{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1134,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.026455026455026454,
      "grad_norm": 7.5692021410659995,
      "learning_rate": 5e-06,
      "loss": 1.0155,
      "step": 10
    },
    {
      "epoch": 0.05291005291005291,
      "grad_norm": 2.6174349537450885,
      "learning_rate": 5e-06,
      "loss": 0.9187,
      "step": 20
    },
    {
      "epoch": 0.07936507936507936,
      "grad_norm": 2.793450551829053,
      "learning_rate": 5e-06,
      "loss": 0.8889,
      "step": 30
    },
    {
      "epoch": 0.10582010582010581,
      "grad_norm": 1.0752828617491306,
      "learning_rate": 5e-06,
      "loss": 0.8728,
      "step": 40
    },
    {
      "epoch": 0.13227513227513227,
      "grad_norm": 0.7890229586623537,
      "learning_rate": 5e-06,
      "loss": 0.8553,
      "step": 50
    },
    {
      "epoch": 0.15873015873015872,
      "grad_norm": 0.8515898940332954,
      "learning_rate": 5e-06,
      "loss": 0.8472,
      "step": 60
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": 0.6739899043810001,
      "learning_rate": 5e-06,
      "loss": 0.8386,
      "step": 70
    },
    {
      "epoch": 0.21164021164021163,
      "grad_norm": 0.7472386351456437,
      "learning_rate": 5e-06,
      "loss": 0.8303,
      "step": 80
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 0.8201164599585603,
      "learning_rate": 5e-06,
      "loss": 0.8296,
      "step": 90
    },
    {
      "epoch": 0.26455026455026454,
      "grad_norm": 0.5974248203060147,
      "learning_rate": 5e-06,
      "loss": 0.8208,
      "step": 100
    },
    {
      "epoch": 0.291005291005291,
      "grad_norm": 0.7418126551089665,
      "learning_rate": 5e-06,
      "loss": 0.817,
      "step": 110
    },
    {
      "epoch": 0.31746031746031744,
      "grad_norm": 0.8971788447098652,
      "learning_rate": 5e-06,
      "loss": 0.816,
      "step": 120
    },
    {
      "epoch": 0.3439153439153439,
      "grad_norm": 0.6659038218094219,
      "learning_rate": 5e-06,
      "loss": 0.8095,
      "step": 130
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.8687534608310304,
      "learning_rate": 5e-06,
      "loss": 0.809,
      "step": 140
    },
    {
      "epoch": 0.3968253968253968,
      "grad_norm": 0.6997183066642144,
      "learning_rate": 5e-06,
      "loss": 0.8116,
      "step": 150
    },
    {
      "epoch": 0.42328042328042326,
      "grad_norm": 0.6784743741082218,
      "learning_rate": 5e-06,
      "loss": 0.8043,
      "step": 160
    },
    {
      "epoch": 0.4497354497354497,
      "grad_norm": 0.8279542209712195,
      "learning_rate": 5e-06,
      "loss": 0.8081,
      "step": 170
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.6448467413029932,
      "learning_rate": 5e-06,
      "loss": 0.806,
      "step": 180
    },
    {
      "epoch": 0.5026455026455027,
      "grad_norm": 0.5414700848521968,
      "learning_rate": 5e-06,
      "loss": 0.8021,
      "step": 190
    },
    {
      "epoch": 0.5291005291005291,
      "grad_norm": 0.6833581882793117,
      "learning_rate": 5e-06,
      "loss": 0.7978,
      "step": 200
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.9216374614283346,
      "learning_rate": 5e-06,
      "loss": 0.7997,
      "step": 210
    },
    {
      "epoch": 0.582010582010582,
      "grad_norm": 0.6017660770757399,
      "learning_rate": 5e-06,
      "loss": 0.7942,
      "step": 220
    },
    {
      "epoch": 0.6084656084656085,
      "grad_norm": 0.6412167679844537,
      "learning_rate": 5e-06,
      "loss": 0.791,
      "step": 230
    },
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 0.6772848668775633,
