{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4782125768967874,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008544087491455912,
      "grad_norm": 11.00854206085205,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.4425,
      "step": 10
    },
    {
      "epoch": 0.0017088174982911825,
      "grad_norm": 4.686764240264893,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 2.1093,
      "step": 20
    },
    {
      "epoch": 0.002563226247436774,
      "grad_norm": 2.6529407501220703,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.7471,
      "step": 30
    },
    {
      "epoch": 0.003417634996582365,
      "grad_norm": 1.363114356994629,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.4501,
      "step": 40
    },
    {
      "epoch": 0.0042720437457279565,
      "grad_norm": 1.054634690284729,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.205,
      "step": 50
    },
    {
      "epoch": 0.005126452494873548,
      "grad_norm": 0.7446718215942383,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.092,
      "step": 60
    },
    {
      "epoch": 0.005980861244019139,
      "grad_norm": 0.7012097835540771,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.0106,
      "step": 70
    },
    {
      "epoch": 0.00683526999316473,
      "grad_norm": 0.6184213757514954,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.9755,
      "step": 80
    },
    {
      "epoch": 0.007689678742310321,
      "grad_norm": 0.6211470365524292,
      "learning_rate": 1.005e-05,
      "loss": 1.0031,
      "step": 90
    },
    {
      "epoch": 0.008544087491455913,
      "grad_norm": 1.4690356254577637,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.9412,
      "step": 100
    },
    {
      "epoch": 0.009398496240601503,
      "grad_norm": 0.6179498434066772,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.9378,
      "step": 110
    },
    {
      "epoch": 0.010252904989747095,
      "grad_norm": 0.651823103427887,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.9391,
      "step": 120
    },
    {
      "epoch": 0.011107313738892685,
      "grad_norm": 0.6029787063598633,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.9291,
      "step": 130
    },
    {
      "epoch": 0.011961722488038277,
      "grad_norm": 0.5876877307891846,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.9231,
      "step": 140
    },
    {
      "epoch": 0.01281613123718387,
      "grad_norm": 0.5466436147689819,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.9285,
      "step": 150
    },
    {
      "epoch": 0.01367053998632946,
      "grad_norm": 0.5822666883468628,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.9159,
      "step": 160
    },
    {
      "epoch": 0.014524948735475052,
      "grad_norm": 0.5883556604385376,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.911,
      "step": 170
    },
    {
      "epoch": 0.015379357484620642,
      "grad_norm": 0.6157915592193604,
      "learning_rate": 2e-05,
      "loss": 0.8931,
      "step": 180
    },
    {
      "epoch": 0.016233766233766232,
      "grad_norm": 0.64449143409729,
      "learning_rate": 2e-05,
      "loss": 0.9024,
      "step": 190
    },
    {
      "epoch": 0.017088174982911826,
      "grad_norm": 0.5318751931190491,
      "learning_rate": 2e-05,
      "loss": 0.9074,
      "step": 200
    },
    {
      "epoch": 0.017942583732057416,
      "grad_norm": 0.5434000492095947,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 210
    },
    {
      "epoch": 0.018796992481203006,
      "grad_norm": 0.5382893681526184,
      "learning_rate": 2e-05,
      "loss": 0.8747,
      "step": 220
    },
    {
      "epoch": 0.0196514012303486,
      "grad_norm": 0.5032609701156616,
      "learning_rate": 2e-05,
      "loss": 0.8862,
      "step": 230
    },
    {
      "epoch": 0.02050580997949419,
      "grad_norm": 0.543346643447876,
      "learning_rate": 2e-05,
      "loss": 0.8872,
      "step": 240
    },
    {
      "epoch": 0.02136021872863978,
      "grad_norm": 0.527274489402771,
      "learning_rate": 2e-05,
      "loss": 0.8818,
      "step": 250
    },
    {
      "epoch": 0.02221462747778537,
      "grad_norm": 0.5765458345413208,
      "learning_rate": 2e-05,
      "loss": 0.8687,
      "step": 260
    },
    {
      "epoch": 0.023069036226930965,
      "grad_norm": 0.5382829904556274,
      "learning_rate": 2e-05,
      "loss": 0.8915,
      "step": 270
    },
    {
      "epoch": 0.023923444976076555,
      "grad_norm": 0.5533940196037292,
      "learning_rate": 2e-05,
      "loss": 0.8891,
      "step": 280
    },
    {
      "epoch": 0.024777853725222145,
      "grad_norm": 0.5417444705963135,
      "learning_rate": 2e-05,
      "loss": 0.8847,
      "step": 290
    },
    {
      "epoch": 0.02563226247436774,
      "grad_norm": 0.5582782626152039,
      "learning_rate": 2e-05,
      "loss": 0.9424,
      "step": 300
    },
    {
      "epoch": 0.02648667122351333,
      "grad_norm": 0.5055362582206726,
      "learning_rate": 2e-05,
      "loss": 0.8781,
      "step": 310
    },
    {
      "epoch": 0.02734107997265892,
      "grad_norm": 0.5508922934532166,
      "learning_rate": 2e-05,
      "loss": 0.8775,
      "step": 320
    },
    {
      "epoch": 0.02819548872180451,
      "grad_norm": 0.5282697677612305,
      "learning_rate": 2e-05,
      "loss": 0.8627,
      "step": 330
    },
    {
      "epoch": 0.029049897470950103,
      "grad_norm": 0.5219830870628357,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 340
    },
    {
      "epoch": 0.029904306220095694,
      "grad_norm": 0.5210185050964355,
      "learning_rate": 2e-05,
      "loss": 0.9101,
      "step": 350
    },
    {
      "epoch": 0.030758714969241284,
      "grad_norm": 0.5474709272384644,
      "learning_rate": 2e-05,
      "loss": 0.8914,
      "step": 360
    },
    {
      "epoch": 0.03161312371838688,
      "grad_norm": 0.5275512337684631,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 370
    },
    {
      "epoch": 0.032467532467532464,
      "grad_norm": 0.5367026925086975,
      "learning_rate": 2e-05,
      "loss": 0.8771,
      "step": 380
    },
    {
      "epoch": 0.03332194121667806,
      "grad_norm": 0.5371636152267456,
      "learning_rate": 2e-05,
      "loss": 0.88,
      "step": 390
    },
    {
      "epoch": 0.03417634996582365,
      "grad_norm": 0.5706266760826111,
      "learning_rate": 2e-05,
      "loss": 0.8772,
      "step": 400
    },
    {
      "epoch": 0.03503075871496924,
      "grad_norm": 0.5139058232307434,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 410
    },
    {
      "epoch": 0.03588516746411483,
      "grad_norm": 0.5124853849411011,
      "learning_rate": 2e-05,
      "loss": 0.8585,
      "step": 420
    },
    {
      "epoch": 0.036739576213260426,
      "grad_norm": 0.4997583329677582,
      "learning_rate": 2e-05,
      "loss": 0.8638,
      "step": 430
    },
    {
      "epoch": 0.03759398496240601,
      "grad_norm": 0.5006685853004456,
      "learning_rate": 2e-05,
      "loss": 0.8651,
      "step": 440
    },
    {
      "epoch": 0.03844839371155161,
      "grad_norm": 0.5022147297859192,
      "learning_rate": 2e-05,
      "loss": 0.856,
      "step": 450
    },
    {
      "epoch": 0.0393028024606972,
      "grad_norm": 0.5060321688652039,
      "learning_rate": 2e-05,
      "loss": 0.8772,
      "step": 460
    },
    {
      "epoch": 0.04015721120984279,
      "grad_norm": 0.5752074718475342,
      "learning_rate": 2e-05,
      "loss": 0.8583,
      "step": 470
    },
    {
      "epoch": 0.04101161995898838,
      "grad_norm": 0.5028130412101746,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 480
    },
    {
      "epoch": 0.041866028708133975,
      "grad_norm": 0.46682220697402954,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 490
    },
    {
      "epoch": 0.04272043745727956,
      "grad_norm": 0.4957294464111328,
      "learning_rate": 2e-05,
      "loss": 0.8593,
      "step": 500
    },
    {
      "epoch": 0.043574846206425155,
      "grad_norm": 0.5031411051750183,
      "learning_rate": 2e-05,
      "loss": 0.8823,
      "step": 510
    },
    {
      "epoch": 0.04442925495557074,
      "grad_norm": 0.572361171245575,
      "learning_rate": 2e-05,
      "loss": 0.8664,
      "step": 520
    },
    {
      "epoch": 0.045283663704716336,
      "grad_norm": 0.4737163484096527,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 530
    },
    {
      "epoch": 0.04613807245386193,
      "grad_norm": 0.5212064981460571,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 540
    },
    {
      "epoch": 0.046992481203007516,
      "grad_norm": 0.4948560297489166,
      "learning_rate": 2e-05,
      "loss": 0.8636,
      "step": 550
    },
    {
      "epoch": 0.04784688995215311,
      "grad_norm": 0.5002424716949463,
      "learning_rate": 2e-05,
      "loss": 0.8632,
      "step": 560
    },
    {
      "epoch": 0.048701298701298704,
      "grad_norm": 0.6509451270103455,
      "learning_rate": 2e-05,
      "loss": 0.8552,
      "step": 570
    },
    {
      "epoch": 0.04955570745044429,
      "grad_norm": 0.4619647264480591,
      "learning_rate": 2e-05,
      "loss": 0.852,
      "step": 580
    },
    {
      "epoch": 0.050410116199589884,
      "grad_norm": 0.5018244385719299,
      "learning_rate": 2e-05,
      "loss": 0.8336,
      "step": 590
    },
    {
      "epoch": 0.05126452494873548,
      "grad_norm": 0.47025132179260254,
      "learning_rate": 2e-05,
      "loss": 0.839,
      "step": 600
    },
    {
      "epoch": 0.052118933697881065,
      "grad_norm": 0.5157033801078796,
      "learning_rate": 2e-05,
      "loss": 0.8513,
      "step": 610
    },
    {
      "epoch": 0.05297334244702666,
      "grad_norm": 0.506414532661438,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 620
    },
    {
      "epoch": 0.05382775119617225,
      "grad_norm": 0.48786959052085876,
      "learning_rate": 2e-05,
      "loss": 0.8747,
      "step": 630
    },
    {
      "epoch": 0.05468215994531784,
      "grad_norm": 0.4625915586948395,
      "learning_rate": 2e-05,
      "loss": 0.8321,
      "step": 640
    },
    {
      "epoch": 0.05553656869446343,
      "grad_norm": 0.49463289976119995,
      "learning_rate": 2e-05,
      "loss": 0.8449,
      "step": 650
    },
    {
      "epoch": 0.05639097744360902,
      "grad_norm": 0.47395265102386475,
      "learning_rate": 2e-05,
      "loss": 0.867,
      "step": 660
    },
    {
      "epoch": 0.05724538619275461,
      "grad_norm": 0.49984267354011536,
      "learning_rate": 2e-05,
      "loss": 0.8622,
      "step": 670
    },
    {
      "epoch": 0.05809979494190021,
      "grad_norm": 0.47769609093666077,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 680
    },
    {
      "epoch": 0.058954203691045794,
      "grad_norm": 0.4874252676963806,
      "learning_rate": 2e-05,
      "loss": 0.8662,
      "step": 690
    },
    {
