{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 1728,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017361111111111112,
      "grad_norm": 2.007795810699463,
      "learning_rate": 5.202312138728324e-06,
      "loss": 1.463,
      "step": 10
    },
    {
      "epoch": 0.034722222222222224,
      "grad_norm": 1.1297450065612793,
      "learning_rate": 1.0982658959537573e-05,
      "loss": 1.2884,
      "step": 20
    },
    {
      "epoch": 0.052083333333333336,
      "grad_norm": 0.774868369102478,
      "learning_rate": 1.676300578034682e-05,
      "loss": 1.0796,
      "step": 30
    },
    {
      "epoch": 0.06944444444444445,
      "grad_norm": 0.5096573233604431,
      "learning_rate": 2.254335260115607e-05,
      "loss": 0.8443,
      "step": 40
    },
    {
      "epoch": 0.08680555555555555,
      "grad_norm": 0.6606957316398621,
      "learning_rate": 2.832369942196532e-05,
      "loss": 0.756,
      "step": 50
    },
    {
      "epoch": 0.10416666666666667,
      "grad_norm": 0.8858688473701477,
      "learning_rate": 3.410404624277457e-05,
      "loss": 0.6759,
      "step": 60
    },
    {
      "epoch": 0.12152777777777778,
      "grad_norm": 1.686498999595642,
      "learning_rate": 3.988439306358382e-05,
      "loss": 0.636,
      "step": 70
    },
    {
      "epoch": 0.1388888888888889,
      "grad_norm": 0.681108295917511,
      "learning_rate": 4.566473988439307e-05,
      "loss": 0.5915,
      "step": 80
    },
    {
      "epoch": 0.15625,
      "grad_norm": 0.657633900642395,
      "learning_rate": 5.1445086705202317e-05,
      "loss": 0.6144,
      "step": 90
    },
    {
      "epoch": 0.1736111111111111,
      "grad_norm": 0.9662413001060486,
      "learning_rate": 5.722543352601156e-05,
      "loss": 0.656,
      "step": 100
    },
    {
      "epoch": 0.1736111111111111,
      "eval_loss": 0.5766947269439697,
      "eval_runtime": 513.1213,
      "eval_samples_per_second": 2.247,
      "eval_steps_per_second": 2.247,
      "step": 100
    },
    {
      "epoch": 0.1909722222222222,
      "grad_norm": 0.6163507699966431,
      "learning_rate": 6.300578034682081e-05,
      "loss": 0.5819,
      "step": 110
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 0.7107399106025696,
      "learning_rate": 6.878612716763007e-05,
      "loss": 0.5718,
      "step": 120
    },
    {
      "epoch": 0.22569444444444445,
      "grad_norm": 0.6259875893592834,
      "learning_rate": 7.456647398843931e-05,
      "loss": 0.658,
      "step": 130
    },
    {
      "epoch": 0.24305555555555555,
      "grad_norm": 0.5605979561805725,
      "learning_rate": 8.034682080924855e-05,
      "loss": 0.549,
      "step": 140
    },
    {
      "epoch": 0.2604166666666667,
      "grad_norm": 0.5297986268997192,
      "learning_rate": 8.612716763005781e-05,
      "loss": 0.5671,
      "step": 150
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 0.5555981397628784,
      "learning_rate": 9.190751445086706e-05,
      "loss": 0.5569,
      "step": 160
    },
    {
      "epoch": 0.2951388888888889,
      "grad_norm": 0.5598844885826111,
      "learning_rate": 9.768786127167631e-05,
      "loss": 0.5862,
      "step": 170
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.5559885501861572,
      "learning_rate": 9.999632653293776e-05,
      "loss": 0.5738,
      "step": 180
    },
    {
      "epoch": 0.3298611111111111,
      "grad_norm": 0.4869542717933655,
      "learning_rate": 9.99738795222662e-05,
      "loss": 0.5789,
      "step": 190
    },
    {
      "epoch": 0.3472222222222222,
      "grad_norm": 0.5302107334136963,
      "learning_rate": 9.993103546669093e-05,
      "loss": 0.5695,
      "step": 200
    },
    {
      "epoch": 0.3472222222222222,
      "eval_loss": 0.5406240224838257,
      "eval_runtime": 515.9751,
      "eval_samples_per_second": 2.235,
      "eval_steps_per_second": 2.235,
      "step": 200
    },
    {
      "epoch": 0.3645833333333333,
      "grad_norm": 0.4969506561756134,
      "learning_rate": 9.98678118531898e-05,
      "loss": 0.5943,
      "step": 210
    },
    {
      "epoch": 0.3819444444444444,
      "grad_norm": 0.5448970794677734,
