{
  "best_global_step": 8815,
  "best_metric": 0.545,
  "best_model_checkpoint": "results_indot5-dapt-finetune-1/checkpoint-8815",
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 10250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04878048780487805,
      "grad_norm": 17.200429916381836,
      "learning_rate": 4.4009779951100247e-07,
      "loss": 7.0618,
      "step": 10
    },
    {
      "epoch": 0.0975609756097561,
      "grad_norm": 25.410673141479492,
      "learning_rate": 9.290953545232274e-07,
      "loss": 7.0349,
      "step": 20
    },
    {
      "epoch": 0.14634146341463414,
      "grad_norm": 19.62042999267578,
      "learning_rate": 1.4180929095354523e-06,
      "loss": 7.0469,
      "step": 30
    },
    {
      "epoch": 0.1951219512195122,
      "grad_norm": 56.81166076660156,
      "learning_rate": 1.9070904645476774e-06,
      "loss": 6.9333,
      "step": 40
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 107.1426773071289,
      "learning_rate": 2.3960880195599024e-06,
      "loss": 6.7783,
      "step": 50
    },
    {
      "epoch": 0.2926829268292683,
      "grad_norm": 12.458768844604492,
      "learning_rate": 2.8850855745721272e-06,
      "loss": 6.445,
      "step": 60
    },
    {
      "epoch": 0.34146341463414637,
      "grad_norm": 8.825554847717285,
      "learning_rate": 3.3740831295843525e-06,
      "loss": 6.2858,
      "step": 70
    },
    {
      "epoch": 0.3902439024390244,
      "grad_norm": 35.15165710449219,
      "learning_rate": 3.863080684596577e-06,
      "loss": 6.1864,
      "step": 80
    },
    {
      "epoch": 0.43902439024390244,
      "grad_norm": 7.6371073722839355,
      "learning_rate": 4.352078239608802e-06,
      "loss": 5.9375,
      "step": 90
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 5.905430793762207,
      "learning_rate": 4.841075794621027e-06,
      "loss": 5.8115,
      "step": 100
    },
    {
      "epoch": 0.5365853658536586,
      "grad_norm": 4.986270427703857,
      "learning_rate": 5.330073349633253e-06,
      "loss": 5.6514,
      "step": 110
    },
    {
      "epoch": 0.5853658536585366,
      "grad_norm": 4.87542724609375,
      "learning_rate": 5.8190709046454775e-06,
      "loss": 5.4786,
      "step": 120
    },
    {
      "epoch": 0.6341463414634146,
      "grad_norm": 4.698776721954346,
      "learning_rate": 6.308068459657702e-06,
      "loss": 5.4382,
      "step": 130
    },
    {
      "epoch": 0.6829268292682927,
      "grad_norm": 3.5507872104644775,
      "learning_rate": 6.797066014669927e-06,
      "loss": 5.3708,
      "step": 140
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 3.769075393676758,
      "learning_rate": 7.286063569682151e-06,
      "loss": 5.1883,
      "step": 150
    },
    {
      "epoch": 0.7804878048780488,
      "grad_norm": 3.371607780456543,
      "learning_rate": 7.775061124694378e-06,
      "loss": 5.16,
      "step": 160
    },
    {
      "epoch": 0.8292682926829268,
      "grad_norm": 7.1413116455078125,
      "learning_rate": 8.264058679706602e-06,
      "loss": 5.0679,
      "step": 170
    },
    {
      "epoch": 0.8780487804878049,
      "grad_norm": 7.872459411621094,
      "learning_rate": 8.753056234718827e-06,
      "loss": 5.0494,
      "step": 180
    },
    {
      "epoch": 0.926829268292683,
      "grad_norm": 4.166257858276367,
      "learning_rate": 9.242053789731052e-06,
      "loss": 4.9679,
      "step": 190
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 23.25122833251953,
      "learning_rate": 9.731051344743277e-06,
      "loss": 4.9343,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2759,
      "eval_gen_len": 211.6429,
      "eval_loss": 4.457103729248047,
      "eval_rouge1": 0.3508,
      "eval_rouge2": 0.0603,
      "eval_rougeL": 0.2235,
      "eval_runtime": 26.6238,
      "eval_samples_per_second": 6.836,
      "eval_steps_per_second": 1.728,
      "step": 205
    },
    {
      "epoch": 1.024390243902439,
      "grad_norm": 15.634258270263672,
      "learning_rate": 1.0220048899755502e-05,
      "loss": 4.9,
      "step": 210
    },
    {
      "epoch": 1.0731707317073171,
      "grad_norm": 3.670888900756836,
      "learning_rate": 1.0709046454767727e-05,
      "loss": 4.7805,
      "step": 220
    },
    {
      "epoch": 1.1219512195121952,
      "grad_norm": 4.5713210105896,
      "learning_rate": 1.1198044009779951e-05,
      "loss": 4.7084,
      "step": 230
    },
    {
      "epoch": 1.170731707317073,
      "grad_norm": 2.022193193435669,
      "learning_rate": 1.1687041564792176e-05,
      "loss": 4.6972,
      "step": 240
    },
    {
      "epoch": 1.2195121951219512,
      "grad_norm": 1.8646818399429321,
      "learning_rate": 1.2176039119804401e-05,
      "loss": 4.6286,
      "step": 250
    },
    {
      "epoch": 1.2682926829268293,
      "grad_norm": 3.063166618347168,
      "learning_rate": 1.2665036674816628e-05,
      "loss": 4.6438,
      "step": 260
    },
    {
      "epoch": 1.3170731707317074,
      "grad_norm": 4.173657417297363,
      "learning_rate": 1.3154034229828852e-05,
      "loss": 4.6001,
      "step": 270
    },
    {
      "epoch": 1.3658536585365852,
      "grad_norm": 7.44461727142334,
      "learning_rate": 1.3643031784841076e-05,
      "loss": 4.5358,
      "step": 280
    },
    {
      "epoch": 1.4146341463414633,
      "grad_norm": 1.7125922441482544,
      "learning_rate": 1.41320293398533e-05,
      "loss": 4.4905,
      "step": 290
    },
    {
      "epoch": 1.4634146341463414,
      "grad_norm": 9.653430938720703,
      "learning_rate": 1.4621026894865527e-05,
      "loss": 4.4497,
      "step": 300
    },
    {
      "epoch": 1.5121951219512195,
      "grad_norm": 2.094055652618408,
      "learning_rate": 1.5110024449877752e-05,
      "loss": 4.3886,
      "step": 310
    },
    {
      "epoch": 1.5609756097560976,
      "grad_norm": 1.2126773595809937,
      "learning_rate": 1.5599022004889977e-05,
      "loss": 4.3007,
      "step": 320
    },
    {
      "epoch": 1.6097560975609757,
      "grad_norm": 2.478563070297241,
      "learning_rate": 1.60880195599022e-05,
      "loss": 4.2896,
      "step": 330
    },
    {
      "epoch": 1.6585365853658538,
      "grad_norm": 3.717766523361206,
      "learning_rate": 1.6577017114914426e-05,
      "loss": 4.2184,
      "step": 340
    },
    {
      "epoch": 1.7073170731707317,
      "grad_norm": 1.308283805847168,
      "learning_rate": 1.706601466992665e-05,
      "loss": 4.2381,
      "step": 350
    },
    {
      "epoch": 1.7560975609756098,
      "grad_norm": 1.2101848125457764,
      "learning_rate": 1.7555012224938876e-05,
      "loss": 4.135,
      "step": 360
    },
    {
      "epoch": 1.8048780487804879,
      "grad_norm": 1.7718238830566406,
      "learning_rate": 1.8044009779951102e-05,
      "loss": 4.1122,
      "step": 370
    },
    {
      "epoch": 1.8536585365853657,
      "grad_norm": 1.8747498989105225,
      "learning_rate": 1.8533007334963325e-05,
      "loss": 4.0616,
      "step": 380
    },
    {
      "epoch": 1.9024390243902438,
      "grad_norm": 1.3740488290786743,
      "learning_rate": 1.9022004889975552e-05,
      "loss": 3.952,
      "step": 390
    },
    {
      "epoch": 1.951219512195122,
      "grad_norm": 4.758138179779053,
      "learning_rate": 1.9511002444987775e-05,
      "loss": 3.9793,
      "step": 400
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1578041315078735,
      "learning_rate": 2e-05,
      "loss": 3.9215,
      "step": 410
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3934,
      "eval_gen_len": 189.9505,
      "eval_loss": 3.538588523864746,
      "eval_rouge1": 0.4765,
      "eval_rouge2": 0.1303,
      "eval_rougeL": 0.3377,
      "eval_runtime": 26.1399,
      "eval_samples_per_second": 6.963,
      "eval_steps_per_second": 1.76,
      "step": 410
    },
    {
      "epoch": 2.048780487804878,
      "grad_norm": 2.019075393676758,
      "learning_rate": 2.0488997555012228e-05,
      "loss": 3.8844,
      "step": 420
    },
    {
      "epoch": 2.097560975609756,
      "grad_norm": 1.2126365900039673,
      "learning_rate": 2.097799511002445e-05,
      "loss": 3.8611,
      "step": 430
    },
    {
      "epoch": 2.1463414634146343,
      "grad_norm": 11.292882919311523,
      "learning_rate": 2.1466992665036674e-05,
      "loss": 3.7623,
      "step": 440
    },
    {
      "epoch": 2.1951219512195124,
      "grad_norm": 1.2274696826934814,
      "learning_rate": 2.19559902200489e-05,
      "loss": 3.7778,
      "step": 450
    },
    {
      "epoch": 2.2439024390243905,
      "grad_norm": 9.684100151062012,
      "learning_rate": 2.2444987775061127e-05,
      "loss": 3.7204,
      "step": 460
    },
    {
      "epoch": 2.292682926829268,
      "grad_norm": 1.1520313024520874,
      "learning_rate": 2.293398533007335e-05,
      "loss": 3.7227,
      "step": 470
    },
    {
      "epoch": 2.341463414634146,
      "grad_norm": 1.2632983922958374,
      "learning_rate": 2.3422982885085577e-05,
      "loss": 3.7362,
      "step": 480
    },
    {
      "epoch": 2.3902439024390243,
      "grad_norm": 0.9752365946769714,
      "learning_rate": 2.39119804400978e-05,
      "loss": 3.6642,
      "step": 490
    },
    {
      "epoch": 2.4390243902439024,
      "grad_norm": 1.1547846794128418,
      "learning_rate": 2.4400977995110023e-05,
      "loss": 3.5886,
      "step": 500
    },
    {
      "epoch": 2.4878048780487805,
      "grad_norm": 1.2422934770584106,
      "learning_rate": 2.488997555012225e-05,
      "loss": 3.6133,
      "step": 510
    },
    {
      "epoch": 2.5365853658536586,
      "grad_norm": 1.253853440284729,
      "learning_rate": 2.5378973105134473e-05,
      "loss": 3.5837,
      "step": 520
    },
    {
      "epoch": 2.5853658536585367,
      "grad_norm": 1.0211130380630493,
      "learning_rate": 2.58679706601467e-05,
      "loss": 3.6163,
      "step": 530
    },
    {
      "epoch": 2.6341463414634148,
      "grad_norm": 1.131115436553955,
      "learning_rate": 2.635696821515893e-05,
      "loss": 3.4982,
      "step": 540
    },
    {
      "epoch": 2.682926829268293,
      "grad_norm": 0.9331684708595276,
      "learning_rate": 2.684596577017115e-05,
      "loss": 3.5171,
      "step": 550
    },
    {
      "epoch": 2.7317073170731705,
      "grad_norm": 1.1529852151870728,
      "learning_rate": 2.7334963325183376e-05,
      "loss": 3.4674,
      "step": 560
    },
    {
      "epoch": 2.7804878048780486,
      "grad_norm": 1.1946120262145996,
      "learning_rate": 2.78239608801956e-05,
      "loss": 3.4866,
      "step": 570
    },
    {
      "epoch": 2.8292682926829267,
      "grad_norm": 1.292737364768982,
      "learning_rate": 2.8312958435207825e-05,
      "loss": 3.4434,
      "step": 580
    },
    {
      "epoch": 2.8780487804878048,
      "grad_norm": 1.0269521474838257,
      "learning_rate": 2.880195599022005e-05,
      "loss": 3.3643,
      "step": 590
    },
    {
      "epoch": 2.926829268292683,
      "grad_norm": 1.1100542545318604,
      "learning_rate": 2.9290953545232275e-05,
      "loss": 3.4197,
      "step": 600
    },
    {
      "epoch": 2.975609756097561,
      "grad_norm": 0.9210990071296692,
      "learning_rate": 2.97799511002445e-05,
      "loss": 3.3427,
      "step": 610
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4734,
      "eval_gen_len": 181.8681,
      "eval_loss": 3.0026919841766357,
      "eval_rouge1": 0.5525,
      "eval_rouge2": 0.1927,
      "eval_rougeL": 0.4214,
      "eval_runtime": 25.9002,
      "eval_samples_per_second": 7.027,
      "eval_steps_per_second": 1.776,
      "step": 615
    },
    {
      "epoch": 3.024390243902439,
      "grad_norm": 0.8961557745933533,
      "learning_rate": 3.0268948655256725e-05,
      "loss": 3.3786,
      "step": 620
    },
    {
      "epoch": 3.073170731707317,
      "grad_norm": 1.2599140405654907,
      "learning_rate": 3.075794621026895e-05,
      "loss": 3.3689,
      "step": 630
    },
    {
      "epoch": 3.1219512195121952,
      "grad_norm": 1.0294504165649414,
      "learning_rate": 3.1246943765281174e-05,
      "loss": 3.354,
      "step": 640
    },
    {
      "epoch": 3.1707317073170733,
      "grad_norm": 0.9114963412284851,
      "learning_rate": 3.17359413202934e-05,
      "loss": 3.2847,
      "step": 650
    },
    {
      "epoch": 3.2195121951219514,
      "grad_norm": 1.0025136470794678,
      "learning_rate": 3.222493887530563e-05,
      "loss": 3.2635,
      "step": 660
    },
    {
      "epoch": 3.2682926829268295,
      "grad_norm": 1.1256715059280396,
      "learning_rate": 3.271393643031785e-05,
      "loss": 3.2903,
      "step": 670
    },
    {
      "epoch": 3.317073170731707,
      "grad_norm": 1.0486799478530884,
      "learning_rate": 3.3202933985330074e-05,
      "loss": 3.2502,
      "step": 680
    },
    {
      "epoch": 3.3658536585365852,
      "grad_norm": 0.9213652014732361,
      "learning_rate": 3.36919315403423e-05,
      "loss": 3.2612,
      "step": 690
    },
    {
      "epoch": 3.4146341463414633,
      "grad_norm": 0.9584261178970337,
      "learning_rate": 3.418092909535453e-05,
      "loss": 3.1981,
      "step": 700
    },
    {
      "epoch": 3.4634146341463414,
      "grad_norm": 0.9650068283081055,
      "learning_rate": 3.466992665036675e-05,
      "loss": 3.2118,
      "step": 710
    },
    {
      "epoch": 3.5121951219512195,
      "grad_norm": 1.228827953338623,
      "learning_rate": 3.515892420537897e-05,
      "loss": 3.2021,
      "step": 720
    },
    {
      "epoch": 3.5609756097560976,
      "grad_norm": 1.1250883340835571,
      "learning_rate": 3.56479217603912e-05,
      "loss": 3.1994,
      "step": 730
    },
    {
      "epoch": 3.6097560975609757,
      "grad_norm": 0.938084602355957,
      "learning_rate": 3.613691931540342e-05,
      "loss": 3.1803,
      "step": 740
    },
    {
      "epoch": 3.658536585365854,
      "grad_norm": 0.9992406964302063,
      "learning_rate": 3.662591687041565e-05,
      "loss": 3.1305,
      "step": 750
    },
    {
      "epoch": 3.7073170731707314,
      "grad_norm": 15.40438175201416,
      "learning_rate": 3.711491442542788e-05,
      "loss": 3.1497,
      "step": 760
    },
    {
      "epoch": 3.7560975609756095,
      "grad_norm": 1.1189182996749878,
      "learning_rate": 3.76039119804401e-05,
      "loss": 3.1439,
      "step": 770
    },
    {
      "epoch": 3.8048780487804876,
      "grad_norm": 0.8453640341758728,
      "learning_rate": 3.8092909535452325e-05,
      "loss": 3.1512,
      "step": 780
    },
    {
      "epoch": 3.8536585365853657,
      "grad_norm": 0.8329293131828308,
      "learning_rate": 3.8581907090464545e-05,
      "loss": 3.0875,
      "step": 790
    },
    {
      "epoch": 3.902439024390244,
      "grad_norm": 1.1350609064102173,
      "learning_rate": 3.907090464547677e-05,
      "loss": 3.1501,
      "step": 800
    },
    {
      "epoch": 3.951219512195122,
      "grad_norm": 0.9956341981887817,
      "learning_rate": 3.9559902200489005e-05,
      "loss": 3.0673,
      "step": 810
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0671465396881104,
      "learning_rate": 4.0048899755501225e-05,
      "loss": 3.0923,
      "step": 820
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5152,
      "eval_gen_len": 177.4011,
      "eval_loss": 2.7666826248168945,
      "eval_rouge1": 0.582,
      "eval_rouge2": 0.233,
      "eval_rougeL": 0.4623,
      "eval_runtime": 26.0386,
      "eval_samples_per_second": 6.99,
      "eval_steps_per_second": 1.767,
      "step": 820
    },
    {
      "epoch": 4.048780487804878,
      "grad_norm": 0.8541052937507629,
      "learning_rate": 4.053789731051345e-05,
      "loss": 3.1169,
      "step": 830
    },
    {
      "epoch": 4.097560975609756,
      "grad_norm": 0.9787116050720215,
      "learning_rate": 4.102689486552567e-05,
      "loss": 3.0454,
      "step": 840
    },
    {
      "epoch": 4.146341463414634,
      "grad_norm": 0.9616861343383789,
      "learning_rate": 4.15158924205379e-05,
      "loss": 3.0472,
      "step": 850
    },
    {
      "epoch": 4.195121951219512,
      "grad_norm": 1.127516269683838,
      "learning_rate": 4.2004889975550124e-05,
      "loss": 3.0318,
      "step": 860
    },
    {
      "epoch": 4.2439024390243905,
      "grad_norm": 1.0883930921554565,
      "learning_rate": 4.249388753056235e-05,
      "loss": 3.0661,
      "step": 870
    },
    {
      "epoch": 4.2926829268292686,
      "grad_norm": 1.149770736694336,
      "learning_rate": 4.298288508557458e-05,
      "loss": 3.0137,
      "step": 880
    },
    {
      "epoch": 4.341463414634147,
      "grad_norm": 0.9378666281700134,
      "learning_rate": 4.34718826405868e-05,
      "loss": 3.0059,
      "step": 890
    },
    {
      "epoch": 4.390243902439025,
      "grad_norm": 0.9855746030807495,
      "learning_rate": 4.396088019559902e-05,
      "loss": 3.0009,
      "step": 900
    },
    {
      "epoch": 4.439024390243903,
      "grad_norm": 1.0716888904571533,
      "learning_rate": 4.444987775061125e-05,
      "loss": 3.0053,
      "step": 910
    },
    {
      "epoch": 4.487804878048781,
      "grad_norm": 0.9853790998458862,
      "learning_rate": 4.4938875305623476e-05,
      "loss": 2.9294,
      "step": 920
    },
    {
      "epoch": 4.536585365853659,
      "grad_norm": 0.9836393594741821,
      "learning_rate": 4.54278728606357e-05,
      "loss": 2.955,
      "step": 930
    },
    {
      "epoch": 4.585365853658536,
      "grad_norm": 1.4114398956298828,
      "learning_rate": 4.591687041564792e-05,
      "loss": 2.9944,
      "step": 940
    },
    {
      "epoch": 4.634146341463414,
      "grad_norm": 0.8007514476776123,
      "learning_rate": 4.640586797066015e-05,
      "loss": 2.9658,
      "step": 950
    },
    {
      "epoch": 4.682926829268292,
      "grad_norm": 0.98356693983078,
      "learning_rate": 4.689486552567237e-05,
      "loss": 2.9314,
      "step": 960
    },
    {
      "epoch": 4.7317073170731705,
      "grad_norm": 1.2858251333236694,
      "learning_rate": 4.7383863080684595e-05,
      "loss": 2.9415,
      "step": 970
    },
    {
      "epoch": 4.780487804878049,
      "grad_norm": 1.033654808998108,
      "learning_rate": 4.787286063569683e-05,
      "loss": 2.9866,
      "step": 980
    },
    {
      "epoch": 4.829268292682927,
      "grad_norm": 0.8810921907424927,
      "learning_rate": 4.836185819070905e-05,
      "loss": 2.9511,
      "step": 990
    },
    {
      "epoch": 4.878048780487805,
      "grad_norm": 1.1121643781661987,
      "learning_rate": 4.8850855745721275e-05,
      "loss": 2.9617,
      "step": 1000
    },
    {
      "epoch": 4.926829268292683,
      "grad_norm": 1.039544701576233,
      "learning_rate": 4.9339853300733495e-05,
      "loss": 2.9242,
      "step": 1010
    },
    {
      "epoch": 4.975609756097561,
      "grad_norm": 0.8879916667938232,
      "learning_rate": 4.982885085574572e-05,
      "loss": 2.8744,
      "step": 1020
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5424,
      "eval_gen_len": 179.2088,
      "eval_loss": 2.6143970489501953,
      "eval_rouge1": 0.6042,
      "eval_rouge2": 0.2613,
      "eval_rougeL": 0.4883,
      "eval_runtime": 25.9288,
      "eval_samples_per_second": 7.019,
      "eval_steps_per_second": 1.774,
      "step": 1025
    },
    {
      "epoch": 5.024390243902439,
      "grad_norm": 1.1612334251403809,
      "learning_rate": 5.031784841075795e-05,
      "loss": 2.9611,
      "step": 1030
    },
    {
      "epoch": 5.073170731707317,
      "grad_norm": 1.5247349739074707,
      "learning_rate": 5.080684596577018e-05,
      "loss": 2.9112,
      "step": 1040
    },
    {
      "epoch": 5.121951219512195,
      "grad_norm": 0.9210406541824341,
      "learning_rate": 5.1295843520782394e-05,
      "loss": 2.8527,
      "step": 1050
    },
    {
      "epoch": 5.170731707317073,
      "grad_norm": 1.1976507902145386,
      "learning_rate": 5.178484107579462e-05,
      "loss": 2.9057,
      "step": 1060
    },
    {
      "epoch": 5.219512195121951,
      "grad_norm": 1.2032570838928223,
      "learning_rate": 5.227383863080685e-05,
      "loss": 2.8742,
      "step": 1070
    },
    {
      "epoch": 5.2682926829268295,
      "grad_norm": 0.9623976945877075,
      "learning_rate": 5.2762836185819073e-05,
      "loss": 2.829,
      "step": 1080
    },
    {
      "epoch": 5.317073170731708,
      "grad_norm": 0.8499730229377747,
      "learning_rate": 5.32518337408313e-05,
      "loss": 2.8517,
      "step": 1090
    },
    {
      "epoch": 5.365853658536586,
      "grad_norm": 1.1849853992462158,
      "learning_rate": 5.374083129584352e-05,
      "loss": 2.8572,
      "step": 1100
    },
    {
      "epoch": 5.414634146341464,
      "grad_norm": 1.0028434991836548,
      "learning_rate": 5.4229828850855746e-05,
      "loss": 2.835,
      "step": 1110
    },
    {
      "epoch": 5.463414634146342,
      "grad_norm": 0.8181015253067017,
      "learning_rate": 5.471882640586797e-05,
      "loss": 2.8557,
      "step": 1120
    },
    {
      "epoch": 5.512195121951219,
      "grad_norm": 1.0572930574417114,
      "learning_rate": 5.52078239608802e-05,
      "loss": 2.8222,
      "step": 1130
    },
    {
      "epoch": 5.560975609756097,
      "grad_norm": 0.8895127177238464,
      "learning_rate": 5.5696821515892426e-05,
      "loss": 2.7869,
      "step": 1140
    },
    {
      "epoch": 5.609756097560975,
      "grad_norm": 1.0056649446487427,
      "learning_rate": 5.6185819070904646e-05,
      "loss": 2.8131,
      "step": 1150
    },
    {
      "epoch": 5.658536585365853,
      "grad_norm": 0.8315693736076355,
      "learning_rate": 5.667481662591687e-05,
      "loss": 2.8587,
      "step": 1160
    },
    {
      "epoch": 5.7073170731707314,
      "grad_norm": 0.8721175789833069,
      "learning_rate": 5.71638141809291e-05,
      "loss": 2.8064,
      "step": 1170
    },
    {
      "epoch": 5.7560975609756095,
      "grad_norm": 0.9493646025657654,
      "learning_rate": 5.7652811735941325e-05,
      "loss": 2.8169,
      "step": 1180
    },
    {
      "epoch": 5.804878048780488,
      "grad_norm": 0.9617358446121216,
      "learning_rate": 5.814180929095355e-05,
      "loss": 2.8168,
      "step": 1190
    },
    {
      "epoch": 5.853658536585366,
      "grad_norm": 0.9313523173332214,
      "learning_rate": 5.863080684596577e-05,
      "loss": 2.771,
      "step": 1200
    },
    {
      "epoch": 5.902439024390244,
      "grad_norm": 0.8505017161369324,
      "learning_rate": 5.9119804400978e-05,
      "loss": 2.7744,
      "step": 1210
    },
    {
      "epoch": 5.951219512195122,
      "grad_norm": 0.9459558129310608,
      "learning_rate": 5.9608801955990224e-05,
      "loss": 2.7718,
      "step": 1220
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.208365797996521,
      "learning_rate": 6.009779951100245e-05,
      "loss": 2.7677,
      "step": 1230
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5575,
      "eval_gen_len": 186.0879,
      "eval_loss": 2.5141470432281494,
      "eval_rouge1": 0.6141,
      "eval_rouge2": 0.2737,
      "eval_rougeL": 0.5028,
      "eval_runtime": 26.0185,
      "eval_samples_per_second": 6.995,
      "eval_steps_per_second": 1.768,
      "step": 1230
    },
    {
      "epoch": 6.048780487804878,
      "grad_norm": 1.1182864904403687,
      "learning_rate": 6.058679706601468e-05,
      "loss": 2.7941,
      "step": 1240
    },
    {
      "epoch": 6.097560975609756,
      "grad_norm": 0.906001627445221,
      "learning_rate": 6.107579462102689e-05,
      "loss": 2.7356,
      "step": 1250
    },
    {
      "epoch": 6.146341463414634,
      "grad_norm": 0.9255037903785706,
      "learning_rate": 6.156479217603912e-05,
      "loss": 2.7293,
      "step": 1260
    },
    {
      "epoch": 6.195121951219512,
      "grad_norm": 0.8442258834838867,
      "learning_rate": 6.205378973105134e-05,
      "loss": 2.7488,
      "step": 1270
    },
    {
      "epoch": 6.2439024390243905,
      "grad_norm": 0.9159016013145447,
      "learning_rate": 6.254278728606357e-05,
      "loss": 2.7394,
      "step": 1280
    },
    {
      "epoch": 6.2926829268292686,
      "grad_norm": 0.909702718257904,
