{
  "best_metric": 0.6827040314674377,
  "best_model_checkpoint": "checkpoints/1a_52k/checkpoint-6200",
  "epoch": 1.9838412926965843,
  "eval_steps": 200,
  "global_step": 6200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.6999999999999996e-05,
      "loss": 2.0256,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.399999999999999e-05,
      "loss": 1.9236,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 7.8e-05,
      "loss": 1.712,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00010799999999999998,
      "loss": 1.2747,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000138,
      "loss": 1.0657,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000168,
      "loss": 0.973,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000198,
      "loss": 0.8178,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.7901,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000258,
      "loss": 0.7289,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.7429,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998840579710145,
      "loss": 0.7631,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002996908212560386,
      "loss": 0.7339,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029949758454106277,
      "loss": 0.7435,
      "step": 130
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00029930434782608696,
      "loss": 0.7333,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002991111111111111,
      "loss": 0.769,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029891787439613525,
      "loss": 0.7644,
      "step": 160
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002987246376811594,
      "loss": 0.7517,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002985314009661836,
      "loss": 0.7212,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029833816425120773,
      "loss": 0.745,
      "step": 190
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029814492753623187,
      "loss": 0.7023,
      "step": 200
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.7244767546653748,
      "eval_runtime": 150.1014,
      "eval_samples_per_second": 13.324,
      "eval_steps_per_second": 1.666,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.000297951690821256,
      "loss": 0.6783,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029775845410628016,
      "loss": 0.7327,
      "step": 220
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029756521739130435,
      "loss": 0.69,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002973719806763285,
      "loss": 0.7069,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029717874396135264,
      "loss": 0.7276,
      "step": 250
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002969855072463768,
      "loss": 0.7356,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.000296792270531401,
      "loss": 0.7103,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002965990338164251,
      "loss": 0.7224,
      "step": 280
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029640579710144926,
      "loss": 0.6898,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002962125603864734,
      "loss": 0.7222,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029601932367149755,
      "loss": 0.685,
      "step": 310
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029582608695652175,
      "loss": 0.7389,
      "step": 320
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002956328502415459,
      "loss": 0.6956,
      "step": 330
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029543961352657003,
      "loss": 0.7191,
      "step": 340
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002952463768115942,
      "loss": 0.6938,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029505314009661837,
      "loss": 0.695,
      "step": 360
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002948599033816425,
      "loss": 0.7169,
      "step": 370
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029466666666666666,
      "loss": 0.7313,
      "step": 380
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002944734299516908,
      "loss": 0.7016,
      "step": 390
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029428019323671494,
      "loss": 0.7149,
      "step": 400
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.7104699611663818,
      "eval_runtime": 150.0015,
      "eval_samples_per_second": 13.333,
      "eval_steps_per_second": 1.667,
      "step": 400
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029408695652173914,
      "loss": 0.7133,
      "step": 410
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002938937198067633,
      "loss": 0.7568,
      "step": 420
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002937004830917874,
      "loss": 0.7159,
      "step": 430
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029350724637681156,
      "loss": 0.7356,
      "step": 440
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029331400966183576,
      "loss": 0.665,
      "step": 450
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002931207729468599,
      "loss": 0.7017,
      "step": 460
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029292753623188405,
      "loss": 0.6979,
      "step": 470
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002927342995169082,
      "loss": 0.7104,
      "step": 480
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00029254106280193233,
      "loss": 0.6907,
      "step": 490
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00029234782608695653,
      "loss": 0.7407,
      "step": 500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00029215458937198067,
      "loss": 0.7028,
      "step": 510
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002919613526570048,
      "loss": 0.7102,
      "step": 520
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00029176811594202896,
      "loss": 0.6956,
      "step": 530
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00029157487922705315,
      "loss": 0.6926,
      "step": 540
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002913816425120773,
      "loss": 0.7114,
      "step": 550
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00029118840579710144,
      "loss": 0.7066,
      "step": 560
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002909951690821256,
      "loss": 0.6853,
      "step": 570
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002908019323671497,
      "loss": 0.7063,
      "step": 580
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002906086956521739,
      "loss": 0.7064,
      "step": 590
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00029041545893719806,
      "loss": 0.729,
      "step": 600
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.7055845856666565,
      "eval_runtime": 149.8446,
      "eval_samples_per_second": 13.347,
      "eval_steps_per_second": 1.668,
      "step": 600
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002902222222222222,
      "loss": 0.7098,
      "step": 610
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00029002898550724635,
      "loss": 0.6363,
      "step": 620
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002898357487922705,
      "loss": 0.7128,
      "step": 630
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002896425120772947,
      "loss": 0.692,
      "step": 640
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00028944927536231883,
      "loss": 0.7068,
      "step": 650
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00028925603864734297,