      "learning_rate": 9.97842344867413e-05,
      "loss": 0.5719,
      "step": 220
    },
    {
      "epoch": 0.3993055555555556,
      "grad_norm": 0.5558710098266602,
      "learning_rate": 9.968033747979195e-05,
      "loss": 0.5303,
      "step": 230
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.5268853902816772,
      "learning_rate": 9.955616323833343e-05,
      "loss": 0.5646,
      "step": 240
    },
    {
      "epoch": 0.4340277777777778,
      "grad_norm": 0.5985867381095886,
      "learning_rate": 9.941176244459418e-05,
      "loss": 0.5558,
      "step": 250
    },
    {
      "epoch": 0.4513888888888889,
      "grad_norm": 0.37411603331565857,
      "learning_rate": 9.924719403635333e-05,
      "loss": 0.5372,
      "step": 260
    },
    {
      "epoch": 0.46875,
      "grad_norm": 0.4203139841556549,
      "learning_rate": 9.906252518288501e-05,
      "loss": 0.5492,
      "step": 270
    },
    {
      "epoch": 0.4861111111111111,
      "grad_norm": 0.5267091393470764,
      "learning_rate": 9.885783125754291e-05,
      "loss": 0.5233,
      "step": 280
    },
    {
      "epoch": 0.5034722222222222,
      "grad_norm": 0.5201379656791687,
      "learning_rate": 9.863319580699635e-05,
      "loss": 0.5909,
      "step": 290
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 0.419812947511673,
      "learning_rate": 9.838871051713033e-05,
      "loss": 0.5849,
      "step": 300
    },
    {
      "epoch": 0.5208333333333334,
      "eval_loss": 0.4985196888446808,
      "eval_runtime": 506.4652,
      "eval_samples_per_second": 2.277,
      "eval_steps_per_second": 2.277,
      "step": 300
    },
    {
      "epoch": 0.5381944444444444,
      "grad_norm": 0.5255069136619568,
      "learning_rate": 9.81244751756236e-05,
      "loss": 0.5894,
      "step": 310
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.5103104114532471,
      "learning_rate": 9.784059763121993e-05,
      "loss": 0.5187,
      "step": 320
    },
    {
      "epoch": 0.5729166666666666,
      "grad_norm": 0.6287491321563721,
      "learning_rate": 9.753719374970908e-05,
      "loss": 0.5181,
      "step": 330
    },
    {
      "epoch": 0.5902777777777778,
      "grad_norm": 0.8683649897575378,
      "learning_rate": 9.721438736663585e-05,
      "loss": 0.5413,
      "step": 340
    },
    {
      "epoch": 0.6076388888888888,
      "grad_norm": 0.7942582964897156,
      "learning_rate": 9.687231023675597e-05,
      "loss": 0.4949,
      "step": 350
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.476675808429718,
      "learning_rate": 9.651110198025997e-05,
      "loss": 0.5139,
      "step": 360
    },
    {
      "epoch": 0.6423611111111112,
      "grad_norm": 0.42043331265449524,
      "learning_rate": 9.613091002578649e-05,
      "loss": 0.5163,
      "step": 370
    },
    {
      "epoch": 0.6597222222222222,
      "grad_norm": 0.4923442602157593,
      "learning_rate": 9.573188955024882e-05,
      "loss": 0.5453,
      "step": 380
    },
    {
      "epoch": 0.6770833333333334,
      "grad_norm": 0.4754851758480072,
      "learning_rate": 9.531420341549865e-05,
      "loss": 0.5674,
      "step": 390
    },
    {
      "epoch": 0.6944444444444444,
      "grad_norm": 0.49470463395118713,
      "learning_rate": 9.48780221018535e-05,
      "loss": 0.527,
      "step": 400
    },
    {
      "epoch": 0.6944444444444444,
      "eval_loss": 0.48567065596580505,
      "eval_runtime": 512.591,
      "eval_samples_per_second": 2.249,
      "eval_steps_per_second": 2.249,
      "step": 400
    },
    {
      "epoch": 0.7118055555555556,
      "grad_norm": 0.5066745281219482,
      "learning_rate": 9.442352363851428e-05,
      "loss": 0.555,
      "step": 410
    },
    {
      "epoch": 0.7291666666666666,
      "grad_norm": 0.3770899772644043,
      "learning_rate": 9.395089353090213e-05,
      "loss": 0.4918,
      "step": 420
    },
    {
      "epoch": 0.7465277777777778,
      "grad_norm": 0.497141569852829,
      "learning_rate": 9.346032468494353e-05,
      "loss": 0.5761,
      "step": 430
    },
    {
      "epoch": 0.7638888888888888,
      "grad_norm": 0.3677353858947754,
      "learning_rate": 9.295201732833487e-05,
