{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5020766035994463,
  "eval_steps": 500,
  "global_step": 68,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007383479464697739,
      "grad_norm": 1.9591929912567139,
      "learning_rate": 0.0,
      "loss": 1.6228,
      "memory/device_mem_reserved(gib)": 21.61,
      "memory/max_mem_active(gib)": 21.2,
      "memory/max_mem_allocated(gib)": 21.2,
      "step": 1
    },
    {
      "epoch": 0.014766958929395477,
      "grad_norm": 1.4523507356643677,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 1.5769,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 2
    },
    {
      "epoch": 0.022150438394093218,
      "grad_norm": 1.1918187141418457,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 1.5435,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 3
    },
    {
      "epoch": 0.029533917858790955,
      "grad_norm": 0.8260876536369324,
      "learning_rate": 4.615384615384616e-05,
      "loss": 1.6523,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 4
    },
    {
      "epoch": 0.03691739732348869,
      "grad_norm": 0.8584926128387451,
      "learning_rate": 6.153846153846155e-05,
      "loss": 1.5745,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 5
    },
    {
      "epoch": 0.044300876788186436,
      "grad_norm": 0.6466429829597473,
      "learning_rate": 7.692307692307693e-05,
      "loss": 1.4759,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 6
    },
    {
      "epoch": 0.05168435625288417,
      "grad_norm": 0.5014482140541077,
      "learning_rate": 9.230769230769232e-05,
      "loss": 1.602,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 7
    },
    {
      "epoch": 0.05906783571758191,
      "grad_norm": 0.6017433404922485,
      "learning_rate": 0.0001076923076923077,
      "loss": 1.4176,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 8
    },
    {
      "epoch": 0.06645131518227965,
      "grad_norm": 0.4612258970737457,
      "learning_rate": 0.0001230769230769231,
      "loss": 1.5819,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 9
    },
    {
      "epoch": 0.07383479464697738,
      "grad_norm": 0.4430214464664459,
      "learning_rate": 0.00013846153846153847,
      "loss": 1.561,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 10
    },
    {
      "epoch": 0.08121827411167512,
      "grad_norm": 0.3746771216392517,
      "learning_rate": 0.00015384615384615385,
      "loss": 1.6744,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 11
    },
    {
      "epoch": 0.08860175357637287,
      "grad_norm": 0.38248857855796814,
      "learning_rate": 0.00016923076923076923,
      "loss": 1.5629,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 12
    },
    {
      "epoch": 0.09598523304107061,
      "grad_norm": 0.515844464302063,
      "learning_rate": 0.00018461538461538463,
      "loss": 1.5264,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 13
    },
    {
      "epoch": 0.10336871250576835,
      "grad_norm": 0.3964424431324005,
      "learning_rate": 0.0002,
      "loss": 1.5398,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 14
    },
    {
      "epoch": 0.11075219197046608,
      "grad_norm": 0.4010593891143799,
      "learning_rate": 0.0001999668467514313,
      "loss": 1.4618,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 15
    },
    {
      "epoch": 0.11813567143516382,
      "grad_norm": 0.3192802965641022,
      "learning_rate": 0.00019986740898848306,
      "loss": 1.6994,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 16
    },
    {
      "epoch": 0.12551915089986157,
      "grad_norm": 0.410099059343338,
      "learning_rate": 0.00019970175264485266,
      "loss": 1.5913,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 17
    },
    {
      "epoch": 0.1329026303645593,
      "grad_norm": 0.312429815530777,
      "learning_rate": 0.0001994699875614589,
      "loss": 1.5701,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 18
    },
    {
      "epoch": 0.14028610982925704,
      "grad_norm": 0.2831230163574219,
      "learning_rate": 0.00019917226741361015,
      "loss": 1.5744,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 19
    },
    {
      "epoch": 0.14766958929395477,
      "grad_norm": 0.3618868291378021,
      "learning_rate": 0.00019880878960910772,
      "loss": 1.5185,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 20
    },
    {
      "epoch": 0.15505306875865252,
      "grad_norm": 0.3151628077030182,
      "learning_rate": 0.00019837979515735166,
      "loss": 1.5086,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 21
    },
    {
      "epoch": 0.16243654822335024,
      "grad_norm": 0.31955838203430176,
      "learning_rate": 0.0001978855685095358,
      "loss": 1.6329,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 22
    },
    {
      "epoch": 0.169820027688048,
      "grad_norm": 0.3030437231063843,
      "learning_rate": 0.00019732643737003827,
