{
  "best_metric": 0.33288896083831787,
  "best_model_checkpoint": "realFake-img/checkpoint-700",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 1560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02564102564102564,
      "grad_norm": 2.928205728530884,
      "learning_rate": 0.00019871794871794874,
      "loss": 1.1209,
      "step": 10
    },
    {
      "epoch": 0.05128205128205128,
      "grad_norm": 3.1282591819763184,
      "learning_rate": 0.00019743589743589744,
      "loss": 0.6435,
      "step": 20
    },
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 2.2042582035064697,
      "learning_rate": 0.00019615384615384615,
      "loss": 0.6513,
      "step": 30
    },
    {
      "epoch": 0.10256410256410256,
      "grad_norm": 4.91236686706543,
      "learning_rate": 0.00019487179487179487,
      "loss": 0.7536,
      "step": 40
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 2.019882917404175,
      "learning_rate": 0.0001935897435897436,
      "loss": 0.6197,
      "step": 50
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 2.4161789417266846,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.5531,
      "step": 60
    },
    {
      "epoch": 0.1794871794871795,
      "grad_norm": 2.189767360687256,
      "learning_rate": 0.00019102564102564104,
      "loss": 0.4985,
      "step": 70
    },
    {
      "epoch": 0.20512820512820512,
      "grad_norm": 3.2690813541412354,
      "learning_rate": 0.00018974358974358974,
      "loss": 0.5499,
      "step": 80
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 1.6541521549224854,
      "learning_rate": 0.00018846153846153847,
      "loss": 0.5365,
      "step": 90
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 1.9506237506866455,
      "learning_rate": 0.0001871794871794872,
      "loss": 0.4892,
      "step": 100
    },
    {
      "epoch": 0.2564102564102564,
      "eval_accuracy": 0.7227272727272728,
      "eval_loss": 0.5756350159645081,
      "eval_runtime": 24.0859,
      "eval_samples_per_second": 45.67,
      "eval_steps_per_second": 5.73,
      "step": 100
    },
    {
      "epoch": 0.28205128205128205,
      "grad_norm": 3.7501442432403564,
      "learning_rate": 0.0001858974358974359,
      "loss": 0.4306,
      "step": 110
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.692314386367798,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.5122,
      "step": 120
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 2.5989458560943604,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.4974,
      "step": 130
    },
    {
      "epoch": 0.358974358974359,
      "grad_norm": 1.8959237337112427,
      "learning_rate": 0.00018205128205128207,
      "loss": 0.4464,
      "step": 140
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.9950543642044067,
      "learning_rate": 0.00018076923076923077,
      "loss": 0.4366,
      "step": 150
    },
    {
      "epoch": 0.41025641025641024,
      "grad_norm": 1.4334654808044434,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.5248,
      "step": 160
    },
    {
      "epoch": 0.4358974358974359,
      "grad_norm": 1.7422298192977905,
      "learning_rate": 0.00017820512820512823,
      "loss": 0.4314,
      "step": 170
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.0945038795471191,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.3464,
      "step": 180
    },
    {
      "epoch": 0.48717948717948717,
      "grad_norm": 2.3801605701446533,
      "learning_rate": 0.00017564102564102566,
      "loss": 0.4059,
      "step": 190
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 1.6624411344528198,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.683,
      "step": 200
    },
    {
      "epoch": 0.5128205128205128,
      "eval_accuracy": 0.6372727272727273,
      "eval_loss": 0.6742109656333923,
      "eval_runtime": 24.1183,
      "eval_samples_per_second": 45.609,
      "eval_steps_per_second": 5.722,
      "step": 200
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 3.1722240447998047,
      "learning_rate": 0.0001730769230769231,
      "loss": 0.4548,
      "step": 210
    },
    {
      "epoch": 0.5641025641025641,
      "grad_norm": 2.0260214805603027,
      "learning_rate": 0.0001717948717948718,