      "learning_rate": 6.30317848410758e-05,
      "loss": 2.7518,
      "step": 1290
    },
    {
      "epoch": 6.341463414634147,
      "grad_norm": 0.8735110759735107,
      "learning_rate": 6.352078239608802e-05,
      "loss": 2.6913,
      "step": 1300
    },
    {
      "epoch": 6.390243902439025,
      "grad_norm": 1.0950109958648682,
      "learning_rate": 6.400977995110025e-05,
      "loss": 2.6841,
      "step": 1310
    },
    {
      "epoch": 6.439024390243903,
      "grad_norm": 0.8980112671852112,
      "learning_rate": 6.449877750611248e-05,
      "loss": 2.7115,
      "step": 1320
    },
    {
      "epoch": 6.487804878048781,
      "grad_norm": 1.0420292615890503,
      "learning_rate": 6.49877750611247e-05,
      "loss": 2.7501,
      "step": 1330
    },
    {
      "epoch": 6.536585365853659,
      "grad_norm": 0.9553576111793518,
      "learning_rate": 6.547677261613693e-05,
      "loss": 2.652,
      "step": 1340
    },
    {
      "epoch": 6.585365853658536,
      "grad_norm": 1.0302420854568481,
      "learning_rate": 6.596577017114914e-05,
      "loss": 2.6845,
      "step": 1350
    },
    {
      "epoch": 6.634146341463414,
      "grad_norm": 1.077214241027832,
      "learning_rate": 6.645476772616137e-05,
      "loss": 2.6935,
      "step": 1360
    },
    {
      "epoch": 6.682926829268292,
      "grad_norm": 0.9967917203903198,
      "learning_rate": 6.69437652811736e-05,
      "loss": 2.7126,
      "step": 1370
    },
    {
      "epoch": 6.7317073170731705,
      "grad_norm": 0.8878841996192932,
      "learning_rate": 6.743276283618582e-05,
      "loss": 2.6744,
      "step": 1380
    },
    {
      "epoch": 6.780487804878049,
      "grad_norm": 0.9185668230056763,
      "learning_rate": 6.792176039119805e-05,
      "loss": 2.6995,
      "step": 1390
    },
    {
      "epoch": 6.829268292682927,
      "grad_norm": 0.9542802572250366,
      "learning_rate": 6.841075794621027e-05,
      "loss": 2.7088,
      "step": 1400
    },
    {
      "epoch": 6.878048780487805,
      "grad_norm": 0.9949384331703186,
      "learning_rate": 6.88997555012225e-05,
      "loss": 2.6895,
      "step": 1410
    },
    {
      "epoch": 6.926829268292683,
      "grad_norm": 0.9762834310531616,
      "learning_rate": 6.938875305623473e-05,
      "loss": 2.6164,
      "step": 1420
    },
    {
      "epoch": 6.975609756097561,
      "grad_norm": 0.9873429536819458,
      "learning_rate": 6.987775061124695e-05,
      "loss": 2.7168,
      "step": 1430
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.57,
      "eval_gen_len": 190.6484,
      "eval_loss": 2.4311938285827637,
      "eval_rouge1": 0.6205,
      "eval_rouge2": 0.2863,
      "eval_rougeL": 0.5136,
      "eval_runtime": 26.1579,
      "eval_samples_per_second": 6.958,
      "eval_steps_per_second": 1.759,
      "step": 1435
    },
    {
      "epoch": 7.024390243902439,
      "grad_norm": 0.8859543204307556,
      "learning_rate": 7.036674816625917e-05,
      "loss": 2.6648,
      "step": 1440
    },
    {
      "epoch": 7.073170731707317,
      "grad_norm": 0.9650115966796875,
      "learning_rate": 7.08557457212714e-05,
      "loss": 2.6306,
      "step": 1450
    },
    {
      "epoch": 7.121951219512195,
      "grad_norm": 0.9108964800834656,
      "learning_rate": 7.134474327628362e-05,
      "loss": 2.6047,
      "step": 1460
    },
    {
      "epoch": 7.170731707317073,
      "grad_norm": 0.8837963342666626,
      "learning_rate": 7.183374083129585e-05,
      "loss": 2.6432,
      "step": 1470
    },
    {
      "epoch": 7.219512195121951,
      "grad_norm": 1.0622878074645996,
      "learning_rate": 7.232273838630807e-05,
      "loss": 2.6387,
      "step": 1480
    },
    {
      "epoch": 7.2682926829268295,
      "grad_norm": 0.9728686213493347,
      "learning_rate": 7.281173594132029e-05,
      "loss": 2.6206,
      "step": 1490
    },
    {
      "epoch": 7.317073170731708,
      "grad_norm": 0.7860050797462463,
      "learning_rate": 7.330073349633251e-05,
      "loss": 2.6196,
      "step": 1500
    },
    {
      "epoch": 7.365853658536586,
      "grad_norm": 0.9671594500541687,
      "learning_rate": 7.378973105134474e-05,
      "loss": 2.6267,
      "step": 1510
    },
    {
      "epoch": 7.414634146341464,
      "grad_norm": 1.1058562994003296,
      "learning_rate": 7.427872860635698e-05,
      "loss": 2.6606,
      "step": 1520
    },
    {
      "epoch": 7.463414634146342,
      "grad_norm": 0.9301013946533203,
      "learning_rate": 7.47677261613692e-05,
      "loss": 2.6526,
      "step": 1530
    },
    {
      "epoch": 7.512195121951219,
      "grad_norm": 1.050396203994751,
      "learning_rate": 7.525672371638142e-05,
      "loss": 2.572,
      "step": 1540
    },
    {
      "epoch": 7.560975609756097,
      "grad_norm": 0.9724476933479309,
      "learning_rate": 7.574572127139365e-05,
      "loss": 2.566,
      "step": 1550
    },
    {
      "epoch": 7.609756097560975,
      "grad_norm": 0.9213927388191223,
      "learning_rate": 7.623471882640587e-05,
      "loss": 2.6139,
      "step": 1560
    },
    {
      "epoch": 7.658536585365853,
      "grad_norm": 1.0309463739395142,
      "learning_rate": 7.67237163814181e-05,
      "loss": 2.6379,
      "step": 1570
    },
    {
      "epoch": 7.7073170731707314,
      "grad_norm": 0.8563573956489563,
      "learning_rate": 7.721271393643032e-05,
      "loss": 2.6191,
      "step": 1580
    },
    {
      "epoch": 7.7560975609756095,
      "grad_norm": 0.9945729970932007,
      "learning_rate": 7.770171149144254e-05,
      "loss": 2.5668,
      "step": 1590
    },
    {
      "epoch": 7.804878048780488,
      "grad_norm": 0.7494526505470276,
      "learning_rate": 7.819070904645476e-05,
      "loss": 2.5047,
      "step": 1600
    },
    {
      "epoch": 7.853658536585366,
      "grad_norm": 0.9642314910888672,
      "learning_rate": 7.867970660146699e-05,
      "loss": 2.592,
      "step": 1610
    },
    {
      "epoch": 7.902439024390244,
      "grad_norm": 0.8688525557518005,
      "learning_rate": 7.916870415647922e-05,
      "loss": 2.5551,
      "step": 1620
    },
    {
      "epoch": 7.951219512195122,
      "grad_norm": 0.8314557075500488,
      "learning_rate": 7.965770171149144e-05,
      "loss": 2.5851,
      "step": 1630
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.000834584236145,
      "learning_rate": 8.014669926650367e-05,
      "loss": 2.5717,
      "step": 1640
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5839,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.3604416847229004,
      "eval_rouge1": 0.5641,
      "eval_rouge2": 0.2668,
      "eval_rougeL": 0.4705,
      "eval_runtime": 26.4515,
      "eval_samples_per_second": 6.881,
      "eval_steps_per_second": 1.739,
      "step": 1640
    },
    {
      "epoch": 8.048780487804878,
      "grad_norm": 1.0702135562896729,
      "learning_rate": 8.06356968215159e-05,
      "loss": 2.5839,
      "step": 1650
    },
    {
      "epoch": 8.097560975609756,
      "grad_norm": 1.0597014427185059,
      "learning_rate": 8.112469437652812e-05,
      "loss": 2.5569,
      "step": 1660
    },
    {
      "epoch": 8.146341463414634,
      "grad_norm": 1.0889874696731567,
      "learning_rate": 8.161369193154035e-05,
      "loss": 2.551,
      "step": 1670
    },
    {
      "epoch": 8.195121951219512,
      "grad_norm": 0.9655137658119202,
      "learning_rate": 8.210268948655258e-05,
      "loss": 2.5546,
      "step": 1680
    },
    {
      "epoch": 8.24390243902439,
      "grad_norm": 0.9861440062522888,
      "learning_rate": 8.259168704156479e-05,
      "loss": 2.4937,
      "step": 1690
    },
    {
      "epoch": 8.292682926829269,
      "grad_norm": 0.9669071435928345,
      "learning_rate": 8.308068459657702e-05,
      "loss": 2.5275,
      "step": 1700
    },
    {
      "epoch": 8.341463414634147,
      "grad_norm": 0.9090121984481812,
      "learning_rate": 8.356968215158924e-05,
      "loss": 2.4987,
      "step": 1710
    },
    {
      "epoch": 8.390243902439025,
      "grad_norm": 1.0994421243667603,
      "learning_rate": 8.405867970660147e-05,
      "loss": 2.5473,
      "step": 1720
    },
    {
      "epoch": 8.439024390243903,
      "grad_norm": 0.8295562267303467,
      "learning_rate": 8.45476772616137e-05,
      "loss": 2.5134,
      "step": 1730
    },
    {
      "epoch": 8.487804878048781,
      "grad_norm": 0.7855349183082581,
      "learning_rate": 8.503667481662592e-05,
      "loss": 2.4448,
      "step": 1740
    },
    {
      "epoch": 8.536585365853659,
      "grad_norm": 1.008330225944519,
      "learning_rate": 8.552567237163815e-05,
      "loss": 2.5522,
      "step": 1750
    },
    {
      "epoch": 8.585365853658537,
      "grad_norm": 1.0320262908935547,
      "learning_rate": 8.601466992665038e-05,
      "loss": 2.5141,
      "step": 1760
    },
    {
      "epoch": 8.634146341463415,
      "grad_norm": 1.0223716497421265,
      "learning_rate": 8.65036674816626e-05,
      "loss": 2.5079,
      "step": 1770
    },
    {
      "epoch": 8.682926829268293,
      "grad_norm": 0.882462203502655,
      "learning_rate": 8.699266503667483e-05,
      "loss": 2.5387,
      "step": 1780
    },
    {
      "epoch": 8.731707317073171,
      "grad_norm": 1.1398268938064575,
      "learning_rate": 8.748166259168704e-05,
      "loss": 2.4915,
      "step": 1790
    },
    {
      "epoch": 8.78048780487805,
      "grad_norm": 1.1037157773971558,
      "learning_rate": 8.797066014669927e-05,
      "loss": 2.5326,
      "step": 1800
    },
    {
      "epoch": 8.829268292682928,
      "grad_norm": 0.7518386840820312,
      "learning_rate": 8.84596577017115e-05,
      "loss": 2.4366,
      "step": 1810
    },
    {
      "epoch": 8.878048780487806,
      "grad_norm": 0.9014624953269958,
      "learning_rate": 8.894865525672372e-05,
      "loss": 2.4906,
      "step": 1820
    },
    {
      "epoch": 8.926829268292684,
      "grad_norm": 1.0596896409988403,
      "learning_rate": 8.943765281173595e-05,
      "loss": 2.541,
      "step": 1830
    },
    {
      "epoch": 8.975609756097562,
      "grad_norm": 0.99559485912323,
      "learning_rate": 8.992665036674816e-05,
      "loss": 2.4663,
      "step": 1840
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5932,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.3033788204193115,
      "eval_rouge1": 0.5701,
      "eval_rouge2": 0.2773,
      "eval_rougeL": 0.4791,
      "eval_runtime": 26.6033,
      "eval_samples_per_second": 6.841,
      "eval_steps_per_second": 1.729,
      "step": 1845
    },
    {
      "epoch": 9.024390243902438,
      "grad_norm": 1.1031367778778076,
      "learning_rate": 9.04156479217604e-05,
      "loss": 2.4692,
      "step": 1850
    },
    {
      "epoch": 9.073170731707316,
      "grad_norm": 1.0542200803756714,
      "learning_rate": 9.090464547677263e-05,
      "loss": 2.4629,
      "step": 1860
    },
    {
      "epoch": 9.121951219512194,
      "grad_norm": 0.8666772246360779,
      "learning_rate": 9.139364303178485e-05,
      "loss": 2.4193,
      "step": 1870
    },
    {
      "epoch": 9.170731707317072,
      "grad_norm": 1.041645884513855,
      "learning_rate": 9.188264058679708e-05,
      "loss": 2.4512,
      "step": 1880
    },
    {
      "epoch": 9.21951219512195,
      "grad_norm": 0.8610559701919556,
      "learning_rate": 9.237163814180929e-05,
      "loss": 2.441,
      "step": 1890
    },
    {
      "epoch": 9.268292682926829,
      "grad_norm": 0.9617129564285278,
      "learning_rate": 9.286063569682152e-05,
      "loss": 2.4903,
      "step": 1900
    },
    {
      "epoch": 9.317073170731707,
      "grad_norm": 0.9784387946128845,
      "learning_rate": 9.334963325183375e-05,
      "loss": 2.4764,
      "step": 1910
    },
    {
      "epoch": 9.365853658536585,
      "grad_norm": 0.8843352794647217,
      "learning_rate": 9.383863080684597e-05,
      "loss": 2.4541,
      "step": 1920
    },
    {
      "epoch": 9.414634146341463,
      "grad_norm": 1.1032359600067139,
      "learning_rate": 9.43276283618582e-05,
      "loss": 2.4355,
      "step": 1930
    },
    {
      "epoch": 9.463414634146341,
      "grad_norm": 1.2064437866210938,
      "learning_rate": 9.481662591687041e-05,
      "loss": 2.4135,
      "step": 1940
    },
    {
      "epoch": 9.512195121951219,
      "grad_norm": 0.810530424118042,
      "learning_rate": 9.530562347188264e-05,
      "loss": 2.3846,
      "step": 1950
    },
    {
      "epoch": 9.560975609756097,
      "grad_norm": 0.7688414454460144,
      "learning_rate": 9.579462102689486e-05,
      "loss": 2.4259,
      "step": 1960
    },
    {
      "epoch": 9.609756097560975,
      "grad_norm": 0.9564357399940491,
      "learning_rate": 9.628361858190709e-05,
      "loss": 2.4354,
      "step": 1970
    },
    {
      "epoch": 9.658536585365853,
      "grad_norm": 0.8629823923110962,
      "learning_rate": 9.677261613691933e-05,
      "loss": 2.4389,
      "step": 1980
    },
    {
      "epoch": 9.707317073170731,
      "grad_norm": 0.8377267122268677,
      "learning_rate": 9.726161369193154e-05,
      "loss": 2.4162,
      "step": 1990
    },
    {
      "epoch": 9.75609756097561,
      "grad_norm": 0.9582370519638062,
      "learning_rate": 9.775061124694377e-05,
      "loss": 2.4044,
      "step": 2000
    },
    {
      "epoch": 9.804878048780488,
      "grad_norm": 0.901448667049408,
      "learning_rate": 9.8239608801956e-05,
      "loss": 2.4627,
      "step": 2010
    },
    {
      "epoch": 9.853658536585366,
      "grad_norm": 1.0475609302520752,
      "learning_rate": 9.872860635696822e-05,
      "loss": 2.4079,
      "step": 2020
    },
    {
      "epoch": 9.902439024390244,
      "grad_norm": 1.0225861072540283,
      "learning_rate": 9.921760391198045e-05,
      "loss": 2.4705,
      "step": 2030
    },
    {
      "epoch": 9.951219512195122,
      "grad_norm": 0.8137247562408447,
      "learning_rate": 9.970660146699266e-05,
      "loss": 2.4109,
      "step": 2040
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.1472225189208984,
      "learning_rate": 9.997826677533279e-05,
      "loss": 2.4236,
      "step": 2050
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6014,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.2605769634246826,
      "eval_rouge1": 0.5749,
      "eval_rouge2": 0.285,
      "eval_rougeL": 0.4872,
      "eval_runtime": 26.4591,
      "eval_samples_per_second": 6.879,
      "eval_steps_per_second": 1.739,
      "step": 2050
    },
    {
      "epoch": 10.048780487804878,
      "grad_norm": 1.0116795301437378,
      "learning_rate": 9.992393371366477e-05,
      "loss": 2.3413,
      "step": 2060
    },
    {
      "epoch": 10.097560975609756,
      "grad_norm": 0.7885662913322449,
      "learning_rate": 9.986960065199674e-05,
      "loss": 2.3437,
      "step": 2070
    },
    {
      "epoch": 10.146341463414634,
      "grad_norm": 0.9145445227622986,
      "learning_rate": 9.981526759032873e-05,
      "loss": 2.3793,
      "step": 2080
    },
    {
      "epoch": 10.195121951219512,
      "grad_norm": 0.8369097113609314,
      "learning_rate": 9.97609345286607e-05,
      "loss": 2.3435,
      "step": 2090
    },
    {
      "epoch": 10.24390243902439,
      "grad_norm": 0.9109391570091248,
      "learning_rate": 9.970660146699266e-05,
      "loss": 2.4217,
      "step": 2100
    },
    {
      "epoch": 10.292682926829269,
      "grad_norm": 0.8721039891242981,
      "learning_rate": 9.965226840532465e-05,
      "loss": 2.3891,
      "step": 2110
    },
    {
      "epoch": 10.341463414634147,
      "grad_norm": 1.0082284212112427,
      "learning_rate": 9.959793534365663e-05,
      "loss": 2.3427,
      "step": 2120
    },
    {
      "epoch": 10.390243902439025,
      "grad_norm": 0.9386345148086548,
      "learning_rate": 9.95436022819886e-05,
      "loss": 2.3686,
      "step": 2130
    },
    {
      "epoch": 10.439024390243903,
      "grad_norm": 0.9664391279220581,
      "learning_rate": 9.948926922032057e-05,
      "loss": 2.4001,
      "step": 2140
    },
    {
      "epoch": 10.487804878048781,
      "grad_norm": 0.9189599752426147,
      "learning_rate": 9.943493615865255e-05,
      "loss": 2.3468,
      "step": 2150
    },
    {
      "epoch": 10.536585365853659,
      "grad_norm": 0.8793349266052246,
      "learning_rate": 9.938060309698452e-05,
      "loss": 2.33,
      "step": 2160
    },
    {
      "epoch": 10.585365853658537,
      "grad_norm": 0.7458771467208862,
      "learning_rate": 9.93262700353165e-05,
      "loss": 2.3779,
      "step": 2170
    },
    {
      "epoch": 10.634146341463415,
      "grad_norm": 1.412410855293274,
      "learning_rate": 9.927193697364847e-05,
      "loss": 2.3591,
      "step": 2180
    },
    {
      "epoch": 10.682926829268293,
      "grad_norm": 0.97608882188797,
      "learning_rate": 9.921760391198045e-05,
      "loss": 2.4116,
      "step": 2190
    },
    {
      "epoch": 10.731707317073171,
      "grad_norm": 0.8573082089424133,
      "learning_rate": 9.916327085031242e-05,
      "loss": 2.3711,
      "step": 2200
    },
    {
      "epoch": 10.78048780487805,
      "grad_norm": 0.9795628786087036,
      "learning_rate": 9.91089377886444e-05,
      "loss": 2.3187,
      "step": 2210
    },
    {
      "epoch": 10.829268292682928,
      "grad_norm": 0.9566768407821655,
      "learning_rate": 9.905460472697637e-05,
      "loss": 2.3801,
      "step": 2220
    },
    {
      "epoch": 10.878048780487806,
      "grad_norm": 0.931747317314148,
      "learning_rate": 9.900027166530834e-05,
      "loss": 2.4118,
      "step": 2230
    },
    {
      "epoch": 10.926829268292684,
      "grad_norm": 1.0201259851455688,
      "learning_rate": 9.894593860364032e-05,
      "loss": 2.395,
      "step": 2240
    },
    {
      "epoch": 10.975609756097562,
      "grad_norm": 0.9366726875305176,
      "learning_rate": 9.88916055419723e-05,
      "loss": 2.3715,
      "step": 2250
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6101,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.2207939624786377,
      "eval_rouge1": 0.5796,
      "eval_rouge2": 0.2939,
      "eval_rougeL": 0.4951,
      "eval_runtime": 26.5531,
      "eval_samples_per_second": 6.854,
      "eval_steps_per_second": 1.732,
      "step": 2255
    },
    {
      "epoch": 11.024390243902438,
      "grad_norm": 0.8451365828514099,
      "learning_rate": 9.883727248030426e-05,
      "loss": 2.3,
      "step": 2260
    },
    {
      "epoch": 11.073170731707316,
      "grad_norm": 0.9864517450332642,
      "learning_rate": 9.878293941863624e-05,
      "loss": 2.3132,
      "step": 2270
    },
    {
      "epoch": 11.121951219512194,
      "grad_norm": 0.9425224661827087,
      "learning_rate": 9.872860635696822e-05,
      "loss": 2.3193,
      "step": 2280
    },
    {
      "epoch": 11.170731707317072,
      "grad_norm": 0.8904969692230225,
      "learning_rate": 9.867427329530019e-05,
      "loss": 2.2679,
      "step": 2290
    },
    {
      "epoch": 11.21951219512195,
      "grad_norm": 0.8168789744377136,
      "learning_rate": 9.861994023363218e-05,
      "loss": 2.3595,
      "step": 2300
    },
    {
      "epoch": 11.268292682926829,
      "grad_norm": 0.8249573707580566,
      "learning_rate": 9.856560717196414e-05,
      "loss": 2.2372,
      "step": 2310
    },
    {
      "epoch": 11.317073170731707,
      "grad_norm": 1.2351031303405762,
      "learning_rate": 9.851127411029611e-05,
      "loss": 2.3489,
      "step": 2320
    },
    {
      "epoch": 11.365853658536585,
      "grad_norm": 0.9290939569473267,
      "learning_rate": 9.84569410486281e-05,
      "loss": 2.2802,
      "step": 2330
    },
    {
      "epoch": 11.414634146341463,
      "grad_norm": 1.1329292058944702,
      "learning_rate": 9.840260798696008e-05,
      "loss": 2.3173,
      "step": 2340
    },
    {
      "epoch": 11.463414634146341,
      "grad_norm": 1.0866820812225342,
      "learning_rate": 9.834827492529205e-05,
      "loss": 2.2682,
      "step": 2350
    },
    {
      "epoch": 11.512195121951219,
      "grad_norm": 0.8205491304397583,
      "learning_rate": 9.829394186362402e-05,
      "loss": 2.3116,
      "step": 2360
    },
    {
      "epoch": 11.560975609756097,
      "grad_norm": 0.8546379804611206,
      "learning_rate": 9.8239608801956e-05,
      "loss": 2.3475,
      "step": 2370
    },
    {
      "epoch": 11.609756097560975,
      "grad_norm": 0.8734157085418701,
      "learning_rate": 9.818527574028798e-05,
      "loss": 2.3206,
      "step": 2380
    },
    {
      "epoch": 11.658536585365853,
      "grad_norm": 0.799018919467926,
      "learning_rate": 9.813094267861994e-05,
      "loss": 2.3096,
      "step": 2390
    },
    {
      "epoch": 11.707317073170731,
      "grad_norm": 0.8234296441078186,
      "learning_rate": 9.807660961695192e-05,
      "loss": 2.32,
      "step": 2400
    },
    {
      "epoch": 11.75609756097561,
      "grad_norm": 0.908701479434967,
      "learning_rate": 9.80222765552839e-05,
      "loss": 2.3284,
      "step": 2410
    },
    {
      "epoch": 11.804878048780488,
      "grad_norm": 0.8176116347312927,
      "learning_rate": 9.796794349361587e-05,
      "loss": 2.3131,
      "step": 2420
    },
    {
      "epoch": 11.853658536585366,
      "grad_norm": 1.1658778190612793,
      "learning_rate": 9.791361043194785e-05,
      "loss": 2.3174,
      "step": 2430
    },
    {
      "epoch": 11.902439024390244,
      "grad_norm": 0.8987638354301453,
      "learning_rate": 9.785927737027982e-05,
      "loss": 2.2697,
      "step": 2440
    },
    {
      "epoch": 11.951219512195122,
      "grad_norm": 1.0862481594085693,
      "learning_rate": 9.780494430861179e-05,
      "loss": 2.3026,
      "step": 2450
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.0809197425842285,
      "learning_rate": 9.775061124694377e-05,
      "loss": 2.2779,
      "step": 2460
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6144,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.1923000812530518,
      "eval_rouge1": 0.584,
      "eval_rouge2": 0.298,
      "eval_rougeL": 0.4986,
      "eval_runtime": 26.6759,
      "eval_samples_per_second": 6.823,
      "eval_steps_per_second": 1.724,
      "step": 2460
    },
    {
      "epoch": 12.048780487804878,
      "grad_norm": 0.9038712978363037,
      "learning_rate": 9.769627818527575e-05,
      "loss": 2.279,
      "step": 2470
    },
    {
      "epoch": 12.097560975609756,
      "grad_norm": 0.9469591975212097,
      "learning_rate": 9.764194512360772e-05,
      "loss": 2.2548,
      "step": 2480
    },
    {
      "epoch": 12.146341463414634,
      "grad_norm": 1.0490235090255737,
      "learning_rate": 9.758761206193969e-05,
      "loss": 2.2762,
      "step": 2490
    },
    {
      "epoch": 12.195121951219512,
      "grad_norm": 0.9397657513618469,
      "learning_rate": 9.753327900027167e-05,
      "loss": 2.2562,
      "step": 2500
    },
    {
      "epoch": 12.24390243902439,
      "grad_norm": 0.903236985206604,
      "learning_rate": 9.747894593860366e-05,
      "loss": 2.2559,
      "step": 2510
    },
    {
      "epoch": 12.292682926829269,
      "grad_norm": 0.8761768937110901,
      "learning_rate": 9.742461287693561e-05,
      "loss": 2.2569,
      "step": 2520
    },
    {
      "epoch": 12.341463414634147,
      "grad_norm": 0.7565495371818542,
      "learning_rate": 9.73702798152676e-05,
      "loss": 2.2202,
      "step": 2530
    },
    {
      "epoch": 12.390243902439025,
      "grad_norm": 0.9713935852050781,
      "learning_rate": 9.731594675359958e-05,
      "loss": 2.2509,
      "step": 2540
    },
    {
      "epoch": 12.439024390243903,
      "grad_norm": 0.8867313861846924,
      "learning_rate": 9.726161369193154e-05,
      "loss": 2.2889,
      "step": 2550
    },
    {
      "epoch": 12.487804878048781,
      "grad_norm": 1.0842514038085938,
      "learning_rate": 9.720728063026351e-05,
      "loss": 2.2381,
      "step": 2560
    },
    {
      "epoch": 12.536585365853659,
      "grad_norm": 0.8155404329299927,
      "learning_rate": 9.71529475685955e-05,
      "loss": 2.2468,
      "step": 2570
    },
    {
      "epoch": 12.585365853658537,
      "grad_norm": 0.7613242864608765,