      "loss": 1.6697,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 23
    },
    {
      "epoch": 0.17720350715274574,
      "grad_norm": 0.41288134455680847,
      "learning_rate": 0.00019670277247913205,
      "loss": 1.7094,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 24
    },
    {
      "epoch": 0.18458698661744347,
      "grad_norm": 0.2887294888496399,
      "learning_rate": 0.00019601498736716017,
      "loss": 1.5554,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 25
    },
    {
      "epoch": 0.19197046608214122,
      "grad_norm": 0.3173791170120239,
      "learning_rate": 0.00019526353808033825,
      "loss": 1.4404,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 26
    },
    {
      "epoch": 0.19935394554683894,
      "grad_norm": 0.2877439558506012,
      "learning_rate": 0.00019444892287836613,
      "loss": 1.4766,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 27
    },
    {
      "epoch": 0.2067374250115367,
      "grad_norm": 0.29286038875579834,
      "learning_rate": 0.00019357168190404936,
      "loss": 1.5156,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 28
    },
    {
      "epoch": 0.2141209044762344,
      "grad_norm": 0.27713659405708313,
      "learning_rate": 0.00019263239682514952,
      "loss": 1.5153,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 29
    },
    {
      "epoch": 0.22150438394093216,
      "grad_norm": 0.29187655448913574,
      "learning_rate": 0.0001916316904487005,
      "loss": 1.6036,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 30
    },
    {
      "epoch": 0.22888786340562992,
      "grad_norm": 0.2671583890914917,
      "learning_rate": 0.00019057022630804716,
      "loss": 1.4675,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 31
    },
    {
      "epoch": 0.23627134287032764,
      "grad_norm": 0.2679831087589264,
      "learning_rate": 0.00018944870822287956,
      "loss": 1.581,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 32
    },
    {
      "epoch": 0.2436548223350254,
      "grad_norm": 0.26359617710113525,
      "learning_rate": 0.00018826787983255473,
      "loss": 1.4674,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 33
    },
    {
      "epoch": 0.25103830179972314,
      "grad_norm": 0.30446046590805054,
      "learning_rate": 0.00018702852410301554,
      "loss": 1.5038,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 34
    },
    {
      "epoch": 0.25842178126442084,
      "grad_norm": 0.3004315197467804,
      "learning_rate": 0.00018573146280763324,
      "loss": 1.4024,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 35
    },
    {
      "epoch": 0.2658052607291186,
      "grad_norm": 0.27353399991989136,
      "learning_rate": 0.00018437755598231856,
      "loss": 1.3652,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 36
    },
    {
      "epoch": 0.27318874019381634,
      "grad_norm": 0.2659265995025635,
      "learning_rate": 0.0001829677013552619,
      "loss": 1.4905,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 37
    },
    {
      "epoch": 0.2805722196585141,
      "grad_norm": 0.2703750431537628,
      "learning_rate": 0.00018150283375168114,
      "loss": 1.3298,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 38
    },
    {
      "epoch": 0.28795569912321184,
      "grad_norm": 0.29322877526283264,
      "learning_rate": 0.00017998392447397197,
      "loss": 1.3639,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 39
    },
    {
      "epoch": 0.29533917858790953,
      "grad_norm": 0.26927265524864197,
      "learning_rate": 0.00017841198065767107,
      "loss": 1.4849,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 40
    },
    {
      "epoch": 0.3027226580526073,
      "grad_norm": 0.26683205366134644,
      "learning_rate": 0.00017678804460366,
      "loss": 1.4523,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 41
    },
    {
      "epoch": 0.31010613751730504,
      "grad_norm": 0.26909339427948,
      "learning_rate": 0.00017511319308705198,
      "loss": 1.6047,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 42
    },
    {
      "epoch": 0.3174896169820028,
      "grad_norm": 0.2938016355037689,
      "learning_rate": 0.00017338853664321992,
      "loss": 1.3858,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 43
    },
    {
      "epoch": 0.3248730964467005,
      "grad_norm": 0.28290048241615295,
      "learning_rate": 0.00017161521883143934,
      "loss": 1.3671,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 44
    },
    {
      "epoch": 0.33225657591139823,
      "grad_norm": 0.2870519161224365,
      "learning_rate": 0.00016979441547663435,
      "loss": 1.5334,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 45
    },
    {
      "epoch": 0.339640055376096,
      "grad_norm": 0.26375913619995117,
      "learning_rate": 0.00016792733388972932,
      "loss": 1.5627,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 46
    },
    {