      "loss": 0.5363,
      "step": 440
    },
    {
      "epoch": 0.78125,
      "grad_norm": 0.5019080638885498,
      "learning_rate": 9.242617892881889e-05,
      "loss": 0.5385,
      "step": 450
    },
    {
      "epoch": 0.7986111111111112,
      "grad_norm": 0.4239552617073059,
      "learning_rate": 9.188302410950565e-05,
      "loss": 0.5026,
      "step": 460
    },
    {
      "epoch": 0.8159722222222222,
      "grad_norm": 0.4331854283809662,
      "learning_rate": 9.13227745612734e-05,
      "loss": 0.5322,
      "step": 470
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.4505394399166107,
      "learning_rate": 9.074565895228442e-05,
      "loss": 0.5145,
      "step": 480
    },
    {
      "epoch": 0.8506944444444444,
      "grad_norm": 0.4377307891845703,
      "learning_rate": 9.015191283465319e-05,
      "loss": 0.5339,
      "step": 490
    },
    {
      "epoch": 0.8680555555555556,
      "grad_norm": 0.5486445426940918,
      "learning_rate": 8.954177854830487e-05,
      "loss": 0.4782,
      "step": 500
    },
    {
      "epoch": 0.8680555555555556,
      "eval_loss": 0.4704386591911316,
      "eval_runtime": 507.1518,
      "eval_samples_per_second": 2.273,
      "eval_steps_per_second": 2.273,
      "step": 500
    },
    {
      "epoch": 0.8854166666666666,
      "grad_norm": 0.505261242389679,
      "learning_rate": 8.891550512206322e-05,
      "loss": 0.51,
      "step": 510
    },
    {
      "epoch": 0.9027777777777778,
      "grad_norm": 0.4194414019584656,
      "learning_rate": 8.827334817200845e-05,
      "loss": 0.5049,
      "step": 520
    },
    {
      "epoch": 0.9201388888888888,
      "grad_norm": 0.5332881212234497,
      "learning_rate": 8.76155697971465e-05,
      "loss": 0.5303,
      "step": 530
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.4372180700302124,
      "learning_rate": 8.694243847243224e-05,
      "loss": 0.5237,
      "step": 540
    },
    {
      "epoch": 0.9548611111111112,
      "grad_norm": 0.5241920948028564,
      "learning_rate": 8.625422893919027e-05,
      "loss": 0.5211,
      "step": 550
    },
    {
      "epoch": 0.9722222222222222,
      "grad_norm": 0.4739208221435547,
      "learning_rate": 8.555122209297817e-05,
      "loss": 0.4999,
      "step": 560
    },
    {
      "epoch": 0.9895833333333334,
      "grad_norm": 0.4937109649181366,
      "learning_rate": 8.483370486893783e-05,
      "loss": 0.5348,
      "step": 570
    },
    {
      "epoch": 1.0069444444444444,
      "grad_norm": 0.40740323066711426,
      "learning_rate": 8.410197012468156e-05,
      "loss": 0.487,
      "step": 580
    },
    {
      "epoch": 1.0243055555555556,
      "grad_norm": 0.4971655607223511,
      "learning_rate": 8.335631652076125e-05,
      "loss": 0.4467,
      "step": 590
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 0.5765105485916138,
      "learning_rate": 8.259704839876871e-05,
      "loss": 0.4357,
      "step": 600
    },
    {
      "epoch": 1.0416666666666667,
      "eval_loss": 0.4711858034133911,
      "eval_runtime": 502.4975,
      "eval_samples_per_second": 2.295,
      "eval_steps_per_second": 2.295,
      "step": 600
    },
    {
      "epoch": 1.0590277777777777,
      "grad_norm": 0.6608824133872986,
      "learning_rate": 8.182447565711737e-05,
      "loss": 0.4109,
      "step": 610
    },
    {
      "epoch": 1.0763888888888888,
      "grad_norm": 0.7043329477310181,
      "learning_rate": 8.103891362455602e-05,
      "loss": 0.425,
      "step": 620
    },
    {
      "epoch": 1.09375,
      "grad_norm": 0.6336780786514282,
      "learning_rate": 8.024068293146612e-05,
      "loss": 0.4188,
      "step": 630
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.6143578886985779,
      "learning_rate": 7.943010937899511e-05,
      "loss": 0.4329,
      "step": 640
    },
    {
      "epoch": 1.1284722222222223,
      "grad_norm": 0.4573313891887665,
      "learning_rate": 7.860752380607936e-05,
      "loss": 0.4484,
      "step": 650
    },
    {
      "epoch": 1.1458333333333333,
      "grad_norm": 0.49324455857276917,
      "learning_rate": 7.777326195441088e-05,
      "loss": 0.4482,
      "step": 660