      "learning_rate": 9.709861450692746e-05,
      "loss": 2.2128,
      "step": 2580
    },
    {
      "epoch": 12.634146341463415,
      "grad_norm": 0.9132261276245117,
      "learning_rate": 9.704428144525945e-05,
      "loss": 2.2659,
      "step": 2590
    },
    {
      "epoch": 12.682926829268293,
      "grad_norm": 1.1883471012115479,
      "learning_rate": 9.698994838359143e-05,
      "loss": 2.2473,
      "step": 2600
    },
    {
      "epoch": 12.731707317073171,
      "grad_norm": 0.7840389013290405,
      "learning_rate": 9.693561532192338e-05,
      "loss": 2.2435,
      "step": 2610
    },
    {
      "epoch": 12.78048780487805,
      "grad_norm": 0.9382815957069397,
      "learning_rate": 9.688128226025537e-05,
      "loss": 2.2278,
      "step": 2620
    },
    {
      "epoch": 12.829268292682928,
      "grad_norm": 0.8595882654190063,
      "learning_rate": 9.682694919858735e-05,
      "loss": 2.27,
      "step": 2630
    },
    {
      "epoch": 12.878048780487806,
      "grad_norm": 0.9498746395111084,
      "learning_rate": 9.677261613691933e-05,
      "loss": 2.2706,
      "step": 2640
    },
    {
      "epoch": 12.926829268292684,
      "grad_norm": 1.005918025970459,
      "learning_rate": 9.671828307525129e-05,
      "loss": 2.2543,
      "step": 2650
    },
    {
      "epoch": 12.975609756097562,
      "grad_norm": 0.8443641662597656,
      "learning_rate": 9.666395001358327e-05,
      "loss": 2.2018,
      "step": 2660
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6211,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.1619982719421387,
      "eval_rouge1": 0.5881,
      "eval_rouge2": 0.3053,
      "eval_rougeL": 0.5057,
      "eval_runtime": 26.4837,
      "eval_samples_per_second": 6.872,
      "eval_steps_per_second": 1.737,
      "step": 2665
    },
    {
      "epoch": 13.024390243902438,
      "grad_norm": 0.9114805459976196,
      "learning_rate": 9.660961695191525e-05,
      "loss": 2.243,
      "step": 2670
    },
    {
      "epoch": 13.073170731707316,
      "grad_norm": 1.0895026922225952,
      "learning_rate": 9.655528389024722e-05,
      "loss": 2.2126,
      "step": 2680
    },
    {
      "epoch": 13.121951219512194,
      "grad_norm": 0.9829836487770081,
      "learning_rate": 9.650095082857919e-05,
      "loss": 2.1898,
      "step": 2690
    },
    {
      "epoch": 13.170731707317072,
      "grad_norm": 0.9027822017669678,
      "learning_rate": 9.644661776691117e-05,
      "loss": 2.2044,
      "step": 2700
    },
    {
      "epoch": 13.21951219512195,
      "grad_norm": 0.8191620111465454,
      "learning_rate": 9.639228470524314e-05,
      "loss": 2.1748,
      "step": 2710
    },
    {
      "epoch": 13.268292682926829,
      "grad_norm": 0.8903685808181763,
      "learning_rate": 9.633795164357512e-05,
      "loss": 2.2767,
      "step": 2720
    },
    {
      "epoch": 13.317073170731707,
      "grad_norm": 0.9027280211448669,
      "learning_rate": 9.628361858190709e-05,
      "loss": 2.1899,
      "step": 2730
    },
    {
      "epoch": 13.365853658536585,
      "grad_norm": 0.8687236309051514,
      "learning_rate": 9.622928552023906e-05,
      "loss": 2.1784,
      "step": 2740
    },
    {
      "epoch": 13.414634146341463,
      "grad_norm": 0.828973650932312,
      "learning_rate": 9.617495245857104e-05,
      "loss": 2.2028,
      "step": 2750
    },
    {
      "epoch": 13.463414634146341,
      "grad_norm": 0.8230794072151184,
      "learning_rate": 9.612061939690302e-05,
      "loss": 2.1924,
      "step": 2760
    },
    {
      "epoch": 13.512195121951219,
      "grad_norm": 0.9190805554389954,
      "learning_rate": 9.6066286335235e-05,
      "loss": 2.2352,
      "step": 2770
    },
    {
      "epoch": 13.560975609756097,
      "grad_norm": 0.7604323029518127,
      "learning_rate": 9.601195327356696e-05,
      "loss": 2.1649,
      "step": 2780
    },
    {
      "epoch": 13.609756097560975,
      "grad_norm": 1.0770535469055176,
      "learning_rate": 9.595762021189894e-05,
      "loss": 2.2197,
      "step": 2790
    },
    {
      "epoch": 13.658536585365853,
      "grad_norm": 0.94279944896698,
      "learning_rate": 9.590328715023093e-05,
      "loss": 2.2216,
      "step": 2800
    },
    {
      "epoch": 13.707317073170731,
      "grad_norm": 0.9158275723457336,
      "learning_rate": 9.58489540885629e-05,
      "loss": 2.215,
      "step": 2810
    },
    {
      "epoch": 13.75609756097561,
      "grad_norm": 0.8379354476928711,
      "learning_rate": 9.579462102689486e-05,
      "loss": 2.1912,
      "step": 2820
    },
    {
      "epoch": 13.804878048780488,
      "grad_norm": 1.0317034721374512,
      "learning_rate": 9.574028796522685e-05,
      "loss": 2.2501,
      "step": 2830
    },
    {
      "epoch": 13.853658536585366,
      "grad_norm": 0.8205541372299194,
      "learning_rate": 9.568595490355882e-05,
      "loss": 2.1945,
      "step": 2840
    },
    {
      "epoch": 13.902439024390244,
      "grad_norm": 0.8240370750427246,
      "learning_rate": 9.56316218418908e-05,
      "loss": 2.1658,
      "step": 2850
    },
    {
      "epoch": 13.951219512195122,
      "grad_norm": 0.8294575214385986,
      "learning_rate": 9.557728878022277e-05,
      "loss": 2.2301,
      "step": 2860
    },
    {
      "epoch": 14.0,
      "grad_norm": 1.1440404653549194,
      "learning_rate": 9.552295571855474e-05,
      "loss": 2.1942,
      "step": 2870
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.625,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.1468594074249268,
      "eval_rouge1": 0.5886,
      "eval_rouge2": 0.3073,
      "eval_rougeL": 0.5089,
      "eval_runtime": 26.4113,
      "eval_samples_per_second": 6.891,
      "eval_steps_per_second": 1.742,
      "step": 2870
    },
    {
      "epoch": 14.048780487804878,
      "grad_norm": 0.8380372524261475,
      "learning_rate": 9.546862265688672e-05,
      "loss": 2.169,
      "step": 2880
    },
    {
      "epoch": 14.097560975609756,
      "grad_norm": 0.8634399771690369,
      "learning_rate": 9.54142895952187e-05,
      "loss": 2.1461,
      "step": 2890
    },
    {
      "epoch": 14.146341463414634,
      "grad_norm": 0.8449454307556152,
      "learning_rate": 9.535995653355067e-05,
      "loss": 2.1429,
      "step": 2900
    },
    {
      "epoch": 14.195121951219512,
      "grad_norm": 1.0804766416549683,
      "learning_rate": 9.530562347188264e-05,
      "loss": 2.1806,
      "step": 2910
    },
    {
      "epoch": 14.24390243902439,
      "grad_norm": 0.8354528546333313,
      "learning_rate": 9.525129041021462e-05,
      "loss": 2.1485,
      "step": 2920
    },
    {
      "epoch": 14.292682926829269,
      "grad_norm": 0.905103325843811,
      "learning_rate": 9.51969573485466e-05,
      "loss": 2.1903,
      "step": 2930
    },
    {
      "epoch": 14.341463414634147,
      "grad_norm": 0.8975040316581726,
      "learning_rate": 9.514262428687857e-05,
      "loss": 2.1938,
      "step": 2940
    },
    {
      "epoch": 14.390243902439025,
      "grad_norm": 0.8443695306777954,
      "learning_rate": 9.508829122521054e-05,
      "loss": 2.1604,
      "step": 2950
    },
    {
      "epoch": 14.439024390243903,
      "grad_norm": 0.8442232608795166,
      "learning_rate": 9.503395816354252e-05,
      "loss": 2.127,
      "step": 2960
    },
    {
      "epoch": 14.487804878048781,
      "grad_norm": 0.7876951098442078,
      "learning_rate": 9.497962510187449e-05,
      "loss": 2.1824,
      "step": 2970
    },
    {
      "epoch": 14.536585365853659,
      "grad_norm": 1.0184301137924194,
      "learning_rate": 9.492529204020647e-05,
      "loss": 2.2021,
      "step": 2980
    },
    {
      "epoch": 14.585365853658537,
      "grad_norm": 0.9558939933776855,
      "learning_rate": 9.487095897853844e-05,
      "loss": 2.1671,
      "step": 2990
    },
    {
      "epoch": 14.634146341463415,
      "grad_norm": 0.8477892875671387,
      "learning_rate": 9.481662591687041e-05,
      "loss": 2.1123,
      "step": 3000
    },
    {
      "epoch": 14.682926829268293,
      "grad_norm": 0.9942978024482727,
      "learning_rate": 9.47622928552024e-05,
      "loss": 2.1388,
      "step": 3010
    },
    {
      "epoch": 14.731707317073171,
      "grad_norm": 0.9383358955383301,
      "learning_rate": 9.470795979353438e-05,
      "loss": 2.1496,
      "step": 3020
    },
    {
      "epoch": 14.78048780487805,
      "grad_norm": 1.062231421470642,
      "learning_rate": 9.465362673186635e-05,
      "loss": 2.1736,
      "step": 3030
    },
    {
      "epoch": 14.829268292682928,
      "grad_norm": 0.9161679148674011,
      "learning_rate": 9.459929367019831e-05,
      "loss": 2.1688,
      "step": 3040
    },
    {
      "epoch": 14.878048780487806,
      "grad_norm": 0.9697257876396179,
      "learning_rate": 9.45449606085303e-05,
      "loss": 2.154,
      "step": 3050
    },
    {
      "epoch": 14.926829268292684,
      "grad_norm": 1.0235304832458496,
      "learning_rate": 9.449062754686227e-05,
      "loss": 2.1676,
      "step": 3060
    },
    {
      "epoch": 14.975609756097562,
      "grad_norm": 0.8928753137588501,
      "learning_rate": 9.443629448519425e-05,
      "loss": 2.1642,
      "step": 3070
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6293,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.1218607425689697,
      "eval_rouge1": 0.5932,
      "eval_rouge2": 0.3125,
      "eval_rougeL": 0.5127,
      "eval_runtime": 26.5151,
      "eval_samples_per_second": 6.864,
      "eval_steps_per_second": 1.735,
      "step": 3075
    },
    {
      "epoch": 15.024390243902438,
      "grad_norm": 0.8534289598464966,
      "learning_rate": 9.438196142352622e-05,
      "loss": 2.1589,
      "step": 3080
    },
    {
      "epoch": 15.073170731707316,
      "grad_norm": 0.846277117729187,
      "learning_rate": 9.43276283618582e-05,
      "loss": 2.0645,
      "step": 3090
    },
    {
      "epoch": 15.121951219512194,
      "grad_norm": 1.1414495706558228,
      "learning_rate": 9.427329530019017e-05,
      "loss": 2.0968,
      "step": 3100
    },
    {
      "epoch": 15.170731707317072,
      "grad_norm": 0.8166375160217285,
      "learning_rate": 9.421896223852215e-05,
      "loss": 2.1028,
      "step": 3110
    },
    {
      "epoch": 15.21951219512195,
      "grad_norm": 0.7828956246376038,
      "learning_rate": 9.416462917685412e-05,
      "loss": 2.1178,
      "step": 3120
    },
    {
      "epoch": 15.268292682926829,
      "grad_norm": 0.8929072022438049,
      "learning_rate": 9.411029611518609e-05,
      "loss": 2.1018,
      "step": 3130
    },
    {
      "epoch": 15.317073170731707,
      "grad_norm": 0.7981455326080322,
      "learning_rate": 9.405596305351807e-05,
      "loss": 2.127,
      "step": 3140
    },
    {
      "epoch": 15.365853658536585,
      "grad_norm": 0.8577933311462402,
      "learning_rate": 9.400162999185005e-05,
      "loss": 2.1227,
      "step": 3150
    },
    {
      "epoch": 15.414634146341463,
      "grad_norm": 0.8780116438865662,
      "learning_rate": 9.394729693018202e-05,
      "loss": 2.1043,
      "step": 3160
    },
    {
      "epoch": 15.463414634146341,
      "grad_norm": 0.8575468063354492,
      "learning_rate": 9.389296386851399e-05,
      "loss": 2.0918,
      "step": 3170
    },
    {
      "epoch": 15.512195121951219,
      "grad_norm": 1.4090367555618286,
      "learning_rate": 9.383863080684597e-05,
      "loss": 2.1102,
      "step": 3180
    },
    {
      "epoch": 15.560975609756097,
      "grad_norm": 1.0460201501846313,
      "learning_rate": 9.378429774517794e-05,
      "loss": 2.1424,
      "step": 3190
    },
    {
      "epoch": 15.609756097560975,
      "grad_norm": 0.9865913391113281,
      "learning_rate": 9.372996468350992e-05,
      "loss": 2.0499,
      "step": 3200
    },
    {
      "epoch": 15.658536585365853,
      "grad_norm": 1.0010690689086914,
      "learning_rate": 9.367563162184189e-05,
      "loss": 2.176,
      "step": 3210
    },
    {
      "epoch": 15.707317073170731,
      "grad_norm": 0.8663599491119385,
      "learning_rate": 9.362129856017387e-05,
      "loss": 2.1513,
      "step": 3220
    },
    {
      "epoch": 15.75609756097561,
      "grad_norm": 0.8186299204826355,
      "learning_rate": 9.356696549850584e-05,
      "loss": 2.1712,
      "step": 3230
    },
    {
      "epoch": 15.804878048780488,
      "grad_norm": 0.8568254709243774,
      "learning_rate": 9.351263243683783e-05,
      "loss": 2.1232,
      "step": 3240
    },
    {
      "epoch": 15.853658536585366,
      "grad_norm": 1.000522255897522,
      "learning_rate": 9.34582993751698e-05,
      "loss": 2.1435,
      "step": 3250
    },
    {
      "epoch": 15.902439024390244,
      "grad_norm": 0.8970864415168762,
      "learning_rate": 9.340396631350176e-05,
      "loss": 2.1366,
      "step": 3260
    },
    {
      "epoch": 15.951219512195122,
      "grad_norm": 0.9909244179725647,
      "learning_rate": 9.334963325183375e-05,
      "loss": 2.1536,
      "step": 3270
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.0475345849990845,
      "learning_rate": 9.329530019016573e-05,
      "loss": 2.0952,
      "step": 3280
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6317,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.1110589504241943,
      "eval_rouge1": 0.5938,
      "eval_rouge2": 0.3169,
      "eval_rougeL": 0.516,
      "eval_runtime": 26.4003,
      "eval_samples_per_second": 6.894,
      "eval_steps_per_second": 1.742,
      "step": 3280
    },
    {
      "epoch": 16.048780487804876,
      "grad_norm": 0.8568356037139893,
      "learning_rate": 9.32409671284977e-05,
      "loss": 2.0625,
      "step": 3290
    },
    {
      "epoch": 16.097560975609756,
      "grad_norm": 0.7888637781143188,
      "learning_rate": 9.318663406682967e-05,
      "loss": 2.0771,
      "step": 3300
    },
    {
      "epoch": 16.146341463414632,
      "grad_norm": 0.8220765590667725,
      "learning_rate": 9.313230100516165e-05,
      "loss": 2.1645,
      "step": 3310
    },
    {
      "epoch": 16.195121951219512,
      "grad_norm": 0.8520556688308716,
      "learning_rate": 9.307796794349362e-05,
      "loss": 2.0747,
      "step": 3320
    },
    {
      "epoch": 16.24390243902439,
      "grad_norm": 0.910944938659668,
      "learning_rate": 9.30236348818256e-05,
      "loss": 2.0778,
      "step": 3330
    },
    {
      "epoch": 16.29268292682927,
      "grad_norm": 0.9317068457603455,
      "learning_rate": 9.296930182015757e-05,
      "loss": 2.1142,
      "step": 3340
    },
    {
      "epoch": 16.341463414634145,
      "grad_norm": 0.8472641706466675,
      "learning_rate": 9.291496875848954e-05,
      "loss": 2.103,
      "step": 3350
    },
    {
      "epoch": 16.390243902439025,
      "grad_norm": 0.9557019472122192,
      "learning_rate": 9.286063569682152e-05,
      "loss": 2.0788,
      "step": 3360
    },
    {
      "epoch": 16.4390243902439,
      "grad_norm": 0.8406639695167542,
      "learning_rate": 9.28063026351535e-05,
      "loss": 2.1061,
      "step": 3370
    },
    {
      "epoch": 16.48780487804878,
      "grad_norm": 1.0637025833129883,
      "learning_rate": 9.275196957348547e-05,
      "loss": 2.039,
      "step": 3380
    },
    {
      "epoch": 16.536585365853657,
      "grad_norm": 0.9092559218406677,
      "learning_rate": 9.269763651181744e-05,
      "loss": 2.0324,
      "step": 3390
    },
    {
      "epoch": 16.585365853658537,
      "grad_norm": 1.031977653503418,
      "learning_rate": 9.264330345014942e-05,
      "loss": 2.1002,
      "step": 3400
    },
    {
      "epoch": 16.634146341463413,
      "grad_norm": 1.0254329442977905,
      "learning_rate": 9.25889703884814e-05,
      "loss": 2.0954,
      "step": 3410
    },
    {
      "epoch": 16.682926829268293,
      "grad_norm": 0.9532445073127747,
      "learning_rate": 9.253463732681337e-05,
      "loss": 2.0529,
      "step": 3420
    },
    {
      "epoch": 16.73170731707317,
      "grad_norm": 1.1140599250793457,
      "learning_rate": 9.248030426514534e-05,
      "loss": 2.0982,
      "step": 3430
    },
    {
      "epoch": 16.78048780487805,
      "grad_norm": 0.8425986766815186,
      "learning_rate": 9.242597120347732e-05,
      "loss": 2.09,
      "step": 3440
    },
    {
      "epoch": 16.829268292682926,
      "grad_norm": 0.9490983486175537,
      "learning_rate": 9.237163814180929e-05,
      "loss": 2.0732,
      "step": 3450
    },
    {
      "epoch": 16.878048780487806,
      "grad_norm": 0.9101831912994385,
      "learning_rate": 9.231730508014128e-05,
      "loss": 2.1161,
      "step": 3460
    },
    {
      "epoch": 16.926829268292682,
      "grad_norm": 0.961280345916748,
      "learning_rate": 9.226297201847324e-05,
      "loss": 2.1085,
      "step": 3470
    },
    {
      "epoch": 16.975609756097562,
      "grad_norm": 1.1503115892410278,
      "learning_rate": 9.220863895680521e-05,
      "loss": 2.0838,
      "step": 3480
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6334,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.098520040512085,
      "eval_rouge1": 0.5944,
      "eval_rouge2": 0.3188,
      "eval_rougeL": 0.5175,
      "eval_runtime": 26.5163,
      "eval_samples_per_second": 6.864,
      "eval_steps_per_second": 1.735,
      "step": 3485
    },
    {
      "epoch": 17.024390243902438,
      "grad_norm": 0.9509657621383667,
      "learning_rate": 9.21543058951372e-05,
      "loss": 2.0309,
      "step": 3490
    },
    {
      "epoch": 17.073170731707318,
      "grad_norm": 0.9084831476211548,
      "learning_rate": 9.209997283346918e-05,
      "loss": 2.0358,
      "step": 3500
    },
    {
      "epoch": 17.121951219512194,
      "grad_norm": 0.8318648338317871,
      "learning_rate": 9.204563977180115e-05,
      "loss": 2.026,
      "step": 3510
    },
    {
      "epoch": 17.170731707317074,
      "grad_norm": 0.9886439442634583,
      "learning_rate": 9.199130671013312e-05,
      "loss": 2.0421,
      "step": 3520
    },
    {
      "epoch": 17.21951219512195,
      "grad_norm": 0.998199462890625,
      "learning_rate": 9.19369736484651e-05,
      "loss": 2.0552,
      "step": 3530
    },
    {
      "epoch": 17.26829268292683,
      "grad_norm": 0.9013775587081909,
      "learning_rate": 9.188264058679708e-05,
      "loss": 2.0296,
      "step": 3540
    },
    {
      "epoch": 17.317073170731707,
      "grad_norm": 0.8572256565093994,
      "learning_rate": 9.182830752512905e-05,
      "loss": 2.0409,
      "step": 3550
    },
    {
      "epoch": 17.365853658536587,
      "grad_norm": 0.9061347246170044,
      "learning_rate": 9.177397446346102e-05,
      "loss": 2.0467,
      "step": 3560
    },
    {
      "epoch": 17.414634146341463,
      "grad_norm": 0.7994931936264038,
      "learning_rate": 9.1719641401793e-05,
      "loss": 2.0645,
      "step": 3570
    },
    {
      "epoch": 17.463414634146343,
      "grad_norm": 0.8628284931182861,
      "learning_rate": 9.166530834012497e-05,
      "loss": 2.0127,
      "step": 3580
    },
    {
      "epoch": 17.51219512195122,
      "grad_norm": 1.0142345428466797,
      "learning_rate": 9.161097527845695e-05,
      "loss": 2.059,
      "step": 3590
    },
    {
      "epoch": 17.5609756097561,
      "grad_norm": 0.9873341917991638,
      "learning_rate": 9.155664221678892e-05,
      "loss": 2.079,
      "step": 3600
    },
    {
      "epoch": 17.609756097560975,
      "grad_norm": 0.9785902500152588,
      "learning_rate": 9.150230915512089e-05,
      "loss": 2.0932,
      "step": 3610
    },
    {
      "epoch": 17.658536585365855,
      "grad_norm": 0.8768711686134338,
      "learning_rate": 9.144797609345287e-05,
      "loss": 2.0169,
      "step": 3620
    },
    {
      "epoch": 17.70731707317073,
      "grad_norm": 0.9447925686836243,
      "learning_rate": 9.139364303178485e-05,
      "loss": 2.0419,
      "step": 3630
    },
    {
      "epoch": 17.75609756097561,
      "grad_norm": 0.8872796297073364,
      "learning_rate": 9.133930997011682e-05,
      "loss": 2.0768,
      "step": 3640
    },
    {
      "epoch": 17.804878048780488,
      "grad_norm": 0.9131309986114502,
      "learning_rate": 9.128497690844879e-05,
      "loss": 2.0619,
      "step": 3650
    },
    {
      "epoch": 17.853658536585368,
      "grad_norm": 0.9417833685874939,
      "learning_rate": 9.123064384678077e-05,
      "loss": 2.0971,
      "step": 3660
    },
    {
      "epoch": 17.902439024390244,
      "grad_norm": 0.9056317806243896,
      "learning_rate": 9.117631078511276e-05,
      "loss": 2.0509,
      "step": 3670
    },
    {
      "epoch": 17.951219512195124,
      "grad_norm": 1.052677869796753,
      "learning_rate": 9.112197772344472e-05,
      "loss": 2.059,
      "step": 3680
    },
    {
      "epoch": 18.0,
      "grad_norm": 1.1331645250320435,
      "learning_rate": 9.10676446617767e-05,
      "loss": 2.0219,
      "step": 3690
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.637,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.086111307144165,
      "eval_rouge1": 0.5969,
      "eval_rouge2": 0.3226,
      "eval_rougeL": 0.521,
      "eval_runtime": 26.4745,
      "eval_samples_per_second": 6.875,
      "eval_steps_per_second": 1.738,
      "step": 3690
    },
    {
      "epoch": 18.048780487804876,
      "grad_norm": 0.8600811958312988,
      "learning_rate": 9.101331160010868e-05,
      "loss": 2.0092,
      "step": 3700
    },
    {
      "epoch": 18.097560975609756,
      "grad_norm": 0.8948452472686768,
      "learning_rate": 9.095897853844064e-05,
      "loss": 2.0124,
      "step": 3710
    },
    {
      "epoch": 18.146341463414632,
      "grad_norm": 0.9096921682357788,
      "learning_rate": 9.090464547677263e-05,
      "loss": 2.016,
      "step": 3720
    },
    {
      "epoch": 18.195121951219512,
      "grad_norm": 0.948293924331665,
      "learning_rate": 9.08503124151046e-05,
      "loss": 1.9863,
      "step": 3730
    },
    {
      "epoch": 18.24390243902439,
      "grad_norm": 0.9195007085800171,
      "learning_rate": 9.079597935343656e-05,
      "loss": 2.0405,
      "step": 3740
    },
    {
      "epoch": 18.29268292682927,
      "grad_norm": 0.9686821699142456,
      "learning_rate": 9.074164629176855e-05,
      "loss": 2.0113,
      "step": 3750
    },
    {
      "epoch": 18.341463414634145,
      "grad_norm": 0.8149988055229187,
      "learning_rate": 9.068731323010053e-05,
      "loss": 1.994,
      "step": 3760
    },
    {
      "epoch": 18.390243902439025,
      "grad_norm": 1.002677083015442,
      "learning_rate": 9.063298016843248e-05,
      "loss": 2.0301,
      "step": 3770
    },
    {
      "epoch": 18.4390243902439,
      "grad_norm": 0.7668618559837341,
      "learning_rate": 9.057864710676447e-05,
      "loss": 1.9885,
      "step": 3780
    },
    {
      "epoch": 18.48780487804878,
      "grad_norm": 1.0023763179779053,
      "learning_rate": 9.052431404509645e-05,
      "loss": 2.0015,
      "step": 3790
    },
    {
      "epoch": 18.536585365853657,
      "grad_norm": 0.8811227679252625,
      "learning_rate": 9.046998098342842e-05,
      "loss": 2.0479,
      "step": 3800
    },
    {
      "epoch": 18.585365853658537,
      "grad_norm": 0.8438715934753418,
      "learning_rate": 9.04156479217604e-05,
      "loss": 2.0093,
      "step": 3810
    },
    {
      "epoch": 18.634146341463413,
      "grad_norm": 0.9621524810791016,
      "learning_rate": 9.036131486009237e-05,
      "loss": 2.0786,
      "step": 3820
    },
    {
      "epoch": 18.682926829268293,
      "grad_norm": 1.0231949090957642,
      "learning_rate": 9.030698179842435e-05,
      "loss": 2.0356,
      "step": 3830
    },
    {
      "epoch": 18.73170731707317,
      "grad_norm": 1.0325591564178467,
      "learning_rate": 9.025264873675632e-05,
      "loss": 2.0383,
      "step": 3840
    },
    {
      "epoch": 18.78048780487805,
      "grad_norm": 1.2348233461380005,
      "learning_rate": 9.01983156750883e-05,
      "loss": 2.0033,
      "step": 3850
    },
    {
      "epoch": 18.829268292682926,
      "grad_norm": 0.9654771685600281,
      "learning_rate": 9.014398261342027e-05,
      "loss": 1.9976,
      "step": 3860
    },
    {
      "epoch": 18.878048780487806,