      "epoch": 0.05980861244019139,
      "grad_norm": 0.4563353955745697,
      "learning_rate": 2e-05,
      "loss": 0.848,
      "step": 700
    },
    {
      "epoch": 0.06066302118933698,
      "grad_norm": 0.461029589176178,
      "learning_rate": 2e-05,
      "loss": 0.8245,
      "step": 710
    },
    {
      "epoch": 0.06151742993848257,
      "grad_norm": 0.4584606885910034,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 720
    },
    {
      "epoch": 0.06237183868762816,
      "grad_norm": 0.5101044774055481,
      "learning_rate": 2e-05,
      "loss": 0.8361,
      "step": 730
    },
    {
      "epoch": 0.06322624743677376,
      "grad_norm": 0.4820401668548584,
      "learning_rate": 2e-05,
      "loss": 0.8452,
      "step": 740
    },
    {
      "epoch": 0.06408065618591935,
      "grad_norm": 0.49986669421195984,
      "learning_rate": 2e-05,
      "loss": 0.8406,
      "step": 750
    },
    {
      "epoch": 0.06493506493506493,
      "grad_norm": 0.4644050598144531,
      "learning_rate": 2e-05,
      "loss": 0.7966,
      "step": 760
    },
    {
      "epoch": 0.06578947368421052,
      "grad_norm": 0.4689406752586365,
      "learning_rate": 2e-05,
      "loss": 0.8434,
      "step": 770
    },
    {
      "epoch": 0.06664388243335612,
      "grad_norm": 0.4809776544570923,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 780
    },
    {
      "epoch": 0.06749829118250171,
      "grad_norm": 0.4512708783149719,
      "learning_rate": 2e-05,
      "loss": 0.8581,
      "step": 790
    },
    {
      "epoch": 0.0683526999316473,
      "grad_norm": 0.45067834854125977,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 800
    },
    {
      "epoch": 0.0692071086807929,
      "grad_norm": 0.45861950516700745,
      "learning_rate": 2e-05,
      "loss": 0.8213,
      "step": 810
    },
    {
      "epoch": 0.07006151742993848,
      "grad_norm": 0.8090688586235046,
      "learning_rate": 2e-05,
      "loss": 0.8438,
      "step": 820
    },
    {
      "epoch": 0.07091592617908407,
      "grad_norm": 0.47155073285102844,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 830
    },
    {
      "epoch": 0.07177033492822966,
      "grad_norm": 0.494607150554657,
      "learning_rate": 2e-05,
      "loss": 0.8342,
      "step": 840
    },
    {
      "epoch": 0.07262474367737526,
      "grad_norm": 0.4732503592967987,
      "learning_rate": 2e-05,
      "loss": 0.8551,
      "step": 850
    },
    {
      "epoch": 0.07347915242652085,
      "grad_norm": 0.45871812105178833,
      "learning_rate": 2e-05,
      "loss": 0.8466,
      "step": 860
    },
    {
      "epoch": 0.07433356117566643,
      "grad_norm": 0.4696637690067291,
      "learning_rate": 2e-05,
      "loss": 0.8548,
      "step": 870
    },
    {
      "epoch": 0.07518796992481203,
      "grad_norm": 0.49322789907455444,
      "learning_rate": 2e-05,
      "loss": 0.8635,
      "step": 880
    },
    {
      "epoch": 0.07604237867395762,
      "grad_norm": 0.4646133780479431,
      "learning_rate": 2e-05,
      "loss": 0.8431,
      "step": 890
    },
    {
      "epoch": 0.07689678742310321,
      "grad_norm": 0.4524765908718109,
      "learning_rate": 2e-05,
      "loss": 0.8164,
      "step": 900
    },
    {
      "epoch": 0.07775119617224881,
      "grad_norm": 0.4888342022895813,
      "learning_rate": 2e-05,
      "loss": 0.8548,
      "step": 910
    },
    {
      "epoch": 0.0786056049213944,
      "grad_norm": 0.43915030360221863,
      "learning_rate": 2e-05,
      "loss": 0.812,
      "step": 920
    },
    {
      "epoch": 0.07946001367053998,
      "grad_norm": 0.4614696800708771,
      "learning_rate": 2e-05,
      "loss": 0.8156,
      "step": 930
    },
    {
      "epoch": 0.08031442241968557,
      "grad_norm": 0.46576476097106934,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 940
    },
    {
      "epoch": 0.08116883116883117,
      "grad_norm": 0.44493210315704346,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 950
    },
    {
      "epoch": 0.08202323991797676,
      "grad_norm": 0.4456934928894043,
      "learning_rate": 2e-05,
      "loss": 0.8421,
      "step": 960
    },
    {
      "epoch": 0.08287764866712236,
      "grad_norm": 0.5433761477470398,
      "learning_rate": 2e-05,
      "loss": 0.87,
      "step": 970
    },
    {
      "epoch": 0.08373205741626795,
      "grad_norm": 0.4579722285270691,
      "learning_rate": 2e-05,
      "loss": 0.8216,
      "step": 980
    },
    {
      "epoch": 0.08458646616541353,
      "grad_norm": 0.452722430229187,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 990
    },
    {
      "epoch": 0.08544087491455912,
      "grad_norm": 0.431753545999527,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 1000
    },
    {
      "epoch": 0.08629528366370472,
      "grad_norm": 0.5925270915031433,
      "learning_rate": 2e-05,
      "loss": 0.8105,
      "step": 1010
    },
    {
      "epoch": 0.08714969241285031,
      "grad_norm": 0.44082966446876526,
      "learning_rate": 2e-05,
      "loss": 0.8191,
      "step": 1020
    },
    {
      "epoch": 0.0880041011619959,
      "grad_norm": 0.47487062215805054,
      "learning_rate": 2e-05,
      "loss": 0.8214,
      "step": 1030
    },
    {
      "epoch": 0.08885850991114148,
      "grad_norm": 0.49121612310409546,
      "learning_rate": 2e-05,
      "loss": 0.8482,
      "step": 1040
    },
    {
      "epoch": 0.08971291866028708,
      "grad_norm": 0.45580044388771057,
      "learning_rate": 2e-05,
      "loss": 0.8376,
      "step": 1050
    },
    {
      "epoch": 0.09056732740943267,
      "grad_norm": 0.4568115174770355,
      "learning_rate": 2e-05,
      "loss": 0.8573,
      "step": 1060
    },
    {
      "epoch": 0.09142173615857827,
      "grad_norm": 0.4428640604019165,
      "learning_rate": 2e-05,
      "loss": 0.8159,
      "step": 1070
    },
    {
      "epoch": 0.09227614490772386,
      "grad_norm": 0.4813441038131714,
      "learning_rate": 2e-05,
      "loss": 0.8493,
      "step": 1080
    },
    {
      "epoch": 0.09313055365686945,
      "grad_norm": 0.4206986129283905,
      "learning_rate": 2e-05,
      "loss": 0.8088,
      "step": 1090
    },
    {
      "epoch": 0.09398496240601503,
      "grad_norm": 0.49628499150276184,
      "learning_rate": 2e-05,
      "loss": 0.815,
      "step": 1100
    },
    {
      "epoch": 0.09483937115516063,
      "grad_norm": 0.4241103529930115,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 1110
    },
    {
      "epoch": 0.09569377990430622,
      "grad_norm": 0.4558090567588806,
      "learning_rate": 2e-05,
      "loss": 0.8052,
      "step": 1120
    },
    {
      "epoch": 0.09654818865345181,
      "grad_norm": 0.4508516192436218,
      "learning_rate": 2e-05,
      "loss": 0.82,
      "step": 1130
    },
    {
      "epoch": 0.09740259740259741,
      "grad_norm": 0.4674357771873474,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 1140
    },
    {
      "epoch": 0.09825700615174299,
      "grad_norm": 0.46306923031806946,
      "learning_rate": 2e-05,
      "loss": 0.8521,
      "step": 1150
    },
    {
      "epoch": 0.09911141490088858,
      "grad_norm": 0.43954190611839294,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 1160
    },
    {
      "epoch": 0.09996582365003417,
      "grad_norm": 0.4370790421962738,
      "learning_rate": 2e-05,
      "loss": 0.8143,
      "step": 1170
    },
    {
      "epoch": 0.10082023239917977,
      "grad_norm": 0.4371121823787689,
      "learning_rate": 2e-05,
      "loss": 0.836,
      "step": 1180
    },
    {
      "epoch": 0.10167464114832536,
      "grad_norm": 0.45741012692451477,
      "learning_rate": 2e-05,
      "loss": 0.83,
      "step": 1190
    },
    {
      "epoch": 0.10252904989747096,
      "grad_norm": 0.4397066831588745,
      "learning_rate": 2e-05,
      "loss": 0.8045,
      "step": 1200
    },
    {
      "epoch": 0.10338345864661654,
      "grad_norm": 0.45700281858444214,
      "learning_rate": 2e-05,
      "loss": 0.812,
      "step": 1210
    },
    {
      "epoch": 0.10423786739576213,
      "grad_norm": 0.4776708781719208,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 1220
    },
    {
      "epoch": 0.10509227614490772,
      "grad_norm": 0.4925238788127899,
      "learning_rate": 2e-05,
      "loss": 0.806,
      "step": 1230
    },
    {
      "epoch": 0.10594668489405332,
      "grad_norm": 0.4427919089794159,
      "learning_rate": 2e-05,
      "loss": 0.8161,
      "step": 1240
    },
    {
      "epoch": 0.10680109364319891,
      "grad_norm": 0.45902568101882935,
      "learning_rate": 2e-05,
      "loss": 0.8254,
      "step": 1250
    },
    {
      "epoch": 0.1076555023923445,
      "grad_norm": 0.4676574468612671,
      "learning_rate": 2e-05,
      "loss": 0.8269,
      "step": 1260
    },
    {
      "epoch": 0.10850991114149008,
      "grad_norm": 0.443413645029068,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 1270
    },
    {
      "epoch": 0.10936431989063568,
      "grad_norm": 0.43236881494522095,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 1280
    },
    {
      "epoch": 0.11021872863978127,
      "grad_norm": 0.4279523193836212,
      "learning_rate": 2e-05,
      "loss": 0.7924,
      "step": 1290
    },
    {
      "epoch": 0.11107313738892687,
      "grad_norm": 0.4617557227611542,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 1300
    },
    {
      "epoch": 0.11192754613807246,
      "grad_norm": 0.4349711537361145,
      "learning_rate": 2e-05,
      "loss": 0.8275,
      "step": 1310
    },
    {
      "epoch": 0.11278195488721804,
      "grad_norm": 0.46123847365379333,
      "learning_rate": 2e-05,
      "loss": 0.808,
      "step": 1320
    },
    {
      "epoch": 0.11363636363636363,
      "grad_norm": 0.46817174553871155,
      "learning_rate": 2e-05,
      "loss": 0.8134,
      "step": 1330
    },
    {
      "epoch": 0.11449077238550923,
      "grad_norm": 0.43027958273887634,
      "learning_rate": 2e-05,
      "loss": 0.8239,
      "step": 1340
    },
    {
      "epoch": 0.11534518113465482,
      "grad_norm": 0.4255163371562958,
      "learning_rate": 2e-05,
      "loss": 0.8354,
      "step": 1350
    },
    {
      "epoch": 0.11619958988380041,
      "grad_norm": 0.42436283826828003,
      "learning_rate": 2e-05,
      "loss": 0.8334,
      "step": 1360
    },
    {
      "epoch": 0.11705399863294601,
      "grad_norm": 0.41989049315452576,
      "learning_rate": 2e-05,
      "loss": 0.8014,
      "step": 1370
    },
    {
      "epoch": 0.11790840738209159,
      "grad_norm": 0.4478135406970978,
      "learning_rate": 2e-05,
      "loss": 0.8041,
      "step": 1380
    },
    {
      "epoch": 0.11876281613123718,
      "grad_norm": 0.45139527320861816,
      "learning_rate": 2e-05,
      "loss": 0.7863,
      "step": 1390
    },
    {
      "epoch": 0.11961722488038277,
      "grad_norm": 0.5576887726783752,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 1400
