learning_rate": 1.6193133047210297e-06,
      "loss": 1.523,
      "step": 3460
    },
    {
      "epoch": 7.45,
      "learning_rate": 1.6066523605150216e-06,
      "loss": 1.5204,
      "step": 3470
    },
    {
      "epoch": 7.47,
      "learning_rate": 1.593991416309013e-06,
      "loss": 1.5226,
      "step": 3480
    },
    {
      "epoch": 7.49,
      "learning_rate": 1.5813304721030043e-06,
      "loss": 1.5178,
      "step": 3490
    },
    {
      "epoch": 7.51,
      "learning_rate": 1.5686695278969958e-06,
      "loss": 1.5184,
      "step": 3500
    },
    {
      "epoch": 7.53,
      "learning_rate": 1.556008583690987e-06,
      "loss": 1.5205,
      "step": 3510
    },
    {
      "epoch": 7.55,
      "learning_rate": 1.543347639484978e-06,
      "loss": 1.5192,
      "step": 3520
    },
    {
      "epoch": 7.55,
      "eval_loss": 1.4141193628311157,
      "eval_runtime": 78.1244,
      "eval_samples_per_second": 512.004,
      "eval_steps_per_second": 8.0,
      "step": 3520
    },
    {
      "epoch": 7.57,
      "learning_rate": 1.53068669527897e-06,
      "loss": 1.5211,
      "step": 3530
    },
    {
      "epoch": 7.6,
      "learning_rate": 1.5180257510729613e-06,
      "loss": 1.5212,
      "step": 3540
    },
    {
      "epoch": 7.62,
      "learning_rate": 1.505364806866953e-06,
      "loss": 1.5226,
      "step": 3550
    },
    {
      "epoch": 7.64,
      "learning_rate": 1.4927038626609444e-06,
      "loss": 1.5204,
      "step": 3560
    },
    {
      "epoch": 7.66,
      "learning_rate": 1.4800429184549353e-06,
      "loss": 1.5258,
      "step": 3570
    },
    {
      "epoch": 7.68,
      "learning_rate": 1.467381974248927e-06,
      "loss": 1.5217,
      "step": 3580
    },
    {
      "epoch": 7.7,
      "learning_rate": 1.4547210300429184e-06,
      "loss": 1.5181,
      "step": 3590
    },
    {
      "epoch": 7.72,
      "learning_rate": 1.4420600858369101e-06,
      "loss": 1.5192,
      "step": 3600
    },
    {
      "epoch": 7.75,
      "learning_rate": 1.4293991416309015e-06,
      "loss": 1.5235,
      "step": 3610
    },
    {
      "epoch": 7.77,
      "learning_rate": 1.4167381974248924e-06,
      "loss": 1.5178,
      "step": 3620
    },
    {
      "epoch": 7.79,
      "learning_rate": 1.4040772532188841e-06,
      "loss": 1.5228,
      "step": 3630
    },
    {
      "epoch": 7.81,
      "learning_rate": 1.3914163090128754e-06,
      "loss": 1.5212,
      "step": 3640
    },
    {
      "epoch": 7.83,
      "learning_rate": 1.3787553648068672e-06,
      "loss": 1.5251,
      "step": 3650
    },
    {
      "epoch": 7.85,
      "learning_rate": 1.3660944206008585e-06,
      "loss": 1.5254,
      "step": 3660
    },
    {
      "epoch": 7.87,
      "learning_rate": 1.3534334763948494e-06,
      "loss": 1.5213,
      "step": 3670
    },
    {
      "epoch": 7.9,
      "learning_rate": 1.3407725321888416e-06,
      "loss": 1.5215,
      "step": 3680
    },
    {
      "epoch": 7.9,
      "eval_loss": 1.4122941493988037,
      "eval_runtime": 78.0947,
      "eval_samples_per_second": 512.199,
      "eval_steps_per_second": 8.003,
      "step": 3680
    },
    {
      "epoch": 7.92,
      "learning_rate": 1.3281115879828325e-06,
      "loss": 1.5199,
      "step": 3690
    },
    {
      "epoch": 7.94,
      "learning_rate": 1.3154506437768242e-06,
      "loss": 1.5236,
      "step": 3700
    },