      "grad_norm": 0.9214688539505005,
      "learning_rate": 9.008964955175224e-05,
      "loss": 2.0259,
      "step": 3870
    },
    {
      "epoch": 18.926829268292682,
      "grad_norm": 0.8998551368713379,
      "learning_rate": 9.003531649008422e-05,
      "loss": 2.015,
      "step": 3880
    },
    {
      "epoch": 18.975609756097562,
      "grad_norm": 0.9731389880180359,
      "learning_rate": 8.99809834284162e-05,
      "loss": 2.0727,
      "step": 3890
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6386,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0800883769989014,
      "eval_rouge1": 0.5985,
      "eval_rouge2": 0.3256,
      "eval_rougeL": 0.5236,
      "eval_runtime": 26.467,
      "eval_samples_per_second": 6.876,
      "eval_steps_per_second": 1.738,
      "step": 3895
    },
    {
      "epoch": 19.024390243902438,
      "grad_norm": 0.9533644318580627,
      "learning_rate": 8.992665036674816e-05,
      "loss": 1.9601,
      "step": 3900
    },
    {
      "epoch": 19.073170731707318,
      "grad_norm": 0.8680891990661621,
      "learning_rate": 8.987231730508014e-05,
      "loss": 1.9617,
      "step": 3910
    },
    {
      "epoch": 19.121951219512194,
      "grad_norm": 0.9162204265594482,
      "learning_rate": 8.981798424341212e-05,
      "loss": 1.9745,
      "step": 3920
    },
    {
      "epoch": 19.170731707317074,
      "grad_norm": 0.9257919788360596,
      "learning_rate": 8.97636511817441e-05,
      "loss": 1.9737,
      "step": 3930
    },
    {
      "epoch": 19.21951219512195,
      "grad_norm": 0.8662911653518677,
      "learning_rate": 8.970931812007606e-05,
      "loss": 1.9627,
      "step": 3940
    },
    {
      "epoch": 19.26829268292683,
      "grad_norm": 0.7983854413032532,
      "learning_rate": 8.965498505840804e-05,
      "loss": 1.9995,
      "step": 3950
    },
    {
      "epoch": 19.317073170731707,
      "grad_norm": 0.9230179190635681,
      "learning_rate": 8.960065199674003e-05,
      "loss": 1.9865,
      "step": 3960
    },
    {
      "epoch": 19.365853658536587,
      "grad_norm": 0.8826920390129089,
      "learning_rate": 8.9546318935072e-05,
      "loss": 1.9957,
      "step": 3970
    },
    {
      "epoch": 19.414634146341463,
      "grad_norm": 0.915490984916687,
      "learning_rate": 8.949198587340398e-05,
      "loss": 1.995,
      "step": 3980
    },
    {
      "epoch": 19.463414634146343,
      "grad_norm": 0.9427376985549927,
      "learning_rate": 8.943765281173595e-05,
      "loss": 1.947,
      "step": 3990
    },
    {
      "epoch": 19.51219512195122,
      "grad_norm": 0.9200916290283203,
      "learning_rate": 8.938331975006792e-05,
      "loss": 1.9974,
      "step": 4000
    },
    {
      "epoch": 19.5609756097561,
      "grad_norm": 0.9528909921646118,
      "learning_rate": 8.93289866883999e-05,
      "loss": 1.9867,
      "step": 4010
    },
    {
      "epoch": 19.609756097560975,
      "grad_norm": 1.02826726436615,
      "learning_rate": 8.927465362673188e-05,
      "loss": 1.9863,
      "step": 4020
    },
    {
      "epoch": 19.658536585365855,
      "grad_norm": 0.8667694926261902,
      "learning_rate": 8.922032056506384e-05,
      "loss": 2.0085,
      "step": 4030
    },
    {
      "epoch": 19.70731707317073,
      "grad_norm": 0.9008286595344543,
      "learning_rate": 8.916598750339582e-05,
      "loss": 1.983,
      "step": 4040
    },
    {
      "epoch": 19.75609756097561,
      "grad_norm": 0.8765814900398254,
      "learning_rate": 8.91116544417278e-05,
      "loss": 1.9919,
      "step": 4050
    },
    {
      "epoch": 19.804878048780488,
      "grad_norm": 0.9815711379051208,
      "learning_rate": 8.905732138005977e-05,
      "loss": 1.9724,
      "step": 4060
    },
    {
      "epoch": 19.853658536585368,
      "grad_norm": 0.8371131420135498,
      "learning_rate": 8.900298831839174e-05,
      "loss": 1.9957,
      "step": 4070
    },
    {
      "epoch": 19.902439024390244,
      "grad_norm": 0.9084226489067078,
      "learning_rate": 8.894865525672372e-05,
      "loss": 2.0002,
      "step": 4080
    },
    {
      "epoch": 19.951219512195124,
      "grad_norm": 1.1448745727539062,
      "learning_rate": 8.889432219505569e-05,
      "loss": 2.0231,
      "step": 4090
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.0898938179016113,
      "learning_rate": 8.883998913338767e-05,
      "loss": 2.0165,
      "step": 4100
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.642,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0671799182891846,
      "eval_rouge1": 0.6005,
      "eval_rouge2": 0.3291,
      "eval_rougeL": 0.5255,
      "eval_runtime": 26.6176,
      "eval_samples_per_second": 6.838,
      "eval_steps_per_second": 1.728,
      "step": 4100
    },
    {
      "epoch": 20.048780487804876,
      "grad_norm": 0.8856220841407776,
      "learning_rate": 8.878565607171964e-05,
      "loss": 1.948,
      "step": 4110
    },
    {
      "epoch": 20.097560975609756,
      "grad_norm": 0.8878317475318909,
      "learning_rate": 8.873132301005162e-05,
      "loss": 1.9741,
      "step": 4120
    },
    {
      "epoch": 20.146341463414632,
      "grad_norm": 0.8628621697425842,
      "learning_rate": 8.867698994838359e-05,
      "loss": 1.9349,
      "step": 4130
    },
    {
      "epoch": 20.195121951219512,
      "grad_norm": 0.8303452134132385,
      "learning_rate": 8.862265688671557e-05,
      "loss": 1.9748,
      "step": 4140
    },
    {
      "epoch": 20.24390243902439,
      "grad_norm": 0.9785374999046326,
      "learning_rate": 8.856832382504756e-05,
      "loss": 1.9411,
      "step": 4150
    },
    {
      "epoch": 20.29268292682927,
      "grad_norm": 1.0943126678466797,
      "learning_rate": 8.851399076337951e-05,
      "loss": 1.9892,
      "step": 4160
    },
    {
      "epoch": 20.341463414634145,
      "grad_norm": 0.8537071347236633,
      "learning_rate": 8.84596577017115e-05,
      "loss": 1.9492,
      "step": 4170
    },
    {
      "epoch": 20.390243902439025,
      "grad_norm": 0.9113747477531433,
      "learning_rate": 8.840532464004348e-05,
      "loss": 2.0174,
      "step": 4180
    },
    {
      "epoch": 20.4390243902439,
      "grad_norm": 0.9659337997436523,
      "learning_rate": 8.835099157837545e-05,
      "loss": 1.9281,
      "step": 4190
    },
    {
      "epoch": 20.48780487804878,
      "grad_norm": 1.0948446989059448,
      "learning_rate": 8.829665851670741e-05,
      "loss": 1.9777,
      "step": 4200
    },
    {
      "epoch": 20.536585365853657,
      "grad_norm": 1.1053972244262695,
      "learning_rate": 8.82423254550394e-05,
      "loss": 1.9583,
      "step": 4210
    },
    {
      "epoch": 20.585365853658537,
      "grad_norm": 0.9370738863945007,
      "learning_rate": 8.818799239337137e-05,
      "loss": 1.9746,
      "step": 4220
    },
    {
      "epoch": 20.634146341463413,
      "grad_norm": 0.8967805504798889,
      "learning_rate": 8.813365933170335e-05,
      "loss": 1.9868,
      "step": 4230
    },
    {
      "epoch": 20.682926829268293,
      "grad_norm": 0.8011646270751953,
      "learning_rate": 8.807932627003532e-05,
      "loss": 1.9357,
      "step": 4240
    },
    {
      "epoch": 20.73170731707317,
      "grad_norm": 0.8785592317581177,
      "learning_rate": 8.80249932083673e-05,
      "loss": 1.9788,
      "step": 4250
    },
    {
      "epoch": 20.78048780487805,
      "grad_norm": 0.8387641310691833,
      "learning_rate": 8.797066014669927e-05,
      "loss": 1.9253,
      "step": 4260
    },
    {
      "epoch": 20.829268292682926,
      "grad_norm": 0.8664023280143738,
      "learning_rate": 8.791632708503125e-05,
      "loss": 1.9868,
      "step": 4270
    },
    {
      "epoch": 20.878048780487806,
      "grad_norm": 0.8770254254341125,
      "learning_rate": 8.786199402336322e-05,
      "loss": 1.9192,
      "step": 4280
    },
    {
      "epoch": 20.926829268292682,
      "grad_norm": 0.8913759589195251,
      "learning_rate": 8.780766096169519e-05,
      "loss": 1.9655,
      "step": 4290
    },
    {
      "epoch": 20.975609756097562,
      "grad_norm": 0.7817836999893188,
      "learning_rate": 8.775332790002717e-05,
      "loss": 1.9613,
      "step": 4300
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.6431,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.057805061340332,
      "eval_rouge1": 0.6017,
      "eval_rouge2": 0.3311,
      "eval_rougeL": 0.5274,
      "eval_runtime": 26.5642,
      "eval_samples_per_second": 6.851,
      "eval_steps_per_second": 1.732,
      "step": 4305
    },
    {
      "epoch": 21.024390243902438,
      "grad_norm": 1.0928252935409546,
      "learning_rate": 8.769899483835915e-05,
      "loss": 1.9218,
      "step": 4310
    },
    {
      "epoch": 21.073170731707318,
      "grad_norm": 1.0891989469528198,
      "learning_rate": 8.764466177669112e-05,
      "loss": 1.9555,
      "step": 4320
    },
    {
      "epoch": 21.121951219512194,
      "grad_norm": 0.8458701372146606,
      "learning_rate": 8.759032871502309e-05,
      "loss": 1.9781,
      "step": 4330
    },
    {
      "epoch": 21.170731707317074,
      "grad_norm": 0.9393008351325989,
      "learning_rate": 8.753599565335507e-05,
      "loss": 1.9089,
      "step": 4340
    },
    {
      "epoch": 21.21951219512195,
      "grad_norm": 0.9594295620918274,
      "learning_rate": 8.748166259168704e-05,
      "loss": 1.9764,
      "step": 4350
    },
    {
      "epoch": 21.26829268292683,
      "grad_norm": 0.9915680289268494,
      "learning_rate": 8.742732953001902e-05,
      "loss": 1.9328,
      "step": 4360
    },
    {
      "epoch": 21.317073170731707,
      "grad_norm": 1.0338279008865356,
      "learning_rate": 8.737299646835099e-05,
      "loss": 1.9124,
      "step": 4370
    },
    {
      "epoch": 21.365853658536587,
      "grad_norm": 0.8570368885993958,
      "learning_rate": 8.731866340668296e-05,
      "loss": 1.9258,
      "step": 4380
    },
    {
      "epoch": 21.414634146341463,
      "grad_norm": 0.9843132495880127,
      "learning_rate": 8.726433034501494e-05,
      "loss": 1.9368,
      "step": 4390
    },
    {
      "epoch": 21.463414634146343,
      "grad_norm": 0.7749292850494385,
      "learning_rate": 8.720999728334693e-05,
      "loss": 1.8731,
      "step": 4400
    },
    {
      "epoch": 21.51219512195122,
      "grad_norm": 0.9112648367881775,
      "learning_rate": 8.71556642216789e-05,
      "loss": 1.8874,
      "step": 4410
    },
    {
      "epoch": 21.5609756097561,
      "grad_norm": 0.9614556431770325,
      "learning_rate": 8.710133116001086e-05,
      "loss": 1.9299,
      "step": 4420
    },
    {
      "epoch": 21.609756097560975,
      "grad_norm": 0.8870623111724854,
      "learning_rate": 8.704699809834285e-05,
      "loss": 1.9401,
      "step": 4430
    },
    {
      "epoch": 21.658536585365855,
      "grad_norm": 0.9557124376296997,
      "learning_rate": 8.699266503667483e-05,
      "loss": 1.9403,
      "step": 4440
    },
    {
      "epoch": 21.70731707317073,
      "grad_norm": 0.8714508414268494,
      "learning_rate": 8.69383319750068e-05,
      "loss": 1.9282,
      "step": 4450
    },
    {
      "epoch": 21.75609756097561,
      "grad_norm": 1.0693939924240112,
      "learning_rate": 8.688399891333877e-05,
      "loss": 1.9715,
      "step": 4460
    },
    {
      "epoch": 21.804878048780488,
      "grad_norm": 0.8669313192367554,
      "learning_rate": 8.682966585167075e-05,
      "loss": 1.9108,
      "step": 4470
    },
    {
      "epoch": 21.853658536585368,
      "grad_norm": 1.013626217842102,
      "learning_rate": 8.677533279000272e-05,
      "loss": 1.92,
      "step": 4480
    },
    {
      "epoch": 21.902439024390244,
      "grad_norm": 1.2305961847305298,
      "learning_rate": 8.67209997283347e-05,
      "loss": 1.9894,
      "step": 4490
    },
    {
      "epoch": 21.951219512195124,
      "grad_norm": 0.8011859655380249,
      "learning_rate": 8.666666666666667e-05,
      "loss": 1.9103,
      "step": 4500
    },
    {
      "epoch": 22.0,
      "grad_norm": 1.023545503616333,
      "learning_rate": 8.661233360499864e-05,
      "loss": 1.9716,
      "step": 4510
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6435,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.056776762008667,
      "eval_rouge1": 0.6021,
      "eval_rouge2": 0.3317,
      "eval_rougeL": 0.5279,
      "eval_runtime": 26.5956,
      "eval_samples_per_second": 6.843,
      "eval_steps_per_second": 1.73,
      "step": 4510
    },
    {
      "epoch": 22.048780487804876,
      "grad_norm": 0.9685917496681213,
      "learning_rate": 8.655800054333062e-05,
      "loss": 1.9143,
      "step": 4520
    },
    {
      "epoch": 22.097560975609756,
      "grad_norm": 0.8598119616508484,
      "learning_rate": 8.65036674816626e-05,
      "loss": 1.883,
      "step": 4530
    },
    {
      "epoch": 22.146341463414632,
      "grad_norm": 0.9003345966339111,
      "learning_rate": 8.644933441999457e-05,
      "loss": 1.8913,
      "step": 4540
    },
    {
      "epoch": 22.195121951219512,
      "grad_norm": 0.8545246720314026,
      "learning_rate": 8.639500135832654e-05,
      "loss": 1.8511,
      "step": 4550
    },
    {
      "epoch": 22.24390243902439,
      "grad_norm": 0.8255957365036011,
      "learning_rate": 8.634066829665852e-05,
      "loss": 1.909,
      "step": 4560
    },
    {
      "epoch": 22.29268292682927,
      "grad_norm": 0.8835028409957886,
      "learning_rate": 8.62863352349905e-05,
      "loss": 1.9223,
      "step": 4570
    },
    {
      "epoch": 22.341463414634145,
      "grad_norm": 0.8599757552146912,
      "learning_rate": 8.623200217332247e-05,
      "loss": 1.8848,
      "step": 4580
    },
    {
      "epoch": 22.390243902439025,
      "grad_norm": 0.8880361318588257,
      "learning_rate": 8.617766911165444e-05,
      "loss": 1.9087,
      "step": 4590
    },
    {
      "epoch": 22.4390243902439,
      "grad_norm": 0.9443928003311157,
      "learning_rate": 8.612333604998642e-05,
      "loss": 1.9419,
      "step": 4600
    },
    {
      "epoch": 22.48780487804878,
      "grad_norm": 0.9929980635643005,
      "learning_rate": 8.606900298831839e-05,
      "loss": 1.9057,
      "step": 4610
    },
    {
      "epoch": 22.536585365853657,
      "grad_norm": 1.2470592260360718,
      "learning_rate": 8.601466992665038e-05,
      "loss": 1.9128,
      "step": 4620
    },
    {
      "epoch": 22.585365853658537,
      "grad_norm": 0.8578996062278748,
      "learning_rate": 8.596033686498234e-05,
      "loss": 1.8991,
      "step": 4630
    },
    {
      "epoch": 22.634146341463413,
      "grad_norm": 1.0579174757003784,
      "learning_rate": 8.590600380331431e-05,
      "loss": 1.9208,
      "step": 4640
    },
    {
      "epoch": 22.682926829268293,
      "grad_norm": 1.0299713611602783,
      "learning_rate": 8.58516707416463e-05,
      "loss": 1.9141,
      "step": 4650
    },
    {
      "epoch": 22.73170731707317,
      "grad_norm": 0.853769838809967,
      "learning_rate": 8.579733767997828e-05,
      "loss": 1.9069,
      "step": 4660
    },
    {
      "epoch": 22.78048780487805,
      "grad_norm": 0.9463319182395935,
      "learning_rate": 8.574300461831025e-05,
      "loss": 1.9162,
      "step": 4670
    },
    {
      "epoch": 22.829268292682926,
      "grad_norm": 0.9902511835098267,
      "learning_rate": 8.568867155664222e-05,
      "loss": 1.9439,
      "step": 4680
    },
    {
      "epoch": 22.878048780487806,
      "grad_norm": 0.9432972073554993,
      "learning_rate": 8.56343384949742e-05,
      "loss": 1.8906,
      "step": 4690
    },
    {
      "epoch": 22.926829268292682,
      "grad_norm": 0.9613229632377625,
      "learning_rate": 8.558000543330618e-05,
      "loss": 1.9144,
      "step": 4700
    },
    {
      "epoch": 22.975609756097562,
      "grad_norm": 0.8384121060371399,
      "learning_rate": 8.552567237163815e-05,
      "loss": 1.9618,
      "step": 4710
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.6467,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.055529832839966,
      "eval_rouge1": 0.603,
      "eval_rouge2": 0.336,
      "eval_rougeL": 0.531,
      "eval_runtime": 26.4562,
      "eval_samples_per_second": 6.879,
      "eval_steps_per_second": 1.739,
      "step": 4715
    },
    {
      "epoch": 23.024390243902438,
      "grad_norm": 0.836834728717804,
      "learning_rate": 8.547133930997012e-05,
      "loss": 1.8823,
      "step": 4720
    },
    {
      "epoch": 23.073170731707318,
      "grad_norm": 1.070541501045227,
      "learning_rate": 8.54170062483021e-05,
      "loss": 1.9168,
      "step": 4730
    },
    {
      "epoch": 23.121951219512194,
      "grad_norm": 0.9768303036689758,
      "learning_rate": 8.536267318663407e-05,
      "loss": 1.8677,
      "step": 4740
    },
    {
      "epoch": 23.170731707317074,
      "grad_norm": 0.9395390748977661,
      "learning_rate": 8.530834012496605e-05,
      "loss": 1.8669,
      "step": 4750
    },
    {
      "epoch": 23.21951219512195,
      "grad_norm": 0.9083912968635559,
      "learning_rate": 8.525400706329802e-05,
      "loss": 1.8952,
      "step": 4760
    },
    {
      "epoch": 23.26829268292683,
      "grad_norm": 0.8809327483177185,
      "learning_rate": 8.519967400162999e-05,
      "loss": 1.884,
      "step": 4770
    },
    {
      "epoch": 23.317073170731707,
      "grad_norm": 1.0371663570404053,
      "learning_rate": 8.514534093996197e-05,
      "loss": 1.889,
      "step": 4780
    },
    {
      "epoch": 23.365853658536587,
      "grad_norm": 0.8325390219688416,
      "learning_rate": 8.509100787829395e-05,
      "loss": 1.9055,
      "step": 4790
    },
    {
      "epoch": 23.414634146341463,
      "grad_norm": 0.9794867038726807,
      "learning_rate": 8.503667481662592e-05,
      "loss": 1.8662,
      "step": 4800
    },
    {
      "epoch": 23.463414634146343,
      "grad_norm": 0.8732627034187317,
      "learning_rate": 8.498234175495789e-05,
      "loss": 1.8775,
      "step": 4810
    },
    {
      "epoch": 23.51219512195122,
      "grad_norm": 0.99794602394104,
      "learning_rate": 8.492800869328987e-05,
      "loss": 1.8718,
      "step": 4820
    },
    {
      "epoch": 23.5609756097561,
      "grad_norm": 0.9115703701972961,
      "learning_rate": 8.487367563162184e-05,
      "loss": 1.9097,
      "step": 4830
    },
    {
      "epoch": 23.609756097560975,
      "grad_norm": 0.835806131362915,
      "learning_rate": 8.481934256995382e-05,
      "loss": 1.8978,
      "step": 4840
    },
    {
      "epoch": 23.658536585365855,
      "grad_norm": 1.145419955253601,
      "learning_rate": 8.476500950828579e-05,
      "loss": 1.8654,
      "step": 4850
    },
    {
      "epoch": 23.70731707317073,
      "grad_norm": 0.9367853999137878,
      "learning_rate": 8.471067644661778e-05,
      "loss": 1.8677,
      "step": 4860
    },
    {
      "epoch": 23.75609756097561,
      "grad_norm": 0.9627723097801208,
      "learning_rate": 8.465634338494974e-05,
      "loss": 1.8869,
      "step": 4870
    },
    {
      "epoch": 23.804878048780488,
      "grad_norm": 0.8921970129013062,
      "learning_rate": 8.460201032328173e-05,
      "loss": 1.8935,
      "step": 4880
    },
    {
      "epoch": 23.853658536585368,
      "grad_norm": 0.8427318334579468,
      "learning_rate": 8.45476772616137e-05,
      "loss": 1.8612,
      "step": 4890
    },
    {
      "epoch": 23.902439024390244,
      "grad_norm": 0.8289673924446106,
      "learning_rate": 8.449334419994566e-05,
      "loss": 1.9018,
      "step": 4900
    },
    {
      "epoch": 23.951219512195124,
      "grad_norm": 0.8206982016563416,
      "learning_rate": 8.443901113827765e-05,
      "loss": 1.8747,
      "step": 4910
    },
    {
      "epoch": 24.0,
      "grad_norm": 1.0877290964126587,
      "learning_rate": 8.438467807660963e-05,
      "loss": 1.916,
      "step": 4920
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6477,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0489752292633057,
      "eval_rouge1": 0.6045,
      "eval_rouge2": 0.3365,
      "eval_rougeL": 0.5323,
      "eval_runtime": 26.6799,
      "eval_samples_per_second": 6.822,
      "eval_steps_per_second": 1.724,
      "step": 4920
    },
    {
      "epoch": 24.048780487804876,
      "grad_norm": 0.9822722673416138,
      "learning_rate": 8.43303450149416e-05,
      "loss": 1.8604,
      "step": 4930
    },
    {
      "epoch": 24.097560975609756,
      "grad_norm": 0.8830099701881409,
      "learning_rate": 8.427601195327357e-05,
      "loss": 1.8229,
      "step": 4940
    },
    {
      "epoch": 24.146341463414632,
      "grad_norm": 0.931056022644043,
      "learning_rate": 8.422167889160555e-05,
      "loss": 1.853,
      "step": 4950
    },
    {
      "epoch": 24.195121951219512,
      "grad_norm": 1.1673840284347534,
      "learning_rate": 8.416734582993752e-05,
      "loss": 1.8008,
      "step": 4960
    },
    {
      "epoch": 24.24390243902439,
      "grad_norm": 0.8437151908874512,
      "learning_rate": 8.41130127682695e-05,
      "loss": 1.8747,
      "step": 4970
    },
    {
      "epoch": 24.29268292682927,
      "grad_norm": 0.8535563945770264,
      "learning_rate": 8.405867970660147e-05,
      "loss": 1.8898,
      "step": 4980
    },
    {
      "epoch": 24.341463414634145,
      "grad_norm": 0.9258223176002502,
      "learning_rate": 8.400434664493345e-05,
      "loss": 1.86,
      "step": 4990
    },
    {
      "epoch": 24.390243902439025,
      "grad_norm": 1.0053988695144653,
      "learning_rate": 8.395001358326542e-05,
      "loss": 1.8101,
      "step": 5000
    },
    {
      "epoch": 24.4390243902439,
      "grad_norm": 0.8608886003494263,
      "learning_rate": 8.38956805215974e-05,
      "loss": 1.8998,
      "step": 5010
    },
    {
      "epoch": 24.48780487804878,
      "grad_norm": 1.111860752105713,
      "learning_rate": 8.384134745992937e-05,
      "loss": 1.9247,
      "step": 5020
    },
    {
      "epoch": 24.536585365853657,
      "grad_norm": 0.9506413340568542,
      "learning_rate": 8.378701439826134e-05,
      "loss": 1.8309,
      "step": 5030
    },
    {
      "epoch": 24.585365853658537,
      "grad_norm": 0.9532819390296936,
      "learning_rate": 8.373268133659332e-05,
      "loss": 1.8863,
      "step": 5040
    },
    {
      "epoch": 24.634146341463413,
      "grad_norm": 1.0726252794265747,
      "learning_rate": 8.36783482749253e-05,
      "loss": 1.8353,
      "step": 5050
    },
    {
      "epoch": 24.682926829268293,
      "grad_norm": 0.8491644263267517,
      "learning_rate": 8.362401521325727e-05,
      "loss": 1.8833,
      "step": 5060
    },
    {
      "epoch": 24.73170731707317,
      "grad_norm": 0.8814260959625244,
      "learning_rate": 8.356968215158924e-05,
      "loss": 1.8765,
      "step": 5070
    },
    {
      "epoch": 24.78048780487805,
      "grad_norm": 1.0011340379714966,
      "learning_rate": 8.351534908992122e-05,
      "loss": 1.8507,
      "step": 5080
    },
    {
      "epoch": 24.829268292682926,
      "grad_norm": 0.819486141204834,
      "learning_rate": 8.34610160282532e-05,
      "loss": 1.8754,
      "step": 5090
    },
    {
      "epoch": 24.878048780487806,
      "grad_norm": 0.797431230545044,
      "learning_rate": 8.340668296658518e-05,
      "loss": 1.8754,
      "step": 5100
    },
    {
      "epoch": 24.926829268292682,
      "grad_norm": 0.8899915814399719,
      "learning_rate": 8.335234990491714e-05,
      "loss": 1.8484,
      "step": 5110
    },
    {
      "epoch": 24.975609756097562,
      "grad_norm": 0.9835275411605835,
      "learning_rate": 8.329801684324911e-05,
      "loss": 1.864,
      "step": 5120
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.649,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0439529418945312,
      "eval_rouge1": 0.6046,
      "eval_rouge2": 0.3378,
      "eval_rougeL": 0.533,
      "eval_runtime": 26.5855,
      "eval_samples_per_second": 6.846,
      "eval_steps_per_second": 1.73,
      "step": 5125
    },
    {
      "epoch": 25.024390243902438,
      "grad_norm": 1.0006405115127563,
      "learning_rate": 8.32436837815811e-05,
      "loss": 1.8371,
      "step": 5130