      "loss": 0.4612,
      "step": 220
    },
    {
      "epoch": 0.5897435897435898,
      "grad_norm": 2.1661155223846436,
      "learning_rate": 0.00017051282051282053,
      "loss": 0.3824,
      "step": 230
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.2094335556030273,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.3815,
      "step": 240
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 2.571754217147827,
      "learning_rate": 0.00016794871794871796,
      "loss": 0.3743,
      "step": 250
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 1.545766830444336,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.3708,
      "step": 260
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.742072343826294,
      "learning_rate": 0.0001653846153846154,
      "loss": 0.3729,
      "step": 270
    },
    {
      "epoch": 0.717948717948718,
      "grad_norm": 2.677527904510498,
      "learning_rate": 0.0001641025641025641,
      "loss": 0.3555,
      "step": 280
    },
    {
      "epoch": 0.7435897435897436,
      "grad_norm": 1.6417704820632935,
      "learning_rate": 0.00016282051282051282,
      "loss": 0.4212,
      "step": 290
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.6961071491241455,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.3737,
      "step": 300
    },
    {
      "epoch": 0.7692307692307693,
      "eval_accuracy": 0.7554545454545455,
      "eval_loss": 0.5462190508842468,
      "eval_runtime": 24.1467,
      "eval_samples_per_second": 45.555,
      "eval_steps_per_second": 5.715,
      "step": 300
    },
    {
      "epoch": 0.7948717948717948,
      "grad_norm": 3.5049915313720703,
      "learning_rate": 0.00016025641025641028,
      "loss": 0.3995,
      "step": 310
    },
    {
      "epoch": 0.8205128205128205,
      "grad_norm": 2.599503517150879,
      "learning_rate": 0.00015897435897435896,
      "loss": 0.3805,
      "step": 320
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.3924107551574707,
      "learning_rate": 0.0001576923076923077,
      "loss": 0.3586,
      "step": 330
    },
    {
      "epoch": 0.8717948717948718,
      "grad_norm": 3.0169456005096436,
      "learning_rate": 0.00015641025641025642,
      "loss": 0.3964,
      "step": 340
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 3.907792091369629,
      "learning_rate": 0.00015512820512820515,
      "loss": 0.3915,
      "step": 350
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 1.2597954273223877,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.3471,
      "step": 360
    },
    {
      "epoch": 0.9487179487179487,
      "grad_norm": 4.323169231414795,
      "learning_rate": 0.00015256410256410255,
      "loss": 0.2476,
      "step": 370
    },
    {
      "epoch": 0.9743589743589743,
      "grad_norm": 3.7260568141937256,
      "learning_rate": 0.00015128205128205128,
      "loss": 0.41,
      "step": 380
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.22700244188308716,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.3773,
      "step": 390
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 1.8512117862701416,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.3554,
      "step": 400
    },
    {
      "epoch": 1.0256410256410255,
      "eval_accuracy": 0.8009090909090909,
      "eval_loss": 0.4354061186313629,
      "eval_runtime": 24.0034,
      "eval_samples_per_second": 45.827,
      "eval_steps_per_second": 5.749,
      "step": 400
    },
    {
      "epoch": 1.0512820512820513,
      "grad_norm": 1.9039109945297241,
      "learning_rate": 0.00014743589743589745,
      "loss": 0.3033,
      "step": 410
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 2.7765700817108154,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.2594,
      "step": 420
    },
    {
      "epoch": 1.1025641025641026,
      "grad_norm": 2.247612953186035,
      "learning_rate": 0.00014487179487179488,
      "loss": 0.3434,
      "step": 430
    },
    {
      "epoch": 1.1282051282051282,
      "grad_norm": 1.161192536354065,
      "learning_rate": 0.0001435897435897436,
      "loss": 0.2653,
      "step": 440
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.9420008659362793,
      "learning_rate": 0.0001423076923076923,