    },
    {
      "epoch": 1.1631944444444444,
      "grad_norm": 0.5778242349624634,
      "learning_rate": 7.692766433140302e-05,
      "loss": 0.4013,
      "step": 670
    },
    {
      "epoch": 1.1805555555555556,
      "grad_norm": 0.515394926071167,
      "learning_rate": 7.607107607121086e-05,
      "loss": 0.4149,
      "step": 680
    },
    {
      "epoch": 1.1979166666666667,
      "grad_norm": 0.5518147349357605,
      "learning_rate": 7.520384679386344e-05,
      "loss": 0.4334,
      "step": 690
    },
    {
      "epoch": 1.2152777777777777,
      "grad_norm": 0.6633225679397583,
      "learning_rate": 7.432633046256472e-05,
      "loss": 0.4033,
      "step": 700
    },
    {
      "epoch": 1.2152777777777777,
      "eval_loss": 0.4653947055339813,
      "eval_runtime": 508.2251,
      "eval_samples_per_second": 2.269,
      "eval_steps_per_second": 2.269,
      "step": 700
    },
    {
      "epoch": 1.2326388888888888,
      "grad_norm": 0.48088809847831726,
      "learning_rate": 7.343888523922219e-05,
      "loss": 0.4328,
      "step": 710
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.6473700404167175,
      "learning_rate": 7.25418733382615e-05,
      "loss": 0.4019,
      "step": 720
    },
    {
      "epoch": 1.2673611111111112,
      "grad_norm": 0.6023727059364319,
      "learning_rate": 7.163566087878726e-05,
      "loss": 0.4238,
      "step": 730
    },
    {
      "epoch": 1.2847222222222223,
      "grad_norm": 0.5473999977111816,
      "learning_rate": 7.072061773514991e-05,
      "loss": 0.4204,
      "step": 740
    },
    {
      "epoch": 1.3020833333333333,
      "grad_norm": 0.5065605640411377,
      "learning_rate": 6.979711738598004e-05,
      "loss": 0.4175,
      "step": 750
    },
    {
      "epoch": 1.3194444444444444,
      "grad_norm": 0.48466363549232483,
      "learning_rate": 6.886553676175155e-05,
      "loss": 0.4559,
      "step": 760
    },
    {
      "epoch": 1.3368055555555556,
      "grad_norm": 0.3937870264053345,
      "learning_rate": 6.792625609093593e-05,
      "loss": 0.4052,
      "step": 770
    },
    {
      "epoch": 1.3541666666666667,
      "grad_norm": 0.5539367198944092,
      "learning_rate": 6.697965874481038e-05,
      "loss": 0.4268,
      "step": 780
    },
    {
      "epoch": 1.3715277777777777,
      "grad_norm": 0.5883862376213074,
      "learning_rate": 6.602613108098327e-05,
      "loss": 0.3801,
      "step": 790
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.5152338743209839,
      "learning_rate": 6.506606228570077e-05,
      "loss": 0.4438,
      "step": 800
    },
    {
      "epoch": 1.3888888888888888,
      "eval_loss": 0.4593656063079834,
      "eval_runtime": 507.6748,
      "eval_samples_per_second": 2.271,
      "eval_steps_per_second": 2.271,
      "step": 800
    },
    {
      "epoch": 1.40625,
      "grad_norm": 0.5437167286872864,
      "learning_rate": 6.409984421499877e-05,
      "loss": 0.4497,
      "step": 810
    },
    {
      "epoch": 1.4236111111111112,
      "grad_norm": 0.560701847076416,
      "learning_rate": 6.312787123476523e-05,
      "loss": 0.4215,
      "step": 820
    },
    {
      "epoch": 1.4409722222222223,
      "grad_norm": 0.4475458860397339,
      "learning_rate": 6.215054005977829e-05,
      "loss": 0.417,
      "step": 830
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 0.6264685988426208,
      "learning_rate": 6.11682495917852e-05,
      "loss": 0.3694,
      "step": 840
    },
    {
      "epoch": 1.4756944444444444,
      "grad_norm": 0.4710831940174103,
      "learning_rate": 6.0181400756689185e-05,
      "loss": 0.4168,
      "step": 850
    },
    {
      "epoch": 1.4930555555555556,
      "grad_norm": 0.4411803185939789,
      "learning_rate": 5.919039634090979e-05,
      "loss": 0.4061,
      "step": 860
    },
    {
      "epoch": 1.5104166666666665,
      "grad_norm": 0.5613175630569458,
      "learning_rate": 5.819564082698392e-05,
      "loss": 0.4011,
      "step": 870
    },
    {
      "epoch": 1.5277777777777777,
      "grad_norm": 0.5883941054344177,
      "learning_rate": 5.7197540228474634e-05,
      "loss": 0.4046,
      "step": 880
    },
    {