      "learning_rate": 5e-06,
      "loss": 0.7948,
      "step": 240
    },
    {
      "epoch": 0.6613756613756614,
      "grad_norm": 0.600355639535383,
      "learning_rate": 5e-06,
      "loss": 0.7914,
      "step": 250
    },
    {
      "epoch": 0.6878306878306878,
      "grad_norm": 0.5606573561606166,
      "learning_rate": 5e-06,
      "loss": 0.7941,
      "step": 260
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.5274683257315145,
      "learning_rate": 5e-06,
      "loss": 0.7934,
      "step": 270
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.5268307259184533,
      "learning_rate": 5e-06,
      "loss": 0.7938,
      "step": 280
    },
    {
      "epoch": 0.7671957671957672,
      "grad_norm": 0.7728039906376213,
      "learning_rate": 5e-06,
      "loss": 0.7917,
      "step": 290
    },
    {
      "epoch": 0.7936507936507936,
      "grad_norm": 0.5113816020074538,
      "learning_rate": 5e-06,
      "loss": 0.7882,
      "step": 300
    },
    {
      "epoch": 0.8201058201058201,
      "grad_norm": 0.5190008909833674,
      "learning_rate": 5e-06,
      "loss": 0.7849,
      "step": 310
    },
    {
      "epoch": 0.8465608465608465,
      "grad_norm": 0.5836368027639713,
      "learning_rate": 5e-06,
      "loss": 0.7884,
      "step": 320
    },
    {
      "epoch": 0.873015873015873,
      "grad_norm": 0.7393083910981807,
      "learning_rate": 5e-06,
      "loss": 0.7832,
      "step": 330
    },
    {
      "epoch": 0.8994708994708994,
      "grad_norm": 0.6790621837196994,
      "learning_rate": 5e-06,
      "loss": 0.7889,
      "step": 340
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 0.7554140140474603,
      "learning_rate": 5e-06,
      "loss": 0.7915,
      "step": 350
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.5810605450609724,
      "learning_rate": 5e-06,
      "loss": 0.7863,
      "step": 360
    },
    {
      "epoch": 0.9788359788359788,
      "grad_norm": 0.6061224991845839,
      "learning_rate": 5e-06,
      "loss": 0.7836,
      "step": 370
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7853822112083435,
      "eval_runtime": 36.7907,
      "eval_samples_per_second": 276.537,
      "eval_steps_per_second": 1.087,
      "step": 378
    },
    {
      "epoch": 1.0052910052910053,
      "grad_norm": 0.8837267831248354,
      "learning_rate": 5e-06,
      "loss": 0.7782,
      "step": 380
    },
    {
      "epoch": 1.0317460317460316,
      "grad_norm": 0.8327957648034358,
      "learning_rate": 5e-06,
      "loss": 0.752,
      "step": 390
    },
    {
      "epoch": 1.0582010582010581,
      "grad_norm": 0.7661996686677544,
      "learning_rate": 5e-06,
      "loss": 0.7504,
      "step": 400
    },
    {
      "epoch": 1.0846560846560847,
      "grad_norm": 0.574247744750553,
      "learning_rate": 5e-06,
      "loss": 0.753,
      "step": 410
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.6270570009050412,
      "learning_rate": 5e-06,
      "loss": 0.7526,
      "step": 420
    },
    {
      "epoch": 1.1375661375661377,
      "grad_norm": 0.7564791753598593,
      "learning_rate": 5e-06,
      "loss": 0.7499,
      "step": 430
    },
    {
      "epoch": 1.164021164021164,
      "grad_norm": 0.6019767178166228,
      "learning_rate": 5e-06,
      "loss": 0.7534,
      "step": 440
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.7045276656881013,
      "learning_rate": 5e-06,
      "loss": 0.7516,
      "step": 450
    },
    {
      "epoch": 1.216931216931217,
      "grad_norm": 0.7151521174595956,
      "learning_rate": 5e-06,
      "loss": 0.7525,
      "step": 460
    },
    {
      "epoch": 1.2433862433862433,