    },
    {
      "epoch": 25.073170731707318,
      "grad_norm": 0.9687209129333496,
      "learning_rate": 8.318935071991308e-05,
      "loss": 1.8286,
      "step": 5140
    },
    {
      "epoch": 25.121951219512194,
      "grad_norm": 0.8135698437690735,
      "learning_rate": 8.313501765824505e-05,
      "loss": 1.8417,
      "step": 5150
    },
    {
      "epoch": 25.170731707317074,
      "grad_norm": 0.9199495911598206,
      "learning_rate": 8.308068459657702e-05,
      "loss": 1.8202,
      "step": 5160
    },
    {
      "epoch": 25.21951219512195,
      "grad_norm": 0.7822583317756653,
      "learning_rate": 8.3026351534909e-05,
      "loss": 1.8535,
      "step": 5170
    },
    {
      "epoch": 25.26829268292683,
      "grad_norm": 0.8207578063011169,
      "learning_rate": 8.297201847324098e-05,
      "loss": 1.8064,
      "step": 5180
    },
    {
      "epoch": 25.317073170731707,
      "grad_norm": 0.8425044417381287,
      "learning_rate": 8.291768541157295e-05,
      "loss": 1.8554,
      "step": 5190
    },
    {
      "epoch": 25.365853658536587,
      "grad_norm": 0.9588025212287903,
      "learning_rate": 8.286335234990492e-05,
      "loss": 1.8297,
      "step": 5200
    },
    {
      "epoch": 25.414634146341463,
      "grad_norm": 0.8000227212905884,
      "learning_rate": 8.28090192882369e-05,
      "loss": 1.856,
      "step": 5210
    },
    {
      "epoch": 25.463414634146343,
      "grad_norm": 0.9651764035224915,
      "learning_rate": 8.275468622656887e-05,
      "loss": 1.8476,
      "step": 5220
    },
    {
      "epoch": 25.51219512195122,
      "grad_norm": 1.183806300163269,
      "learning_rate": 8.270035316490085e-05,
      "loss": 1.8133,
      "step": 5230
    },
    {
      "epoch": 25.5609756097561,
      "grad_norm": 0.9591472744941711,
      "learning_rate": 8.264602010323282e-05,
      "loss": 1.8158,
      "step": 5240
    },
    {
      "epoch": 25.609756097560975,
      "grad_norm": 1.062137484550476,
      "learning_rate": 8.259168704156479e-05,
      "loss": 1.8258,
      "step": 5250
    },
    {
      "epoch": 25.658536585365855,
      "grad_norm": 1.1689926385879517,
      "learning_rate": 8.253735397989677e-05,
      "loss": 1.8838,
      "step": 5260
    },
    {
      "epoch": 25.70731707317073,
      "grad_norm": 1.1229288578033447,
      "learning_rate": 8.248302091822875e-05,
      "loss": 1.8473,
      "step": 5270
    },
    {
      "epoch": 25.75609756097561,
      "grad_norm": 0.8929149508476257,
      "learning_rate": 8.242868785656071e-05,
      "loss": 1.7918,
      "step": 5280
    },
    {
      "epoch": 25.804878048780488,
      "grad_norm": 0.9470261931419373,
      "learning_rate": 8.237435479489269e-05,
      "loss": 1.8462,
      "step": 5290
    },
    {
      "epoch": 25.853658536585368,
      "grad_norm": 0.8769372701644897,
      "learning_rate": 8.232002173322467e-05,
      "loss": 1.8893,
      "step": 5300
    },
    {
      "epoch": 25.902439024390244,
      "grad_norm": 0.9617382287979126,
      "learning_rate": 8.226568867155666e-05,
      "loss": 1.8635,
      "step": 5310
    },
    {
      "epoch": 25.951219512195124,
      "grad_norm": 0.9043958187103271,
      "learning_rate": 8.221135560988861e-05,
      "loss": 1.8432,
      "step": 5320
    },
    {
      "epoch": 26.0,
      "grad_norm": 1.0180726051330566,
      "learning_rate": 8.21570225482206e-05,
      "loss": 1.8356,
      "step": 5330
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.65,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.047593832015991,
      "eval_rouge1": 0.6044,
      "eval_rouge2": 0.3385,
      "eval_rougeL": 0.534,
      "eval_runtime": 26.7369,
      "eval_samples_per_second": 6.807,
      "eval_steps_per_second": 1.72,
      "step": 5330
    },
    {
      "epoch": 26.048780487804876,
      "grad_norm": 2.064523696899414,
      "learning_rate": 8.210268948655258e-05,
      "loss": 1.8343,
      "step": 5340
    },
    {
      "epoch": 26.097560975609756,
      "grad_norm": 0.8723379969596863,
      "learning_rate": 8.204835642488455e-05,
      "loss": 1.83,
      "step": 5350
    },
    {
      "epoch": 26.146341463414632,
      "grad_norm": 0.9224991798400879,
      "learning_rate": 8.199402336321653e-05,
      "loss": 1.8117,
      "step": 5360
    },
    {
      "epoch": 26.195121951219512,
      "grad_norm": 1.009584903717041,
      "learning_rate": 8.19396903015485e-05,
      "loss": 1.7873,
      "step": 5370
    },
    {
      "epoch": 26.24390243902439,
      "grad_norm": 1.431678295135498,
      "learning_rate": 8.188535723988047e-05,
      "loss": 1.8388,
      "step": 5380
    },
    {
      "epoch": 26.29268292682927,
      "grad_norm": 1.0047580003738403,
      "learning_rate": 8.183102417821245e-05,
      "loss": 1.8312,
      "step": 5390
    },
    {
      "epoch": 26.341463414634145,
      "grad_norm": 0.9655563235282898,
      "learning_rate": 8.177669111654443e-05,
      "loss": 1.7934,
      "step": 5400
    },
    {
      "epoch": 26.390243902439025,
      "grad_norm": 0.8792780637741089,
      "learning_rate": 8.172235805487639e-05,
      "loss": 1.82,
      "step": 5410
    },
    {
      "epoch": 26.4390243902439,
      "grad_norm": 1.029196858406067,
      "learning_rate": 8.166802499320837e-05,
      "loss": 1.8188,
      "step": 5420
    },
    {
      "epoch": 26.48780487804878,
      "grad_norm": 0.8266960382461548,
      "learning_rate": 8.161369193154035e-05,
      "loss": 1.816,
      "step": 5430
    },
    {
      "epoch": 26.536585365853657,
      "grad_norm": 0.9967761039733887,
      "learning_rate": 8.155935886987233e-05,
      "loss": 1.8426,
      "step": 5440
    },
    {
      "epoch": 26.585365853658537,
      "grad_norm": 1.188904881477356,
      "learning_rate": 8.150502580820429e-05,
      "loss": 1.8239,
      "step": 5450
    },
    {
      "epoch": 26.634146341463413,
      "grad_norm": 0.8830356001853943,
      "learning_rate": 8.145069274653627e-05,
      "loss": 1.8242,
      "step": 5460
    },
    {
      "epoch": 26.682926829268293,
      "grad_norm": 0.9936667084693909,
      "learning_rate": 8.139635968486825e-05,
      "loss": 1.8068,
      "step": 5470
    },
    {
      "epoch": 26.73170731707317,
      "grad_norm": 1.1022076606750488,
      "learning_rate": 8.134202662320022e-05,
      "loss": 1.7986,
      "step": 5480
    },
    {
      "epoch": 26.78048780487805,
      "grad_norm": 1.1271874904632568,
      "learning_rate": 8.128769356153219e-05,
      "loss": 1.7993,
      "step": 5490
    },
    {
      "epoch": 26.829268292682926,
      "grad_norm": 0.9428896903991699,
      "learning_rate": 8.123336049986417e-05,
      "loss": 1.8519,
      "step": 5500
    },
    {
      "epoch": 26.878048780487806,
      "grad_norm": 0.874235212802887,
      "learning_rate": 8.117902743819614e-05,
      "loss": 1.7874,
      "step": 5510
    },
    {
      "epoch": 26.926829268292682,
      "grad_norm": 0.8641676902770996,
      "learning_rate": 8.112469437652812e-05,
      "loss": 1.8449,
      "step": 5520
    },
    {
      "epoch": 26.975609756097562,
      "grad_norm": 1.411012887954712,
      "learning_rate": 8.10703613148601e-05,
      "loss": 1.8203,
      "step": 5530
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.6519,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.039573907852173,
      "eval_rouge1": 0.6066,
      "eval_rouge2": 0.3403,
      "eval_rougeL": 0.5359,
      "eval_runtime": 26.6606,
      "eval_samples_per_second": 6.827,
      "eval_steps_per_second": 1.725,
      "step": 5535
    },
    {
      "epoch": 27.024390243902438,
      "grad_norm": 0.8869510293006897,
      "learning_rate": 8.101602825319206e-05,
      "loss": 1.8272,
      "step": 5540
    },
    {
      "epoch": 27.073170731707318,
      "grad_norm": 0.8785194158554077,
      "learning_rate": 8.096169519152404e-05,
      "loss": 1.8282,
      "step": 5550
    },
    {
      "epoch": 27.121951219512194,
      "grad_norm": 0.9073692560195923,
      "learning_rate": 8.090736212985603e-05,
      "loss": 1.7835,
      "step": 5560
    },
    {
      "epoch": 27.170731707317074,
      "grad_norm": 0.9559656977653503,
      "learning_rate": 8.0853029068188e-05,
      "loss": 1.8084,
      "step": 5570
    },
    {
      "epoch": 27.21951219512195,
      "grad_norm": 0.9089655876159668,
      "learning_rate": 8.079869600651996e-05,
      "loss": 1.8016,
      "step": 5580
    },
    {
      "epoch": 27.26829268292683,
      "grad_norm": 0.9023314714431763,
      "learning_rate": 8.074436294485195e-05,
      "loss": 1.8039,
      "step": 5590
    },
    {
      "epoch": 27.317073170731707,
      "grad_norm": 0.9001786708831787,
      "learning_rate": 8.069002988318393e-05,
      "loss": 1.8007,
      "step": 5600
    },
    {
      "epoch": 27.365853658536587,
      "grad_norm": 0.9051687121391296,
      "learning_rate": 8.06356968215159e-05,
      "loss": 1.8289,
      "step": 5610
    },
    {
      "epoch": 27.414634146341463,
      "grad_norm": 1.1516135931015015,
      "learning_rate": 8.058136375984787e-05,
      "loss": 1.8031,
      "step": 5620
    },
    {
      "epoch": 27.463414634146343,
      "grad_norm": 0.8888604640960693,
      "learning_rate": 8.052703069817985e-05,
      "loss": 1.7895,
      "step": 5630
    },
    {
      "epoch": 27.51219512195122,
      "grad_norm": 0.905109703540802,
      "learning_rate": 8.047269763651182e-05,
      "loss": 1.7975,
      "step": 5640
    },
    {
      "epoch": 27.5609756097561,
      "grad_norm": 0.9893642067909241,
      "learning_rate": 8.04183645748438e-05,
      "loss": 1.8248,
      "step": 5650
    },
    {
      "epoch": 27.609756097560975,
      "grad_norm": 0.8948488831520081,
      "learning_rate": 8.036403151317577e-05,
      "loss": 1.8219,
      "step": 5660
    },
    {
      "epoch": 27.658536585365855,
      "grad_norm": 0.871070921421051,
      "learning_rate": 8.030969845150774e-05,
      "loss": 1.7988,
      "step": 5670
    },
    {
      "epoch": 27.70731707317073,
      "grad_norm": 0.9689252376556396,
      "learning_rate": 8.025536538983972e-05,
      "loss": 1.7638,
      "step": 5680
    },
    {
      "epoch": 27.75609756097561,
      "grad_norm": 0.9327222108840942,
      "learning_rate": 8.02010323281717e-05,
      "loss": 1.7726,
      "step": 5690
    },
    {
      "epoch": 27.804878048780488,
      "grad_norm": 0.9302574396133423,
      "learning_rate": 8.014669926650367e-05,
      "loss": 1.7918,
      "step": 5700
    },
    {
      "epoch": 27.853658536585368,
      "grad_norm": 0.840606153011322,
      "learning_rate": 8.009236620483564e-05,
      "loss": 1.8006,
      "step": 5710
    },
    {
      "epoch": 27.902439024390244,
      "grad_norm": 0.894504725933075,
      "learning_rate": 8.003803314316762e-05,
      "loss": 1.7624,
      "step": 5720
    },
    {
      "epoch": 27.951219512195124,
      "grad_norm": 0.861795961856842,
      "learning_rate": 7.99837000814996e-05,
      "loss": 1.8129,
      "step": 5730
    },
    {
      "epoch": 28.0,
      "grad_norm": 1.6097631454467773,
      "learning_rate": 7.992936701983157e-05,
      "loss": 1.7592,
      "step": 5740
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6507,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.048104763031006,
      "eval_rouge1": 0.6057,
      "eval_rouge2": 0.3405,
      "eval_rougeL": 0.5354,
      "eval_runtime": 26.6142,
      "eval_samples_per_second": 6.838,
      "eval_steps_per_second": 1.728,
      "step": 5740
    },
    {
      "epoch": 28.048780487804876,
      "grad_norm": 0.8586742877960205,
      "learning_rate": 7.987503395816354e-05,
      "loss": 1.76,
      "step": 5750
    },
    {
      "epoch": 28.097560975609756,
      "grad_norm": 1.0001939535140991,
      "learning_rate": 7.982070089649552e-05,
      "loss": 1.8019,
      "step": 5760
    },
    {
      "epoch": 28.146341463414632,
      "grad_norm": 0.9395818710327148,
      "learning_rate": 7.976636783482749e-05,
      "loss": 1.7472,
      "step": 5770
    },
    {
      "epoch": 28.195121951219512,
      "grad_norm": 0.9348780512809753,
      "learning_rate": 7.971203477315948e-05,
      "loss": 1.7761,
      "step": 5780
    },
    {
      "epoch": 28.24390243902439,
      "grad_norm": 0.9028440713882446,
      "learning_rate": 7.965770171149144e-05,
      "loss": 1.7789,
      "step": 5790
    },
    {
      "epoch": 28.29268292682927,
      "grad_norm": 0.8583874702453613,
      "learning_rate": 7.960336864982341e-05,
      "loss": 1.7558,
      "step": 5800
    },
    {
      "epoch": 28.341463414634145,
      "grad_norm": 0.9990294575691223,
      "learning_rate": 7.95490355881554e-05,
      "loss": 1.8019,
      "step": 5810
    },
    {
      "epoch": 28.390243902439025,
      "grad_norm": 1.0208673477172852,
      "learning_rate": 7.949470252648738e-05,
      "loss": 1.783,
      "step": 5820
    },
    {
      "epoch": 28.4390243902439,
      "grad_norm": 0.9775688052177429,
      "learning_rate": 7.944036946481935e-05,
      "loss": 1.7554,
      "step": 5830
    },
    {
      "epoch": 28.48780487804878,
      "grad_norm": 0.9302669763565063,
      "learning_rate": 7.938603640315132e-05,
      "loss": 1.7615,
      "step": 5840
    },
    {
      "epoch": 28.536585365853657,
      "grad_norm": 0.8069590926170349,
      "learning_rate": 7.93317033414833e-05,
      "loss": 1.7985,
      "step": 5850
    },
    {
      "epoch": 28.585365853658537,
      "grad_norm": 0.979270339012146,
      "learning_rate": 7.927737027981527e-05,
      "loss": 1.7564,
      "step": 5860
    },
    {
      "epoch": 28.634146341463413,
      "grad_norm": 0.9459352493286133,
      "learning_rate": 7.922303721814725e-05,
      "loss": 1.7706,
      "step": 5870
    },
    {
      "epoch": 28.682926829268293,
      "grad_norm": 0.9745155572891235,
      "learning_rate": 7.916870415647922e-05,
      "loss": 1.7843,
      "step": 5880
    },
    {
      "epoch": 28.73170731707317,
      "grad_norm": 0.8070019483566284,
      "learning_rate": 7.91143710948112e-05,
      "loss": 1.7472,
      "step": 5890
    },
    {
      "epoch": 28.78048780487805,
      "grad_norm": 0.7862805128097534,
      "learning_rate": 7.906003803314317e-05,
      "loss": 1.7924,
      "step": 5900
    },
    {
      "epoch": 28.829268292682926,
      "grad_norm": 0.8524170517921448,
      "learning_rate": 7.900570497147515e-05,
      "loss": 1.7843,
      "step": 5910
    },
    {
      "epoch": 28.878048780487806,
      "grad_norm": 0.989010751247406,
      "learning_rate": 7.895137190980712e-05,
      "loss": 1.8014,
      "step": 5920
    },
    {
      "epoch": 28.926829268292682,
      "grad_norm": 0.9778967499732971,
      "learning_rate": 7.889703884813909e-05,
      "loss": 1.7958,
      "step": 5930
    },
    {
      "epoch": 28.975609756097562,
      "grad_norm": 0.9401681423187256,
      "learning_rate": 7.884270578647107e-05,
      "loss": 1.8091,
      "step": 5940
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.6501,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.045945405960083,
      "eval_rouge1": 0.6045,
      "eval_rouge2": 0.3397,
      "eval_rougeL": 0.5346,
      "eval_runtime": 26.6264,
      "eval_samples_per_second": 6.835,
      "eval_steps_per_second": 1.728,
      "step": 5945
    },
    {
      "epoch": 29.024390243902438,
      "grad_norm": 0.9820492267608643,
      "learning_rate": 7.878837272480305e-05,
      "loss": 1.7651,
      "step": 5950
    },
    {
      "epoch": 29.073170731707318,
      "grad_norm": 1.0031334161758423,
      "learning_rate": 7.873403966313502e-05,
      "loss": 1.7359,
      "step": 5960
    },
    {
      "epoch": 29.121951219512194,
      "grad_norm": 0.8762280344963074,
      "learning_rate": 7.867970660146699e-05,
      "loss": 1.7908,
      "step": 5970
    },
    {
      "epoch": 29.170731707317074,
      "grad_norm": 0.8959635496139526,
      "learning_rate": 7.862537353979897e-05,
      "loss": 1.7316,
      "step": 5980
    },
    {
      "epoch": 29.21951219512195,
      "grad_norm": 1.0365235805511475,
      "learning_rate": 7.857104047813094e-05,
      "loss": 1.7899,
      "step": 5990
    },
    {
      "epoch": 29.26829268292683,
      "grad_norm": 0.9004618525505066,
      "learning_rate": 7.851670741646292e-05,
      "loss": 1.7315,
      "step": 6000
    },
    {
      "epoch": 29.317073170731707,
      "grad_norm": 1.1213397979736328,
      "learning_rate": 7.846237435479489e-05,
      "loss": 1.7577,
      "step": 6010
    },
    {
      "epoch": 29.365853658536587,
      "grad_norm": 0.8381146788597107,
      "learning_rate": 7.840804129312686e-05,
      "loss": 1.7488,
      "step": 6020
    },
    {
      "epoch": 29.414634146341463,
      "grad_norm": 1.0437678098678589,
      "learning_rate": 7.835370823145884e-05,
      "loss": 1.7583,
      "step": 6030
    },
    {
      "epoch": 29.463414634146343,
      "grad_norm": 1.0147404670715332,
      "learning_rate": 7.829937516979083e-05,
      "loss": 1.7358,
      "step": 6040
    },
    {
      "epoch": 29.51219512195122,
      "grad_norm": 1.1309133768081665,
      "learning_rate": 7.82450421081228e-05,
      "loss": 1.7431,
      "step": 6050
    },
    {
      "epoch": 29.5609756097561,
      "grad_norm": 1.2266108989715576,
      "learning_rate": 7.819070904645476e-05,
      "loss": 1.7526,
      "step": 6060
    },
    {
      "epoch": 29.609756097560975,
      "grad_norm": 0.9693206548690796,
      "learning_rate": 7.813637598478675e-05,
      "loss": 1.7633,
      "step": 6070
    },
    {
      "epoch": 29.658536585365855,
      "grad_norm": 1.009680151939392,
      "learning_rate": 7.808204292311873e-05,
      "loss": 1.7504,
      "step": 6080
    },
    {
      "epoch": 29.70731707317073,
      "grad_norm": 0.9043843746185303,
      "learning_rate": 7.80277098614507e-05,
      "loss": 1.7796,
      "step": 6090
    },
    {
      "epoch": 29.75609756097561,
      "grad_norm": 0.7989161610603333,
      "learning_rate": 7.797337679978267e-05,
      "loss": 1.7718,
      "step": 6100
    },
    {
      "epoch": 29.804878048780488,
      "grad_norm": 0.9764060974121094,
      "learning_rate": 7.791904373811465e-05,
      "loss": 1.7213,
      "step": 6110
    },
    {
      "epoch": 29.853658536585368,
      "grad_norm": 1.0457733869552612,
      "learning_rate": 7.786471067644662e-05,
      "loss": 1.7723,
      "step": 6120
    },
    {
      "epoch": 29.902439024390244,
      "grad_norm": 1.0509238243103027,
      "learning_rate": 7.78103776147786e-05,
      "loss": 1.7622,
      "step": 6130
    },
    {
      "epoch": 29.951219512195124,
      "grad_norm": 1.1585285663604736,
      "learning_rate": 7.775604455311057e-05,
      "loss": 1.765,
      "step": 6140
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.9969220757484436,
      "learning_rate": 7.770171149144254e-05,
      "loss": 1.7445,
      "step": 6150
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6529,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.041071653366089,
      "eval_rouge1": 0.6081,
      "eval_rouge2": 0.3424,
      "eval_rougeL": 0.5373,
      "eval_runtime": 26.8722,
      "eval_samples_per_second": 6.773,
      "eval_steps_per_second": 1.712,
      "step": 6150
    },
    {
      "epoch": 30.048780487804876,
      "grad_norm": 0.8964574933052063,
      "learning_rate": 7.764737842977452e-05,
      "loss": 1.7597,
      "step": 6160
    },
    {
      "epoch": 30.097560975609756,
      "grad_norm": 1.152288794517517,
      "learning_rate": 7.75930453681065e-05,
      "loss": 1.7119,
      "step": 6170
    },
    {
      "epoch": 30.146341463414632,
      "grad_norm": 0.9494566917419434,
      "learning_rate": 7.753871230643847e-05,
      "loss": 1.7019,
      "step": 6180
    },
    {
      "epoch": 30.195121951219512,
      "grad_norm": 0.9420191049575806,
      "learning_rate": 7.748437924477044e-05,
      "loss": 1.759,
      "step": 6190
    },
    {
      "epoch": 30.24390243902439,
      "grad_norm": 1.0232353210449219,
      "learning_rate": 7.743004618310242e-05,
      "loss": 1.7417,
      "step": 6200
    },
    {
      "epoch": 30.29268292682927,
      "grad_norm": 0.9363576769828796,
      "learning_rate": 7.73757131214344e-05,
      "loss": 1.7557,
      "step": 6210
    },
    {
      "epoch": 30.341463414634145,
      "grad_norm": 0.8861675262451172,
      "learning_rate": 7.732138005976637e-05,
      "loss": 1.7179,
      "step": 6220
    },
    {
      "epoch": 30.390243902439025,
      "grad_norm": 1.0272518396377563,
      "learning_rate": 7.726704699809834e-05,
      "loss": 1.7583,
      "step": 6230
    },
    {
      "epoch": 30.4390243902439,
      "grad_norm": 0.9605531096458435,
      "learning_rate": 7.721271393643032e-05,
      "loss": 1.7494,
      "step": 6240
    },
    {
      "epoch": 30.48780487804878,
      "grad_norm": 0.9756708741188049,
      "learning_rate": 7.71583808747623e-05,
      "loss": 1.7332,
      "step": 6250
    },
    {
      "epoch": 30.536585365853657,
      "grad_norm": 1.0306313037872314,
      "learning_rate": 7.710404781309428e-05,
      "loss": 1.726,
      "step": 6260
    },
    {
      "epoch": 30.585365853658537,
      "grad_norm": 1.0111984014511108,
      "learning_rate": 7.704971475142624e-05,
      "loss": 1.7343,
      "step": 6270
    },
    {
      "epoch": 30.634146341463413,
      "grad_norm": 0.9273544549942017,
      "learning_rate": 7.699538168975821e-05,
      "loss": 1.747,
      "step": 6280
    },
    {
      "epoch": 30.682926829268293,
      "grad_norm": 0.8815616369247437,
      "learning_rate": 7.69410486280902e-05,
      "loss": 1.7264,
      "step": 6290
    },
    {
      "epoch": 30.73170731707317,
      "grad_norm": 0.8531066179275513,
      "learning_rate": 7.688671556642218e-05,
      "loss": 1.7417,
      "step": 6300
    },
    {
      "epoch": 30.78048780487805,
      "grad_norm": 0.964625358581543,
      "learning_rate": 7.683238250475415e-05,
      "loss": 1.7882,
      "step": 6310
    },
    {
      "epoch": 30.829268292682926,
      "grad_norm": 1.0142056941986084,
      "learning_rate": 7.677804944308612e-05,
      "loss": 1.7422,
      "step": 6320
    },
    {
      "epoch": 30.878048780487806,
      "grad_norm": 0.9176868200302124,
      "learning_rate": 7.67237163814181e-05,
      "loss": 1.7422,
      "step": 6330
    },
    {
      "epoch": 30.926829268292682,
      "grad_norm": 1.126758337020874,
      "learning_rate": 7.666938331975008e-05,
      "loss": 1.7626,
      "step": 6340
    },
    {
      "epoch": 30.975609756097562,
      "grad_norm": 1.0797853469848633,
      "learning_rate": 7.661505025808205e-05,
      "loss": 1.7303,
      "step": 6350
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.6559,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0429515838623047,
      "eval_rouge1": 0.6091,
      "eval_rouge2": 0.3452,
      "eval_rougeL": 0.5398,
      "eval_runtime": 26.8129,
      "eval_samples_per_second": 6.788,
      "eval_steps_per_second": 1.716,
      "step": 6355
    },
    {
      "epoch": 31.024390243902438,
      "grad_norm": 1.0031147003173828,
      "learning_rate": 7.656071719641402e-05,
      "loss": 1.6963,
      "step": 6360
    },
    {
      "epoch": 31.073170731707318,
      "grad_norm": 1.0235823392868042,
      "learning_rate": 7.6506384134746e-05,
      "loss": 1.6978,
      "step": 6370
    },
    {
      "epoch": 31.121951219512194,
      "grad_norm": 0.9943031072616577,
      "learning_rate": 7.645205107307797e-05,
      "loss": 1.7137,
      "step": 6380
    },
    {
      "epoch": 31.170731707317074,
      "grad_norm": 0.9972837567329407,
      "learning_rate": 7.639771801140995e-05,
      "loss": 1.7383,
      "step": 6390
    },
    {
      "epoch": 31.21951219512195,
      "grad_norm": 0.9345105290412903,
      "learning_rate": 7.634338494974192e-05,
      "loss": 1.6891,
      "step": 6400
    },
    {
      "epoch": 31.26829268292683,
      "grad_norm": 0.9205601215362549,
      "learning_rate": 7.628905188807389e-05,
      "loss": 1.7028,
      "step": 6410
    },
    {
      "epoch": 31.317073170731707,
      "grad_norm": 0.8768946528434753,
      "learning_rate": 7.623471882640587e-05,