      "epoch": 1.5451388888888888,
      "grad_norm": 0.7705323100090027,
      "learning_rate": 5.6196501924255083e-05,
      "loss": 0.4447,
      "step": 890
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.4412948489189148,
      "learning_rate": 5.5192934492235116e-05,
      "loss": 0.4297,
      "step": 900
    },
    {
      "epoch": 1.5625,
      "eval_loss": 0.45315322279930115,
      "eval_runtime": 509.3304,
      "eval_samples_per_second": 2.264,
      "eval_steps_per_second": 2.264,
      "step": 900
    },
    {
      "epoch": 1.5798611111111112,
      "grad_norm": 0.4683547914028168,
      "learning_rate": 5.4187247542598495e-05,
      "loss": 0.4094,
      "step": 910
    },
    {
      "epoch": 1.5972222222222223,
      "grad_norm": 0.5470554232597351,
      "learning_rate": 5.317985155061882e-05,
      "loss": 0.413,
      "step": 920
    },
    {
      "epoch": 1.6145833333333335,
      "grad_norm": 0.476554274559021,
      "learning_rate": 5.217115768912242e-05,
      "loss": 0.3978,
      "step": 930
    },
    {
      "epoch": 1.6319444444444444,
      "grad_norm": 0.4833614230155945,
      "learning_rate": 5.116157766066637e-05,
      "loss": 0.4377,
      "step": 940
    },
    {
      "epoch": 1.6493055555555556,
      "grad_norm": 0.5570272207260132,
      "learning_rate": 5.015152352950045e-05,
      "loss": 0.4299,
      "step": 950
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.37459680438041687,
      "learning_rate": 4.914140755338144e-05,
      "loss": 0.4285,
      "step": 960
    },
    {
      "epoch": 1.6840277777777777,
      "grad_norm": 0.462679922580719,
      "learning_rate": 4.813164201530838e-05,
      "loss": 0.4038,
      "step": 970
    },
    {
      "epoch": 1.7013888888888888,
      "grad_norm": 0.5221256017684937,
      "learning_rate": 4.712263905524754e-05,
      "loss": 0.4016,
      "step": 980
    },
    {
      "epoch": 1.71875,
      "grad_norm": 0.6251868009567261,
      "learning_rate": 4.611481050191587e-05,
      "loss": 0.4188,
      "step": 990
    },
    {
      "epoch": 1.7361111111111112,
      "grad_norm": 0.5684912204742432,
      "learning_rate": 4.5108567704691405e-05,
      "loss": 0.4012,
      "step": 1000
    },
    {
      "epoch": 1.7361111111111112,
      "eval_loss": 0.4480222463607788,
      "eval_runtime": 505.0708,
      "eval_samples_per_second": 2.283,
      "eval_steps_per_second": 2.283,
      "step": 1000
    },
    {
      "epoch": 1.7534722222222223,
      "grad_norm": 0.44916579127311707,
      "learning_rate": 4.4104321365719216e-05,
      "loss": 0.3071,
      "step": 1010
    },
    {
      "epoch": 1.7708333333333335,
      "grad_norm": 0.4596291482448578,
      "learning_rate": 4.310248137228176e-05,
      "loss": 0.302,
      "step": 1020
    },
    {
      "epoch": 1.7881944444444444,
      "grad_norm": 0.44650307297706604,
      "learning_rate": 4.210345662950161e-05,
      "loss": 0.2602,
      "step": 1030
    },
    {
      "epoch": 1.8055555555555556,
      "grad_norm": 0.4998283088207245,
      "learning_rate": 4.11076548934452e-05,
      "loss": 0.3086,
      "step": 1040
    },
    {
      "epoch": 1.8229166666666665,
      "grad_norm": 0.4060577154159546,
      "learning_rate": 4.011548260469552e-05,
      "loss": 0.2801,
      "step": 1050
    },
    {
      "epoch": 1.8402777777777777,
      "grad_norm": 0.8512741327285767,
      "learning_rate": 3.9127344722461615e-05,
      "loss": 0.3004,
      "step": 1060
    },
    {
      "epoch": 1.8576388888888888,
      "grad_norm": 0.6027821898460388,
      "learning_rate": 3.8143644559292927e-05,
      "loss": 0.2837,
      "step": 1070
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.5069570541381836,
      "learning_rate": 3.716478361646555e-05,
      "loss": 0.2604,
      "step": 1080
    },
    {
      "epoch": 1.8923611111111112,
      "grad_norm": 0.56328284740448,
      "learning_rate": 3.6191161420107864e-05,
      "loss": 0.304,
      "step": 1090
    },
    {
      "epoch": 1.9097222222222223,
      "grad_norm": 0.5753824710845947,
      "learning_rate": 3.5223175358132256e-05,
      "loss": 0.2866,
      "step": 1100
    },
    {