      "loss": 0.7191,
      "step": 660
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002890628019323671,
      "loss": 0.6865,
      "step": 670
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002888695652173913,
      "loss": 0.7331,
      "step": 680
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00028867632850241545,
      "loss": 0.652,
      "step": 690
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002884830917874396,
      "loss": 0.7188,
      "step": 700
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00028828985507246374,
      "loss": 0.6834,
      "step": 710
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002880966183574879,
      "loss": 0.693,
      "step": 720
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002879033816425121,
      "loss": 0.7445,
      "step": 730
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002877101449275362,
      "loss": 0.6904,
      "step": 740
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00028751690821256036,
      "loss": 0.7547,
      "step": 750
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002873236714975845,
      "loss": 0.7068,
      "step": 760
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002871304347826087,
      "loss": 0.6677,
      "step": 770
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00028693719806763285,
      "loss": 0.6808,
      "step": 780
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000286743961352657,
      "loss": 0.7142,
      "step": 790
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00028655072463768113,
      "loss": 0.7126,
      "step": 800
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.7015364766120911,
      "eval_runtime": 149.6299,
      "eval_samples_per_second": 13.366,
      "eval_steps_per_second": 1.671,
      "step": 800
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002863574879227053,
      "loss": 0.69,
      "step": 810
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00028616425120772947,
      "loss": 0.6772,
      "step": 820
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002859710144927536,
      "loss": 0.6881,
      "step": 830
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00028577777777777776,
      "loss": 0.6509,
      "step": 840
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002855845410628019,
      "loss": 0.6924,
      "step": 850
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00028539130434782604,
      "loss": 0.7116,
      "step": 860
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00028519806763285024,
      "loss": 0.711,
      "step": 870
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002850048309178744,
      "loss": 0.6897,
      "step": 880
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002848115942028985,
      "loss": 0.6689,
      "step": 890
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00028461835748792266,
      "loss": 0.6891,
      "step": 900
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00028442512077294686,
      "loss": 0.6985,
      "step": 910
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000284231884057971,
      "loss": 0.6643,
      "step": 920
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028403864734299515,
      "loss": 0.7277,
      "step": 930
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002838454106280193,
      "loss": 0.703,
      "step": 940
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028365217391304343,
      "loss": 0.6678,
      "step": 950
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00028345893719806763,
      "loss": 0.6836,
      "step": 960
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00028326570048309177,
      "loss": 0.7164,
      "step": 970
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002830724637681159,
      "loss": 0.6382,
      "step": 980
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028287922705314006,
      "loss": 0.7,
      "step": 990
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002826859903381642,
      "loss": 0.6974,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.6978325843811035,
      "eval_runtime": 149.6863,
      "eval_samples_per_second": 13.361,
      "eval_steps_per_second": 1.67,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002824927536231884,
      "loss": 0.6644,
      "step": 1010
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00028229951690821254,
      "loss": 0.672,
      "step": 1020
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002821062801932367,
      "loss": 0.7036,
      "step": 1030
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002819130434782608,
      "loss": 0.7049,
      "step": 1040
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000281719806763285,
      "loss": 0.7033,
      "step": 1050
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00028152657004830916,
      "loss": 0.7059,
      "step": 1060
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002813333333333333,
      "loss": 0.6915,
      "step": 1070
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00028114009661835745,
      "loss": 0.7609,
      "step": 1080
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002809468599033816,
      "loss": 0.6756,
      "step": 1090
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002807536231884058,
      "loss": 0.7127,
      "step": 1100
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00028056038647342993,
      "loss": 0.7212,
      "step": 1110
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00028036714975845407,
      "loss": 0.6974,
      "step": 1120
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002801739130434782,
      "loss": 0.6948,
      "step": 1130
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002799806763285024,
      "loss": 0.73,
      "step": 1140
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00027978743961352655,
      "loss": 0.7154,
      "step": 1150
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002795942028985507,
      "loss": 0.7007,
      "step": 1160
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00027940096618357484,
      "loss": 0.6854,
      "step": 1170
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.000279207729468599,
      "loss": 0.7075,
      "step": 1180
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002790144927536232,
      "loss": 0.6942,
      "step": 1190
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002788212560386473,
      "loss": 0.7389,
      "step": 1200
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.6969788670539856,
      "eval_runtime": 149.7859,
      "eval_samples_per_second": 13.352,
      "eval_steps_per_second": 1.669,
      "step": 1200
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00027862801932367146,
      "loss": 0.6633,
      "step": 1210
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002784347826086956,
      "loss": 0.6741,
      "step": 1220
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00027824154589371975,
      "loss": 0.702,
      "step": 1230
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00027804830917874395,
      "loss": 0.691,
      "step": 1240
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002778550724637681,
      "loss": 0.7049,
      "step": 1250
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00027766183574879223,
      "loss": 0.7254,
      "step": 1260
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002774685990338164,
      "loss": 0.6732,