      "loss": 0.3184,
      "step": 450
    },
    {
      "epoch": 1.1794871794871795,
      "grad_norm": 2.359160900115967,
      "learning_rate": 0.00014102564102564104,
      "loss": 0.4049,
      "step": 460
    },
    {
      "epoch": 1.205128205128205,
      "grad_norm": 1.5929157733917236,
      "learning_rate": 0.00013974358974358974,
      "loss": 0.371,
      "step": 470
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 3.8561315536499023,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.2864,
      "step": 480
    },
    {
      "epoch": 1.2564102564102564,
      "grad_norm": 2.111147403717041,
      "learning_rate": 0.00013717948717948718,
      "loss": 0.3173,
      "step": 490
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 1.6825300455093384,
      "learning_rate": 0.0001358974358974359,
      "loss": 0.2368,
      "step": 500
    },
    {
      "epoch": 1.282051282051282,
      "eval_accuracy": 0.8309090909090909,
      "eval_loss": 0.4046396017074585,
      "eval_runtime": 24.1395,
      "eval_samples_per_second": 45.568,
      "eval_steps_per_second": 5.717,
      "step": 500
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 1.7883163690567017,
      "learning_rate": 0.00013461538461538464,
      "loss": 0.2616,
      "step": 510
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 3.3474502563476562,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.292,
      "step": 520
    },
    {
      "epoch": 1.358974358974359,
      "grad_norm": 1.9872941970825195,
      "learning_rate": 0.00013205128205128204,
      "loss": 0.3284,
      "step": 530
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 1.508928894996643,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.3132,
      "step": 540
    },
    {
      "epoch": 1.4102564102564101,
      "grad_norm": 2.3678171634674072,
      "learning_rate": 0.0001294871794871795,
      "loss": 0.2698,
      "step": 550
    },
    {
      "epoch": 1.435897435897436,
      "grad_norm": 3.961099147796631,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.2522,
      "step": 560
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 1.7161500453948975,
      "learning_rate": 0.00012692307692307693,
      "loss": 0.3514,
      "step": 570
    },
    {
      "epoch": 1.4871794871794872,
      "grad_norm": 2.0210063457489014,
      "learning_rate": 0.00012564102564102564,
      "loss": 0.2064,
      "step": 580
    },
    {
      "epoch": 1.5128205128205128,
      "grad_norm": 1.9867080450057983,
      "learning_rate": 0.00012435897435897437,
      "loss": 0.3796,
      "step": 590
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 5.6288981437683105,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.3696,
      "step": 600
    },
    {
      "epoch": 1.5384615384615383,
      "eval_accuracy": 0.7809090909090909,
      "eval_loss": 0.5547047257423401,
      "eval_runtime": 22.8194,
      "eval_samples_per_second": 48.205,
      "eval_steps_per_second": 6.047,
      "step": 600
    },
    {
      "epoch": 1.564102564102564,
      "grad_norm": 2.3821280002593994,
      "learning_rate": 0.00012179487179487179,
      "loss": 0.169,
      "step": 610
    },
    {
      "epoch": 1.5897435897435899,
      "grad_norm": 2.7712907791137695,
      "learning_rate": 0.00012051282051282052,
      "loss": 0.294,
      "step": 620
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 2.803150177001953,
      "learning_rate": 0.00011923076923076923,
      "loss": 0.3544,
      "step": 630
    },
    {
      "epoch": 1.641025641025641,
      "grad_norm": 2.660898447036743,
      "learning_rate": 0.00011794871794871796,
      "loss": 0.2811,
      "step": 640
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 1.4722263813018799,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.26,
      "step": 650
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 2.1373977661132812,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.2282,
      "step": 660
    },
    {
      "epoch": 1.717948717948718,
      "grad_norm": 4.172289848327637,
      "learning_rate": 0.0001141025641025641,
      "loss": 0.2673,
      "step": 670
    },
    {
      "epoch": 1.7435897435897436,
      "grad_norm": 1.547676682472229,
      "learning_rate": 0.00011282051282051283,