    },
    {
      "epoch": 0.12047163362952837,
      "grad_norm": 0.5432232618331909,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 1410
    },
    {
      "epoch": 0.12132604237867396,
      "grad_norm": 0.4304180145263672,
      "learning_rate": 2e-05,
      "loss": 0.8182,
      "step": 1420
    },
    {
      "epoch": 0.12218045112781954,
      "grad_norm": 0.4250909388065338,
      "learning_rate": 2e-05,
      "loss": 0.7976,
      "step": 1430
    },
    {
      "epoch": 0.12303485987696514,
      "grad_norm": 0.4376416802406311,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 1440
    },
    {
      "epoch": 0.12388926862611073,
      "grad_norm": 0.5384215712547302,
      "learning_rate": 2e-05,
      "loss": 0.8283,
      "step": 1450
    },
    {
      "epoch": 0.12474367737525632,
      "grad_norm": 0.44483479857444763,
      "learning_rate": 2e-05,
      "loss": 0.82,
      "step": 1460
    },
    {
      "epoch": 0.1255980861244019,
      "grad_norm": 0.4774000942707062,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 1470
    },
    {
      "epoch": 0.1264524948735475,
      "grad_norm": 0.45032012462615967,
      "learning_rate": 2e-05,
      "loss": 0.8084,
      "step": 1480
    },
    {
      "epoch": 0.1273069036226931,
      "grad_norm": 0.43089014291763306,
      "learning_rate": 2e-05,
      "loss": 0.8098,
      "step": 1490
    },
    {
      "epoch": 0.1281613123718387,
      "grad_norm": 0.44045063853263855,
      "learning_rate": 2e-05,
      "loss": 0.802,
      "step": 1500
    },
    {
      "epoch": 0.12901572112098428,
      "grad_norm": 0.441500723361969,
      "learning_rate": 2e-05,
      "loss": 0.8227,
      "step": 1510
    },
    {
      "epoch": 0.12987012987012986,
      "grad_norm": 0.43266189098358154,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 1520
    },
    {
      "epoch": 0.13072453861927547,
      "grad_norm": 0.4659614562988281,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 1530
    },
    {
      "epoch": 0.13157894736842105,
      "grad_norm": 0.42092642188072205,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 1540
    },
    {
      "epoch": 0.13243335611756665,
      "grad_norm": 0.44579121470451355,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 1550
    },
    {
      "epoch": 0.13328776486671223,
      "grad_norm": 0.6033672094345093,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 1560
    },
    {
      "epoch": 0.1341421736158578,
      "grad_norm": 0.44670525193214417,
      "learning_rate": 2e-05,
      "loss": 0.7947,
      "step": 1570
    },
    {
      "epoch": 0.13499658236500342,
      "grad_norm": 0.45696425437927246,
      "learning_rate": 2e-05,
      "loss": 0.8021,
      "step": 1580
    },
    {
      "epoch": 0.135850991114149,
      "grad_norm": 0.4262278079986572,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 1590
    },
    {
      "epoch": 0.1367053998632946,
      "grad_norm": 0.4282379746437073,
      "learning_rate": 2e-05,
      "loss": 0.8398,
      "step": 1600
    },
    {
      "epoch": 0.1375598086124402,
      "grad_norm": 0.4352816343307495,
      "learning_rate": 2e-05,
      "loss": 0.7778,
      "step": 1610
    },
    {
      "epoch": 0.1384142173615858,
      "grad_norm": 0.44685494899749756,
      "learning_rate": 2e-05,
      "loss": 0.788,
      "step": 1620
    },
    {
      "epoch": 0.13926862611073137,
      "grad_norm": 0.43920254707336426,
      "learning_rate": 2e-05,
      "loss": 0.8238,
      "step": 1630
    },
    {
      "epoch": 0.14012303485987695,
      "grad_norm": 0.4376579821109772,
      "learning_rate": 2e-05,
      "loss": 0.8061,
      "step": 1640
    },
    {
      "epoch": 0.14097744360902256,
      "grad_norm": 0.43427619338035583,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 1650
    },
    {
      "epoch": 0.14183185235816814,
      "grad_norm": 0.478322833776474,
      "learning_rate": 2e-05,
      "loss": 0.8192,
      "step": 1660
    },
    {
      "epoch": 0.14268626110731375,
      "grad_norm": 0.43004366755485535,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 1670
    },
    {
      "epoch": 0.14354066985645933,
      "grad_norm": 0.45207637548446655,
      "learning_rate": 2e-05,
      "loss": 0.8166,
      "step": 1680
    },
    {
      "epoch": 0.1443950786056049,
      "grad_norm": 0.4198301434516907,
      "learning_rate": 2e-05,
      "loss": 0.8011,
      "step": 1690
    },
    {
      "epoch": 0.14524948735475052,
      "grad_norm": 0.4201711118221283,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 1700
    },
    {
      "epoch": 0.1461038961038961,
      "grad_norm": 0.4392196536064148,
      "learning_rate": 2e-05,
      "loss": 0.7841,
      "step": 1710
    },
    {
      "epoch": 0.1469583048530417,
      "grad_norm": 0.43777990341186523,
      "learning_rate": 2e-05,
      "loss": 0.7922,
      "step": 1720
    },
    {
      "epoch": 0.14781271360218728,
      "grad_norm": 0.41786715388298035,
      "learning_rate": 2e-05,
      "loss": 0.8177,
      "step": 1730
    },
    {
      "epoch": 0.14866712235133286,
      "grad_norm": 0.45031750202178955,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 1740
    },
    {
      "epoch": 0.14952153110047847,
      "grad_norm": 0.4286364018917084,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 1750
    },
    {
      "epoch": 0.15037593984962405,
      "grad_norm": 0.4539773166179657,
      "learning_rate": 2e-05,
      "loss": 0.8298,
      "step": 1760
    },
    {
      "epoch": 0.15123034859876966,
      "grad_norm": 0.41946274042129517,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 1770
    },
    {
      "epoch": 0.15208475734791524,
      "grad_norm": 0.44184574484825134,
      "learning_rate": 2e-05,
      "loss": 0.8141,
      "step": 1780
    },
    {
      "epoch": 0.15293916609706085,
      "grad_norm": 0.453799843788147,
      "learning_rate": 2e-05,
      "loss": 0.8356,
      "step": 1790
    },
    {
      "epoch": 0.15379357484620643,
      "grad_norm": 0.46639925241470337,
      "learning_rate": 2e-05,
      "loss": 0.7995,
      "step": 1800
    },
    {
      "epoch": 0.154647983595352,
      "grad_norm": 0.42801088094711304,
      "learning_rate": 2e-05,
      "loss": 0.7778,
      "step": 1810
    },
    {
      "epoch": 0.15550239234449761,
      "grad_norm": 0.4495405852794647,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 1820
    },
    {
      "epoch": 0.1563568010936432,
      "grad_norm": 0.4354593753814697,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 1830
    },
    {
      "epoch": 0.1572112098427888,
      "grad_norm": 0.4067724347114563,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 1840
    },
    {
      "epoch": 0.15806561859193438,
      "grad_norm": 0.4517872631549835,
      "learning_rate": 2e-05,
      "loss": 0.779,
      "step": 1850
    },
    {
      "epoch": 0.15892002734107996,
      "grad_norm": 0.4226350784301758,
      "learning_rate": 2e-05,
      "loss": 0.7816,
      "step": 1860
    },
    {
      "epoch": 0.15977443609022557,
      "grad_norm": 0.43251317739486694,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 1870
    },
    {
      "epoch": 0.16062884483937115,
      "grad_norm": 0.4463696777820587,
      "learning_rate": 2e-05,
      "loss": 0.816,
      "step": 1880
    },
    {
      "epoch": 0.16148325358851676,
      "grad_norm": 0.41246628761291504,
      "learning_rate": 2e-05,
      "loss": 0.7909,
      "step": 1890
    },
    {
      "epoch": 0.16233766233766234,
      "grad_norm": 0.4619990885257721,
      "learning_rate": 2e-05,
      "loss": 0.794,
      "step": 1900
    },
    {
      "epoch": 0.16319207108680792,
      "grad_norm": 0.6987309455871582,
      "learning_rate": 2e-05,
      "loss": 0.8339,
      "step": 1910
    },
    {
      "epoch": 0.16404647983595352,
      "grad_norm": 0.4621772766113281,
      "learning_rate": 2e-05,
      "loss": 0.8005,
      "step": 1920
    },
    {
      "epoch": 0.1649008885850991,
      "grad_norm": 0.4086742401123047,
      "learning_rate": 2e-05,
      "loss": 0.7964,
      "step": 1930
    },
    {
      "epoch": 0.1657552973342447,
      "grad_norm": 0.4323413670063019,
      "learning_rate": 2e-05,
      "loss": 0.7759,
      "step": 1940
    },
    {
      "epoch": 0.1666097060833903,
      "grad_norm": 0.4169401526451111,
      "learning_rate": 2e-05,
      "loss": 0.7965,
      "step": 1950
    },
    {
      "epoch": 0.1674641148325359,
      "grad_norm": 0.44284898042678833,
      "learning_rate": 2e-05,
      "loss": 0.8008,
      "step": 1960
    },
    {
      "epoch": 0.16831852358168148,
      "grad_norm": 0.44360917806625366,
      "learning_rate": 2e-05,
      "loss": 0.8244,
      "step": 1970
    },
    {
      "epoch": 0.16917293233082706,
      "grad_norm": 0.43291980028152466,
      "learning_rate": 2e-05,
      "loss": 0.7864,
      "step": 1980
    },
    {
      "epoch": 0.17002734107997267,
      "grad_norm": 0.43684226274490356,
      "learning_rate": 2e-05,
      "loss": 0.8172,
      "step": 1990
    },
    {
      "epoch": 0.17088174982911825,
      "grad_norm": 0.4126867651939392,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 2000
    },
    {
      "epoch": 0.17173615857826385,
      "grad_norm": 0.4156489968299866,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 2010
    },
    {
      "epoch": 0.17259056732740943,
      "grad_norm": 0.4190804362297058,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 2020
    },
    {
      "epoch": 0.173444976076555,
      "grad_norm": 0.4342236816883087,
      "learning_rate": 2e-05,
      "loss": 0.7805,
      "step": 2030
    },
    {
      "epoch": 0.17429938482570062,
      "grad_norm": 0.4375854730606079,
      "learning_rate": 2e-05,
      "loss": 0.8016,
      "step": 2040
    },
    {
      "epoch": 0.1751537935748462,
      "grad_norm": 0.43170157074928284,
      "learning_rate": 2e-05,
      "loss": 0.7991,
      "step": 2050
    },
    {
      "epoch": 0.1760082023239918,
      "grad_norm": 0.4225408136844635,
      "learning_rate": 2e-05,
      "loss": 0.8297,
      "step": 2060
    },
    {
      "epoch": 0.1768626110731374,
      "grad_norm": 0.4088500142097473,
      "learning_rate": 2e-05,
      "loss": 0.8231,
      "step": 2070
    },
    {
      "epoch": 0.17771701982228297,
      "grad_norm": 0.39183762669563293,
      "learning_rate": 2e-05,
      "loss": 0.7654,
      "step": 2080
    },
    {
      "epoch": 0.17857142857142858,
      "grad_norm": 0.4299352169036865,
      "learning_rate": 2e-05,
      "loss": 0.7944,
      "step": 2090
    },
    {
      "epoch": 0.17942583732057416,
      "grad_norm": 0.41311776638031006,
      "learning_rate": 2e-05,
      "loss": 0.8065,
      "step": 2100
    },
    {
      "epoch": 0.18028024606971976,
      "grad_norm": 0.4197043180465698,
      "learning_rate": 2e-05,
      "loss": 0.7967,
      "step": 2110
    },
    {