      "epoch": 1.9097222222222223,
      "eval_loss": 0.48055773973464966,
      "eval_runtime": 491.0229,
      "eval_samples_per_second": 2.348,
      "eval_steps_per_second": 2.348,
      "step": 1100
    },
    {
      "epoch": 1.9270833333333335,
      "grad_norm": 0.6144798398017883,
      "learning_rate": 3.426122051803969e-05,
      "loss": 0.2847,
      "step": 1110
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 0.7388680577278137,
      "learning_rate": 3.330568952566302e-05,
      "loss": 0.278,
      "step": 1120
    },
    {
      "epoch": 1.9618055555555556,
      "grad_norm": 0.7205446362495422,
      "learning_rate": 3.235697238491516e-05,
      "loss": 0.2883,
      "step": 1130
    },
    {
      "epoch": 1.9791666666666665,
      "grad_norm": 0.6098833680152893,
      "learning_rate": 3.141545631860746e-05,
      "loss": 0.2924,
      "step": 1140
    },
    {
      "epoch": 1.9965277777777777,
      "grad_norm": 0.5834055542945862,
      "learning_rate": 3.0481525610402978e-05,
      "loss": 0.3073,
      "step": 1150
    },
    {
      "epoch": 2.013888888888889,
      "grad_norm": 0.6533185839653015,
      "learning_rate": 2.9555561447969683e-05,
      "loss": 0.304,
      "step": 1160
    },
    {
      "epoch": 2.03125,
      "grad_norm": 0.6724961996078491,
      "learning_rate": 2.8637941767396954e-05,
      "loss": 0.2801,
      "step": 1170
    },
    {
      "epoch": 2.048611111111111,
      "grad_norm": 0.771094024181366,
      "learning_rate": 2.7729041098939585e-05,
      "loss": 0.287,
      "step": 1180
    },
    {
      "epoch": 2.0659722222222223,
      "grad_norm": 0.6162275075912476,
      "learning_rate": 2.682923041415163e-05,
      "loss": 0.3276,
      "step": 1190
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 0.7060827016830444,
      "learning_rate": 2.5938876974472802e-05,
      "loss": 0.2718,
      "step": 1200
    },
    {
      "epoch": 2.0833333333333335,
      "eval_loss": 0.47958508133888245,
      "eval_runtime": 491.4469,
      "eval_samples_per_second": 2.346,
      "eval_steps_per_second": 2.346,
      "step": 1200
    },
    {
      "epoch": 2.1006944444444446,
      "grad_norm": 0.7561212182044983,
      "learning_rate": 2.5058344181329273e-05,
      "loss": 0.2929,
      "step": 1210
    },
    {
      "epoch": 2.1180555555555554,
      "grad_norm": 0.5527466535568237,
      "learning_rate": 2.418799142780982e-05,
      "loss": 0.3054,
      "step": 1220
    },
    {
      "epoch": 2.1354166666666665,
      "grad_norm": 0.6940686702728271,
      "learning_rate": 2.332817395197813e-05,
      "loss": 0.3087,
      "step": 1230
    },
    {
      "epoch": 2.1527777777777777,
      "grad_norm": 0.7222234010696411,
      "learning_rate": 2.2479242691880887e-05,
      "loss": 0.286,
      "step": 1240
    },
    {
      "epoch": 2.170138888888889,
      "grad_norm": 0.6128625273704529,
      "learning_rate": 2.164154414231091e-05,
      "loss": 0.2902,
      "step": 1250
    },
    {
      "epoch": 2.1875,
      "grad_norm": 0.672198474407196,
      "learning_rate": 2.0815420213383936e-05,
      "loss": 0.2631,
      "step": 1260
    },
    {
      "epoch": 2.204861111111111,
      "grad_norm": 0.5412243008613586,
      "learning_rate": 2.0001208090986557e-05,
      "loss": 0.3163,
      "step": 1270
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.8546826839447021,
      "learning_rate": 1.9199240099152498e-05,
      "loss": 0.3257,
      "step": 1280
    },
    {
      "epoch": 2.2395833333333335,
      "grad_norm": 0.7093095183372498,
      "learning_rate": 1.8409843564423056e-05,
      "loss": 0.2952,
      "step": 1290
    },
    {
      "epoch": 2.2569444444444446,
      "grad_norm": 0.6084073781967163,
      "learning_rate": 1.7633340682247558e-05,
      "loss": 0.3169,
      "step": 1300
    },
    {
      "epoch": 2.2569444444444446,
      "eval_loss": 0.468641072511673,
      "eval_runtime": 486.238,
      "eval_samples_per_second": 2.371,
      "eval_steps_per_second": 2.371,
      "step": 1300
    },
    {
      "epoch": 2.2743055555555554,
      "grad_norm": 0.7556958198547363,
      "learning_rate": 1.6870048385477848e-05,