      "loss": 0.234,
      "step": 680
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 1.9279000759124756,
      "learning_rate": 0.00011153846153846154,
      "loss": 0.4459,
      "step": 690
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 2.7669806480407715,
      "learning_rate": 0.00011025641025641027,
      "loss": 0.2824,
      "step": 700
    },
    {
      "epoch": 1.7948717948717947,
      "eval_accuracy": 0.8518181818181818,
      "eval_loss": 0.33288896083831787,
      "eval_runtime": 23.8646,
      "eval_samples_per_second": 46.093,
      "eval_steps_per_second": 5.783,
      "step": 700
    },
    {
      "epoch": 1.8205128205128205,
      "grad_norm": 1.626448154449463,
      "learning_rate": 0.00010897435897435896,
      "loss": 0.2789,
      "step": 710
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 2.5008246898651123,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.2939,
      "step": 720
    },
    {
      "epoch": 1.8717948717948718,
      "grad_norm": 1.4484879970550537,
      "learning_rate": 0.00010641025641025641,
      "loss": 0.3107,
      "step": 730
    },
    {
      "epoch": 1.8974358974358974,
      "grad_norm": 2.9797451496124268,
      "learning_rate": 0.00010512820512820514,
      "loss": 0.1989,
      "step": 740
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 2.551682710647583,
      "learning_rate": 0.00010384615384615386,
      "loss": 0.3133,
      "step": 750
    },
    {
      "epoch": 1.9487179487179487,
      "grad_norm": 3.318741798400879,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.2384,
      "step": 760
    },
    {
      "epoch": 1.9743589743589745,
      "grad_norm": 0.9309015274047852,
      "learning_rate": 0.00010128205128205129,
      "loss": 0.2314,
      "step": 770
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.10331392288208,
      "learning_rate": 0.0001,
      "loss": 0.2341,
      "step": 780
    },
    {
      "epoch": 2.0256410256410255,
      "grad_norm": 2.6792232990264893,
      "learning_rate": 9.871794871794872e-05,
      "loss": 0.1841,
      "step": 790
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 2.074448585510254,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.2366,
      "step": 800
    },
    {
      "epoch": 2.051282051282051,
      "eval_accuracy": 0.8254545454545454,
      "eval_loss": 0.45823609828948975,
      "eval_runtime": 24.0195,
      "eval_samples_per_second": 45.796,
      "eval_steps_per_second": 5.745,
      "step": 800
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.9467771649360657,
      "learning_rate": 9.615384615384617e-05,
      "loss": 0.1895,
      "step": 810
    },
    {
      "epoch": 2.1025641025641026,
      "grad_norm": 3.1332082748413086,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.2665,
      "step": 820
    },
    {
      "epoch": 2.128205128205128,
      "grad_norm": 3.9276282787323,
      "learning_rate": 9.35897435897436e-05,
      "loss": 0.2388,
      "step": 830
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 2.7033755779266357,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.1917,
      "step": 840
    },
    {
      "epoch": 2.1794871794871793,
      "grad_norm": 1.5250920057296753,
      "learning_rate": 9.102564102564103e-05,
      "loss": 0.3245,
      "step": 850
    },
    {
      "epoch": 2.2051282051282053,
      "grad_norm": 1.5101457834243774,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.1377,
      "step": 860
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 0.743198573589325,
      "learning_rate": 8.846153846153847e-05,
      "loss": 0.2243,
      "step": 870
    },
    {
      "epoch": 2.2564102564102564,
      "grad_norm": 5.429717540740967,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.2683,
      "step": 880
    },
    {
      "epoch": 2.282051282051282,
      "grad_norm": 2.3276283740997314,
      "learning_rate": 8.58974358974359e-05,
      "loss": 0.2083,
      "step": 890
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 1.3464454412460327,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.2212,
      "step": 900
    },
    {
      "epoch": 2.3076923076923075,
      "eval_accuracy": 0.8254545454545454,
      "eval_loss": 0.4885379374027252,