      "epoch": 0.18113465481886534,
      "grad_norm": 0.4080440104007721,
      "learning_rate": 2e-05,
      "loss": 0.7941,
      "step": 2120
    },
    {
      "epoch": 0.18198906356801095,
      "grad_norm": 0.4393218159675598,
      "learning_rate": 2e-05,
      "loss": 0.7926,
      "step": 2130
    },
    {
      "epoch": 0.18284347231715653,
      "grad_norm": 0.4172094464302063,
      "learning_rate": 2e-05,
      "loss": 0.7831,
      "step": 2140
    },
    {
      "epoch": 0.1836978810663021,
      "grad_norm": 0.4084227979183197,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 2150
    },
    {
      "epoch": 0.18455228981544772,
      "grad_norm": 0.4330205023288727,
      "learning_rate": 2e-05,
      "loss": 0.7875,
      "step": 2160
    },
    {
      "epoch": 0.1854066985645933,
      "grad_norm": 0.3870986998081207,
      "learning_rate": 2e-05,
      "loss": 0.7813,
      "step": 2170
    },
    {
      "epoch": 0.1862611073137389,
      "grad_norm": 0.4238833785057068,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 2180
    },
    {
      "epoch": 0.18711551606288448,
      "grad_norm": 0.4221448004245758,
      "learning_rate": 2e-05,
      "loss": 0.8056,
      "step": 2190
    },
    {
      "epoch": 0.18796992481203006,
      "grad_norm": 0.4315853416919708,
      "learning_rate": 2e-05,
      "loss": 0.7838,
      "step": 2200
    },
    {
      "epoch": 0.18882433356117567,
      "grad_norm": 0.4521797001361847,
      "learning_rate": 2e-05,
      "loss": 0.798,
      "step": 2210
    },
    {
      "epoch": 0.18967874231032125,
      "grad_norm": 0.42917776107788086,
      "learning_rate": 2e-05,
      "loss": 0.8113,
      "step": 2220
    },
    {
      "epoch": 0.19053315105946686,
      "grad_norm": 0.4716418385505676,
      "learning_rate": 2e-05,
      "loss": 0.8062,
      "step": 2230
    },
    {
      "epoch": 0.19138755980861244,
      "grad_norm": 0.4381401836872101,
      "learning_rate": 2e-05,
      "loss": 0.7671,
      "step": 2240
    },
    {
      "epoch": 0.19224196855775802,
      "grad_norm": 0.42370718717575073,
      "learning_rate": 2e-05,
      "loss": 0.7853,
      "step": 2250
    },
    {
      "epoch": 0.19309637730690363,
      "grad_norm": 0.4564455449581146,
      "learning_rate": 2e-05,
      "loss": 0.7809,
      "step": 2260
    },
    {
      "epoch": 0.1939507860560492,
      "grad_norm": 0.4647822976112366,
      "learning_rate": 2e-05,
      "loss": 0.8156,
      "step": 2270
    },
    {
      "epoch": 0.19480519480519481,
      "grad_norm": 0.43348145484924316,
      "learning_rate": 2e-05,
      "loss": 0.7993,
      "step": 2280
    },
    {
      "epoch": 0.1956596035543404,
      "grad_norm": 0.40158313512802124,
      "learning_rate": 2e-05,
      "loss": 0.7929,
      "step": 2290
    },
    {
      "epoch": 0.19651401230348597,
      "grad_norm": 0.43731987476348877,
      "learning_rate": 2e-05,
      "loss": 0.8001,
      "step": 2300
    },
    {
      "epoch": 0.19736842105263158,
      "grad_norm": 0.42699843645095825,
      "learning_rate": 2e-05,
      "loss": 0.7947,
      "step": 2310
    },
    {
      "epoch": 0.19822282980177716,
      "grad_norm": 0.4112282693386078,
      "learning_rate": 2e-05,
      "loss": 0.7848,
      "step": 2320
    },
    {
      "epoch": 0.19907723855092277,
      "grad_norm": 0.4258609116077423,
      "learning_rate": 2e-05,
      "loss": 0.7714,
      "step": 2330
    },
    {
      "epoch": 0.19993164730006835,
      "grad_norm": 0.41939055919647217,
      "learning_rate": 2e-05,
      "loss": 0.7786,
      "step": 2340
    },
    {
      "epoch": 0.20078605604921396,
      "grad_norm": 0.42235198616981506,
      "learning_rate": 2e-05,
      "loss": 0.8058,
      "step": 2350
    },
    {
      "epoch": 0.20164046479835954,
      "grad_norm": 0.4100799262523651,
      "learning_rate": 2e-05,
      "loss": 0.7895,
      "step": 2360
    },
    {
      "epoch": 0.20249487354750512,
      "grad_norm": 0.4226270020008087,
      "learning_rate": 2e-05,
      "loss": 0.8104,
      "step": 2370
    },
    {
      "epoch": 0.20334928229665072,
      "grad_norm": 0.41681861877441406,
      "learning_rate": 2e-05,
      "loss": 0.7823,
      "step": 2380
    },
    {
      "epoch": 0.2042036910457963,
      "grad_norm": 0.41445469856262207,
      "learning_rate": 2e-05,
      "loss": 0.7792,
      "step": 2390
    },
    {
      "epoch": 0.2050580997949419,
      "grad_norm": 0.4336954951286316,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 2400
    },
    {
      "epoch": 0.2059125085440875,
      "grad_norm": 0.4223636984825134,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 2410
    },
    {
      "epoch": 0.20676691729323307,
      "grad_norm": 0.4034973084926605,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 2420
    },
    {
      "epoch": 0.20762132604237868,
      "grad_norm": 0.42055076360702515,
      "learning_rate": 2e-05,
      "loss": 0.7701,
      "step": 2430
    },
    {
      "epoch": 0.20847573479152426,
      "grad_norm": 0.4145602285861969,
      "learning_rate": 2e-05,
      "loss": 0.7973,
      "step": 2440
    },
    {
      "epoch": 0.20933014354066987,
      "grad_norm": 0.4494640827178955,
      "learning_rate": 2e-05,
      "loss": 0.7956,
      "step": 2450
    },
    {
      "epoch": 0.21018455228981545,
      "grad_norm": 0.4173319339752197,
      "learning_rate": 2e-05,
      "loss": 0.832,
      "step": 2460
    },
    {
      "epoch": 0.21103896103896103,
      "grad_norm": 0.4216117560863495,
      "learning_rate": 2e-05,
      "loss": 0.7925,
      "step": 2470
    },
    {
      "epoch": 0.21189336978810663,
      "grad_norm": 0.9099227786064148,
      "learning_rate": 2e-05,
      "loss": 0.811,
      "step": 2480
    },
    {
      "epoch": 0.2127477785372522,
      "grad_norm": 0.41247525811195374,
      "learning_rate": 2e-05,
      "loss": 0.7879,
      "step": 2490
    },
    {
      "epoch": 0.21360218728639782,
      "grad_norm": 0.40090277791023254,
      "learning_rate": 2e-05,
      "loss": 0.7759,
      "step": 2500
    },
    {
      "epoch": 0.2144565960355434,
      "grad_norm": 0.43528005480766296,
      "learning_rate": 2e-05,
      "loss": 0.742,
      "step": 2510
    },
    {
      "epoch": 0.215311004784689,
      "grad_norm": 0.40456560254096985,
      "learning_rate": 2e-05,
      "loss": 0.7851,
      "step": 2520
    },
    {
      "epoch": 0.2161654135338346,
      "grad_norm": 0.41995969414711,
      "learning_rate": 2e-05,
      "loss": 0.7725,
      "step": 2530
    },
    {
      "epoch": 0.21701982228298017,
      "grad_norm": 0.43686243891716003,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 2540
    },
    {
      "epoch": 0.21787423103212578,
      "grad_norm": 0.4204292893409729,
      "learning_rate": 2e-05,
      "loss": 0.8399,
      "step": 2550
    },
    {
      "epoch": 0.21872863978127136,
      "grad_norm": 0.4287063777446747,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 2560
    },
    {
      "epoch": 0.21958304853041696,
      "grad_norm": 0.4153030812740326,
      "learning_rate": 2e-05,
      "loss": 0.7573,
      "step": 2570
    },
    {
      "epoch": 0.22043745727956254,
      "grad_norm": 0.43453824520111084,
      "learning_rate": 2e-05,
      "loss": 0.7872,
      "step": 2580
    },
    {
      "epoch": 0.22129186602870812,
      "grad_norm": 0.42367810010910034,
      "learning_rate": 2e-05,
      "loss": 0.7772,
      "step": 2590
    },
    {
      "epoch": 0.22214627477785373,
      "grad_norm": 0.3947020173072815,
      "learning_rate": 2e-05,
      "loss": 0.7789,
      "step": 2600
    },
    {
      "epoch": 0.2230006835269993,
      "grad_norm": 0.39880385994911194,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 2610
    },
    {
      "epoch": 0.22385509227614492,
      "grad_norm": 0.41851526498794556,
      "learning_rate": 2e-05,
      "loss": 0.7848,
      "step": 2620
    },
    {
      "epoch": 0.2247095010252905,
      "grad_norm": 0.49217864871025085,
      "learning_rate": 2e-05,
      "loss": 0.7955,
      "step": 2630
    },
    {
      "epoch": 0.22556390977443608,
      "grad_norm": 0.45744040608406067,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 2640
    },
    {
      "epoch": 0.22641831852358169,
      "grad_norm": 0.41490212082862854,
      "learning_rate": 2e-05,
      "loss": 0.7772,
      "step": 2650
    },
    {
      "epoch": 0.22727272727272727,
      "grad_norm": 0.4131002128124237,
      "learning_rate": 2e-05,
      "loss": 0.8039,
      "step": 2660
    },
    {
      "epoch": 0.22812713602187287,
      "grad_norm": 0.43832796812057495,
      "learning_rate": 2e-05,
      "loss": 0.8124,
      "step": 2670
    },
    {
      "epoch": 0.22898154477101845,
      "grad_norm": 0.432599812746048,
      "learning_rate": 2e-05,
      "loss": 0.8155,
      "step": 2680
    },
    {
      "epoch": 0.22983595352016406,
      "grad_norm": 0.4112294912338257,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 2690
    },
    {
      "epoch": 0.23069036226930964,
      "grad_norm": 0.4199526607990265,
      "learning_rate": 2e-05,
      "loss": 0.7806,
      "step": 2700
    },
    {
      "epoch": 0.23154477101845522,
      "grad_norm": 0.45168495178222656,
      "learning_rate": 2e-05,
      "loss": 0.7692,
      "step": 2710
    },
    {
      "epoch": 0.23239917976760083,
      "grad_norm": 0.40396353602409363,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 2720
    },
    {
      "epoch": 0.2332535885167464,
      "grad_norm": 0.41315266489982605,
      "learning_rate": 2e-05,
      "loss": 0.7728,
      "step": 2730
    },
    {
      "epoch": 0.23410799726589201,
      "grad_norm": 0.41976848244667053,
      "learning_rate": 2e-05,
      "loss": 0.7715,
      "step": 2740
    },
    {
      "epoch": 0.2349624060150376,
      "grad_norm": 0.4033108949661255,
      "learning_rate": 2e-05,
      "loss": 0.7765,
      "step": 2750
    },
    {
      "epoch": 0.23581681476418317,
      "grad_norm": 0.4449901580810547,
      "learning_rate": 2e-05,
      "loss": 0.7929,
      "step": 2760
    },
    {
      "epoch": 0.23667122351332878,
      "grad_norm": 0.4080033302307129,
      "learning_rate": 2e-05,
      "loss": 0.7744,
      "step": 2770
    },
    {
      "epoch": 0.23752563226247436,
      "grad_norm": 0.40643179416656494,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 2780
    },
    {
      "epoch": 0.23838004101161997,
      "grad_norm": 0.41164934635162354,
      "learning_rate": 2e-05,
      "loss": 0.8,
      "step": 2790
    },
    {
      "epoch": 0.23923444976076555,
      "grad_norm": 0.447868674993515,
      "learning_rate": 2e-05,
      "loss": 0.8174,
      "step": 2800
    },
    {
      "epoch": 0.24008885850991113,
      "grad_norm": 0.39664238691329956,
      "learning_rate": 2e-05,
      "loss": 0.7784,
      "step": 2810
    },
    {
      "epoch": 0.24094326725905674,