      "loss": 0.3121,
      "step": 1310
    },
    {
      "epoch": 2.2916666666666665,
      "grad_norm": 0.6000959873199463,
      "learning_rate": 1.6120278215010886e-05,
      "loss": 0.2928,
      "step": 1320
    },
    {
      "epoch": 2.3090277777777777,
      "grad_norm": 0.6133815050125122,
      "learning_rate": 1.5384336192632104e-05,
      "loss": 0.2578,
      "step": 1330
    },
    {
      "epoch": 2.326388888888889,
      "grad_norm": 0.7813517451286316,
      "learning_rate": 1.4662522696111225e-05,
      "loss": 0.2856,
      "step": 1340
    },
    {
      "epoch": 2.34375,
      "grad_norm": 0.738869309425354,
      "learning_rate": 1.395513233660199e-05,
      "loss": 0.2829,
      "step": 1350
    },
    {
      "epoch": 2.361111111111111,
      "grad_norm": 0.7364549040794373,
      "learning_rate": 1.3262453838395205e-05,
      "loss": 0.3188,
      "step": 1360
    },
    {
      "epoch": 2.3784722222222223,
      "grad_norm": 0.5563517212867737,
      "learning_rate": 1.2584769921074962e-05,
      "loss": 0.295,
      "step": 1370
    },
    {
      "epoch": 2.3958333333333335,
      "grad_norm": 0.6138696074485779,
      "learning_rate": 1.1922357184125327e-05,
      "loss": 0.3112,
      "step": 1380
    },
    {
      "epoch": 2.4131944444444446,
      "grad_norm": 0.6921333074569702,
      "learning_rate": 1.1275485994035107e-05,
      "loss": 0.2837,
      "step": 1390
    },
    {
      "epoch": 2.4305555555555554,
      "grad_norm": 0.6158069968223572,
      "learning_rate": 1.064442037394679e-05,
      "loss": 0.276,
      "step": 1400
    },
    {
      "epoch": 2.4305555555555554,
      "eval_loss": 0.46923181414604187,
      "eval_runtime": 491.6516,
      "eval_samples_per_second": 2.345,
      "eval_steps_per_second": 2.345,
      "step": 1400
    },
    {
      "epoch": 2.4479166666666665,
      "grad_norm": 0.5451982021331787,
      "learning_rate": 1.0029417895894211e-05,
      "loss": 0.305,
      "step": 1410
    },
    {
      "epoch": 2.4652777777777777,
      "grad_norm": 0.5262075662612915,
      "learning_rate": 9.430729575673658e-06,
      "loss": 0.263,
      "step": 1420
    },
    {
      "epoch": 2.482638888888889,
      "grad_norm": 0.5978283286094666,
      "learning_rate": 8.848599770390653e-06,
      "loss": 0.2709,
      "step": 1430
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.7257421612739563,
      "learning_rate": 8.283266078724571e-06,
      "loss": 0.2834,
      "step": 1440
    },
    {
      "epoch": 2.517361111111111,
      "grad_norm": 0.682624101638794,
      "learning_rate": 7.734959243951783e-06,
      "loss": 0.2785,
      "step": 1450
    },
    {
      "epoch": 2.5347222222222223,
      "grad_norm": 0.5846174955368042,
      "learning_rate": 7.203903059766848e-06,
      "loss": 0.2671,
      "step": 1460
    },
    {
      "epoch": 2.5520833333333335,
      "grad_norm": 0.7986878752708435,
      "learning_rate": 6.690314278940207e-06,
      "loss": 0.2646,
      "step": 1470
    },
    {
      "epoch": 2.5694444444444446,
      "grad_norm": 0.6198819875717163,
      "learning_rate": 6.194402524849585e-06,
      "loss": 0.2905,
      "step": 1480
    },
    {
      "epoch": 2.5868055555555554,
      "grad_norm": 0.5760586261749268,
      "learning_rate": 5.716370205921434e-06,
      "loss": 0.2948,
      "step": 1490
    },
    {
      "epoch": 2.6041666666666665,
      "grad_norm": 0.707608163356781,
      "learning_rate": 5.256412433017005e-06,
      "loss": 0.2616,
      "step": 1500
    },
    {
      "epoch": 2.6041666666666665,
      "eval_loss": 0.4693165123462677,
      "eval_runtime": 489.2257,
      "eval_samples_per_second": 2.357,
      "eval_steps_per_second": 2.357,
      "step": 1500
    },
    {
      "epoch": 2.6215277777777777,
      "grad_norm": 0.43746083974838257,
      "learning_rate": 4.814716939797109e-06,
      "loss": 0.2411,
      "step": 1510
    },
    {
      "epoch": 2.638888888888889,
      "grad_norm": 0.6086592674255371,
      "learning_rate": 4.391464006097845e-06,
      "loss": 0.2909,
      "step": 1520
    },
    {
      "epoch": 2.65625,
      "grad_norm": 0.6783342957496643,
      "learning_rate": 3.986826384348608e-06,