      "grad_norm": 0.8589885709685776,
      "learning_rate": 5e-06,
      "loss": 0.7506,
      "step": 470
    },
    {
      "epoch": 1.2698412698412698,
      "grad_norm": 0.8448779205925924,
      "learning_rate": 5e-06,
      "loss": 0.7529,
      "step": 480
    },
    {
      "epoch": 1.2962962962962963,
      "grad_norm": 0.698173785850267,
      "learning_rate": 5e-06,
      "loss": 0.7523,
      "step": 490
    },
    {
      "epoch": 1.3227513227513228,
      "grad_norm": 0.5810999760014718,
      "learning_rate": 5e-06,
      "loss": 0.7493,
      "step": 500
    },
    {
      "epoch": 1.3492063492063493,
      "grad_norm": 0.6992760350263512,
      "learning_rate": 5e-06,
      "loss": 0.7468,
      "step": 510
    },
    {
      "epoch": 1.3756613756613756,
      "grad_norm": 0.6741207559975058,
      "learning_rate": 5e-06,
      "loss": 0.7498,
      "step": 520
    },
    {
      "epoch": 1.402116402116402,
      "grad_norm": 0.5507423746197616,
      "learning_rate": 5e-06,
      "loss": 0.7523,
      "step": 530
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.6210883025743585,
      "learning_rate": 5e-06,
      "loss": 0.7491,
      "step": 540
    },
    {
      "epoch": 1.455026455026455,
      "grad_norm": 0.5352886813564971,
      "learning_rate": 5e-06,
      "loss": 0.7501,
      "step": 550
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 0.6312193974549839,
      "learning_rate": 5e-06,
      "loss": 0.7523,
      "step": 560
    },
    {
      "epoch": 1.507936507936508,
      "grad_norm": 0.5240777963700096,
      "learning_rate": 5e-06,
      "loss": 0.7506,
      "step": 570
    },
    {
      "epoch": 1.5343915343915344,
      "grad_norm": 0.6713658440893504,
      "learning_rate": 5e-06,
      "loss": 0.7498,
      "step": 580
    },
    {
      "epoch": 1.560846560846561,
      "grad_norm": 0.5581636551389646,
      "learning_rate": 5e-06,
      "loss": 0.749,
      "step": 590
    },
    {
      "epoch": 1.5873015873015874,
      "grad_norm": 0.5510394311493787,
      "learning_rate": 5e-06,
      "loss": 0.7507,
      "step": 600
    },
    {
      "epoch": 1.6137566137566137,
      "grad_norm": 0.7490035349717284,
      "learning_rate": 5e-06,
      "loss": 0.751,
      "step": 610
    },
    {
      "epoch": 1.6402116402116402,
      "grad_norm": 0.7019577853473584,
      "learning_rate": 5e-06,
      "loss": 0.7452,
      "step": 620
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.5188517209828246,
      "learning_rate": 5e-06,
      "loss": 0.747,
      "step": 630
    },
    {
      "epoch": 1.693121693121693,
      "grad_norm": 0.5852910158185847,
      "learning_rate": 5e-06,
      "loss": 0.7478,
      "step": 640
    },
    {
      "epoch": 1.7195767195767195,
      "grad_norm": 0.5163509954560109,
      "learning_rate": 5e-06,
      "loss": 0.7493,
      "step": 650
    },
    {
      "epoch": 1.746031746031746,
      "grad_norm": 0.5259629378155002,
      "learning_rate": 5e-06,
      "loss": 0.7488,
      "step": 660
    },
    {
      "epoch": 1.7724867724867726,
      "grad_norm": 0.5847699777528805,
      "learning_rate": 5e-06,
      "loss": 0.7484,
      "step": 670
    },
    {
      "epoch": 1.798941798941799,
      "grad_norm": 0.6909699114073832,
      "learning_rate": 5e-06,
      "loss": 0.7425,
      "step": 680
    },
    {
      "epoch": 1.8253968253968254,
      "grad_norm": 0.5026266608814028,
      "learning_rate": 5e-06,
      "loss": 0.7481,
      "step": 690
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.4905863322036327,
      "learning_rate": 5e-06,
      "loss": 0.7427,
      "step": 700
    },
    {
      "epoch": 1.8783068783068781,