      "grad_norm": 0.4246390759944916,
      "learning_rate": 2e-05,
      "loss": 0.7671,
      "step": 2820
    },
    {
      "epoch": 0.24179767600820232,
      "grad_norm": 0.3972958028316498,
      "learning_rate": 2e-05,
      "loss": 0.7902,
      "step": 2830
    },
    {
      "epoch": 0.24265208475734792,
      "grad_norm": 0.4145045578479767,
      "learning_rate": 2e-05,
      "loss": 0.7589,
      "step": 2840
    },
    {
      "epoch": 0.2435064935064935,
      "grad_norm": 0.39986804127693176,
      "learning_rate": 2e-05,
      "loss": 0.7951,
      "step": 2850
    },
    {
      "epoch": 0.24436090225563908,
      "grad_norm": 0.4382905066013336,
      "learning_rate": 2e-05,
      "loss": 0.7945,
      "step": 2860
    },
    {
      "epoch": 0.2452153110047847,
      "grad_norm": 0.4199586808681488,
      "learning_rate": 2e-05,
      "loss": 0.7943,
      "step": 2870
    },
    {
      "epoch": 0.24606971975393027,
      "grad_norm": 0.6010065078735352,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 2880
    },
    {
      "epoch": 0.24692412850307588,
      "grad_norm": 0.39158105850219727,
      "learning_rate": 2e-05,
      "loss": 0.7849,
      "step": 2890
    },
    {
      "epoch": 0.24777853725222146,
      "grad_norm": 0.4009363651275635,
      "learning_rate": 2e-05,
      "loss": 0.765,
      "step": 2900
    },
    {
      "epoch": 0.24863294600136707,
      "grad_norm": 0.4049187898635864,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 2910
    },
    {
      "epoch": 0.24948735475051265,
      "grad_norm": 0.4128693640232086,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 2920
    },
    {
      "epoch": 0.25034176349965825,
      "grad_norm": 0.39886823296546936,
      "learning_rate": 2e-05,
      "loss": 0.7797,
      "step": 2930
    },
    {
      "epoch": 0.2511961722488038,
      "grad_norm": 0.42408713698387146,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 2940
    },
    {
      "epoch": 0.2520505809979494,
      "grad_norm": 0.3976152837276459,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 2950
    },
    {
      "epoch": 0.252904989747095,
      "grad_norm": 0.4111250638961792,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 2960
    },
    {
      "epoch": 0.25375939849624063,
      "grad_norm": 0.39206910133361816,
      "learning_rate": 2e-05,
      "loss": 0.7619,
      "step": 2970
    },
    {
      "epoch": 0.2546138072453862,
      "grad_norm": 0.41485005617141724,
      "learning_rate": 2e-05,
      "loss": 0.7976,
      "step": 2980
    },
    {
      "epoch": 0.2554682159945318,
      "grad_norm": 0.4331391751766205,
      "learning_rate": 2e-05,
      "loss": 0.7777,
      "step": 2990
    },
    {
      "epoch": 0.2563226247436774,
      "grad_norm": 0.4180612862110138,
      "learning_rate": 2e-05,
      "loss": 0.7743,
      "step": 3000
    },
    {
      "epoch": 0.25717703349282295,
      "grad_norm": 0.4284307062625885,
      "learning_rate": 2e-05,
      "loss": 0.7744,
      "step": 3010
    },
    {
      "epoch": 0.25803144224196856,
      "grad_norm": 0.40192970633506775,
      "learning_rate": 2e-05,
      "loss": 0.749,
      "step": 3020
    },
    {
      "epoch": 0.25888585099111416,
      "grad_norm": 0.42463406920433044,
      "learning_rate": 2e-05,
      "loss": 0.7861,
      "step": 3030
    },
    {
      "epoch": 0.2597402597402597,
      "grad_norm": 0.3968360126018524,
      "learning_rate": 2e-05,
      "loss": 0.7678,
      "step": 3040
    },
    {
      "epoch": 0.2605946684894053,
      "grad_norm": 0.4153085947036743,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 3050
    },
    {
      "epoch": 0.26144907723855093,
      "grad_norm": 0.4490017294883728,
      "learning_rate": 2e-05,
      "loss": 0.7871,
      "step": 3060
    },
    {
      "epoch": 0.26230348598769654,
      "grad_norm": 0.4165320098400116,
      "learning_rate": 2e-05,
      "loss": 0.7662,
      "step": 3070
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 0.38397857546806335,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 3080
    },
    {
      "epoch": 0.2640123034859877,
      "grad_norm": 0.39957764744758606,
      "learning_rate": 2e-05,
      "loss": 0.772,
      "step": 3090
    },
    {
      "epoch": 0.2648667122351333,
      "grad_norm": 0.4143160879611969,
      "learning_rate": 2e-05,
      "loss": 0.7798,
      "step": 3100
    },
    {
      "epoch": 0.26572112098427886,
      "grad_norm": 0.3915466070175171,
      "learning_rate": 2e-05,
      "loss": 0.7539,
      "step": 3110
    },
    {
      "epoch": 0.26657552973342447,
      "grad_norm": 0.4126955568790436,
      "learning_rate": 2e-05,
      "loss": 0.778,
      "step": 3120
    },
    {
      "epoch": 0.2674299384825701,
      "grad_norm": 0.39375099539756775,
      "learning_rate": 2e-05,
      "loss": 0.7826,
      "step": 3130
    },
    {
      "epoch": 0.2682843472317156,
      "grad_norm": 0.421099454164505,
      "learning_rate": 2e-05,
      "loss": 0.752,
      "step": 3140
    },
    {
      "epoch": 0.26913875598086123,
      "grad_norm": 0.4043879508972168,
      "learning_rate": 2e-05,
      "loss": 0.7853,
      "step": 3150
    },
    {
      "epoch": 0.26999316473000684,
      "grad_norm": 0.4344448149204254,
      "learning_rate": 2e-05,
      "loss": 0.7788,
      "step": 3160
    },
    {
      "epoch": 0.27084757347915245,
      "grad_norm": 0.40575578808784485,
      "learning_rate": 2e-05,
      "loss": 0.7801,
      "step": 3170
    },
    {
      "epoch": 0.271701982228298,
      "grad_norm": 0.40416884422302246,
      "learning_rate": 2e-05,
      "loss": 0.8032,
      "step": 3180
    },
    {
      "epoch": 0.2725563909774436,
      "grad_norm": 0.4253685176372528,
      "learning_rate": 2e-05,
      "loss": 0.813,
      "step": 3190
    },
    {
      "epoch": 0.2734107997265892,
      "grad_norm": 0.3963514268398285,
      "learning_rate": 2e-05,
      "loss": 0.7774,
      "step": 3200
    },
    {
      "epoch": 0.27426520847573477,
      "grad_norm": 0.4417586624622345,
      "learning_rate": 2e-05,
      "loss": 0.7773,
      "step": 3210
    },
    {
      "epoch": 0.2751196172248804,
      "grad_norm": 0.39846548438072205,
      "learning_rate": 2e-05,
      "loss": 0.7927,
      "step": 3220
    },
    {
      "epoch": 0.275974025974026,
      "grad_norm": 0.40777456760406494,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 3230
    },
    {
      "epoch": 0.2768284347231716,
      "grad_norm": 0.3965834081172943,
      "learning_rate": 2e-05,
      "loss": 0.7803,
      "step": 3240
    },
    {
      "epoch": 0.27768284347231714,
      "grad_norm": 0.41601282358169556,
      "learning_rate": 2e-05,
      "loss": 0.7635,
      "step": 3250
    },
    {
      "epoch": 0.27853725222146275,
      "grad_norm": 0.39050528407096863,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 3260
    },
    {
      "epoch": 0.27939166097060836,
      "grad_norm": 0.5213533639907837,
      "learning_rate": 2e-05,
      "loss": 0.78,
      "step": 3270
    },
    {
      "epoch": 0.2802460697197539,
      "grad_norm": 0.43881452083587646,
      "learning_rate": 2e-05,
      "loss": 0.8004,
      "step": 3280
    },
    {
      "epoch": 0.2811004784688995,
      "grad_norm": 0.42476263642311096,
      "learning_rate": 2e-05,
      "loss": 0.799,
      "step": 3290
    },
    {
      "epoch": 0.2819548872180451,
      "grad_norm": 0.399971604347229,
      "learning_rate": 2e-05,
      "loss": 0.7809,
      "step": 3300
    },
    {
      "epoch": 0.2828092959671907,
      "grad_norm": 0.3991090953350067,
      "learning_rate": 2e-05,
      "loss": 0.7819,
      "step": 3310
    },
    {
      "epoch": 0.2836637047163363,
      "grad_norm": 0.4399102032184601,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 3320
    },
    {
      "epoch": 0.2845181134654819,
      "grad_norm": 0.4094748795032501,
      "learning_rate": 2e-05,
      "loss": 0.773,
      "step": 3330
    },
    {
      "epoch": 0.2853725222146275,
      "grad_norm": 0.4769653081893921,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 3340
    },
    {
      "epoch": 0.28622693096377305,
      "grad_norm": 0.39396145939826965,
      "learning_rate": 2e-05,
      "loss": 0.7676,
      "step": 3350
    },
    {
      "epoch": 0.28708133971291866,
      "grad_norm": 0.47275757789611816,
      "learning_rate": 2e-05,
      "loss": 0.8053,
      "step": 3360
    },
    {
      "epoch": 0.28793574846206427,
      "grad_norm": 0.4360111653804779,
      "learning_rate": 2e-05,
      "loss": 0.7803,
      "step": 3370
    },
    {
      "epoch": 0.2887901572112098,
      "grad_norm": 0.4465491771697998,
      "learning_rate": 2e-05,
      "loss": 0.7902,
      "step": 3380
    },
    {
      "epoch": 0.2896445659603554,
      "grad_norm": 0.4282015562057495,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 3390
    },
    {
      "epoch": 0.29049897470950103,
      "grad_norm": 0.4026396572589874,
      "learning_rate": 2e-05,
      "loss": 0.7732,
      "step": 3400
    },
    {
      "epoch": 0.29135338345864664,
      "grad_norm": 0.4072575867176056,
      "learning_rate": 2e-05,
      "loss": 0.7583,
      "step": 3410
    },
    {
      "epoch": 0.2922077922077922,
      "grad_norm": 0.3923650085926056,
      "learning_rate": 2e-05,
      "loss": 0.7572,
      "step": 3420
    },
    {
      "epoch": 0.2930622009569378,
      "grad_norm": 0.37525373697280884,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 3430
    },
    {
      "epoch": 0.2939166097060834,
      "grad_norm": 0.4097144603729248,
      "learning_rate": 2e-05,
      "loss": 0.7752,
      "step": 3440
    },
    {
      "epoch": 0.29477101845522896,
      "grad_norm": 0.4037517309188843,
      "learning_rate": 2e-05,
      "loss": 0.797,
      "step": 3450
    },
    {
      "epoch": 0.29562542720437457,
      "grad_norm": 0.39452654123306274,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 3460
    },
    {
      "epoch": 0.2964798359535202,
      "grad_norm": 0.46122249960899353,
      "learning_rate": 2e-05,
      "loss": 0.8029,
      "step": 3470
    },
    {
      "epoch": 0.29733424470266573,
      "grad_norm": 0.38137784600257874,
      "learning_rate": 2e-05,
      "loss": 0.7395,
      "step": 3480
    },
    {
      "epoch": 0.29818865345181134,
      "grad_norm": 0.4483271837234497,
      "learning_rate": 2e-05,
      "loss": 0.8225,
      "step": 3490
    },
    {
      "epoch": 0.29904306220095694,
      "grad_norm": 0.4055141806602478,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 3500
    },
    {
      "epoch": 0.29989747095010255,
      "grad_norm": 0.43829309940338135,
      "learning_rate": 2e-05,
      "loss": 0.7737,
      "step": 3510
    },
    {
      "epoch": 0.3007518796992481,
      "grad_norm": 0.3994957506656647,
      "learning_rate": 2e-05,
      "loss": 0.747,
      "step": 3520
    },
    {
      "epoch": 0.3016062884483937,