      "eval_runtime": 24.2062,
      "eval_samples_per_second": 45.443,
      "eval_steps_per_second": 5.701,
      "step": 900
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 2.0947823524475098,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.1469,
      "step": 910
    },
    {
      "epoch": 2.358974358974359,
      "grad_norm": 2.0833053588867188,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.1637,
      "step": 920
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 3.5598833560943604,
      "learning_rate": 8.076923076923078e-05,
      "loss": 0.2127,
      "step": 930
    },
    {
      "epoch": 2.41025641025641,
      "grad_norm": 5.30457878112793,
      "learning_rate": 7.948717948717948e-05,
      "loss": 0.1561,
      "step": 940
    },
    {
      "epoch": 2.435897435897436,
      "grad_norm": 2.163148880004883,
      "learning_rate": 7.820512820512821e-05,
      "loss": 0.1539,
      "step": 950
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 4.815582752227783,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.1802,
      "step": 960
    },
    {
      "epoch": 2.4871794871794872,
      "grad_norm": 1.3422257900238037,
      "learning_rate": 7.564102564102564e-05,
      "loss": 0.2059,
      "step": 970
    },
    {
      "epoch": 2.5128205128205128,
      "grad_norm": 2.441047430038452,
      "learning_rate": 7.435897435897436e-05,
      "loss": 0.2389,
      "step": 980
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 3.386566162109375,
      "learning_rate": 7.307692307692307e-05,
      "loss": 0.1882,
      "step": 990
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 3.123497247695923,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.2031,
      "step": 1000
    },
    {
      "epoch": 2.564102564102564,
      "eval_accuracy": 0.8563636363636363,
      "eval_loss": 0.42818111181259155,
      "eval_runtime": 24.1992,
      "eval_samples_per_second": 45.456,
      "eval_steps_per_second": 5.703,
      "step": 1000
    },
    {
      "epoch": 2.58974358974359,
      "grad_norm": 2.518524646759033,
      "learning_rate": 7.051282051282052e-05,
      "loss": 0.2551,
      "step": 1010
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 2.376194715499878,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.1853,
      "step": 1020
    },
    {
      "epoch": 2.641025641025641,
      "grad_norm": 0.8307498097419739,
      "learning_rate": 6.794871794871795e-05,
      "loss": 0.1643,
      "step": 1030
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 2.050661087036133,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.105,
      "step": 1040
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 2.984266996383667,
      "learning_rate": 6.538461538461539e-05,
      "loss": 0.1774,
      "step": 1050
    },
    {
      "epoch": 2.717948717948718,
      "grad_norm": 3.933162212371826,
      "learning_rate": 6.410256410256412e-05,
      "loss": 0.1079,
      "step": 1060
    },
    {
      "epoch": 2.7435897435897436,
      "grad_norm": 4.650693893432617,
      "learning_rate": 6.282051282051282e-05,
      "loss": 0.1542,
      "step": 1070
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 2.796116828918457,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.4463,
      "step": 1080
    },
    {
      "epoch": 2.7948717948717947,
      "grad_norm": 3.1699883937835693,
      "learning_rate": 6.025641025641026e-05,
      "loss": 0.1348,
      "step": 1090
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 2.0635199546813965,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.1717,
      "step": 1100
    },
    {
      "epoch": 2.8205128205128203,
      "eval_accuracy": 0.85,
      "eval_loss": 0.4373130798339844,
      "eval_runtime": 23.6976,
      "eval_samples_per_second": 46.418,
      "eval_steps_per_second": 5.823,
      "step": 1100
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 3.7501204013824463,
      "learning_rate": 5.769230769230769e-05,
      "loss": 0.1505,
      "step": 1110
    },
    {
      "epoch": 2.871794871794872,
      "grad_norm": 0.866908609867096,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.131,
      "step": 1120
    },
    {
      "epoch": 2.8974358974358974,