      "grad_norm": 0.4940567635795736,
      "learning_rate": 5e-06,
      "loss": 0.7468,
      "step": 710
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.5002276189718439,
      "learning_rate": 5e-06,
      "loss": 0.7411,
      "step": 720
    },
    {
      "epoch": 1.9312169312169312,
      "grad_norm": 0.5990930220626486,
      "learning_rate": 5e-06,
      "loss": 0.7445,
      "step": 730
    },
    {
      "epoch": 1.9576719576719577,
      "grad_norm": 0.575362618033918,
      "learning_rate": 5e-06,
      "loss": 0.746,
      "step": 740
    },
    {
      "epoch": 1.9841269841269842,
      "grad_norm": 0.5971817140113224,
      "learning_rate": 5e-06,
      "loss": 0.7439,
      "step": 750
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7724016904830933,
      "eval_runtime": 36.4294,
      "eval_samples_per_second": 279.28,
      "eval_steps_per_second": 1.098,
      "step": 756
    },
    {
      "epoch": 2.0105820105820107,
      "grad_norm": 0.7946153532333404,
      "learning_rate": 5e-06,
      "loss": 0.7315,
      "step": 760
    },
    {
      "epoch": 2.037037037037037,
      "grad_norm": 0.6230352829055962,
      "learning_rate": 5e-06,
      "loss": 0.7084,
      "step": 770
    },
    {
      "epoch": 2.0634920634920633,
      "grad_norm": 0.6511688564166924,
      "learning_rate": 5e-06,
      "loss": 0.7103,
      "step": 780
    },
    {
      "epoch": 2.0899470899470898,
      "grad_norm": 0.6162972908978676,
      "learning_rate": 5e-06,
      "loss": 0.7106,
      "step": 790
    },
    {
      "epoch": 2.1164021164021163,
      "grad_norm": 0.5483322240242068,
      "learning_rate": 5e-06,
      "loss": 0.7103,
      "step": 800
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.5167695346041687,
      "learning_rate": 5e-06,
      "loss": 0.7118,
      "step": 810
    },
    {
      "epoch": 2.1693121693121693,
      "grad_norm": 0.5731147019555838,
      "learning_rate": 5e-06,
      "loss": 0.7117,
      "step": 820
    },
    {
      "epoch": 2.195767195767196,
      "grad_norm": 0.5992340825814363,
      "learning_rate": 5e-06,
      "loss": 0.7115,
      "step": 830
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.8204374260544979,
      "learning_rate": 5e-06,
      "loss": 0.7116,
      "step": 840
    },
    {
      "epoch": 2.248677248677249,
      "grad_norm": 0.6452436861380945,
      "learning_rate": 5e-06,
      "loss": 0.7132,
      "step": 850
    },
    {
      "epoch": 2.2751322751322753,
      "grad_norm": 0.7905676999430032,
      "learning_rate": 5e-06,
      "loss": 0.7156,
      "step": 860
    },
    {
      "epoch": 2.3015873015873014,
      "grad_norm": 0.6453047673638057,
      "learning_rate": 5e-06,
      "loss": 0.7134,
      "step": 870
    },
    {
      "epoch": 2.328042328042328,
      "grad_norm": 0.6161389768819914,
      "learning_rate": 5e-06,
      "loss": 0.7182,
      "step": 880
    },
    {
      "epoch": 2.3544973544973544,
      "grad_norm": 0.5925752010088033,
      "learning_rate": 5e-06,
      "loss": 0.7152,
      "step": 890
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.5465043223339282,
      "learning_rate": 5e-06,
      "loss": 0.7164,
      "step": 900
    },
    {
      "epoch": 2.4074074074074074,
      "grad_norm": 0.5991686328257099,
      "learning_rate": 5e-06,
      "loss": 0.7169,
      "step": 910
    },
    {
      "epoch": 2.433862433862434,
      "grad_norm": 0.5529948466057861,
      "learning_rate": 5e-06,
      "loss": 0.7134,
      "step": 920
    },
    {
      "epoch": 2.4603174603174605,
      "grad_norm": 0.5700704291829093,
      "learning_rate": 5e-06,
      "loss": 0.714,
      "step": 930