      "grad_norm": 0.4170013964176178,
      "learning_rate": 2e-05,
      "loss": 0.7602,
      "step": 3530
    },
    {
      "epoch": 0.3024606971975393,
      "grad_norm": 0.3832671046257019,
      "learning_rate": 2e-05,
      "loss": 0.7741,
      "step": 3540
    },
    {
      "epoch": 0.30331510594668487,
      "grad_norm": 0.4173208773136139,
      "learning_rate": 2e-05,
      "loss": 0.7798,
      "step": 3550
    },
    {
      "epoch": 0.3041695146958305,
      "grad_norm": 0.4196033179759979,
      "learning_rate": 2e-05,
      "loss": 0.8139,
      "step": 3560
    },
    {
      "epoch": 0.3050239234449761,
      "grad_norm": 0.39927202463150024,
      "learning_rate": 2e-05,
      "loss": 0.7526,
      "step": 3570
    },
    {
      "epoch": 0.3058783321941217,
      "grad_norm": 0.40381738543510437,
      "learning_rate": 2e-05,
      "loss": 0.7661,
      "step": 3580
    },
    {
      "epoch": 0.30673274094326725,
      "grad_norm": 0.4107179641723633,
      "learning_rate": 2e-05,
      "loss": 0.7839,
      "step": 3590
    },
    {
      "epoch": 0.30758714969241285,
      "grad_norm": 0.413437157869339,
      "learning_rate": 2e-05,
      "loss": 0.7456,
      "step": 3600
    },
    {
      "epoch": 0.30844155844155846,
      "grad_norm": 0.41528069972991943,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 3610
    },
    {
      "epoch": 0.309295967190704,
      "grad_norm": 0.40742027759552,
      "learning_rate": 2e-05,
      "loss": 0.7509,
      "step": 3620
    },
    {
      "epoch": 0.3101503759398496,
      "grad_norm": 0.42556872963905334,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 3630
    },
    {
      "epoch": 0.31100478468899523,
      "grad_norm": 0.4212917685508728,
      "learning_rate": 2e-05,
      "loss": 0.7723,
      "step": 3640
    },
    {
      "epoch": 0.3118591934381408,
      "grad_norm": 0.405527800321579,
      "learning_rate": 2e-05,
      "loss": 0.7633,
      "step": 3650
    },
    {
      "epoch": 0.3127136021872864,
      "grad_norm": 0.3918667733669281,
      "learning_rate": 2e-05,
      "loss": 0.794,
      "step": 3660
    },
    {
      "epoch": 0.313568010936432,
      "grad_norm": 0.4197263717651367,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 3670
    },
    {
      "epoch": 0.3144224196855776,
      "grad_norm": 0.3946523666381836,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 3680
    },
    {
      "epoch": 0.31527682843472316,
      "grad_norm": 0.4318191111087799,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 3690
    },
    {
      "epoch": 0.31613123718386876,
      "grad_norm": 0.38150814175605774,
      "learning_rate": 2e-05,
      "loss": 0.7465,
      "step": 3700
    },
    {
      "epoch": 0.31698564593301437,
      "grad_norm": 0.4278201162815094,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 3710
    },
    {
      "epoch": 0.3178400546821599,
      "grad_norm": 0.4148176610469818,
      "learning_rate": 2e-05,
      "loss": 0.7857,
      "step": 3720
    },
    {
      "epoch": 0.31869446343130553,
      "grad_norm": 0.39185822010040283,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 3730
    },
    {
      "epoch": 0.31954887218045114,
      "grad_norm": 0.38302555680274963,
      "learning_rate": 2e-05,
      "loss": 0.7717,
      "step": 3740
    },
    {
      "epoch": 0.32040328092959675,
      "grad_norm": 0.4142118990421295,
      "learning_rate": 2e-05,
      "loss": 0.7814,
      "step": 3750
    },
    {
      "epoch": 0.3212576896787423,
      "grad_norm": 0.4014316201210022,
      "learning_rate": 2e-05,
      "loss": 0.7482,
      "step": 3760
    },
    {
      "epoch": 0.3221120984278879,
      "grad_norm": 0.4141567349433899,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 3770
    },
    {
      "epoch": 0.3229665071770335,
      "grad_norm": 0.4212769567966461,
      "learning_rate": 2e-05,
      "loss": 0.8012,
      "step": 3780
    },
    {
      "epoch": 0.32382091592617906,
      "grad_norm": 0.3971374034881592,
      "learning_rate": 2e-05,
      "loss": 0.7733,
      "step": 3790
    },
    {
      "epoch": 0.3246753246753247,
      "grad_norm": 0.4019508957862854,
      "learning_rate": 2e-05,
      "loss": 0.7804,
      "step": 3800
    },
    {
      "epoch": 0.3255297334244703,
      "grad_norm": 0.40491899847984314,
      "learning_rate": 2e-05,
      "loss": 0.7368,
      "step": 3810
    },
    {
      "epoch": 0.32638414217361583,
      "grad_norm": 0.38166218996047974,
      "learning_rate": 2e-05,
      "loss": 0.78,
      "step": 3820
    },
    {
      "epoch": 0.32723855092276144,
      "grad_norm": 0.422429621219635,
      "learning_rate": 2e-05,
      "loss": 0.77,
      "step": 3830
    },
    {
      "epoch": 0.32809295967190705,
      "grad_norm": 0.4231925308704376,
      "learning_rate": 2e-05,
      "loss": 0.7855,
      "step": 3840
    },
    {
      "epoch": 0.32894736842105265,
      "grad_norm": 0.42817559838294983,
      "learning_rate": 2e-05,
      "loss": 0.7794,
      "step": 3850
    },
    {
      "epoch": 0.3298017771701982,
      "grad_norm": 0.38504835963249207,
      "learning_rate": 2e-05,
      "loss": 0.7562,
      "step": 3860
    },
    {
      "epoch": 0.3306561859193438,
      "grad_norm": 0.3995634913444519,
      "learning_rate": 2e-05,
      "loss": 0.772,
      "step": 3870
    },
    {
      "epoch": 0.3315105946684894,
      "grad_norm": 0.44168439507484436,
      "learning_rate": 2e-05,
      "loss": 0.7601,
      "step": 3880
    },
    {
      "epoch": 0.332365003417635,
      "grad_norm": 0.4031580984592438,
      "learning_rate": 2e-05,
      "loss": 0.7689,
      "step": 3890
    },
    {
      "epoch": 0.3332194121667806,
      "grad_norm": 0.40626052021980286,
      "learning_rate": 2e-05,
      "loss": 0.7444,
      "step": 3900
    },
    {
      "epoch": 0.3340738209159262,
      "grad_norm": 0.3823252022266388,
      "learning_rate": 2e-05,
      "loss": 0.776,
      "step": 3910
    },
    {
      "epoch": 0.3349282296650718,
      "grad_norm": 0.3728090226650238,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 3920
    },
    {
      "epoch": 0.33578263841421735,
      "grad_norm": 0.3929399251937866,
      "learning_rate": 2e-05,
      "loss": 0.7718,
      "step": 3930
    },
    {
      "epoch": 0.33663704716336296,
      "grad_norm": 0.42269980907440186,
      "learning_rate": 2e-05,
      "loss": 0.7844,
      "step": 3940
    },
    {
      "epoch": 0.33749145591250856,
      "grad_norm": 0.4141886234283447,
      "learning_rate": 2e-05,
      "loss": 0.7867,
      "step": 3950
    },
    {
      "epoch": 0.3383458646616541,
      "grad_norm": 0.4118608236312866,
      "learning_rate": 2e-05,
      "loss": 0.7751,
      "step": 3960
    },
    {
      "epoch": 0.3392002734107997,
      "grad_norm": 0.4049041271209717,
      "learning_rate": 2e-05,
      "loss": 0.7861,
      "step": 3970
    },
    {
      "epoch": 0.34005468215994533,
      "grad_norm": 0.41091078519821167,
      "learning_rate": 2e-05,
      "loss": 0.7629,
      "step": 3980
    },
    {
      "epoch": 0.3409090909090909,
      "grad_norm": 0.39125552773475647,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 3990
    },
    {
      "epoch": 0.3417634996582365,
      "grad_norm": 0.4104257822036743,
      "learning_rate": 2e-05,
      "loss": 0.7845,
      "step": 4000
    },
    {
      "epoch": 0.3426179084073821,
      "grad_norm": 0.3986048400402069,
      "learning_rate": 2e-05,
      "loss": 0.7619,
      "step": 4010
    },
    {
      "epoch": 0.3434723171565277,
      "grad_norm": 0.42920932173728943,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 4020
    },
    {
      "epoch": 0.34432672590567326,
      "grad_norm": 0.3850744366645813,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 4030
    },
    {
      "epoch": 0.34518113465481887,
      "grad_norm": 0.41486093401908875,
      "learning_rate": 2e-05,
      "loss": 0.7811,
      "step": 4040
    },
    {
      "epoch": 0.3460355434039645,
      "grad_norm": 0.40918493270874023,
      "learning_rate": 2e-05,
      "loss": 0.7733,
      "step": 4050
    },
    {
      "epoch": 0.34688995215311,
      "grad_norm": 0.5060675144195557,
      "learning_rate": 2e-05,
      "loss": 0.7638,
      "step": 4060
    },
    {
      "epoch": 0.34774436090225563,
      "grad_norm": 0.4126531481742859,
      "learning_rate": 2e-05,
      "loss": 0.7624,
      "step": 4070
    },
    {
      "epoch": 0.34859876965140124,
      "grad_norm": 1.0571564435958862,
      "learning_rate": 2e-05,
      "loss": 0.769,
      "step": 4080
    },
    {
      "epoch": 0.34945317840054685,
      "grad_norm": 0.3744198679924011,
      "learning_rate": 2e-05,
      "loss": 0.7678,
      "step": 4090
    },
    {
      "epoch": 0.3503075871496924,
      "grad_norm": 0.4054426848888397,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 4100
    },
    {
      "epoch": 0.351161995898838,
      "grad_norm": 0.40937623381614685,
      "learning_rate": 2e-05,
      "loss": 0.7734,
      "step": 4110
    },
    {
      "epoch": 0.3520164046479836,
      "grad_norm": 0.4030979871749878,
      "learning_rate": 2e-05,
      "loss": 0.7792,
      "step": 4120
    },
    {
      "epoch": 0.35287081339712917,
      "grad_norm": 0.40039506554603577,
      "learning_rate": 2e-05,
      "loss": 0.7747,
      "step": 4130
    },
    {
      "epoch": 0.3537252221462748,
      "grad_norm": 0.4101240336894989,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 4140
    },
    {
      "epoch": 0.3545796308954204,
      "grad_norm": 0.3959333896636963,
      "learning_rate": 2e-05,
      "loss": 0.7566,
      "step": 4150
    },
    {
      "epoch": 0.35543403964456594,
      "grad_norm": 0.4016161262989044,
      "learning_rate": 2e-05,
      "loss": 0.7862,
      "step": 4160
    },
    {
      "epoch": 0.35628844839371154,
      "grad_norm": 0.404707670211792,
      "learning_rate": 2e-05,
      "loss": 0.7766,
      "step": 4170
    },
    {
      "epoch": 0.35714285714285715,
      "grad_norm": 0.3973741829395294,
      "learning_rate": 2e-05,
      "loss": 0.7662,
      "step": 4180
    },
    {
      "epoch": 0.35799726589200276,
      "grad_norm": 0.4071394205093384,
      "learning_rate": 2e-05,
      "loss": 0.7484,
      "step": 4190
    },
    {
      "epoch": 0.3588516746411483,
      "grad_norm": 0.8778959512710571,
      "learning_rate": 2e-05,
      "loss": 0.751,
      "step": 4200
    },
    {
      "epoch": 0.3597060833902939,
      "grad_norm": 0.4312654435634613,
      "learning_rate": 2e-05,
      "loss": 0.7591,
      "step": 4210
    },
    {
      "epoch": 0.3605604921394395,
      "grad_norm": 0.39795437455177307,
      "learning_rate": 2e-05,
      "loss": 0.7498,
      "step": 4220
    },
    {
      "epoch": 0.3614149008885851,
      "grad_norm": 0.4243643879890442,
      "learning_rate": 2e-05,
      "loss": 0.777,
      "step": 4230