    {
      "epoch": 7.96,
      "learning_rate": 1.3027896995708156e-06,
      "loss": 1.5216,
      "step": 3710
    },
    {
      "epoch": 7.98,
      "learning_rate": 1.2901287553648069e-06,
      "loss": 1.5219,
      "step": 3720
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.2774678111587986e-06,
      "loss": 1.664,
      "step": 3730
    },
    {
      "epoch": 8.03,
      "learning_rate": 1.2648068669527895e-06,
      "loss": 1.5229,
      "step": 3740
    },
    {
      "epoch": 8.05,
      "learning_rate": 1.2521459227467809e-06,
      "loss": 1.5241,
      "step": 3750
    },
    {
      "epoch": 8.07,
      "learning_rate": 1.2394849785407726e-06,
      "loss": 1.5221,
      "step": 3760
    },
    {
      "epoch": 8.09,
      "learning_rate": 1.226824034334764e-06,
      "loss": 1.5236,
      "step": 3770
    },
    {
      "epoch": 8.11,
      "learning_rate": 1.2141630901287557e-06,
      "loss": 1.5215,
      "step": 3780
    },
    {
      "epoch": 8.13,
      "learning_rate": 1.201502145922747e-06,
      "loss": 1.5207,
      "step": 3790
    },
    {
      "epoch": 8.15,
      "learning_rate": 1.188841201716738e-06,
      "loss": 1.5206,
      "step": 3800
    },
    {
      "epoch": 8.18,
      "learning_rate": 1.1761802575107297e-06,
      "loss": 1.5198,
      "step": 3810
    },
    {
      "epoch": 8.2,
      "learning_rate": 1.163519313304721e-06,
      "loss": 1.518,
      "step": 3820
    },
    {
      "epoch": 8.22,
      "learning_rate": 1.1508583690987127e-06,
      "loss": 1.521,
      "step": 3830
    },
    {
      "epoch": 8.24,
      "learning_rate": 1.138197424892704e-06,
      "loss": 1.5228,
      "step": 3840
    },
    {
      "epoch": 8.24,
      "eval_loss": 1.4147212505340576,
      "eval_runtime": 78.136,
      "eval_samples_per_second": 511.928,
      "eval_steps_per_second": 7.999,
      "step": 3840
    },
    {
      "epoch": 8.26,
      "learning_rate": 1.125536480686695e-06,
      "loss": 1.5228,
      "step": 3850
    },
    {
      "epoch": 8.28,
      "learning_rate": 1.1128755364806867e-06,
      "loss": 1.5232,
      "step": 3860
    },
    {
      "epoch": 8.3,
      "learning_rate": 1.100214592274678e-06,
      "loss": 1.5213,
      "step": 3870
    },
    {
      "epoch": 8.33,
      "learning_rate": 1.0875536480686698e-06,
      "loss": 1.5173,
      "step": 3880
    },
    {
      "epoch": 8.35,
      "learning_rate": 1.074892703862661e-06,
      "loss": 1.5206,
      "step": 3890
    },
    {
      "epoch": 8.37,
      "learning_rate": 1.062231759656652e-06,
      "loss": 1.5224,
      "step": 3900
    },
    {
      "epoch": 8.39,
      "learning_rate": 1.049570815450644e-06,
      "loss": 1.5215,
      "step": 3910
    },
    {
      "epoch": 8.41,
      "learning_rate": 1.036909871244635e-06,
      "loss": 1.5211,
      "step": 3920
    },
    {
      "epoch": 8.43,
      "learning_rate": 1.0242489270386264e-06,
      "loss": 1.5201,
      "step": 3930
    },
    {
      "epoch": 8.45,
      "learning_rate": 1.0115879828326181e-06,
      "loss": 1.5248,
      "step": 3940
    },
    {
      "epoch": 8.48,
      "learning_rate": 9.989270386266093e-07,
      "loss": 1.5194,
      "step": 3950
    },
    {
      "epoch": 8.5,
      "learning_rate": 9.862660944206012e-07,
      "loss": 1.5202,
      "step": 3960
    },
    {
      "epoch": 8.52,
      "learning_rate": 9.736051502145921e-07,