      "epoch": 0.34702353484079373,
      "grad_norm": 0.26975446939468384,
      "learning_rate": 0.00016601521206712318,
      "loss": 1.4987,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 47
    },
    {
      "epoch": 0.3544070143054915,
      "grad_norm": 0.30166226625442505,
      "learning_rate": 0.00016405931786981755,
      "loss": 1.5128,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 48
    },
    {
      "epoch": 0.3617904937701892,
      "grad_norm": 0.29190436005592346,
      "learning_rate": 0.00016206094818274229,
      "loss": 1.4914,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 49
    },
    {
      "epoch": 0.36917397323488693,
      "grad_norm": 0.2847207486629486,
      "learning_rate": 0.00016002142805483685,
      "loss": 1.4755,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 50
    },
    {
      "epoch": 0.3765574526995847,
      "grad_norm": 0.2877140939235687,
      "learning_rate": 0.00015794210982045636,
      "loss": 1.422,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 51
    },
    {
      "epoch": 0.38394093216428243,
      "grad_norm": 0.2705255150794983,
      "learning_rate": 0.00015582437220268647,
      "loss": 1.4822,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 52
    },
    {
      "epoch": 0.3913244116289802,
      "grad_norm": 0.2597866654396057,
      "learning_rate": 0.00015366961939916008,
      "loss": 1.5088,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 53
    },
    {
      "epoch": 0.3987078910936779,
      "grad_norm": 0.2965547740459442,
      "learning_rate": 0.0001514792801509831,
      "loss": 1.5346,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 54
    },
    {
      "epoch": 0.40609137055837563,
      "grad_norm": 0.2670862078666687,
      "learning_rate": 0.00014925480679538647,
      "loss": 1.5695,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 55
    },
    {
      "epoch": 0.4134748500230734,
      "grad_norm": 0.30679062008857727,
      "learning_rate": 0.000146997674302732,
      "loss": 1.6911,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 56
    },
    {
      "epoch": 0.42085832948777113,
      "grad_norm": 0.2812528908252716,
      "learning_rate": 0.0001447093792985114,
      "loss": 1.664,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 57
    },
    {
      "epoch": 0.4282418089524688,
      "grad_norm": 0.26079821586608887,
      "learning_rate": 0.0001423914390709861,
      "loss": 1.3911,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 58
    },
    {
      "epoch": 0.4356252884171666,
      "grad_norm": 0.3026648461818695,
      "learning_rate": 0.00014004539056512667,
      "loss": 1.5685,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 59
    },
    {
      "epoch": 0.44300876788186433,
      "grad_norm": 0.28435277938842773,
      "learning_rate": 0.00013767278936351854,
      "loss": 1.3599,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 60
    },
    {
      "epoch": 0.4503922473465621,
      "grad_norm": 0.29877325892448425,
      "learning_rate": 0.0001352752086549095,
      "loss": 1.5173,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 61
    },
    {
      "epoch": 0.45777572681125983,
      "grad_norm": 0.29666033387184143,
      "learning_rate": 0.0001328542381910835,
      "loss": 1.5017,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 62
    },
    {
      "epoch": 0.4651592062759575,
      "grad_norm": 0.26934438943862915,
      "learning_rate": 0.0001304114832327518,
      "loss": 1.4674,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 63
    },
    {
      "epoch": 0.4725426857406553,
      "grad_norm": 0.30038923025131226,
      "learning_rate": 0.00012794856348516095,
      "loss": 1.5306,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 64
    },
    {
      "epoch": 0.47992616520535303,
      "grad_norm": 0.2799171805381775,
      "learning_rate": 0.00012546711202412287,
      "loss": 1.4696,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 65
    },
    {
      "epoch": 0.4873096446700508,
      "grad_norm": 0.313919335603714,
      "learning_rate": 0.0001229687742131796,
      "loss": 1.5927,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 66
    },
    {
      "epoch": 0.4946931241347485,
      "grad_norm": 0.27579808235168457,
      "learning_rate": 0.0001204552066126201,
      "loss": 1.4654,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 67
    },
    {
      "epoch": 0.5020766035994463,
      "grad_norm": 0.28434908390045166,
      "learning_rate": 0.00011792807588107357,
      "loss": 1.5039,
      "memory/device_mem_reserved(gib)": 21.8,
      "memory/max_mem_active(gib)": 21.55,
      "memory/max_mem_allocated(gib)": 21.55,
      "step": 68
    }
  ],
  "logging_steps": 1,
  "max_steps": 135,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 34,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.234585444763566e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}