    },
    {
      "epoch": 0.3622693096377307,
      "grad_norm": 0.393503874540329,
      "learning_rate": 2e-05,
      "loss": 0.762,
      "step": 4240
    },
    {
      "epoch": 0.3631237183868763,
      "grad_norm": 0.4122103452682495,
      "learning_rate": 2e-05,
      "loss": 0.7737,
      "step": 4250
    },
    {
      "epoch": 0.3639781271360219,
      "grad_norm": 0.43337583541870117,
      "learning_rate": 2e-05,
      "loss": 0.7699,
      "step": 4260
    },
    {
      "epoch": 0.36483253588516745,
      "grad_norm": 0.38475391268730164,
      "learning_rate": 2e-05,
      "loss": 0.7711,
      "step": 4270
    },
    {
      "epoch": 0.36568694463431306,
      "grad_norm": 0.40655094385147095,
      "learning_rate": 2e-05,
      "loss": 0.7812,
      "step": 4280
    },
    {
      "epoch": 0.36654135338345867,
      "grad_norm": 0.39318129420280457,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 4290
    },
    {
      "epoch": 0.3673957621326042,
      "grad_norm": 0.39770469069480896,
      "learning_rate": 2e-05,
      "loss": 0.7746,
      "step": 4300
    },
    {
      "epoch": 0.3682501708817498,
      "grad_norm": 0.3997613489627838,
      "learning_rate": 2e-05,
      "loss": 0.7684,
      "step": 4310
    },
    {
      "epoch": 0.36910457963089544,
      "grad_norm": 0.3889233469963074,
      "learning_rate": 2e-05,
      "loss": 0.7703,
      "step": 4320
    },
    {
      "epoch": 0.369958988380041,
      "grad_norm": 0.40742871165275574,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 4330
    },
    {
      "epoch": 0.3708133971291866,
      "grad_norm": 0.4064273536205292,
      "learning_rate": 2e-05,
      "loss": 0.749,
      "step": 4340
    },
    {
      "epoch": 0.3716678058783322,
      "grad_norm": 0.4193180203437805,
      "learning_rate": 2e-05,
      "loss": 0.7609,
      "step": 4350
    },
    {
      "epoch": 0.3725222146274778,
      "grad_norm": 0.4050132632255554,
      "learning_rate": 2e-05,
      "loss": 0.7693,
      "step": 4360
    },
    {
      "epoch": 0.37337662337662336,
      "grad_norm": 0.39398810267448425,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 4370
    },
    {
      "epoch": 0.37423103212576897,
      "grad_norm": 0.422772079706192,
      "learning_rate": 2e-05,
      "loss": 0.7697,
      "step": 4380
    },
    {
      "epoch": 0.3750854408749146,
      "grad_norm": 0.43367302417755127,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 4390
    },
    {
      "epoch": 0.37593984962406013,
      "grad_norm": 0.4362691044807434,
      "learning_rate": 2e-05,
      "loss": 0.7866,
      "step": 4400
    },
    {
      "epoch": 0.37679425837320574,
      "grad_norm": 0.38281482458114624,
      "learning_rate": 2e-05,
      "loss": 0.7399,
      "step": 4410
    },
    {
      "epoch": 0.37764866712235134,
      "grad_norm": 0.42784583568573,
      "learning_rate": 2e-05,
      "loss": 0.7846,
      "step": 4420
    },
    {
      "epoch": 0.3785030758714969,
      "grad_norm": 0.39173761010169983,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 4430
    },
    {
      "epoch": 0.3793574846206425,
      "grad_norm": 0.4276053011417389,
      "learning_rate": 2e-05,
      "loss": 0.7594,
      "step": 4440
    },
    {
      "epoch": 0.3802118933697881,
      "grad_norm": 0.5029833912849426,
      "learning_rate": 2e-05,
      "loss": 0.7539,
      "step": 4450
    },
    {
      "epoch": 0.3810663021189337,
      "grad_norm": 0.4133250415325165,
      "learning_rate": 2e-05,
      "loss": 0.7607,
      "step": 4460
    },
    {
      "epoch": 0.38192071086807927,
      "grad_norm": 0.3917803466320038,
      "learning_rate": 2e-05,
      "loss": 0.736,
      "step": 4470
    },
    {
      "epoch": 0.3827751196172249,
      "grad_norm": 0.3776458203792572,
      "learning_rate": 2e-05,
      "loss": 0.7621,
      "step": 4480
    },
    {
      "epoch": 0.3836295283663705,
      "grad_norm": 0.40102991461753845,
      "learning_rate": 2e-05,
      "loss": 0.7746,
      "step": 4490
    },
    {
      "epoch": 0.38448393711551604,
      "grad_norm": 0.3935014009475708,
      "learning_rate": 2e-05,
      "loss": 0.7691,
      "step": 4500
    },
    {
      "epoch": 0.38533834586466165,
      "grad_norm": 0.38191425800323486,
      "learning_rate": 2e-05,
      "loss": 0.7568,
      "step": 4510
    },
    {
      "epoch": 0.38619275461380725,
      "grad_norm": 0.40427932143211365,
      "learning_rate": 2e-05,
      "loss": 0.7555,
      "step": 4520
    },
    {
      "epoch": 0.38704716336295286,
      "grad_norm": 0.3987480700016022,
      "learning_rate": 2e-05,
      "loss": 0.7637,
      "step": 4530
    },
    {
      "epoch": 0.3879015721120984,
      "grad_norm": 0.39704596996307373,
      "learning_rate": 2e-05,
      "loss": 0.7692,
      "step": 4540
    },
    {
      "epoch": 0.388755980861244,
      "grad_norm": 0.38936182856559753,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 4550
    },
    {
      "epoch": 0.38961038961038963,
      "grad_norm": 0.392251193523407,
      "learning_rate": 2e-05,
      "loss": 0.7739,
      "step": 4560
    },
    {
      "epoch": 0.3904647983595352,
      "grad_norm": 0.413920134305954,
      "learning_rate": 2e-05,
      "loss": 0.7672,
      "step": 4570
    },
    {
      "epoch": 0.3913192071086808,
      "grad_norm": 0.3814871609210968,
      "learning_rate": 2e-05,
      "loss": 0.747,
      "step": 4580
    },
    {
      "epoch": 0.3921736158578264,
      "grad_norm": 0.4402540624141693,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 4590
    },
    {
      "epoch": 0.39302802460697195,
      "grad_norm": 0.45276889204978943,
      "learning_rate": 2e-05,
      "loss": 0.7493,
      "step": 4600
    },
    {
      "epoch": 0.39388243335611756,
      "grad_norm": 0.4238435626029968,
      "learning_rate": 2e-05,
      "loss": 0.754,
      "step": 4610
    },
    {
      "epoch": 0.39473684210526316,
      "grad_norm": 0.42151281237602234,
      "learning_rate": 2e-05,
      "loss": 0.7797,
      "step": 4620
    },
    {
      "epoch": 0.39559125085440877,
      "grad_norm": 0.38528621196746826,
      "learning_rate": 2e-05,
      "loss": 0.786,
      "step": 4630
    },
    {
      "epoch": 0.3964456596035543,
      "grad_norm": 0.4101194739341736,
      "learning_rate": 2e-05,
      "loss": 0.7767,
      "step": 4640
    },
    {
      "epoch": 0.39730006835269993,
      "grad_norm": 0.4006814658641815,
      "learning_rate": 2e-05,
      "loss": 0.7976,
      "step": 4650
    },
    {
      "epoch": 0.39815447710184554,
      "grad_norm": 0.4116314947605133,
      "learning_rate": 2e-05,
      "loss": 0.7525,
      "step": 4660
    },
    {
      "epoch": 0.3990088858509911,
      "grad_norm": 0.3980487883090973,
      "learning_rate": 2e-05,
      "loss": 0.7573,
      "step": 4670
    },
    {
      "epoch": 0.3998632946001367,
      "grad_norm": 0.38866686820983887,
      "learning_rate": 2e-05,
      "loss": 0.7487,
      "step": 4680
    },
    {
      "epoch": 0.4007177033492823,
      "grad_norm": 0.4014759957790375,
      "learning_rate": 2e-05,
      "loss": 0.7326,
      "step": 4690
    },
    {
      "epoch": 0.4015721120984279,
      "grad_norm": 0.403236448764801,
      "learning_rate": 2e-05,
      "loss": 0.7768,
      "step": 4700
    },
    {
      "epoch": 0.40242652084757347,
      "grad_norm": 0.3917689919471741,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 4710
    },
    {
      "epoch": 0.4032809295967191,
      "grad_norm": 0.37933942675590515,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 4720
    },
    {
      "epoch": 0.4041353383458647,
      "grad_norm": 0.38117894530296326,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 4730
    },
    {
      "epoch": 0.40498974709501023,
      "grad_norm": 0.4002052843570709,
      "learning_rate": 2e-05,
      "loss": 0.7398,
      "step": 4740
    },
    {
      "epoch": 0.40584415584415584,
      "grad_norm": 0.4108046293258667,
      "learning_rate": 2e-05,
      "loss": 0.7403,
      "step": 4750
    },
    {
      "epoch": 0.40669856459330145,
      "grad_norm": 0.3776397705078125,
      "learning_rate": 2e-05,
      "loss": 0.7707,
      "step": 4760
    },
    {
      "epoch": 0.407552973342447,
      "grad_norm": 0.3892161250114441,
      "learning_rate": 2e-05,
      "loss": 0.7654,
      "step": 4770
    },
    {
      "epoch": 0.4084073820915926,
      "grad_norm": 0.40108174085617065,
      "learning_rate": 2e-05,
      "loss": 0.7858,
      "step": 4780
    },
    {
      "epoch": 0.4092617908407382,
      "grad_norm": 0.40284615755081177,
      "learning_rate": 2e-05,
      "loss": 0.7419,
      "step": 4790
    },
    {
      "epoch": 0.4101161995898838,
      "grad_norm": 0.4094385504722595,
      "learning_rate": 2e-05,
      "loss": 0.7518,
      "step": 4800
    },
    {
      "epoch": 0.4109706083390294,
      "grad_norm": 0.382321834564209,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 4810
    },
    {
      "epoch": 0.411825017088175,
      "grad_norm": 0.39105379581451416,
      "learning_rate": 2e-05,
      "loss": 0.7836,
      "step": 4820
    },
    {
      "epoch": 0.4126794258373206,
      "grad_norm": 0.38638919591903687,
      "learning_rate": 2e-05,
      "loss": 0.7618,
      "step": 4830
    },
    {
      "epoch": 0.41353383458646614,
      "grad_norm": 0.42078790068626404,
      "learning_rate": 2e-05,
      "loss": 0.7587,
      "step": 4840
    },
    {
      "epoch": 0.41438824333561175,
      "grad_norm": 0.4311010539531708,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 4850
    },
    {
      "epoch": 0.41524265208475736,
      "grad_norm": 0.40017813444137573,
      "learning_rate": 2e-05,
      "loss": 0.7443,
      "step": 4860
    },
    {
      "epoch": 0.41609706083390297,
      "grad_norm": 0.41783082485198975,
      "learning_rate": 2e-05,
      "loss": 0.7504,
      "step": 4870
    },
    {
      "epoch": 0.4169514695830485,
      "grad_norm": 0.3885749876499176,
      "learning_rate": 2e-05,
      "loss": 0.7798,
      "step": 4880
    },
    {
      "epoch": 0.4178058783321941,
      "grad_norm": 0.3940373957157135,
      "learning_rate": 2e-05,
      "loss": 0.7614,
      "step": 4890
    },
    {
      "epoch": 0.41866028708133973,
      "grad_norm": 0.41718438267707825,
      "learning_rate": 2e-05,
      "loss": 0.744,
      "step": 4900
    },
    {
      "epoch": 0.4195146958304853,
      "grad_norm": 0.4303945004940033,
      "learning_rate": 2e-05,
      "loss": 0.7831,
      "step": 4910
    },
    {
      "epoch": 0.4203691045796309,
      "grad_norm": 0.3983951508998871,
      "learning_rate": 2e-05,
      "loss": 0.7625,
      "step": 4920
    },
    {
      "epoch": 0.4212235133287765,
      "grad_norm": 0.39108559489250183,
      "learning_rate": 2e-05,
      "loss": 0.7602,
      "step": 4930
    },
    {
      "epoch": 0.42207792207792205,