      "loss": 1.518,
      "step": 3970
    },
    {
      "epoch": 8.54,
      "learning_rate": 9.609442060085834e-07,
      "loss": 1.5199,
      "step": 3980
    },
    {
      "epoch": 8.56,
      "learning_rate": 9.482832618025752e-07,
      "loss": 1.5221,
      "step": 3990
    },
    {
      "epoch": 8.58,
      "learning_rate": 9.356223175965663e-07,
      "loss": 1.5222,
      "step": 4000
    },
    {
      "epoch": 8.58,
      "eval_loss": 1.4144057035446167,
      "eval_runtime": 78.0968,
      "eval_samples_per_second": 512.185,
      "eval_steps_per_second": 8.003,
      "step": 4000
    },
    {
      "epoch": 8.6,
      "learning_rate": 9.229613733905583e-07,
      "loss": 1.5217,
      "step": 4010
    },
    {
      "epoch": 8.63,
      "learning_rate": 9.103004291845494e-07,
      "loss": 1.5225,
      "step": 4020
    },
    {
      "epoch": 8.65,
      "learning_rate": 8.976394849785405e-07,
      "loss": 1.5232,
      "step": 4030
    },
    {
      "epoch": 8.67,
      "learning_rate": 8.849785407725323e-07,
      "loss": 1.5248,
      "step": 4040
    },
    {
      "epoch": 8.69,
      "learning_rate": 8.723175965665236e-07,
      "loss": 1.5213,
      "step": 4050
    },
    {
      "epoch": 8.71,
      "learning_rate": 8.596566523605154e-07,
      "loss": 1.5182,
      "step": 4060
    },
    {
      "epoch": 8.73,
      "learning_rate": 8.469957081545065e-07,
      "loss": 1.5203,
      "step": 4070
    },
    {
      "epoch": 8.75,
      "learning_rate": 8.343347639484976e-07,
      "loss": 1.523,
      "step": 4080
    },
    {
      "epoch": 8.78,
      "learning_rate": 8.216738197424896e-07,
      "loss": 1.5213,
      "step": 4090
    },
    {
      "epoch": 8.8,
      "learning_rate": 8.090128755364807e-07,
      "loss": 1.5235,
      "step": 4100
    },
    {
      "epoch": 8.82,
      "learning_rate": 7.963519313304725e-07,
      "loss": 1.5236,
      "step": 4110
    },
    {
      "epoch": 8.84,
      "learning_rate": 7.836909871244636e-07,
      "loss": 1.5238,
      "step": 4120
    },
    {
      "epoch": 8.86,
      "learning_rate": 7.710300429184549e-07,
      "loss": 1.5228,
      "step": 4130
    },
    {
      "epoch": 8.88,
      "learning_rate": 7.583690987124464e-07,
      "loss": 1.5228,
      "step": 4140
    },
    {
      "epoch": 8.9,
      "learning_rate": 7.457081545064378e-07,
      "loss": 1.5223,
      "step": 4150
    },
    {
      "epoch": 8.93,
      "learning_rate": 7.330472103004289e-07,
      "loss": 1.5201,
      "step": 4160
    },
    {
      "epoch": 8.93,
      "eval_loss": 1.4172754287719727,
      "eval_runtime": 78.1438,
      "eval_samples_per_second": 511.877,
      "eval_steps_per_second": 7.998,
      "step": 4160
    },
    {
      "epoch": 8.95,
      "learning_rate": 7.203862660944207e-07,
      "loss": 1.5214,
      "step": 4170
    },
    {
      "epoch": 8.97,
      "learning_rate": 7.077253218884118e-07,
      "loss": 1.5222,
      "step": 4180
    },
    {
      "epoch": 8.99,
      "learning_rate": 6.950643776824037e-07,
      "loss": 1.5215,
      "step": 4190
    },
    {
      "epoch": 9.01,
      "learning_rate": 6.824034334763948e-07,
      "loss": 1.6641,
      "step": 4200
    },
    {
      "epoch": 9.03,
      "learning_rate": 6.69742489270386e-07,
      "loss": 1.5223,
      "step": 4210
    },
    {
      "epoch": 9.06,