      "loss": 1.7453,
      "step": 6420
    },
    {
      "epoch": 31.365853658536587,
      "grad_norm": 0.7679451107978821,
      "learning_rate": 7.618038576473785e-05,
      "loss": 1.739,
      "step": 6430
    },
    {
      "epoch": 31.414634146341463,
      "grad_norm": 0.8392273783683777,
      "learning_rate": 7.612605270306982e-05,
      "loss": 1.7096,
      "step": 6440
    },
    {
      "epoch": 31.463414634146343,
      "grad_norm": 1.02446448802948,
      "learning_rate": 7.607171964140179e-05,
      "loss": 1.7051,
      "step": 6450
    },
    {
      "epoch": 31.51219512195122,
      "grad_norm": 0.9734623432159424,
      "learning_rate": 7.601738657973377e-05,
      "loss": 1.7371,
      "step": 6460
    },
    {
      "epoch": 31.5609756097561,
      "grad_norm": 0.8957915306091309,
      "learning_rate": 7.596305351806576e-05,
      "loss": 1.7308,
      "step": 6470
    },
    {
      "epoch": 31.609756097560975,
      "grad_norm": 0.969406008720398,
      "learning_rate": 7.590872045639773e-05,
      "loss": 1.7053,
      "step": 6480
    },
    {
      "epoch": 31.658536585365855,
      "grad_norm": 0.9478543996810913,
      "learning_rate": 7.58543873947297e-05,
      "loss": 1.7717,
      "step": 6490
    },
    {
      "epoch": 31.70731707317073,
      "grad_norm": 0.8509706854820251,
      "learning_rate": 7.580005433306168e-05,
      "loss": 1.7209,
      "step": 6500
    },
    {
      "epoch": 31.75609756097561,
      "grad_norm": 0.9139954447746277,
      "learning_rate": 7.574572127139365e-05,
      "loss": 1.7463,
      "step": 6510
    },
    {
      "epoch": 31.804878048780488,
      "grad_norm": 1.0609540939331055,
      "learning_rate": 7.569138820972563e-05,
      "loss": 1.7234,
      "step": 6520
    },
    {
      "epoch": 31.853658536585368,
      "grad_norm": 0.9388116002082825,
      "learning_rate": 7.56370551480576e-05,
      "loss": 1.7286,
      "step": 6530
    },
    {
      "epoch": 31.902439024390244,
      "grad_norm": 0.9279087781906128,
      "learning_rate": 7.558272208638957e-05,
      "loss": 1.6966,
      "step": 6540
    },
    {
      "epoch": 31.951219512195124,
      "grad_norm": 0.9830589294433594,
      "learning_rate": 7.552838902472155e-05,
      "loss": 1.7369,
      "step": 6550
    },
    {
      "epoch": 32.0,
      "grad_norm": 1.2547836303710938,
      "learning_rate": 7.547405596305353e-05,
      "loss": 1.7659,
      "step": 6560
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6549,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0383822917938232,
      "eval_rouge1": 0.6098,
      "eval_rouge2": 0.3453,
      "eval_rougeL": 0.5391,
      "eval_runtime": 26.6337,
      "eval_samples_per_second": 6.833,
      "eval_steps_per_second": 1.727,
      "step": 6560
    },
    {
      "epoch": 32.048780487804876,
      "grad_norm": 1.1577900648117065,
      "learning_rate": 7.54197229013855e-05,
      "loss": 1.693,
      "step": 6570
    },
    {
      "epoch": 32.09756097560975,
      "grad_norm": 1.0532095432281494,
      "learning_rate": 7.536538983971747e-05,
      "loss": 1.6854,
      "step": 6580
    },
    {
      "epoch": 32.146341463414636,
      "grad_norm": 1.2690882682800293,
      "learning_rate": 7.531105677804945e-05,
      "loss": 1.7024,
      "step": 6590
    },
    {
      "epoch": 32.19512195121951,
      "grad_norm": 0.9884124994277954,
      "learning_rate": 7.525672371638142e-05,
      "loss": 1.7099,
      "step": 6600
    },
    {
      "epoch": 32.24390243902439,
      "grad_norm": 0.9226039052009583,
      "learning_rate": 7.52023906547134e-05,
      "loss": 1.6841,
      "step": 6610
    },
    {
      "epoch": 32.292682926829265,
      "grad_norm": 0.8764020800590515,
      "learning_rate": 7.514805759304537e-05,
      "loss": 1.7114,
      "step": 6620
    },
    {
      "epoch": 32.34146341463415,
      "grad_norm": 0.9021767973899841,
      "learning_rate": 7.509372453137735e-05,
      "loss": 1.7112,
      "step": 6630
    },
    {
      "epoch": 32.390243902439025,
      "grad_norm": 0.9591324925422668,
      "learning_rate": 7.503939146970932e-05,
      "loss": 1.7037,
      "step": 6640
    },
    {
      "epoch": 32.4390243902439,
      "grad_norm": 0.8968470096588135,
      "learning_rate": 7.49850584080413e-05,
      "loss": 1.7042,
      "step": 6650
    },
    {
      "epoch": 32.48780487804878,
      "grad_norm": 0.8518949151039124,
      "learning_rate": 7.493072534637327e-05,
      "loss": 1.7031,
      "step": 6660
    },
    {
      "epoch": 32.53658536585366,
      "grad_norm": 1.0725502967834473,
      "learning_rate": 7.487639228470524e-05,
      "loss": 1.6743,
      "step": 6670
    },
    {
      "epoch": 32.58536585365854,
      "grad_norm": 0.92808997631073,
      "learning_rate": 7.482205922303722e-05,
      "loss": 1.7228,
      "step": 6680
    },
    {
      "epoch": 32.63414634146341,
      "grad_norm": 0.856819748878479,
      "learning_rate": 7.47677261613692e-05,
      "loss": 1.7069,
      "step": 6690
    },
    {
      "epoch": 32.68292682926829,
      "grad_norm": 1.0233670473098755,
      "learning_rate": 7.471339309970116e-05,
      "loss": 1.7207,
      "step": 6700
    },
    {
      "epoch": 32.73170731707317,
      "grad_norm": 0.8869268298149109,
      "learning_rate": 7.465906003803314e-05,
      "loss": 1.6835,
      "step": 6710
    },
    {
      "epoch": 32.78048780487805,
      "grad_norm": 0.9873183369636536,
      "learning_rate": 7.460472697636513e-05,
      "loss": 1.6548,
      "step": 6720
    },
    {
      "epoch": 32.829268292682926,
      "grad_norm": 1.2351232767105103,
      "learning_rate": 7.45503939146971e-05,
      "loss": 1.7437,
      "step": 6730
    },
    {
      "epoch": 32.8780487804878,
      "grad_norm": 0.9190348982810974,
      "learning_rate": 7.449606085302908e-05,
      "loss": 1.7203,
      "step": 6740
    },
    {
      "epoch": 32.926829268292686,
      "grad_norm": 1.0856812000274658,
      "learning_rate": 7.444172779136105e-05,
      "loss": 1.7515,
      "step": 6750
    },
    {
      "epoch": 32.97560975609756,
      "grad_norm": 0.9336557388305664,
      "learning_rate": 7.438739472969301e-05,
      "loss": 1.7392,
      "step": 6760
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.6555,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0462191104888916,
      "eval_rouge1": 0.6078,
      "eval_rouge2": 0.3444,
      "eval_rougeL": 0.5387,
      "eval_runtime": 26.6235,
      "eval_samples_per_second": 6.836,
      "eval_steps_per_second": 1.728,
      "step": 6765
    },
    {
      "epoch": 33.02439024390244,
      "grad_norm": 0.9959226250648499,
      "learning_rate": 7.4333061668025e-05,
      "loss": 1.6647,
      "step": 6770
    },
    {
      "epoch": 33.073170731707314,
      "grad_norm": 0.9365916848182678,
      "learning_rate": 7.427872860635698e-05,
      "loss": 1.6814,
      "step": 6780
    },
    {
      "epoch": 33.1219512195122,
      "grad_norm": 0.9217053651809692,
      "learning_rate": 7.422439554468895e-05,
      "loss": 1.6695,
      "step": 6790
    },
    {
      "epoch": 33.170731707317074,
      "grad_norm": 1.186452031135559,
      "learning_rate": 7.417006248302092e-05,
      "loss": 1.6631,
      "step": 6800
    },
    {
      "epoch": 33.21951219512195,
      "grad_norm": 0.8331051468849182,
      "learning_rate": 7.41157294213529e-05,
      "loss": 1.6808,
      "step": 6810
    },
    {
      "epoch": 33.26829268292683,
      "grad_norm": 0.9306808114051819,
      "learning_rate": 7.406139635968488e-05,
      "loss": 1.7146,
      "step": 6820
    },
    {
      "epoch": 33.31707317073171,
      "grad_norm": 1.0423853397369385,
      "learning_rate": 7.400706329801684e-05,
      "loss": 1.7025,
      "step": 6830
    },
    {
      "epoch": 33.36585365853659,
      "grad_norm": 0.974290132522583,
      "learning_rate": 7.395273023634882e-05,
      "loss": 1.7215,
      "step": 6840
    },
    {
      "epoch": 33.41463414634146,
      "grad_norm": 0.9810996055603027,
      "learning_rate": 7.38983971746808e-05,
      "loss": 1.6833,
      "step": 6850
    },
    {
      "epoch": 33.46341463414634,
      "grad_norm": 1.0704784393310547,
      "learning_rate": 7.384406411301277e-05,
      "loss": 1.6692,
      "step": 6860
    },
    {
      "epoch": 33.51219512195122,
      "grad_norm": 0.9080216884613037,
      "learning_rate": 7.378973105134474e-05,
      "loss": 1.7025,
      "step": 6870
    },
    {
      "epoch": 33.5609756097561,
      "grad_norm": 0.9012085795402527,
      "learning_rate": 7.373539798967672e-05,
      "loss": 1.7092,
      "step": 6880
    },
    {
      "epoch": 33.609756097560975,
      "grad_norm": 1.106162667274475,
      "learning_rate": 7.368106492800869e-05,
      "loss": 1.673,
      "step": 6890
    },
    {
      "epoch": 33.65853658536585,
      "grad_norm": 0.9788945317268372,
      "learning_rate": 7.362673186634067e-05,
      "loss": 1.6947,
      "step": 6900
    },
    {
      "epoch": 33.707317073170735,
      "grad_norm": 0.968518853187561,
      "learning_rate": 7.357239880467266e-05,
      "loss": 1.6818,
      "step": 6910
    },
    {
      "epoch": 33.75609756097561,
      "grad_norm": 0.9239228963851929,
      "learning_rate": 7.351806574300462e-05,
      "loss": 1.6944,
      "step": 6920
    },
    {
      "epoch": 33.80487804878049,
      "grad_norm": 1.005051851272583,
      "learning_rate": 7.346373268133659e-05,
      "loss": 1.7169,
      "step": 6930
    },
    {
      "epoch": 33.853658536585364,
      "grad_norm": 0.8998588919639587,
      "learning_rate": 7.340939961966858e-05,
      "loss": 1.6913,
      "step": 6940
    },
    {
      "epoch": 33.90243902439025,
      "grad_norm": 0.918029248714447,
      "learning_rate": 7.335506655800056e-05,
      "loss": 1.6925,
      "step": 6950
    },
    {
      "epoch": 33.951219512195124,
      "grad_norm": 0.8755961656570435,
      "learning_rate": 7.330073349633251e-05,
      "loss": 1.6859,
      "step": 6960
    },
    {
      "epoch": 34.0,
      "grad_norm": 1.1943050622940063,
      "learning_rate": 7.32464004346645e-05,
      "loss": 1.7102,
      "step": 6970
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6567,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.041721820831299,
      "eval_rouge1": 0.6102,
      "eval_rouge2": 0.3486,
      "eval_rougeL": 0.5411,
      "eval_runtime": 26.6556,
      "eval_samples_per_second": 6.828,
      "eval_steps_per_second": 1.726,
      "step": 6970
    },
    {
      "epoch": 34.048780487804876,
      "grad_norm": 0.9968337416648865,
      "learning_rate": 7.319206737299648e-05,
      "loss": 1.6624,
      "step": 6980
    },
    {
      "epoch": 34.09756097560975,
      "grad_norm": 0.837787389755249,
      "learning_rate": 7.313773431132845e-05,
      "loss": 1.6608,
      "step": 6990
    },
    {
      "epoch": 34.146341463414636,
      "grad_norm": 0.9373610615730286,
      "learning_rate": 7.308340124966042e-05,
      "loss": 1.6851,
      "step": 7000
    },
    {
      "epoch": 34.19512195121951,
      "grad_norm": 0.8893011808395386,
      "learning_rate": 7.30290681879924e-05,
      "loss": 1.6436,
      "step": 7010
    },
    {
      "epoch": 34.24390243902439,
      "grad_norm": 1.1527223587036133,
      "learning_rate": 7.297473512632437e-05,
      "loss": 1.6833,
      "step": 7020
    },
    {
      "epoch": 34.292682926829265,
      "grad_norm": 0.9689236879348755,
      "learning_rate": 7.292040206465635e-05,
      "loss": 1.6533,
      "step": 7030
    },
    {
      "epoch": 34.34146341463415,
      "grad_norm": 1.0654863119125366,
      "learning_rate": 7.286606900298833e-05,
      "loss": 1.6665,
      "step": 7040
    },
    {
      "epoch": 34.390243902439025,
      "grad_norm": 1.0611974000930786,
      "learning_rate": 7.281173594132029e-05,
      "loss": 1.6584,
      "step": 7050
    },
    {
      "epoch": 34.4390243902439,
      "grad_norm": 0.9801427721977234,
      "learning_rate": 7.275740287965227e-05,
      "loss": 1.7052,
      "step": 7060
    },
    {
      "epoch": 34.48780487804878,
      "grad_norm": 1.125786542892456,
      "learning_rate": 7.270306981798425e-05,
      "loss": 1.6948,
      "step": 7070
    },
    {
      "epoch": 34.53658536585366,
      "grad_norm": 0.8822893500328064,
      "learning_rate": 7.264873675631623e-05,
      "loss": 1.6732,
      "step": 7080
    },
    {
      "epoch": 34.58536585365854,
      "grad_norm": 0.8989709615707397,
      "learning_rate": 7.259440369464819e-05,
      "loss": 1.6392,
      "step": 7090
    },
    {
      "epoch": 34.63414634146341,
      "grad_norm": 0.8959782719612122,
      "learning_rate": 7.254007063298017e-05,
      "loss": 1.6664,
      "step": 7100
    },
    {
      "epoch": 34.68292682926829,
      "grad_norm": 0.9004738330841064,
      "learning_rate": 7.248573757131215e-05,
      "loss": 1.6846,
      "step": 7110
    },
    {
      "epoch": 34.73170731707317,
      "grad_norm": 1.095853567123413,
      "learning_rate": 7.243140450964412e-05,
      "loss": 1.6757,
      "step": 7120
    },
    {
      "epoch": 34.78048780487805,
      "grad_norm": 1.0721943378448486,
      "learning_rate": 7.237707144797609e-05,
      "loss": 1.6562,
      "step": 7130
    },
    {
      "epoch": 34.829268292682926,
      "grad_norm": 0.9506831169128418,
      "learning_rate": 7.232273838630807e-05,
      "loss": 1.6834,
      "step": 7140
    },
    {
      "epoch": 34.8780487804878,
      "grad_norm": 1.006168007850647,
      "learning_rate": 7.226840532464004e-05,
      "loss": 1.6787,
      "step": 7150
    },
    {
      "epoch": 34.926829268292686,
      "grad_norm": 0.990227460861206,
      "learning_rate": 7.221407226297202e-05,
      "loss": 1.6531,
      "step": 7160
    },
    {
      "epoch": 34.97560975609756,
      "grad_norm": 0.9547024965286255,
      "learning_rate": 7.215973920130399e-05,
      "loss": 1.6717,
      "step": 7170
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.6541,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0457675457000732,
      "eval_rouge1": 0.608,
      "eval_rouge2": 0.3446,
      "eval_rougeL": 0.5386,
      "eval_runtime": 26.6803,
      "eval_samples_per_second": 6.822,
      "eval_steps_per_second": 1.724,
      "step": 7175
    },
    {
      "epoch": 35.02439024390244,
      "grad_norm": 0.9323550462722778,
      "learning_rate": 7.210540613963596e-05,
      "loss": 1.6773,
      "step": 7180
    },
    {
      "epoch": 35.073170731707314,
      "grad_norm": 0.9128536581993103,
      "learning_rate": 7.205107307796794e-05,
      "loss": 1.6005,
      "step": 7190
    },
    {
      "epoch": 35.1219512195122,
      "grad_norm": 0.9445728659629822,
      "learning_rate": 7.199674001629993e-05,
      "loss": 1.6501,
      "step": 7200
    },
    {
      "epoch": 35.170731707317074,
      "grad_norm": 0.8394743204116821,
      "learning_rate": 7.194240695463191e-05,
      "loss": 1.6709,
      "step": 7210
    },
    {
      "epoch": 35.21951219512195,
      "grad_norm": 0.8431241512298584,
      "learning_rate": 7.188807389296386e-05,
      "loss": 1.6454,
      "step": 7220
    },
    {
      "epoch": 35.26829268292683,
      "grad_norm": 1.0416409969329834,
      "learning_rate": 7.183374083129585e-05,
      "loss": 1.6869,
      "step": 7230
    },
    {
      "epoch": 35.31707317073171,
      "grad_norm": 1.0885652303695679,
      "learning_rate": 7.177940776962783e-05,
      "loss": 1.6383,
      "step": 7240
    },
    {
      "epoch": 35.36585365853659,
      "grad_norm": 1.0008647441864014,
      "learning_rate": 7.17250747079598e-05,
      "loss": 1.6154,
      "step": 7250
    },
    {
      "epoch": 35.41463414634146,
      "grad_norm": 1.1045302152633667,
      "learning_rate": 7.167074164629177e-05,
      "loss": 1.65,
      "step": 7260
    },
    {
      "epoch": 35.46341463414634,
      "grad_norm": 0.8903409838676453,
      "learning_rate": 7.161640858462375e-05,
      "loss": 1.6746,
      "step": 7270
    },
    {
      "epoch": 35.51219512195122,
      "grad_norm": 1.0800362825393677,
      "learning_rate": 7.156207552295572e-05,
      "loss": 1.6472,
      "step": 7280
    },
    {
      "epoch": 35.5609756097561,
      "grad_norm": 0.8663403391838074,
      "learning_rate": 7.15077424612877e-05,
      "loss": 1.6474,
      "step": 7290
    },
    {
      "epoch": 35.609756097560975,
      "grad_norm": 1.0188908576965332,
      "learning_rate": 7.145340939961967e-05,
      "loss": 1.6916,
      "step": 7300
    },
    {
      "epoch": 35.65853658536585,
      "grad_norm": 0.9418292045593262,
      "learning_rate": 7.139907633795164e-05,
      "loss": 1.6662,
      "step": 7310
    },
    {
      "epoch": 35.707317073170735,
      "grad_norm": 0.963869035243988,
      "learning_rate": 7.134474327628362e-05,
      "loss": 1.6341,
      "step": 7320
    },
    {
      "epoch": 35.75609756097561,
      "grad_norm": 0.908099353313446,
      "learning_rate": 7.12904102146156e-05,
      "loss": 1.6545,
      "step": 7330
    },
    {
      "epoch": 35.80487804878049,
      "grad_norm": 0.9877302646636963,
      "learning_rate": 7.123607715294757e-05,
      "loss": 1.6664,
      "step": 7340
    },
    {
      "epoch": 35.853658536585364,
      "grad_norm": 0.8455460667610168,
      "learning_rate": 7.118174409127954e-05,
      "loss": 1.6903,
      "step": 7350
    },
    {
      "epoch": 35.90243902439025,
      "grad_norm": 1.126727819442749,
      "learning_rate": 7.112741102961152e-05,
      "loss": 1.7045,
      "step": 7360
    },
    {
      "epoch": 35.951219512195124,
      "grad_norm": 0.9532201290130615,
      "learning_rate": 7.10730779679435e-05,
      "loss": 1.6644,
      "step": 7370
    },
    {
      "epoch": 36.0,
      "grad_norm": 1.0697003602981567,
      "learning_rate": 7.101874490627547e-05,
      "loss": 1.6716,
      "step": 7380
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6569,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0504531860351562,
      "eval_rouge1": 0.6102,
      "eval_rouge2": 0.347,
      "eval_rougeL": 0.5419,
      "eval_runtime": 26.7626,
      "eval_samples_per_second": 6.801,
      "eval_steps_per_second": 1.719,
      "step": 7380
    },
    {
      "epoch": 36.048780487804876,
      "grad_norm": 1.00981605052948,
      "learning_rate": 7.096441184460744e-05,
      "loss": 1.6188,
      "step": 7390
    },
    {
      "epoch": 36.09756097560975,
      "grad_norm": 0.9147917032241821,
      "learning_rate": 7.091007878293942e-05,
      "loss": 1.6757,
      "step": 7400
    },
    {
      "epoch": 36.146341463414636,
      "grad_norm": 0.8071935176849365,
      "learning_rate": 7.08557457212714e-05,
      "loss": 1.6582,
      "step": 7410
    },
    {
      "epoch": 36.19512195121951,
      "grad_norm": 1.0185075998306274,
      "learning_rate": 7.080141265960338e-05,
      "loss": 1.643,
      "step": 7420
    },
    {
      "epoch": 36.24390243902439,
      "grad_norm": 1.0392359495162964,
      "learning_rate": 7.074707959793534e-05,
      "loss": 1.6333,
      "step": 7430
    },
    {
      "epoch": 36.292682926829265,
      "grad_norm": 1.3190245628356934,
      "learning_rate": 7.069274653626731e-05,
      "loss": 1.6432,
      "step": 7440
    },
    {
      "epoch": 36.34146341463415,
      "grad_norm": 0.9551993608474731,
      "learning_rate": 7.06384134745993e-05,
      "loss": 1.6192,
      "step": 7450
    },
    {
      "epoch": 36.390243902439025,
      "grad_norm": 0.9931862354278564,
      "learning_rate": 7.058408041293128e-05,
      "loss": 1.6457,
      "step": 7460
    },
    {
      "epoch": 36.4390243902439,
      "grad_norm": 1.0751144886016846,
      "learning_rate": 7.052974735126325e-05,
      "loss": 1.6484,
      "step": 7470
    },
    {
      "epoch": 36.48780487804878,
      "grad_norm": 0.9190940260887146,
      "learning_rate": 7.047541428959522e-05,
      "loss": 1.6115,
      "step": 7480
    },
    {
      "epoch": 36.53658536585366,
      "grad_norm": 0.8716026544570923,
      "learning_rate": 7.04210812279272e-05,
      "loss": 1.6337,
      "step": 7490
    },
    {
      "epoch": 36.58536585365854,
      "grad_norm": 1.0738297700881958,
      "learning_rate": 7.036674816625917e-05,
      "loss": 1.6441,
      "step": 7500
    },
    {
      "epoch": 36.63414634146341,
      "grad_norm": 0.9632681608200073,
      "learning_rate": 7.031241510459115e-05,
      "loss": 1.639,
      "step": 7510
    },
    {
      "epoch": 36.68292682926829,
      "grad_norm": 1.0181065797805786,
      "learning_rate": 7.025808204292312e-05,
      "loss": 1.6603,
      "step": 7520
    },
    {
      "epoch": 36.73170731707317,
      "grad_norm": 0.9177279472351074,
      "learning_rate": 7.02037489812551e-05,
      "loss": 1.6373,
      "step": 7530
    },
    {
      "epoch": 36.78048780487805,
      "grad_norm": 0.9398072361946106,
      "learning_rate": 7.014941591958707e-05,
      "loss": 1.646,
      "step": 7540
    },
    {
      "epoch": 36.829268292682926,
      "grad_norm": 0.8873163461685181,
      "learning_rate": 7.009508285791905e-05,
      "loss": 1.6427,
      "step": 7550
    },
    {
      "epoch": 36.8780487804878,
      "grad_norm": 1.026986837387085,
      "learning_rate": 7.004074979625102e-05,
      "loss": 1.6429,
      "step": 7560
    },
    {
      "epoch": 36.926829268292686,
      "grad_norm": 0.9548583626747131,
      "learning_rate": 6.998641673458299e-05,
      "loss": 1.6611,
      "step": 7570
    },
    {
      "epoch": 36.97560975609756,
      "grad_norm": 0.9584435224533081,
      "learning_rate": 6.993208367291497e-05,
      "loss": 1.649,
      "step": 7580
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.6581,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.046879768371582,
      "eval_rouge1": 0.6126,
      "eval_rouge2": 0.3496,
      "eval_rougeL": 0.5429,
      "eval_runtime": 26.6906,
      "eval_samples_per_second": 6.819,
      "eval_steps_per_second": 1.723,
      "step": 7585
    },
    {
      "epoch": 37.02439024390244,
      "grad_norm": 0.9119429588317871,
      "learning_rate": 6.987775061124695e-05,
      "loss": 1.6352,
      "step": 7590
    },
    {
      "epoch": 37.073170731707314,
      "grad_norm": 1.1885426044464111,
      "learning_rate": 6.982341754957892e-05,
      "loss": 1.6066,
      "step": 7600
    },
    {
      "epoch": 37.1219512195122,
      "grad_norm": 0.9667913317680359,
      "learning_rate": 6.976908448791089e-05,
      "loss": 1.6495,
      "step": 7610
    },
    {
      "epoch": 37.170731707317074,
      "grad_norm": 0.9809781312942505,
      "learning_rate": 6.971475142624287e-05,
      "loss": 1.6541,
      "step": 7620
    },
    {
      "epoch": 37.21951219512195,
      "grad_norm": 0.8620357513427734,
      "learning_rate": 6.966041836457484e-05,
      "loss": 1.6428,
      "step": 7630
    },
    {
      "epoch": 37.26829268292683,
      "grad_norm": 0.8469148278236389,
      "learning_rate": 6.960608530290683e-05,
      "loss": 1.5949,
      "step": 7640
    },
    {
      "epoch": 37.31707317073171,
      "grad_norm": 0.930397629737854,
      "learning_rate": 6.95517522412388e-05,
      "loss": 1.6256,
      "step": 7650
    },
    {
      "epoch": 37.36585365853659,
      "grad_norm": 0.9291563630104065,
      "learning_rate": 6.949741917957078e-05,
      "loss": 1.6236,
      "step": 7660
    },
    {
      "epoch": 37.41463414634146,
      "grad_norm": 0.9306030869483948,
      "learning_rate": 6.944308611790275e-05,
      "loss": 1.6182,
      "step": 7670
    },
    {
      "epoch": 37.46341463414634,
      "grad_norm": 0.8049061298370361,
      "learning_rate": 6.938875305623473e-05,
      "loss": 1.638,
      "step": 7680
    },
    {
      "epoch": 37.51219512195122,
      "grad_norm": 0.8694387674331665,
      "learning_rate": 6.93344199945667e-05,
      "loss": 1.6102,
      "step": 7690
    },
    {
      "epoch": 37.5609756097561,
      "grad_norm": 1.0936464071273804,
      "learning_rate": 6.928008693289867e-05,
      "loss": 1.6482,
      "step": 7700
    },
    {
      "epoch": 37.609756097560975,
      "grad_norm": 0.9273491501808167,
      "learning_rate": 6.922575387123065e-05,