      "grad_norm": 0.3912990093231201,
      "learning_rate": 2e-05,
      "loss": 0.7241,
      "step": 4940
    },
    {
      "epoch": 0.42293233082706766,
      "grad_norm": 0.4086664915084839,
      "learning_rate": 2e-05,
      "loss": 0.7543,
      "step": 4950
    },
    {
      "epoch": 0.42378673957621327,
      "grad_norm": 0.43198081851005554,
      "learning_rate": 2e-05,
      "loss": 0.7426,
      "step": 4960
    },
    {
      "epoch": 0.4246411483253589,
      "grad_norm": 0.37144365906715393,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 4970
    },
    {
      "epoch": 0.4254955570745044,
      "grad_norm": 0.3948076367378235,
      "learning_rate": 2e-05,
      "loss": 0.7499,
      "step": 4980
    },
    {
      "epoch": 0.42634996582365003,
      "grad_norm": 0.4113522469997406,
      "learning_rate": 2e-05,
      "loss": 0.7657,
      "step": 4990
    },
    {
      "epoch": 0.42720437457279564,
      "grad_norm": 0.3880317211151123,
      "learning_rate": 2e-05,
      "loss": 0.7534,
      "step": 5000
    },
    {
      "epoch": 0.4280587833219412,
      "grad_norm": 0.3677981495857239,
      "learning_rate": 2e-05,
      "loss": 0.752,
      "step": 5010
    },
    {
      "epoch": 0.4289131920710868,
      "grad_norm": 0.4228217601776123,
      "learning_rate": 2e-05,
      "loss": 0.7637,
      "step": 5020
    },
    {
      "epoch": 0.4297676008202324,
      "grad_norm": 0.38919076323509216,
      "learning_rate": 2e-05,
      "loss": 0.7473,
      "step": 5030
    },
    {
      "epoch": 0.430622009569378,
      "grad_norm": 0.4081915616989136,
      "learning_rate": 2e-05,
      "loss": 0.7675,
      "step": 5040
    },
    {
      "epoch": 0.43147641831852357,
      "grad_norm": 0.4186146855354309,
      "learning_rate": 2e-05,
      "loss": 0.7491,
      "step": 5050
    },
    {
      "epoch": 0.4323308270676692,
      "grad_norm": 0.4249284267425537,
      "learning_rate": 2e-05,
      "loss": 0.7577,
      "step": 5060
    },
    {
      "epoch": 0.4331852358168148,
      "grad_norm": 0.5370435118675232,
      "learning_rate": 2e-05,
      "loss": 0.7321,
      "step": 5070
    },
    {
      "epoch": 0.43403964456596034,
      "grad_norm": 0.40723684430122375,
      "learning_rate": 2e-05,
      "loss": 0.7369,
      "step": 5080
    },
    {
      "epoch": 0.43489405331510594,
      "grad_norm": 0.3775256872177124,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.7352,
      "step": 5090
    },
    {
      "epoch": 0.43574846206425155,
      "grad_norm": 0.4152348041534424,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.7645,
      "step": 5100
    },
    {
      "epoch": 0.4366028708133971,
      "grad_norm": 0.408880352973938,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7746,
      "step": 5110
    },
    {
      "epoch": 0.4374572795625427,
      "grad_norm": 0.3951629102230072,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.7266,
      "step": 5120
    },
    {
      "epoch": 0.4383116883116883,
      "grad_norm": 0.41564100980758667,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.7356,
      "step": 5130
    },
    {
      "epoch": 0.4391660970608339,
      "grad_norm": 0.40281832218170166,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.7454,
      "step": 5140
    },
    {
      "epoch": 0.4400205058099795,
      "grad_norm": 0.42460203170776367,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.7624,
      "step": 5150
    },
    {
      "epoch": 0.4408749145591251,
      "grad_norm": 0.40546995401382446,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.742,
      "step": 5160
    },
    {
      "epoch": 0.4417293233082707,
      "grad_norm": 0.4504067599773407,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.754,
      "step": 5170
    },
    {
      "epoch": 0.44258373205741625,
      "grad_norm": 0.39395347237586975,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7342,
      "step": 5180
    },
    {
      "epoch": 0.44343814080656185,
      "grad_norm": 0.650147020816803,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.7669,
      "step": 5190
    },
    {
      "epoch": 0.44429254955570746,
      "grad_norm": 0.42760539054870605,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.7573,
      "step": 5200
    },
    {
      "epoch": 0.44514695830485307,
      "grad_norm": 0.4003875255584717,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.7583,
      "step": 5210
    },
    {
      "epoch": 0.4460013670539986,
      "grad_norm": 0.4100761413574219,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.7691,
      "step": 5220
    },
    {
      "epoch": 0.44685577580314423,
      "grad_norm": 0.3920421302318573,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.7549,
      "step": 5230
    },
    {
      "epoch": 0.44771018455228984,
      "grad_norm": 0.4117676317691803,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.7428,
      "step": 5240
    },
    {
      "epoch": 0.4485645933014354,
      "grad_norm": 0.39280110597610474,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.7458,
      "step": 5250
    },
    {
      "epoch": 0.449419002050581,
      "grad_norm": 0.38863030076026917,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.7372,
      "step": 5260
    },
    {
      "epoch": 0.4502734107997266,
      "grad_norm": 0.3958275020122528,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.7262,
      "step": 5270
    },
    {
      "epoch": 0.45112781954887216,
      "grad_norm": 0.403012752532959,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.771,
      "step": 5280
    },
    {
      "epoch": 0.45198222829801776,
      "grad_norm": 0.41234296560287476,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.7314,
      "step": 5290
    },
    {
      "epoch": 0.45283663704716337,
      "grad_norm": 0.39876899123191833,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.7495,
      "step": 5300
    },
    {
      "epoch": 0.453691045796309,
      "grad_norm": 0.40820083022117615,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.7482,
      "step": 5310
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 0.39455652236938477,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.7553,
      "step": 5320
    },
    {
      "epoch": 0.45539986329460014,
      "grad_norm": 0.40043115615844727,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.7521,
      "step": 5330
    },
    {
      "epoch": 0.45625427204374575,
      "grad_norm": 0.38866814970970154,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.7523,
      "step": 5340
    },
    {
      "epoch": 0.4571086807928913,
      "grad_norm": 0.392313688993454,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.763,
      "step": 5350
    },
    {
      "epoch": 0.4579630895420369,
      "grad_norm": 0.41387996077537537,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.7364,
      "step": 5360
    },
    {
      "epoch": 0.4588174982911825,
      "grad_norm": 0.3698216676712036,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.7412,
      "step": 5370
    },
    {
      "epoch": 0.4596719070403281,
      "grad_norm": 0.3841533362865448,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.7321,
      "step": 5380
    },
    {
      "epoch": 0.4605263157894737,
      "grad_norm": 0.3749816417694092,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.7343,
      "step": 5390
    },
    {
      "epoch": 0.4613807245386193,
      "grad_norm": 0.3746258616447449,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.7392,
      "step": 5400
    },
    {
      "epoch": 0.4622351332877649,
      "grad_norm": 0.42499133944511414,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7381,
      "step": 5410
    },
    {
      "epoch": 0.46308954203691044,
      "grad_norm": 0.42077338695526123,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.7617,
      "step": 5420
    },
    {
      "epoch": 0.46394395078605605,
      "grad_norm": 0.38609129190444946,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.694,
      "step": 5430
    },
    {
      "epoch": 0.46479835953520166,
      "grad_norm": 0.41250157356262207,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.7514,
      "step": 5440
    },
    {
      "epoch": 0.4656527682843472,
      "grad_norm": 0.3996208906173706,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.7304,
      "step": 5450
    },
    {
      "epoch": 0.4665071770334928,
      "grad_norm": 0.3829016387462616,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.7545,
      "step": 5460
    },
    {
      "epoch": 0.4673615857826384,
      "grad_norm": 0.37041059136390686,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.7529,
      "step": 5470
    },
    {
      "epoch": 0.46821599453178403,
      "grad_norm": 0.38161271810531616,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.7154,
      "step": 5480
    },
    {
      "epoch": 0.4690704032809296,
      "grad_norm": 0.3718547523021698,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.7439,
      "step": 5490
    },
    {
      "epoch": 0.4699248120300752,
      "grad_norm": 0.4723457098007202,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.7369,
      "step": 5500
    },
    {
      "epoch": 0.4707792207792208,
      "grad_norm": 0.3703615963459015,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.7354,
      "step": 5510
    },
    {
      "epoch": 0.47163362952836635,
      "grad_norm": 0.4209197759628296,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.7486,
      "step": 5520
    },
    {
      "epoch": 0.47248803827751196,
      "grad_norm": 0.3584165573120117,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.722,
      "step": 5530
    },
    {
      "epoch": 0.47334244702665756,
      "grad_norm": 0.3540111184120178,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.7488,
      "step": 5540
    },
    {
      "epoch": 0.4741968557758031,
      "grad_norm": 0.37436020374298096,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.7382,
      "step": 5550
    },
    {
      "epoch": 0.4750512645249487,
      "grad_norm": 0.3658936023712158,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.7018,
      "step": 5560
    },
    {
      "epoch": 0.47590567327409433,
      "grad_norm": 0.37405404448509216,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.7524,
      "step": 5570
    },
    {
      "epoch": 0.47676008202323994,
      "grad_norm": 0.35756152868270874,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6975,
      "step": 5580
    },
    {
      "epoch": 0.4776144907723855,
      "grad_norm": 0.36947324872016907,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.7471,
      "step": 5590
    },
    {
      "epoch": 0.4782125768967874,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.025095393705819577,
      "train_runtime": 1778.2869,
      "train_samples_per_second": 37.769,
      "train_steps_per_second": 3.147
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}