      "learning_rate": 6.570815450643778e-07,
      "loss": 1.5223,
      "step": 4220
    },
    {
      "epoch": 9.08,
      "learning_rate": 6.44420600858369e-07,
      "loss": 1.5227,
      "step": 4230
    },
    {
      "epoch": 9.1,
      "learning_rate": 6.317596566523608e-07,
      "loss": 1.5217,
      "step": 4240
    },
    {
      "epoch": 9.12,
      "learning_rate": 6.19098712446352e-07,
      "loss": 1.5217,
      "step": 4250
    },
    {
      "epoch": 9.14,
      "learning_rate": 6.064377682403431e-07,
      "loss": 1.5196,
      "step": 4260
    },
    {
      "epoch": 9.16,
      "learning_rate": 5.937768240343349e-07,
      "loss": 1.5198,
      "step": 4270
    },
    {
      "epoch": 9.18,
      "learning_rate": 5.81115879828326e-07,
      "loss": 1.5228,
      "step": 4280
    },
    {
      "epoch": 9.21,
      "learning_rate": 5.684549356223179e-07,
      "loss": 1.5192,
      "step": 4290
    },
    {
      "epoch": 9.23,
      "learning_rate": 5.55793991416309e-07,
      "loss": 1.5203,
      "step": 4300
    },
    {
      "epoch": 9.25,
      "learning_rate": 5.431330472103002e-07,
      "loss": 1.5235,
      "step": 4310
    },
    {
      "epoch": 9.27,
      "learning_rate": 5.304721030042921e-07,
      "loss": 1.523,
      "step": 4320
    },
    {
      "epoch": 9.27,
      "eval_loss": 1.4170712232589722,
      "eval_runtime": 78.1196,
      "eval_samples_per_second": 512.035,
      "eval_steps_per_second": 8.001,
      "step": 4320
    },
    {
      "epoch": 9.29,
      "learning_rate": 5.178111587982832e-07,
      "loss": 1.5227,
      "step": 4330
    },
    {
      "epoch": 9.31,
      "learning_rate": 5.05150214592275e-07,
      "loss": 1.5211,
      "step": 4340
    },
    {
      "epoch": 9.33,
      "learning_rate": 4.924892703862662e-07,
      "loss": 1.5211,
      "step": 4350
    },
    {
      "epoch": 9.36,
      "learning_rate": 4.798283261802574e-07,
      "loss": 1.5206,
      "step": 4360
    },
    {
      "epoch": 9.38,
      "learning_rate": 4.6716738197424914e-07,
      "loss": 1.5224,
      "step": 4370
    },
    {
      "epoch": 9.4,
      "learning_rate": 4.545064377682403e-07,
      "loss": 1.5221,
      "step": 4380
    },
    {
      "epoch": 9.42,
      "learning_rate": 4.4184549356223147e-07,
      "loss": 1.5203,
      "step": 4390
    },
    {
      "epoch": 9.44,
      "learning_rate": 4.291845493562233e-07,
      "loss": 1.5229,
      "step": 4400
    },
    {
      "epoch": 9.46,
      "learning_rate": 4.1652360515021444e-07,
      "loss": 1.5219,
      "step": 4410
    },
    {
      "epoch": 9.48,
      "learning_rate": 4.038626609442062e-07,
      "loss": 1.5188,
      "step": 4420
    },
    {
      "epoch": 9.51,
      "learning_rate": 3.912017167381974e-07,
      "loss": 1.52,
      "step": 4430
    },
    {
      "epoch": 9.53,
      "learning_rate": 3.785407725321886e-07,
      "loss": 1.5204,
      "step": 4440
    },
    {
      "epoch": 9.55,
      "learning_rate": 3.6587982832618037e-07,
      "loss": 1.5176,
      "step": 4450
    },
    {
      "epoch": 9.57,
      "learning_rate": 3.5321888412017154e-07,
      "loss": 1.522,
      "step": 4460
    },
    {
      "epoch": 9.59,
      "learning_rate": 3.405579399141634e-07,
      "loss": 1.5204,
      "step": 4470
    },
    {
      "epoch": 9.61,
      "learning_rate": 3.2789699570815456e-07,
      "loss": 1.5212,
      "step": 4480