      "loss": 1.6184,
      "step": 7710
    },
    {
      "epoch": 37.65853658536585,
      "grad_norm": 0.8881824016571045,
      "learning_rate": 6.917142080956263e-05,
      "loss": 1.633,
      "step": 7720
    },
    {
      "epoch": 37.707317073170735,
      "grad_norm": 1.1889209747314453,
      "learning_rate": 6.91170877478946e-05,
      "loss": 1.6263,
      "step": 7730
    },
    {
      "epoch": 37.75609756097561,
      "grad_norm": 0.9249302744865417,
      "learning_rate": 6.906275468622657e-05,
      "loss": 1.5986,
      "step": 7740
    },
    {
      "epoch": 37.80487804878049,
      "grad_norm": 0.8693262338638306,
      "learning_rate": 6.900842162455855e-05,
      "loss": 1.6218,
      "step": 7750
    },
    {
      "epoch": 37.853658536585364,
      "grad_norm": 0.9223681688308716,
      "learning_rate": 6.895408856289052e-05,
      "loss": 1.6129,
      "step": 7760
    },
    {
      "epoch": 37.90243902439025,
      "grad_norm": 0.9245777130126953,
      "learning_rate": 6.88997555012225e-05,
      "loss": 1.6467,
      "step": 7770
    },
    {
      "epoch": 37.951219512195124,
      "grad_norm": 1.1191091537475586,
      "learning_rate": 6.884542243955447e-05,
      "loss": 1.6185,
      "step": 7780
    },
    {
      "epoch": 38.0,
      "grad_norm": 1.2376272678375244,
      "learning_rate": 6.879108937788644e-05,
      "loss": 1.6635,
      "step": 7790
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6581,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.046464681625366,
      "eval_rouge1": 0.6116,
      "eval_rouge2": 0.3489,
      "eval_rougeL": 0.5422,
      "eval_runtime": 26.6682,
      "eval_samples_per_second": 6.825,
      "eval_steps_per_second": 1.725,
      "step": 7790
    },
    {
      "epoch": 38.048780487804876,
      "grad_norm": 1.0961906909942627,
      "learning_rate": 6.873675631621842e-05,
      "loss": 1.6007,
      "step": 7800
    },
    {
      "epoch": 38.09756097560975,
      "grad_norm": 0.9302559494972229,
      "learning_rate": 6.86824232545504e-05,
      "loss": 1.5941,
      "step": 7810
    },
    {
      "epoch": 38.146341463414636,
      "grad_norm": 0.8833104968070984,
      "learning_rate": 6.862809019288237e-05,
      "loss": 1.587,
      "step": 7820
    },
    {
      "epoch": 38.19512195121951,
      "grad_norm": 0.9067671895027161,
      "learning_rate": 6.857375713121434e-05,
      "loss": 1.5975,
      "step": 7830
    },
    {
      "epoch": 38.24390243902439,
      "grad_norm": 1.0321314334869385,
      "learning_rate": 6.851942406954632e-05,
      "loss": 1.6184,
      "step": 7840
    },
    {
      "epoch": 38.292682926829265,
      "grad_norm": 1.0822004079818726,
      "learning_rate": 6.84650910078783e-05,
      "loss": 1.608,
      "step": 7850
    },
    {
      "epoch": 38.34146341463415,
      "grad_norm": 1.154619812965393,
      "learning_rate": 6.841075794621027e-05,
      "loss": 1.6111,
      "step": 7860
    },
    {
      "epoch": 38.390243902439025,
      "grad_norm": 0.8337762951850891,
      "learning_rate": 6.835642488454224e-05,
      "loss": 1.6151,
      "step": 7870
    },
    {
      "epoch": 38.4390243902439,
      "grad_norm": 16.136899948120117,
      "learning_rate": 6.830209182287423e-05,
      "loss": 1.638,
      "step": 7880
    },
    {
      "epoch": 38.48780487804878,
      "grad_norm": 1.3192740678787231,
      "learning_rate": 6.82477587612062e-05,
      "loss": 1.6336,
      "step": 7890
    },
    {
      "epoch": 38.53658536585366,
      "grad_norm": 0.9884423613548279,
      "learning_rate": 6.819342569953818e-05,
      "loss": 1.5949,
      "step": 7900
    },
    {
      "epoch": 38.58536585365854,
      "grad_norm": 0.9335576891899109,
      "learning_rate": 6.813909263787015e-05,
      "loss": 1.627,
      "step": 7910
    },
    {
      "epoch": 38.63414634146341,
      "grad_norm": 0.7936056852340698,
      "learning_rate": 6.808475957620211e-05,
      "loss": 1.6181,
      "step": 7920
    },
    {
      "epoch": 38.68292682926829,
      "grad_norm": 0.8893793225288391,
      "learning_rate": 6.80304265145341e-05,
      "loss": 1.6484,
      "step": 7930
    },
    {
      "epoch": 38.73170731707317,
      "grad_norm": 0.9549230933189392,
      "learning_rate": 6.797609345286608e-05,
      "loss": 1.6268,
      "step": 7940
    },
    {
      "epoch": 38.78048780487805,
      "grad_norm": 0.8394344449043274,
      "learning_rate": 6.792176039119805e-05,
      "loss": 1.6217,
      "step": 7950
    },
    {
      "epoch": 38.829268292682926,
      "grad_norm": 0.9312503337860107,
      "learning_rate": 6.786742732953002e-05,
      "loss": 1.6401,
      "step": 7960
    },
    {
      "epoch": 38.8780487804878,
      "grad_norm": 1.1489685773849487,
      "learning_rate": 6.7813094267862e-05,
      "loss": 1.6342,
      "step": 7970
    },
    {
      "epoch": 38.926829268292686,
      "grad_norm": 0.9154052138328552,
      "learning_rate": 6.775876120619398e-05,
      "loss": 1.6165,
      "step": 7980
    },
    {
      "epoch": 38.97560975609756,
      "grad_norm": 0.8178635239601135,
      "learning_rate": 6.770442814452595e-05,
      "loss": 1.6019,
      "step": 7990
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.6571,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.057765483856201,
      "eval_rouge1": 0.6095,
      "eval_rouge2": 0.3467,
      "eval_rougeL": 0.5406,
      "eval_runtime": 26.6397,
      "eval_samples_per_second": 6.832,
      "eval_steps_per_second": 1.727,
      "step": 7995
    },
    {
      "epoch": 39.02439024390244,
      "grad_norm": 1.0003865957260132,
      "learning_rate": 6.765009508285792e-05,
      "loss": 1.6608,
      "step": 8000
    },
    {
      "epoch": 39.073170731707314,
      "grad_norm": 0.8248067498207092,
      "learning_rate": 6.75957620211899e-05,
      "loss": 1.5807,
      "step": 8010
    },
    {
      "epoch": 39.1219512195122,
      "grad_norm": 0.9207936525344849,
      "learning_rate": 6.754142895952187e-05,
      "loss": 1.5651,
      "step": 8020
    },
    {
      "epoch": 39.170731707317074,
      "grad_norm": 0.9343249797821045,
      "learning_rate": 6.748709589785385e-05,
      "loss": 1.5866,
      "step": 8030
    },
    {
      "epoch": 39.21951219512195,
      "grad_norm": 0.9275141954421997,
      "learning_rate": 6.743276283618582e-05,
      "loss": 1.6036,
      "step": 8040
    },
    {
      "epoch": 39.26829268292683,
      "grad_norm": 0.859588086605072,
      "learning_rate": 6.737842977451779e-05,
      "loss": 1.591,
      "step": 8050
    },
    {
      "epoch": 39.31707317073171,
      "grad_norm": 1.0168163776397705,
      "learning_rate": 6.732409671284977e-05,
      "loss": 1.6141,
      "step": 8060
    },
    {
      "epoch": 39.36585365853659,
      "grad_norm": 0.9833230972290039,
      "learning_rate": 6.726976365118176e-05,
      "loss": 1.5999,
      "step": 8070
    },
    {
      "epoch": 39.41463414634146,
      "grad_norm": 0.9076850414276123,
      "learning_rate": 6.721543058951371e-05,
      "loss": 1.5879,
      "step": 8080
    },
    {
      "epoch": 39.46341463414634,
      "grad_norm": 0.9242377877235413,
      "learning_rate": 6.716109752784569e-05,
      "loss": 1.6272,
      "step": 8090
    },
    {
      "epoch": 39.51219512195122,
      "grad_norm": 1.1659233570098877,
      "learning_rate": 6.710676446617768e-05,
      "loss": 1.6131,
      "step": 8100
    },
    {
      "epoch": 39.5609756097561,
      "grad_norm": 1.0663799047470093,
      "learning_rate": 6.705243140450966e-05,
      "loss": 1.626,
      "step": 8110
    },
    {
      "epoch": 39.609756097560975,
      "grad_norm": 1.089972734451294,
      "learning_rate": 6.699809834284163e-05,
      "loss": 1.609,
      "step": 8120
    },
    {
      "epoch": 39.65853658536585,
      "grad_norm": 0.8596948981285095,
      "learning_rate": 6.69437652811736e-05,
      "loss": 1.6103,
      "step": 8130
    },
    {
      "epoch": 39.707317073170735,
      "grad_norm": 0.9379211664199829,
      "learning_rate": 6.688943221950558e-05,
      "loss": 1.6289,
      "step": 8140
    },
    {
      "epoch": 39.75609756097561,
      "grad_norm": 0.9608719944953918,
      "learning_rate": 6.683509915783755e-05,
      "loss": 1.5971,
      "step": 8150
    },
    {
      "epoch": 39.80487804878049,
      "grad_norm": 0.9829626083374023,
      "learning_rate": 6.678076609616953e-05,
      "loss": 1.6082,
      "step": 8160
    },
    {
      "epoch": 39.853658536585364,
      "grad_norm": 1.1987886428833008,
      "learning_rate": 6.67264330345015e-05,
      "loss": 1.5899,
      "step": 8170
    },
    {
      "epoch": 39.90243902439025,
      "grad_norm": 0.9772175550460815,
      "learning_rate": 6.667209997283347e-05,
      "loss": 1.6133,
      "step": 8180
    },
    {
      "epoch": 39.951219512195124,
      "grad_norm": 0.9270687103271484,
      "learning_rate": 6.661776691116545e-05,
      "loss": 1.5937,
      "step": 8190
    },
    {
      "epoch": 40.0,
      "grad_norm": 1.2685972452163696,
      "learning_rate": 6.656343384949743e-05,
      "loss": 1.5987,
      "step": 8200
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6576,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0566084384918213,
      "eval_rouge1": 0.6098,
      "eval_rouge2": 0.3483,
      "eval_rougeL": 0.5419,
      "eval_runtime": 26.6474,
      "eval_samples_per_second": 6.83,
      "eval_steps_per_second": 1.726,
      "step": 8200
    },
    {
      "epoch": 40.048780487804876,
      "grad_norm": 0.9441468119621277,
      "learning_rate": 6.650910078782939e-05,
      "loss": 1.5862,
      "step": 8210
    },
    {
      "epoch": 40.09756097560975,
      "grad_norm": 0.9096714854240417,
      "learning_rate": 6.645476772616137e-05,
      "loss": 1.552,
      "step": 8220
    },
    {
      "epoch": 40.146341463414636,
      "grad_norm": 0.980735719203949,
      "learning_rate": 6.640043466449335e-05,
      "loss": 1.5896,
      "step": 8230
    },
    {
      "epoch": 40.19512195121951,
      "grad_norm": 1.0142563581466675,
      "learning_rate": 6.634610160282532e-05,
      "loss": 1.6244,
      "step": 8240
    },
    {
      "epoch": 40.24390243902439,
      "grad_norm": 0.9127105474472046,
      "learning_rate": 6.62917685411573e-05,
      "loss": 1.5819,
      "step": 8250
    },
    {
      "epoch": 40.292682926829265,
      "grad_norm": 0.9310650825500488,
      "learning_rate": 6.623743547948927e-05,
      "loss": 1.6063,
      "step": 8260
    },
    {
      "epoch": 40.34146341463415,
      "grad_norm": 1.0724411010742188,
      "learning_rate": 6.618310241782125e-05,
      "loss": 1.5862,
      "step": 8270
    },
    {
      "epoch": 40.390243902439025,
      "grad_norm": 0.8463386297225952,
      "learning_rate": 6.612876935615322e-05,
      "loss": 1.5425,
      "step": 8280
    },
    {
      "epoch": 40.4390243902439,
      "grad_norm": 0.9898768067359924,
      "learning_rate": 6.60744362944852e-05,
      "loss": 1.5725,
      "step": 8290
    },
    {
      "epoch": 40.48780487804878,
      "grad_norm": 0.9412344694137573,
      "learning_rate": 6.602010323281717e-05,
      "loss": 1.5738,
      "step": 8300
    },
    {
      "epoch": 40.53658536585366,
      "grad_norm": 1.1098183393478394,
      "learning_rate": 6.596577017114914e-05,
      "loss": 1.6052,
      "step": 8310
    },
    {
      "epoch": 40.58536585365854,
      "grad_norm": 1.029123067855835,
      "learning_rate": 6.591143710948112e-05,
      "loss": 1.6101,
      "step": 8320
    },
    {
      "epoch": 40.63414634146341,
      "grad_norm": 1.137770175933838,
      "learning_rate": 6.58571040478131e-05,
      "loss": 1.6024,
      "step": 8330
    },
    {
      "epoch": 40.68292682926829,
      "grad_norm": 1.009940505027771,
      "learning_rate": 6.580277098614506e-05,
      "loss": 1.5886,
      "step": 8340
    },
    {
      "epoch": 40.73170731707317,
      "grad_norm": 0.953379213809967,
      "learning_rate": 6.574843792447704e-05,
      "loss": 1.6252,
      "step": 8350
    },
    {
      "epoch": 40.78048780487805,
      "grad_norm": 0.9754422903060913,
      "learning_rate": 6.569410486280903e-05,
      "loss": 1.5888,
      "step": 8360
    },
    {
      "epoch": 40.829268292682926,
      "grad_norm": 1.1647788286209106,
      "learning_rate": 6.5639771801141e-05,
      "loss": 1.5829,
      "step": 8370
    },
    {
      "epoch": 40.8780487804878,
      "grad_norm": 1.0146311521530151,
      "learning_rate": 6.558543873947296e-05,
      "loss": 1.606,
      "step": 8380
    },
    {
      "epoch": 40.926829268292686,
      "grad_norm": 0.9312621355056763,
      "learning_rate": 6.553110567780495e-05,
      "loss": 1.5791,
      "step": 8390
    },
    {
      "epoch": 40.97560975609756,
      "grad_norm": 0.8976138830184937,
      "learning_rate": 6.547677261613693e-05,
      "loss": 1.5977,
      "step": 8400
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.6604,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.050206184387207,
      "eval_rouge1": 0.6125,
      "eval_rouge2": 0.3508,
      "eval_rougeL": 0.5448,
      "eval_runtime": 26.7993,
      "eval_samples_per_second": 6.791,
      "eval_steps_per_second": 1.716,
      "step": 8405
    },
    {
      "epoch": 41.02439024390244,
      "grad_norm": 0.9545244574546814,
      "learning_rate": 6.54224395544689e-05,
      "loss": 1.5558,
      "step": 8410
    },
    {
      "epoch": 41.073170731707314,
      "grad_norm": 0.8932915925979614,
      "learning_rate": 6.536810649280088e-05,
      "loss": 1.5799,
      "step": 8420
    },
    {
      "epoch": 41.1219512195122,
      "grad_norm": 0.8935472965240479,
      "learning_rate": 6.531377343113285e-05,
      "loss": 1.5513,
      "step": 8430
    },
    {
      "epoch": 41.170731707317074,
      "grad_norm": 0.9940924048423767,
      "learning_rate": 6.525944036946482e-05,
      "loss": 1.5354,
      "step": 8440
    },
    {
      "epoch": 41.21951219512195,
      "grad_norm": 0.9507209062576294,
      "learning_rate": 6.52051073077968e-05,
      "loss": 1.5698,
      "step": 8450
    },
    {
      "epoch": 41.26829268292683,
      "grad_norm": 0.9492302536964417,
      "learning_rate": 6.515077424612878e-05,
      "loss": 1.5804,
      "step": 8460
    },
    {
      "epoch": 41.31707317073171,
      "grad_norm": 0.840816855430603,
      "learning_rate": 6.509644118446074e-05,
      "loss": 1.5849,
      "step": 8470
    },
    {
      "epoch": 41.36585365853659,
      "grad_norm": 0.9821751713752747,
      "learning_rate": 6.504210812279272e-05,
      "loss": 1.5754,
      "step": 8480
    },
    {
      "epoch": 41.41463414634146,
      "grad_norm": 0.9300991892814636,
      "learning_rate": 6.49877750611247e-05,
      "loss": 1.5818,
      "step": 8490
    },
    {
      "epoch": 41.46341463414634,
      "grad_norm": 1.0535385608673096,
      "learning_rate": 6.493344199945667e-05,
      "loss": 1.5699,
      "step": 8500
    },
    {
      "epoch": 41.51219512195122,
      "grad_norm": 0.8814225196838379,
      "learning_rate": 6.487910893778864e-05,
      "loss": 1.5778,
      "step": 8510
    },
    {
      "epoch": 41.5609756097561,
      "grad_norm": 1.0759248733520508,
      "learning_rate": 6.482477587612062e-05,
      "loss": 1.5656,
      "step": 8520
    },
    {
      "epoch": 41.609756097560975,
      "grad_norm": 1.0118745565414429,
      "learning_rate": 6.477044281445259e-05,
      "loss": 1.5969,
      "step": 8530
    },
    {
      "epoch": 41.65853658536585,
      "grad_norm": 0.9958849549293518,
      "learning_rate": 6.471610975278457e-05,
      "loss": 1.5702,
      "step": 8540
    },
    {
      "epoch": 41.707317073170735,
      "grad_norm": 0.8765369057655334,
      "learning_rate": 6.466177669111654e-05,
      "loss": 1.5907,
      "step": 8550
    },
    {
      "epoch": 41.75609756097561,
      "grad_norm": 0.9264476299285889,
      "learning_rate": 6.460744362944852e-05,
      "loss": 1.5897,
      "step": 8560
    },
    {
      "epoch": 41.80487804878049,
      "grad_norm": 1.0101897716522217,
      "learning_rate": 6.45531105677805e-05,
      "loss": 1.5912,
      "step": 8570
    },
    {
      "epoch": 41.853658536585364,
      "grad_norm": 1.125571608543396,
      "learning_rate": 6.449877750611248e-05,
      "loss": 1.5849,
      "step": 8580
    },
    {
      "epoch": 41.90243902439025,
      "grad_norm": 0.9828618764877319,
      "learning_rate": 6.444444444444446e-05,
      "loss": 1.6028,
      "step": 8590
    },
    {
      "epoch": 41.951219512195124,
      "grad_norm": 1.0148931741714478,
      "learning_rate": 6.439011138277641e-05,
      "loss": 1.5818,
      "step": 8600
    },
    {
      "epoch": 42.0,
      "grad_norm": 1.1765875816345215,
      "learning_rate": 6.43357783211084e-05,
      "loss": 1.6187,
      "step": 8610
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.6561,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0694971084594727,
      "eval_rouge1": 0.61,
      "eval_rouge2": 0.3467,
      "eval_rougeL": 0.5411,
      "eval_runtime": 26.635,
      "eval_samples_per_second": 6.833,
      "eval_steps_per_second": 1.727,
      "step": 8610
    },
    {
      "epoch": 42.048780487804876,
      "grad_norm": 0.9942975640296936,
      "learning_rate": 6.428144525944038e-05,
      "loss": 1.5533,
      "step": 8620
    },
    {
      "epoch": 42.09756097560975,
      "grad_norm": 0.9553746581077576,
      "learning_rate": 6.422711219777235e-05,
      "loss": 1.5592,
      "step": 8630
    },
    {
      "epoch": 42.146341463414636,
      "grad_norm": 0.9394342303276062,
      "learning_rate": 6.417277913610432e-05,
      "loss": 1.5817,
      "step": 8640
    },
    {
      "epoch": 42.19512195121951,
      "grad_norm": 0.9168606400489807,
      "learning_rate": 6.41184460744363e-05,
      "loss": 1.5574,
      "step": 8650
    },
    {
      "epoch": 42.24390243902439,
      "grad_norm": 0.9688276052474976,
      "learning_rate": 6.406411301276827e-05,
      "loss": 1.5462,
      "step": 8660
    },
    {
      "epoch": 42.292682926829265,
      "grad_norm": 1.0756962299346924,
      "learning_rate": 6.400977995110025e-05,
      "loss": 1.5723,
      "step": 8670
    },
    {
      "epoch": 42.34146341463415,
      "grad_norm": 0.9371064901351929,
      "learning_rate": 6.395544688943222e-05,
      "loss": 1.5391,
      "step": 8680
    },
    {
      "epoch": 42.390243902439025,
      "grad_norm": 1.1569350957870483,
      "learning_rate": 6.39011138277642e-05,
      "loss": 1.5705,
      "step": 8690
    },
    {
      "epoch": 42.4390243902439,
      "grad_norm": 0.8656958341598511,
      "learning_rate": 6.384678076609617e-05,
      "loss": 1.5384,
      "step": 8700
    },
    {
      "epoch": 42.48780487804878,
      "grad_norm": 0.8878985643386841,
      "learning_rate": 6.379244770442815e-05,
      "loss": 1.5634,
      "step": 8710
    },
    {
      "epoch": 42.53658536585366,
      "grad_norm": 0.9635202288627625,
      "learning_rate": 6.373811464276012e-05,
      "loss": 1.5564,
      "step": 8720
    },
    {
      "epoch": 42.58536585365854,
      "grad_norm": 0.9100601077079773,
      "learning_rate": 6.368378158109209e-05,
      "loss": 1.5908,
      "step": 8730
    },
    {
      "epoch": 42.63414634146341,
      "grad_norm": 0.8707578778266907,
      "learning_rate": 6.362944851942407e-05,
      "loss": 1.5661,
      "step": 8740
    },
    {
      "epoch": 42.68292682926829,
      "grad_norm": 1.0029586553573608,
      "learning_rate": 6.357511545775605e-05,
      "loss": 1.5594,
      "step": 8750
    },
    {
      "epoch": 42.73170731707317,
      "grad_norm": 0.9894896745681763,
      "learning_rate": 6.352078239608802e-05,
      "loss": 1.5755,
      "step": 8760
    },
    {
      "epoch": 42.78048780487805,
      "grad_norm": 1.100540280342102,
      "learning_rate": 6.346644933441999e-05,
      "loss": 1.5841,
      "step": 8770
    },
    {
      "epoch": 42.829268292682926,
      "grad_norm": 0.9192934632301331,
      "learning_rate": 6.341211627275197e-05,
      "loss": 1.5381,
      "step": 8780
    },
    {
      "epoch": 42.8780487804878,
      "grad_norm": 0.964634120464325,
      "learning_rate": 6.335778321108394e-05,
      "loss": 1.5509,
      "step": 8790
    },
    {
      "epoch": 42.926829268292686,
      "grad_norm": 0.9472265839576721,
      "learning_rate": 6.330345014941593e-05,
      "loss": 1.5761,
      "step": 8800
    },
    {
      "epoch": 42.97560975609756,
      "grad_norm": 1.0623375177383423,
      "learning_rate": 6.32491170877479e-05,
      "loss": 1.564,
      "step": 8810
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.6605,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0630738735198975,
      "eval_rouge1": 0.6122,
      "eval_rouge2": 0.3516,
      "eval_rougeL": 0.545,
      "eval_runtime": 26.7171,
      "eval_samples_per_second": 6.812,
      "eval_steps_per_second": 1.722,
      "step": 8815
    },
    {
      "epoch": 43.02439024390244,
      "grad_norm": 0.9268796443939209,
      "learning_rate": 6.319478402607986e-05,
      "loss": 1.5838,
      "step": 8820
    },
    {
      "epoch": 43.073170731707314,
      "grad_norm": 0.9908514022827148,
      "learning_rate": 6.314045096441185e-05,
      "loss": 1.5476,
      "step": 8830
    },
    {
      "epoch": 43.1219512195122,
      "grad_norm": 0.9008921980857849,
      "learning_rate": 6.308611790274383e-05,
      "loss": 1.5401,
      "step": 8840
    },
    {
      "epoch": 43.170731707317074,
      "grad_norm": 0.9314485788345337,
      "learning_rate": 6.30317848410758e-05,
      "loss": 1.5596,
      "step": 8850
    },
    {
      "epoch": 43.21951219512195,
      "grad_norm": 1.072072982788086,
      "learning_rate": 6.297745177940777e-05,
      "loss": 1.5731,
      "step": 8860
    },
    {
      "epoch": 43.26829268292683,
      "grad_norm": 0.9419031739234924,
      "learning_rate": 6.292311871773975e-05,
      "loss": 1.5651,
      "step": 8870
    },
    {
      "epoch": 43.31707317073171,
      "grad_norm": 0.9719817042350769,
      "learning_rate": 6.286878565607173e-05,
      "loss": 1.5481,
      "step": 8880
    },
    {
      "epoch": 43.36585365853659,
      "grad_norm": 0.9506922960281372,
      "learning_rate": 6.28144525944037e-05,
      "loss": 1.5288,
      "step": 8890
    },
    {
      "epoch": 43.41463414634146,
      "grad_norm": 0.8473609089851379,
      "learning_rate": 6.276011953273567e-05,
      "loss": 1.5502,
      "step": 8900
    },
    {
      "epoch": 43.46341463414634,
      "grad_norm": 0.8621916770935059,
      "learning_rate": 6.270578647106765e-05,
      "loss": 1.532,
      "step": 8910
    },
    {
      "epoch": 43.51219512195122,
      "grad_norm": 0.94191974401474,
      "learning_rate": 6.265145340939962e-05,
      "loss": 1.5396,
      "step": 8920
    },
    {
      "epoch": 43.5609756097561,
      "grad_norm": 0.983315110206604,
      "learning_rate": 6.25971203477316e-05,
      "loss": 1.602,
      "step": 8930
    },
    {
      "epoch": 43.609756097560975,
      "grad_norm": 0.9320430755615234,
      "learning_rate": 6.254278728606357e-05,
      "loss": 1.5246,
      "step": 8940
    },
    {
      "epoch": 43.65853658536585,
      "grad_norm": 0.859078049659729,
      "learning_rate": 6.248845422439554e-05,
      "loss": 1.5346,
      "step": 8950
    },
    {
      "epoch": 43.707317073170735,
      "grad_norm": 1.0229722261428833,
      "learning_rate": 6.243412116272752e-05,
      "loss": 1.5397,
      "step": 8960
    },
    {
      "epoch": 43.75609756097561,
      "grad_norm": 0.9220072031021118,
      "learning_rate": 6.23797881010595e-05,
      "loss": 1.5728,
      "step": 8970
    },
    {
      "epoch": 43.80487804878049,
      "grad_norm": 0.9612592458724976,
      "learning_rate": 6.232545503939147e-05,
      "loss": 1.5375,
      "step": 8980
    },
    {
      "epoch": 43.853658536585364,
      "grad_norm": 0.9555723071098328,
      "learning_rate": 6.227112197772344e-05,
      "loss": 1.5689,
      "step": 8990
    },
    {
      "epoch": 43.90243902439025,
      "grad_norm": 0.9540090560913086,
      "learning_rate": 6.221678891605542e-05,
      "loss": 1.5672,