      "grad_norm": 2.7631490230560303,
      "learning_rate": 5.512820512820514e-05,
      "loss": 0.1067,
      "step": 1130
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.8835192918777466,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.2449,
      "step": 1140
    },
    {
      "epoch": 2.948717948717949,
      "grad_norm": 0.17269015312194824,
      "learning_rate": 5.256410256410257e-05,
      "loss": 0.1235,
      "step": 1150
    },
    {
      "epoch": 2.9743589743589745,
      "grad_norm": 2.5380775928497314,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.1214,
      "step": 1160
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.240225791931152,
      "learning_rate": 5e-05,
      "loss": 0.2372,
      "step": 1170
    },
    {
      "epoch": 3.0256410256410255,
      "grad_norm": 0.454428106546402,
      "learning_rate": 4.871794871794872e-05,
      "loss": 0.1121,
      "step": 1180
    },
    {
      "epoch": 3.051282051282051,
      "grad_norm": 3.3110735416412354,
      "learning_rate": 4.7435897435897435e-05,
      "loss": 0.1121,
      "step": 1190
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.4833953380584717,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.1303,
      "step": 1200
    },
    {
      "epoch": 3.076923076923077,
      "eval_accuracy": 0.8718181818181818,
      "eval_loss": 0.36585894227027893,
      "eval_runtime": 24.2959,
      "eval_samples_per_second": 45.275,
      "eval_steps_per_second": 5.68,
      "step": 1200
    },
    {
      "epoch": 3.1025641025641026,
      "grad_norm": 0.08100098371505737,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.0911,
      "step": 1210
    },
    {
      "epoch": 3.128205128205128,
      "grad_norm": 0.30585813522338867,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.0834,
      "step": 1220
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 4.129181385040283,
      "learning_rate": 4.230769230769231e-05,
      "loss": 0.1144,
      "step": 1230
    },
    {
      "epoch": 3.1794871794871793,
      "grad_norm": 0.367727667093277,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.0808,
      "step": 1240
    },
    {
      "epoch": 3.2051282051282053,
      "grad_norm": 0.10303868353366852,
      "learning_rate": 3.974358974358974e-05,
      "loss": 0.1758,
      "step": 1250
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 2.300645589828491,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.227,
      "step": 1260
    },
    {
      "epoch": 3.2564102564102564,
      "grad_norm": 1.345780372619629,
      "learning_rate": 3.717948717948718e-05,
      "loss": 0.1345,
      "step": 1270
    },
    {
      "epoch": 3.282051282051282,
      "grad_norm": 2.5391829013824463,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.0496,
      "step": 1280
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 0.31912463903427124,
      "learning_rate": 3.461538461538462e-05,
      "loss": 0.1165,
      "step": 1290
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.5431106686592102,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0889,
      "step": 1300
    },
    {
      "epoch": 3.3333333333333335,
      "eval_accuracy": 0.8736363636363637,
      "eval_loss": 0.3662668764591217,
      "eval_runtime": 23.4444,
      "eval_samples_per_second": 46.92,
      "eval_steps_per_second": 5.886,
      "step": 1300
    },
    {
      "epoch": 3.358974358974359,
      "grad_norm": 2.443268299102783,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.1256,
      "step": 1310
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 2.0804026126861572,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0973,
      "step": 1320
    },
    {
      "epoch": 3.41025641025641,
      "grad_norm": 10.397607803344727,
      "learning_rate": 2.948717948717949e-05,
      "loss": 0.1183,
      "step": 1330
    },
    {
      "epoch": 3.435897435897436,
      "grad_norm": 3.746250867843628,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.046,
      "step": 1340
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 0.7118757367134094,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 0.1611,
      "step": 1350
    },
    {
      "epoch": 3.4871794871794872,