    },
    {
      "epoch": 9.61,
      "eval_loss": 1.4148975610733032,
      "eval_runtime": 78.114,
      "eval_samples_per_second": 512.072,
      "eval_steps_per_second": 8.001,
      "step": 4480
    },
    {
      "epoch": 9.63,
      "learning_rate": 3.1523605150214567e-07,
      "loss": 1.5188,
      "step": 4490
    },
    {
      "epoch": 9.66,
      "learning_rate": 3.0257510729613753e-07,
      "loss": 1.5224,
      "step": 4500
    },
    {
      "epoch": 9.68,
      "learning_rate": 2.8991416309012864e-07,
      "loss": 1.5266,
      "step": 4510
    },
    {
      "epoch": 9.7,
      "learning_rate": 2.772532188841205e-07,
      "loss": 1.5218,
      "step": 4520
    },
    {
      "epoch": 9.72,
      "learning_rate": 2.645922746781116e-07,
      "loss": 1.5205,
      "step": 4530
    },
    {
      "epoch": 9.74,
      "learning_rate": 2.5193133047210283e-07,
      "loss": 1.5228,
      "step": 4540
    },
    {
      "epoch": 9.76,
      "learning_rate": 2.3927038626609463e-07,
      "loss": 1.5214,
      "step": 4550
    },
    {
      "epoch": 9.78,
      "learning_rate": 2.266094420600858e-07,
      "loss": 1.5233,
      "step": 4560
    },
    {
      "epoch": 9.81,
      "learning_rate": 2.1394849785407693e-07,
      "loss": 1.5209,
      "step": 4570
    },
    {
      "epoch": 9.83,
      "learning_rate": 2.0128755364806876e-07,
      "loss": 1.5241,
      "step": 4580
    },
    {
      "epoch": 9.85,
      "learning_rate": 1.8862660944205993e-07,
      "loss": 1.521,
      "step": 4590
    },
    {
      "epoch": 9.87,
      "learning_rate": 1.7596566523605173e-07,
      "loss": 1.5219,
      "step": 4600
    },
    {
      "epoch": 9.89,
      "learning_rate": 1.633047210300429e-07,
      "loss": 1.5209,
      "step": 4610
    },
    {
      "epoch": 9.91,
      "learning_rate": 1.5064377682403406e-07,
      "loss": 1.5249,
      "step": 4620
    },
    {
      "epoch": 9.93,
      "learning_rate": 1.3798283261802586e-07,
      "loss": 1.5231,
      "step": 4630
    },
    {
      "epoch": 9.96,
      "learning_rate": 1.2532188841201703e-07,
      "loss": 1.522,
      "step": 4640
    },
    {
      "epoch": 9.96,
      "eval_loss": 1.4186652898788452,
      "eval_runtime": 78.1323,
      "eval_samples_per_second": 511.952,
      "eval_steps_per_second": 7.999,
      "step": 4640
    },
    {
      "epoch": 9.98,
      "learning_rate": 1.1266094420600883e-07,
      "loss": 1.5239,
      "step": 4650
    },
    {
      "epoch": 10.0,
      "learning_rate": 1e-07,
      "loss": 1.5179,
      "step": 4660
    },
    {
      "before_init_mem_cpu": 933711872,
      "before_init_mem_gpu": 0,
      "epoch": 10.0,
      "init_mem_cpu_alloc_delta": 2791841792,
      "init_mem_cpu_peaked_delta": 413900800,
      "init_mem_gpu_alloc_delta": 497994240,
      "init_mem_gpu_peaked_delta": 512,
      "step": 4660,
      "total_flos": 1.0068024259071967e+19,
      "train_loss": 1.525319737733178,
      "train_mem_cpu_alloc_delta": 8646197248,
      "train_mem_cpu_peaked_delta": 402849792,
      "train_mem_gpu_alloc_delta": 2010885120,
      "train_mem_gpu_peaked_delta": 15468135424,
      "train_runtime": 218631.5085,
      "train_samples_per_second": 174.954,
      "train_steps_per_second": 0.021
    }
  ],
  "max_steps": 4660,
  "num_train_epochs": 10,
  "total_flos": 1.0068024259071967e+19,
  "trial_name": null,
  "trial_params": null
}