    },
    {
      "epoch": 2.4867724867724865,
      "grad_norm": 0.7107367519714186,
      "learning_rate": 5e-06,
      "loss": 0.7119,
      "step": 940
    },
    {
      "epoch": 2.5132275132275135,
      "grad_norm": 0.5145994918143386,
      "learning_rate": 5e-06,
      "loss": 0.7164,
      "step": 950
    },
    {
      "epoch": 2.5396825396825395,
      "grad_norm": 0.8526143229072503,
      "learning_rate": 5e-06,
      "loss": 0.7176,
      "step": 960
    },
    {
      "epoch": 2.566137566137566,
      "grad_norm": 0.6460058195221838,
      "learning_rate": 5e-06,
      "loss": 0.718,
      "step": 970
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 0.5722175187803165,
      "learning_rate": 5e-06,
      "loss": 0.7158,
      "step": 980
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.6199806409264405,
      "learning_rate": 5e-06,
      "loss": 0.7152,
      "step": 990
    },
    {
      "epoch": 2.6455026455026456,
      "grad_norm": 0.6295374166579548,
      "learning_rate": 5e-06,
      "loss": 0.7167,
      "step": 1000
    },
    {
      "epoch": 2.671957671957672,
      "grad_norm": 0.5562186015106506,
      "learning_rate": 5e-06,
      "loss": 0.7143,
      "step": 1010
    },
    {
      "epoch": 2.6984126984126986,
      "grad_norm": 0.5802340156160444,
      "learning_rate": 5e-06,
      "loss": 0.7201,
      "step": 1020
    },
    {
      "epoch": 2.7248677248677247,
      "grad_norm": 0.5674240362878424,
      "learning_rate": 5e-06,
      "loss": 0.7151,
      "step": 1030
    },
    {
      "epoch": 2.751322751322751,
      "grad_norm": 0.5579240804629998,
      "learning_rate": 5e-06,
      "loss": 0.7113,
      "step": 1040
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.5511568456212458,
      "learning_rate": 5e-06,
      "loss": 0.7153,
      "step": 1050
    },
    {
      "epoch": 2.804232804232804,
      "grad_norm": 0.53261750815606,
      "learning_rate": 5e-06,
      "loss": 0.7147,
      "step": 1060
    },
    {
      "epoch": 2.8306878306878307,
      "grad_norm": 0.601473528074606,
      "learning_rate": 5e-06,
      "loss": 0.7165,
      "step": 1070
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.4922870741419976,
      "learning_rate": 5e-06,
      "loss": 0.7147,
      "step": 1080
    },
    {
      "epoch": 2.8835978835978837,
      "grad_norm": 0.5647388229398058,
      "learning_rate": 5e-06,
      "loss": 0.7146,
      "step": 1090
    },
    {
      "epoch": 2.91005291005291,
      "grad_norm": 0.6422541209526703,
      "learning_rate": 5e-06,
      "loss": 0.7175,
      "step": 1100
    },
    {
      "epoch": 2.9365079365079367,
      "grad_norm": 0.6977559447017284,
      "learning_rate": 5e-06,
      "loss": 0.7162,
      "step": 1110
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.7941392765996913,
      "learning_rate": 5e-06,
      "loss": 0.714,
      "step": 1120
    },
    {
      "epoch": 2.9894179894179893,
      "grad_norm": 0.6766160989486909,
      "learning_rate": 5e-06,
      "loss": 0.7146,
      "step": 1130
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7703806757926941,
      "eval_runtime": 35.7626,
      "eval_samples_per_second": 284.487,
      "eval_steps_per_second": 1.118,
      "step": 1134
    },
    {
      "epoch": 3.0,
      "step": 1134,
      "total_flos": 1899492236328960.0,
      "train_loss": 0.7598387925923397,
      "train_runtime": 7118.4294,
      "train_samples_per_second": 81.465,
      "train_steps_per_second": 0.159
    }
  ],
  "logging_steps": 10,
  "max_steps": 1134,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1899492236328960.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}