      "loss": 0.2828,
      "step": 1530
    },
    {
      "epoch": 2.673611111111111,
      "grad_norm": 2.342499017715454,
      "learning_rate": 3.600969229062523e-06,
      "loss": 0.3004,
      "step": 1540
    },
    {
      "epoch": 2.6909722222222223,
      "grad_norm": 0.7391747832298279,
      "learning_rate": 3.2340500294278876e-06,
      "loss": 0.3031,
      "step": 1550
    },
    {
      "epoch": 2.7083333333333335,
      "grad_norm": 0.8236761093139648,
      "learning_rate": 2.8862185450283572e-06,
      "loss": 0.2654,
      "step": 1560
    },
    {
      "epoch": 2.7256944444444446,
      "grad_norm": 0.5708310604095459,
      "learning_rate": 2.55761674471795e-06,
      "loss": 0.2812,
      "step": 1570
    },
    {
      "epoch": 2.7430555555555554,
      "grad_norm": 0.6445372104644775,
      "learning_rate": 2.24837874867585e-06,
      "loss": 0.2854,
      "step": 1580
    },
    {
      "epoch": 2.7604166666666665,
      "grad_norm": 0.7138336300849915,
      "learning_rate": 1.9586307736647323e-06,
      "loss": 0.2975,
      "step": 1590
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.7076891660690308,
      "learning_rate": 1.6884910815148425e-06,
      "loss": 0.2545,
      "step": 1600
    },
    {
      "epoch": 2.7777777777777777,
      "eval_loss": 0.4670693278312683,
      "eval_runtime": 490.7784,
      "eval_samples_per_second": 2.349,
      "eval_steps_per_second": 2.349,
      "step": 1600
    },
    {
      "epoch": 2.795138888888889,
      "grad_norm": 0.6267815828323364,
      "learning_rate": 1.4380699308549572e-06,
      "loss": 0.2886,
      "step": 1610
    },
    {
      "epoch": 2.8125,
      "grad_norm": 0.6017192602157593,
      "learning_rate": 1.2074695321098895e-06,
      "loss": 0.2791,
      "step": 1620
    },
    {
      "epoch": 2.829861111111111,
      "grad_norm": 0.6211732029914856,
      "learning_rate": 9.96784005782836e-07,
      "loss": 0.2814,
      "step": 1630
    },
    {
      "epoch": 2.8472222222222223,
      "grad_norm": 0.6479988694190979,
      "learning_rate": 8.060993440397624e-07,
      "loss": 0.2814,
      "step": 1640
    },
    {
      "epoch": 2.8645833333333335,
      "grad_norm": 0.6914340257644653,
      "learning_rate": 6.354933756112902e-07,
      "loss": 0.2736,
      "step": 1650
    },
    {
      "epoch": 2.8819444444444446,
      "grad_norm": 0.5987943410873413,
      "learning_rate": 4.850357340266231e-07,
      "loss": 0.286,
      "step": 1660
    },
    {
      "epoch": 2.8993055555555554,
      "grad_norm": 0.8642510771751404,
      "learning_rate": 3.5478782919230457e-07,
      "loss": 0.2604,
      "step": 1670
    },
    {
      "epoch": 2.9166666666666665,
      "grad_norm": 0.6662091016769409,
      "learning_rate": 2.4480282232754445e-07,
      "loss": 0.288,
      "step": 1680
    },
    {
      "epoch": 2.9340277777777777,
      "grad_norm": 0.6560271978378296,
      "learning_rate": 1.5512560426621526e-07,
      "loss": 0.2588,
      "step": 1690
    },
    {
      "epoch": 2.951388888888889,
      "grad_norm": 0.5207160115242004,
      "learning_rate": 8.579277713448574e-08,
      "loss": 0.2673,
      "step": 1700
    },
    {
      "epoch": 2.951388888888889,
      "eval_loss": 0.46653616428375244,
      "eval_runtime": 493.8981,
      "eval_samples_per_second": 2.334,
      "eval_steps_per_second": 2.334,
      "step": 1700
    },
    {
      "epoch": 2.96875,
      "grad_norm": 0.6704832315444946,
      "learning_rate": 3.6832639411521485e-08,
      "loss": 0.2919,
      "step": 1710
    },
    {
      "epoch": 2.986111111111111,
      "grad_norm": 0.6270363330841064,
      "learning_rate": 8.26517437933494e-09,
      "loss": 0.2505,
      "step": 1720
    },
    {
      "epoch": 3.0,
      "step": 1728,
      "total_flos": 2.0991618521684705e+18,
      "train_loss": 0.12052025701160785,
      "train_runtime": 11479.8644,
      "train_samples_per_second": 1.204,
      "train_steps_per_second": 0.151
    }
  ],
  "logging_steps": 10,
  "max_steps": 1728,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.0991618521684705e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}