      "grad_norm": 0.34771645069122314,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.1974,
      "step": 1360
    },
    {
      "epoch": 3.5128205128205128,
      "grad_norm": 6.590170860290527,
      "learning_rate": 2.435897435897436e-05,
      "loss": 0.1392,
      "step": 1370
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 3.6979663372039795,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.1153,
      "step": 1380
    },
    {
      "epoch": 3.564102564102564,
      "grad_norm": 0.12197946012020111,
      "learning_rate": 2.1794871794871795e-05,
      "loss": 0.1027,
      "step": 1390
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 2.5246639251708984,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.1157,
      "step": 1400
    },
    {
      "epoch": 3.58974358974359,
      "eval_accuracy": 0.8436363636363636,
      "eval_loss": 0.4587700366973877,
      "eval_runtime": 22.8201,
      "eval_samples_per_second": 48.203,
      "eval_steps_per_second": 6.047,
      "step": 1400
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 0.37446674704551697,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.0839,
      "step": 1410
    },
    {
      "epoch": 3.641025641025641,
      "grad_norm": 0.7361642718315125,
      "learning_rate": 1.794871794871795e-05,
      "loss": 0.0541,
      "step": 1420
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.11162062734365463,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1791,
      "step": 1430
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 3.2151377201080322,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0597,
      "step": 1440
    },
    {
      "epoch": 3.717948717948718,
      "grad_norm": 0.853471040725708,
      "learning_rate": 1.4102564102564104e-05,
      "loss": 0.1246,
      "step": 1450
    },
    {
      "epoch": 3.7435897435897436,
      "grad_norm": 0.2989501953125,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.0747,
      "step": 1460
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 0.4194205403327942,
      "learning_rate": 1.153846153846154e-05,
      "loss": 0.078,
      "step": 1470
    },
    {
      "epoch": 3.7948717948717947,
      "grad_norm": 0.2623525857925415,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.064,
      "step": 1480
    },
    {
      "epoch": 3.8205128205128203,
      "grad_norm": 1.1962109804153442,
      "learning_rate": 8.974358974358976e-06,
      "loss": 0.0955,
      "step": 1490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 2.009432792663574,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.1215,
      "step": 1500
    },
    {
      "epoch": 3.8461538461538463,
      "eval_accuracy": 0.8654545454545455,
      "eval_loss": 0.43503817915916443,
      "eval_runtime": 23.6622,
      "eval_samples_per_second": 46.488,
      "eval_steps_per_second": 5.832,
      "step": 1500
    },
    {
      "epoch": 3.871794871794872,
      "grad_norm": 3.284787178039551,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.0614,
      "step": 1510
    },
    {
      "epoch": 3.8974358974358974,
      "grad_norm": 0.1390266716480255,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.0795,
      "step": 1520
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 3.4633984565734863,
      "learning_rate": 3.846153846153847e-06,
      "loss": 0.1268,
      "step": 1530
    },
    {
      "epoch": 3.948717948717949,
      "grad_norm": 3.78682804107666,
      "learning_rate": 2.564102564102564e-06,
      "loss": 0.1049,
      "step": 1540
    },
    {
      "epoch": 3.9743589743589745,
      "grad_norm": 3.6551170349121094,
      "learning_rate": 1.282051282051282e-06,
      "loss": 0.0924,
      "step": 1550
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1017698347568512,
      "learning_rate": 0.0,
      "loss": 0.1392,
      "step": 1560
    },
    {
      "epoch": 4.0,
      "step": 1560,
      "total_flos": 1.9301704773202575e+18,
      "train_loss": 0.266804637053074,
      "train_runtime": 1157.865,
      "train_samples_per_second": 21.512,
      "train_steps_per_second": 1.347
    }
  ],
  "logging_steps": 10,
  "max_steps": 1560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9301704773202575e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}