      "step": 9000
    },
    {
      "epoch": 43.951219512195124,
      "grad_norm": 1.0556234121322632,
      "learning_rate": 6.21624558543874e-05,
      "loss": 1.555,
      "step": 9010
    },
    {
      "epoch": 44.0,
      "grad_norm": 1.2409921884536743,
      "learning_rate": 6.210812279271937e-05,
      "loss": 1.5469,
      "step": 9020
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.6587,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0725717544555664,
      "eval_rouge1": 0.6119,
      "eval_rouge2": 0.3496,
      "eval_rougeL": 0.5434,
      "eval_runtime": 26.7294,
      "eval_samples_per_second": 6.809,
      "eval_steps_per_second": 1.721,
      "step": 9020
    },
    {
      "epoch": 44.048780487804876,
      "grad_norm": 0.9860386252403259,
      "learning_rate": 6.205378973105134e-05,
      "loss": 1.5527,
      "step": 9030
    },
    {
      "epoch": 44.09756097560975,
      "grad_norm": 0.9774211049079895,
      "learning_rate": 6.199945666938333e-05,
      "loss": 1.5325,
      "step": 9040
    },
    {
      "epoch": 44.146341463414636,
      "grad_norm": 1.153135895729065,
      "learning_rate": 6.19451236077153e-05,
      "loss": 1.5533,
      "step": 9050
    },
    {
      "epoch": 44.19512195121951,
      "grad_norm": 0.8926750421524048,
      "learning_rate": 6.189079054604728e-05,
      "loss": 1.534,
      "step": 9060
    },
    {
      "epoch": 44.24390243902439,
      "grad_norm": 0.9813703894615173,
      "learning_rate": 6.183645748437925e-05,
      "loss": 1.555,
      "step": 9070
    },
    {
      "epoch": 44.292682926829265,
      "grad_norm": 0.9816197156906128,
      "learning_rate": 6.178212442271121e-05,
      "loss": 1.5252,
      "step": 9080
    },
    {
      "epoch": 44.34146341463415,
      "grad_norm": 0.9646711349487305,
      "learning_rate": 6.17277913610432e-05,
      "loss": 1.5246,
      "step": 9090
    },
    {
      "epoch": 44.390243902439025,
      "grad_norm": 0.9579485654830933,
      "learning_rate": 6.167345829937518e-05,
      "loss": 1.5301,
      "step": 9100
    },
    {
      "epoch": 44.4390243902439,
      "grad_norm": 0.9191059470176697,
      "learning_rate": 6.161912523770715e-05,
      "loss": 1.5144,
      "step": 9110
    },
    {
      "epoch": 44.48780487804878,
      "grad_norm": 0.9516372680664062,
      "learning_rate": 6.156479217603912e-05,
      "loss": 1.5256,
      "step": 9120
    },
    {
      "epoch": 44.53658536585366,
      "grad_norm": 0.9884464740753174,
      "learning_rate": 6.15104591143711e-05,
      "loss": 1.5442,
      "step": 9130
    },
    {
      "epoch": 44.58536585365854,
      "grad_norm": 0.9394478797912598,
      "learning_rate": 6.145612605270308e-05,
      "loss": 1.5154,
      "step": 9140
    },
    {
      "epoch": 44.63414634146341,
      "grad_norm": 1.0759013891220093,
      "learning_rate": 6.140179299103505e-05,
      "loss": 1.5659,
      "step": 9150
    },
    {
      "epoch": 44.68292682926829,
      "grad_norm": 1.0106854438781738,
      "learning_rate": 6.134745992936702e-05,
      "loss": 1.5532,
      "step": 9160
    },
    {
      "epoch": 44.73170731707317,
      "grad_norm": 0.9932792782783508,
      "learning_rate": 6.1293126867699e-05,
      "loss": 1.5544,
      "step": 9170
    },
    {
      "epoch": 44.78048780487805,
      "grad_norm": 0.9034538865089417,
      "learning_rate": 6.123879380603097e-05,
      "loss": 1.5535,
      "step": 9180
    },
    {
      "epoch": 44.829268292682926,
      "grad_norm": 0.8972671031951904,
      "learning_rate": 6.118446074436295e-05,
      "loss": 1.503,
      "step": 9190
    },
    {
      "epoch": 44.8780487804878,
      "grad_norm": 1.041559100151062,
      "learning_rate": 6.113012768269492e-05,
      "loss": 1.5716,
      "step": 9200
    },
    {
      "epoch": 44.926829268292686,
      "grad_norm": 1.0077598094940186,
      "learning_rate": 6.107579462102689e-05,
      "loss": 1.5548,
      "step": 9210
    },
    {
      "epoch": 44.97560975609756,
      "grad_norm": 0.9414793848991394,
      "learning_rate": 6.102146155935887e-05,
      "loss": 1.5317,
      "step": 9220
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.6567,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.076894521713257,
      "eval_rouge1": 0.6108,
      "eval_rouge2": 0.3478,
      "eval_rougeL": 0.5417,
      "eval_runtime": 26.7192,
      "eval_samples_per_second": 6.812,
      "eval_steps_per_second": 1.722,
      "step": 9225
    },
    {
      "epoch": 45.02439024390244,
      "grad_norm": 1.0775814056396484,
      "learning_rate": 6.096712849769085e-05,
      "loss": 1.5471,
      "step": 9230
    },
    {
      "epoch": 45.073170731707314,
      "grad_norm": 0.9766963124275208,
      "learning_rate": 6.091279543602282e-05,
      "loss": 1.4864,
      "step": 9240
    },
    {
      "epoch": 45.1219512195122,
      "grad_norm": 0.9466918706893921,
      "learning_rate": 6.08584623743548e-05,
      "loss": 1.5288,
      "step": 9250
    },
    {
      "epoch": 45.170731707317074,
      "grad_norm": 1.0558148622512817,
      "learning_rate": 6.0804129312686775e-05,
      "loss": 1.5422,
      "step": 9260
    },
    {
      "epoch": 45.21951219512195,
      "grad_norm": 1.0389299392700195,
      "learning_rate": 6.0749796251018744e-05,
      "loss": 1.5205,
      "step": 9270
    },
    {
      "epoch": 45.26829268292683,
      "grad_norm": 0.779005765914917,
      "learning_rate": 6.069546318935072e-05,
      "loss": 1.5206,
      "step": 9280
    },
    {
      "epoch": 45.31707317073171,
      "grad_norm": 0.9916539192199707,
      "learning_rate": 6.06411301276827e-05,
      "loss": 1.5342,
      "step": 9290
    },
    {
      "epoch": 45.36585365853659,
      "grad_norm": 0.9707236886024475,
      "learning_rate": 6.058679706601468e-05,
      "loss": 1.516,
      "step": 9300
    },
    {
      "epoch": 45.41463414634146,
      "grad_norm": 1.099131464958191,
      "learning_rate": 6.0532464004346646e-05,
      "loss": 1.5047,
      "step": 9310
    },
    {
      "epoch": 45.46341463414634,
      "grad_norm": 0.967131495475769,
      "learning_rate": 6.047813094267862e-05,
      "loss": 1.5228,
      "step": 9320
    },
    {
      "epoch": 45.51219512195122,
      "grad_norm": 1.0093587636947632,
      "learning_rate": 6.0423797881010604e-05,
      "loss": 1.5409,
      "step": 9330
    },
    {
      "epoch": 45.5609756097561,
      "grad_norm": 0.8259481191635132,
      "learning_rate": 6.0369464819342566e-05,
      "loss": 1.5176,
      "step": 9340
    },
    {
      "epoch": 45.609756097560975,
      "grad_norm": 0.9468622803688049,
      "learning_rate": 6.031513175767455e-05,
      "loss": 1.5416,
      "step": 9350
    },
    {
      "epoch": 45.65853658536585,
      "grad_norm": 0.9278303980827332,
      "learning_rate": 6.0260798696006524e-05,
      "loss": 1.5471,
      "step": 9360
    },
    {
      "epoch": 45.707317073170735,
      "grad_norm": 0.8681600689888,
      "learning_rate": 6.020646563433849e-05,
      "loss": 1.5393,
      "step": 9370
    },
    {
      "epoch": 45.75609756097561,
      "grad_norm": 0.863831639289856,
      "learning_rate": 6.0152132572670475e-05,
      "loss": 1.5301,
      "step": 9380
    },
    {
      "epoch": 45.80487804878049,
      "grad_norm": 0.9817381501197815,
      "learning_rate": 6.009779951100245e-05,
      "loss": 1.5353,
      "step": 9390
    },
    {
      "epoch": 45.853658536585364,
      "grad_norm": 0.9569991827011108,
      "learning_rate": 6.004346644933442e-05,
      "loss": 1.5303,
      "step": 9400
    },
    {
      "epoch": 45.90243902439025,
      "grad_norm": 0.8829732537269592,
      "learning_rate": 5.9989133387666395e-05,
      "loss": 1.544,
      "step": 9410
    },
    {
      "epoch": 45.951219512195124,
      "grad_norm": 0.989835798740387,
      "learning_rate": 5.993480032599838e-05,
      "loss": 1.5373,
      "step": 9420
    },
    {
      "epoch": 46.0,
      "grad_norm": 1.108426570892334,
      "learning_rate": 5.988046726433035e-05,
      "loss": 1.51,
      "step": 9430
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.6558,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.081815004348755,
      "eval_rouge1": 0.6101,
      "eval_rouge2": 0.347,
      "eval_rougeL": 0.5414,
      "eval_runtime": 26.6895,
      "eval_samples_per_second": 6.819,
      "eval_steps_per_second": 1.724,
      "step": 9430
    },
    {
      "epoch": 46.048780487804876,
      "grad_norm": 0.9329161643981934,
      "learning_rate": 5.982613420266232e-05,
      "loss": 1.5037,
      "step": 9440
    },
    {
      "epoch": 46.09756097560975,
      "grad_norm": 0.9613210558891296,
      "learning_rate": 5.97718011409943e-05,
      "loss": 1.5273,
      "step": 9450
    },
    {
      "epoch": 46.146341463414636,
      "grad_norm": 1.005125880241394,
      "learning_rate": 5.971746807932628e-05,
      "loss": 1.512,
      "step": 9460
    },
    {
      "epoch": 46.19512195121951,
      "grad_norm": 0.9762298464775085,
      "learning_rate": 5.966313501765824e-05,
      "loss": 1.4876,
      "step": 9470
    },
    {
      "epoch": 46.24390243902439,
      "grad_norm": 1.0202457904815674,
      "learning_rate": 5.9608801955990224e-05,
      "loss": 1.5143,
      "step": 9480
    },
    {
      "epoch": 46.292682926829265,
      "grad_norm": 0.8822451829910278,
      "learning_rate": 5.95544688943222e-05,
      "loss": 1.5268,
      "step": 9490
    },
    {
      "epoch": 46.34146341463415,
      "grad_norm": 0.9687830805778503,
      "learning_rate": 5.950013583265417e-05,
      "loss": 1.5161,
      "step": 9500
    },
    {
      "epoch": 46.390243902439025,
      "grad_norm": 0.8915550708770752,
      "learning_rate": 5.9445802770986144e-05,
      "loss": 1.5014,
      "step": 9510
    },
    {
      "epoch": 46.4390243902439,
      "grad_norm": 0.9499079585075378,
      "learning_rate": 5.939146970931813e-05,
      "loss": 1.5192,
      "step": 9520
    },
    {
      "epoch": 46.48780487804878,
      "grad_norm": 0.9280877113342285,
      "learning_rate": 5.9337136647650096e-05,
      "loss": 1.5136,
      "step": 9530
    },
    {
      "epoch": 46.53658536585366,
      "grad_norm": 0.9028041958808899,
      "learning_rate": 5.928280358598207e-05,
      "loss": 1.5314,
      "step": 9540
    },
    {
      "epoch": 46.58536585365854,
      "grad_norm": 1.0268795490264893,
      "learning_rate": 5.9228470524314053e-05,
      "loss": 1.5047,
      "step": 9550
    },
    {
      "epoch": 46.63414634146341,
      "grad_norm": 1.0093835592269897,
      "learning_rate": 5.9174137462646015e-05,
      "loss": 1.5284,
      "step": 9560
    },
    {
      "epoch": 46.68292682926829,
      "grad_norm": 1.0665066242218018,
      "learning_rate": 5.9119804400978e-05,
      "loss": 1.5319,
      "step": 9570
    },
    {
      "epoch": 46.73170731707317,
      "grad_norm": 0.9907817244529724,
      "learning_rate": 5.9065471339309973e-05,
      "loss": 1.5348,
      "step": 9580
    },
    {
      "epoch": 46.78048780487805,
      "grad_norm": 0.8871275186538696,
      "learning_rate": 5.9011138277641956e-05,
      "loss": 1.5202,
      "step": 9590
    },
    {
      "epoch": 46.829268292682926,
      "grad_norm": 0.8575066328048706,
      "learning_rate": 5.895680521597392e-05,
      "loss": 1.5259,
      "step": 9600
    },
    {
      "epoch": 46.8780487804878,
      "grad_norm": 0.9315159320831299,
      "learning_rate": 5.89024721543059e-05,
      "loss": 1.5551,
      "step": 9610
    },
    {
      "epoch": 46.926829268292686,
      "grad_norm": 0.9398794770240784,
      "learning_rate": 5.8848139092637876e-05,
      "loss": 1.4994,
      "step": 9620
    },
    {
      "epoch": 46.97560975609756,
      "grad_norm": 1.1729868650436401,
      "learning_rate": 5.8793806030969845e-05,
      "loss": 1.5174,
      "step": 9630
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.6592,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.078704595565796,
      "eval_rouge1": 0.6124,
      "eval_rouge2": 0.3514,
      "eval_rougeL": 0.5439,
      "eval_runtime": 26.7177,
      "eval_samples_per_second": 6.812,
      "eval_steps_per_second": 1.722,
      "step": 9635
    },
    {
      "epoch": 47.02439024390244,
      "grad_norm": 0.9085572957992554,
      "learning_rate": 5.873947296930182e-05,
      "loss": 1.5013,
      "step": 9640
    },
    {
      "epoch": 47.073170731707314,
      "grad_norm": 1.0357688665390015,
      "learning_rate": 5.86851399076338e-05,
      "loss": 1.4971,
      "step": 9650
    },
    {
      "epoch": 47.1219512195122,
      "grad_norm": 0.9050779342651367,
      "learning_rate": 5.863080684596577e-05,
      "loss": 1.5163,
      "step": 9660
    },
    {
      "epoch": 47.170731707317074,
      "grad_norm": 0.9679107666015625,
      "learning_rate": 5.857647378429775e-05,
      "loss": 1.4896,
      "step": 9670
    },
    {
      "epoch": 47.21951219512195,
      "grad_norm": 0.9459844827651978,
      "learning_rate": 5.852214072262972e-05,
      "loss": 1.51,
      "step": 9680
    },
    {
      "epoch": 47.26829268292683,
      "grad_norm": 0.8712072372436523,
      "learning_rate": 5.846780766096169e-05,
      "loss": 1.5141,
      "step": 9690
    },
    {
      "epoch": 47.31707317073171,
      "grad_norm": 1.0053471326828003,
      "learning_rate": 5.8413474599293674e-05,
      "loss": 1.5439,
      "step": 9700
    },
    {
      "epoch": 47.36585365853659,
      "grad_norm": 1.0147236585617065,
      "learning_rate": 5.835914153762565e-05,
      "loss": 1.5161,
      "step": 9710
    },
    {
      "epoch": 47.41463414634146,
      "grad_norm": 0.9281237125396729,
      "learning_rate": 5.830480847595763e-05,
      "loss": 1.4859,
      "step": 9720
    },
    {
      "epoch": 47.46341463414634,
      "grad_norm": 0.9387296438217163,
      "learning_rate": 5.8250475414289594e-05,
      "loss": 1.5253,
      "step": 9730
    },
    {
      "epoch": 47.51219512195122,
      "grad_norm": 1.0199874639511108,
      "learning_rate": 5.8196142352621576e-05,
      "loss": 1.5278,
      "step": 9740
    },
    {
      "epoch": 47.5609756097561,
      "grad_norm": 0.9802501797676086,
      "learning_rate": 5.814180929095355e-05,
      "loss": 1.5018,
      "step": 9750
    },
    {
      "epoch": 47.609756097560975,
      "grad_norm": 0.8037899732589722,
      "learning_rate": 5.808747622928552e-05,
      "loss": 1.5121,
      "step": 9760
    },
    {
      "epoch": 47.65853658536585,
      "grad_norm": 1.0279395580291748,
      "learning_rate": 5.8033143167617496e-05,
      "loss": 1.4819,
      "step": 9770
    },
    {
      "epoch": 47.707317073170735,
      "grad_norm": 1.021676778793335,
      "learning_rate": 5.797881010594948e-05,
      "loss": 1.5011,
      "step": 9780
    },
    {
      "epoch": 47.75609756097561,
      "grad_norm": 1.1812975406646729,
      "learning_rate": 5.792447704428145e-05,
      "loss": 1.4819,
      "step": 9790
    },
    {
      "epoch": 47.80487804878049,
      "grad_norm": 0.9932900071144104,
      "learning_rate": 5.787014398261342e-05,
      "loss": 1.5001,
      "step": 9800
    },
    {
      "epoch": 47.853658536585364,
      "grad_norm": 1.0043509006500244,
      "learning_rate": 5.78158109209454e-05,
      "loss": 1.5159,
      "step": 9810
    },
    {
      "epoch": 47.90243902439025,
      "grad_norm": 0.9138625860214233,
      "learning_rate": 5.776147785927737e-05,
      "loss": 1.5267,
      "step": 9820
    },
    {
      "epoch": 47.951219512195124,
      "grad_norm": 0.9451781511306763,
      "learning_rate": 5.770714479760935e-05,
      "loss": 1.5051,
      "step": 9830
    },
    {
      "epoch": 48.0,
      "grad_norm": 1.0806002616882324,
      "learning_rate": 5.7652811735941325e-05,
      "loss": 1.5357,
      "step": 9840
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.6595,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.079742908477783,
      "eval_rouge1": 0.6128,
      "eval_rouge2": 0.351,
      "eval_rougeL": 0.5445,
      "eval_runtime": 26.732,
      "eval_samples_per_second": 6.808,
      "eval_steps_per_second": 1.721,
      "step": 9840
    },
    {
      "epoch": 48.048780487804876,
      "grad_norm": 0.8942461609840393,
      "learning_rate": 5.7598478674273294e-05,
      "loss": 1.5143,
      "step": 9850
    },
    {
      "epoch": 48.09756097560975,
      "grad_norm": 0.9439877867698669,
      "learning_rate": 5.754414561260527e-05,
      "loss": 1.4611,
      "step": 9860
    },
    {
      "epoch": 48.146341463414636,
      "grad_norm": 0.9229232668876648,
      "learning_rate": 5.748981255093725e-05,
      "loss": 1.4988,
      "step": 9870
    },
    {
      "epoch": 48.19512195121951,
      "grad_norm": 0.9288315176963806,
      "learning_rate": 5.743547948926923e-05,
      "loss": 1.4753,
      "step": 9880
    },
    {
      "epoch": 48.24390243902439,
      "grad_norm": 1.0723158121109009,
      "learning_rate": 5.7381146427601196e-05,
      "loss": 1.5116,
      "step": 9890
    },
    {
      "epoch": 48.292682926829265,
      "grad_norm": 0.9865169525146484,
      "learning_rate": 5.732681336593317e-05,
      "loss": 1.4924,
      "step": 9900
    },
    {
      "epoch": 48.34146341463415,
      "grad_norm": 1.02310311794281,
      "learning_rate": 5.7272480304265154e-05,
      "loss": 1.4992,
      "step": 9910
    },
    {
      "epoch": 48.390243902439025,
      "grad_norm": 1.032989263534546,
      "learning_rate": 5.721814724259712e-05,
      "loss": 1.5065,
      "step": 9920
    },
    {
      "epoch": 48.4390243902439,
      "grad_norm": 0.9401241540908813,
      "learning_rate": 5.71638141809291e-05,
      "loss": 1.4799,
      "step": 9930
    },
    {
      "epoch": 48.48780487804878,
      "grad_norm": 1.178525447845459,
      "learning_rate": 5.7109481119261074e-05,
      "loss": 1.5085,
      "step": 9940
    },
    {
      "epoch": 48.53658536585366,
      "grad_norm": 0.8683713674545288,
      "learning_rate": 5.705514805759304e-05,
      "loss": 1.5123,
      "step": 9950
    },
    {
      "epoch": 48.58536585365854,
      "grad_norm": 1.0342472791671753,
      "learning_rate": 5.7000814995925025e-05,
      "loss": 1.5292,
      "step": 9960
    },
    {
      "epoch": 48.63414634146341,
      "grad_norm": 0.9925521016120911,
      "learning_rate": 5.6946481934257e-05,
      "loss": 1.4806,
      "step": 9970
    },
    {
      "epoch": 48.68292682926829,
      "grad_norm": 0.9937178492546082,
      "learning_rate": 5.689214887258897e-05,
      "loss": 1.4674,
      "step": 9980
    },
    {
      "epoch": 48.73170731707317,
      "grad_norm": 0.8783002495765686,
      "learning_rate": 5.6837815810920945e-05,
      "loss": 1.4889,
      "step": 9990
    },
    {
      "epoch": 48.78048780487805,
      "grad_norm": 0.9060630798339844,
      "learning_rate": 5.678348274925293e-05,
      "loss": 1.4731,
      "step": 10000
    },
    {
      "epoch": 48.829268292682926,
      "grad_norm": 0.9229355454444885,
      "learning_rate": 5.672914968758489e-05,
      "loss": 1.5158,
      "step": 10010
    },
    {
      "epoch": 48.8780487804878,
      "grad_norm": 1.007168173789978,
      "learning_rate": 5.667481662591687e-05,
      "loss": 1.5212,
      "step": 10020
    },
    {
      "epoch": 48.926829268292686,
      "grad_norm": 0.9485666155815125,
      "learning_rate": 5.662048356424885e-05,
      "loss": 1.5356,
      "step": 10030
    },
    {
      "epoch": 48.97560975609756,
      "grad_norm": 1.065011978149414,
      "learning_rate": 5.656615050258083e-05,
      "loss": 1.4857,
      "step": 10040
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.6598,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.087491273880005,
      "eval_rouge1": 0.6112,
      "eval_rouge2": 0.3522,
      "eval_rougeL": 0.5444,
      "eval_runtime": 26.6917,
      "eval_samples_per_second": 6.819,
      "eval_steps_per_second": 1.723,
      "step": 10045
    },
    {
      "epoch": 49.02439024390244,
      "grad_norm": 0.9155561327934265,
      "learning_rate": 5.65118174409128e-05,
      "loss": 1.5131,
      "step": 10050
    },
    {
      "epoch": 49.073170731707314,
      "grad_norm": 0.9302152991294861,
      "learning_rate": 5.6457484379244774e-05,
      "loss": 1.5032,
      "step": 10060
    },
    {
      "epoch": 49.1219512195122,
      "grad_norm": 0.9523373246192932,
      "learning_rate": 5.640315131757675e-05,
      "loss": 1.4814,
      "step": 10070
    },
    {
      "epoch": 49.170731707317074,
      "grad_norm": 1.1650649309158325,
      "learning_rate": 5.634881825590872e-05,
      "loss": 1.4777,
      "step": 10080
    },
    {
      "epoch": 49.21951219512195,
      "grad_norm": 0.9776322245597839,
      "learning_rate": 5.62944851942407e-05,
      "loss": 1.4791,
      "step": 10090
    },
    {
      "epoch": 49.26829268292683,
      "grad_norm": 1.0491796731948853,
      "learning_rate": 5.624015213257268e-05,
      "loss": 1.5079,
      "step": 10100
    },
    {
      "epoch": 49.31707317073171,
      "grad_norm": 0.9312990307807922,
      "learning_rate": 5.6185819070904646e-05,
      "loss": 1.486,
      "step": 10110
    },
    {
      "epoch": 49.36585365853659,
      "grad_norm": 1.1731209754943848,
      "learning_rate": 5.613148600923662e-05,
      "loss": 1.4841,
      "step": 10120
    },
    {
      "epoch": 49.41463414634146,
      "grad_norm": 0.9213290810585022,
      "learning_rate": 5.6077152947568603e-05,
      "loss": 1.4767,
      "step": 10130
    },
    {
      "epoch": 49.46341463414634,
      "grad_norm": 1.035176157951355,
      "learning_rate": 5.6022819885900565e-05,
      "loss": 1.4872,
      "step": 10140
    },
    {
      "epoch": 49.51219512195122,
      "grad_norm": 1.0663214921951294,
      "learning_rate": 5.596848682423255e-05,
      "loss": 1.4926,
      "step": 10150
    },
    {
      "epoch": 49.5609756097561,
      "grad_norm": 0.9391820430755615,
      "learning_rate": 5.5914153762564523e-05,
      "loss": 1.4914,
      "step": 10160
    },
    {
      "epoch": 49.609756097560975,
      "grad_norm": 0.9385082721710205,
      "learning_rate": 5.5859820700896506e-05,
      "loss": 1.4833,
      "step": 10170
    },
    {
      "epoch": 49.65853658536585,
      "grad_norm": 0.8373975157737732,
      "learning_rate": 5.580548763922847e-05,
      "loss": 1.4758,
      "step": 10180
    },
    {
      "epoch": 49.707317073170735,
      "grad_norm": 0.9798086881637573,
      "learning_rate": 5.575115457756045e-05,
      "loss": 1.5062,
      "step": 10190
    },
    {
      "epoch": 49.75609756097561,
      "grad_norm": 0.9505453109741211,
      "learning_rate": 5.5696821515892426e-05,
      "loss": 1.4867,
      "step": 10200
    },
    {
      "epoch": 49.80487804878049,
      "grad_norm": 0.8706372976303101,
      "learning_rate": 5.5642488454224395e-05,
      "loss": 1.4573,
      "step": 10210
    },
    {
      "epoch": 49.853658536585364,
      "grad_norm": 1.0630508661270142,
      "learning_rate": 5.558815539255638e-05,
      "loss": 1.4822,
      "step": 10220
    },
    {
      "epoch": 49.90243902439025,
      "grad_norm": 0.9813632369041443,
      "learning_rate": 5.553382233088835e-05,
      "loss": 1.5155,
      "step": 10230
    },
    {
      "epoch": 49.951219512195124,
      "grad_norm": 1.0495411157608032,
      "learning_rate": 5.547948926922032e-05,
      "loss": 1.4857,
      "step": 10240
    },
    {
      "epoch": 50.0,
      "grad_norm": 1.2372345924377441,
      "learning_rate": 5.54251562075523e-05,
      "loss": 1.5093,
      "step": 10250
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.66,
      "eval_gen_len": 211.6593,
      "eval_loss": 2.0857725143432617,
      "eval_rouge1": 0.6128,
      "eval_rouge2": 0.3524,
      "eval_rougeL": 0.545,
      "eval_runtime": 26.6761,
      "eval_samples_per_second": 6.823,
      "eval_steps_per_second": 1.724,
      "step": 10250
    }
  ],
  "logging_steps": 10,
  "max_steps": 10250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 10,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 7
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3138571725673472e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}