      "step": 1270
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00027727536231884057,
      "loss": 0.6995,
      "step": 1280
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002770821256038647,
      "loss": 0.687,
      "step": 1290
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00027688888888888885,
      "loss": 0.6831,
      "step": 1300
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.000276695652173913,
      "loss": 0.7189,
      "step": 1310
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00027650241545893714,
      "loss": 0.6996,
      "step": 1320
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00027630917874396134,
      "loss": 0.696,
      "step": 1330
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002761159420289855,
      "loss": 0.7237,
      "step": 1340
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002759227053140096,
      "loss": 0.7266,
      "step": 1350
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00027572946859903376,
      "loss": 0.6745,
      "step": 1360
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002755362318840579,
      "loss": 0.7102,
      "step": 1370
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002753429951690821,
      "loss": 0.6922,
      "step": 1380
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00027514975845410625,
      "loss": 0.7059,
      "step": 1390
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002749565217391304,
      "loss": 0.7198,
      "step": 1400
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.6949622631072998,
      "eval_runtime": 150.1806,
      "eval_samples_per_second": 13.317,
      "eval_steps_per_second": 1.665,
      "step": 1400
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027476328502415453,
      "loss": 0.6608,
      "step": 1410
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027457004830917873,
      "loss": 0.6699,
      "step": 1420
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027437681159420287,
      "loss": 0.6817,
      "step": 1430
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.000274183574879227,
      "loss": 0.7019,
      "step": 1440
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027399033816425116,
      "loss": 0.6839,
      "step": 1450
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002737971014492753,
      "loss": 0.6725,
      "step": 1460
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002736038647342995,
      "loss": 0.7065,
      "step": 1470
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00027341062801932364,
      "loss": 0.6728,
      "step": 1480
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002732173913043478,
      "loss": 0.6449,
      "step": 1490
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002730241545893719,
      "loss": 0.7094,
      "step": 1500
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002728309178743961,
      "loss": 0.6881,
      "step": 1510
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00027263768115942026,
      "loss": 0.6804,
      "step": 1520
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002724444444444444,
      "loss": 0.6822,
      "step": 1530
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00027225120772946855,
      "loss": 0.6816,
      "step": 1540
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002720579710144927,
      "loss": 0.6615,
      "step": 1550
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002718647342995169,
      "loss": 0.6945,
      "step": 1560
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00027167149758454103,
      "loss": 0.7249,
      "step": 1570
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00027147826086956517,
      "loss": 0.7061,
      "step": 1580
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002712850241545893,
      "loss": 0.6909,
      "step": 1590
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002710917874396135,
      "loss": 0.7214,
      "step": 1600
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.6923746466636658,
      "eval_runtime": 150.1003,
      "eval_samples_per_second": 13.324,
      "eval_steps_per_second": 1.666,
      "step": 1600
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00027089855072463765,
      "loss": 0.7448,
      "step": 1610
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002707053140096618,
      "loss": 0.6746,
      "step": 1620
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00027051207729468594,
      "loss": 0.6952,
      "step": 1630
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002703188405797101,
      "loss": 0.6985,
      "step": 1640
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002701256038647343,
      "loss": 0.706,
      "step": 1650
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002699323671497584,
      "loss": 0.6838,
      "step": 1660
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00026973913043478256,
      "loss": 0.6809,
      "step": 1670
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002695458937198067,
      "loss": 0.7066,
      "step": 1680
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002693526570048309,
      "loss": 0.6828,
      "step": 1690
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00026915942028985505,
      "loss": 0.6653,
      "step": 1700
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002689661835748792,
      "loss": 0.6772,
      "step": 1710
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00026877294685990333,
      "loss": 0.6798,
      "step": 1720
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00026857971014492753,
      "loss": 0.6838,
      "step": 1730
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00026838647342995167,
      "loss": 0.7115,
      "step": 1740
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002681932367149758,
      "loss": 0.6907,
      "step": 1750
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00026799999999999995,
      "loss": 0.6587,
      "step": 1760
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002678067632850241,
      "loss": 0.7089,
      "step": 1770
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002676135265700483,
      "loss": 0.6947,
      "step": 1780
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00026742028985507244,
      "loss": 0.698,
      "step": 1790
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002672270531400966,
      "loss": 0.7183,
      "step": 1800
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.6911550164222717,
      "eval_runtime": 368.3262,
      "eval_samples_per_second": 5.43,
      "eval_steps_per_second": 0.679,
      "step": 1800
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002670338164251207,
      "loss": 0.7501,
      "step": 1810
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002668405797101449,
      "loss": 0.6472,
      "step": 1820
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00026664734299516906,
      "loss": 0.6773,
      "step": 1830
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002664541062801932,
      "loss": 0.7195,
      "step": 1840
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00026626086956521735,
      "loss": 0.7111,
      "step": 1850
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002660676328502415,
      "loss": 0.6921,
      "step": 1860
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002658743961352657,
      "loss": 0.6776,
      "step": 1870
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026568115942028983,
      "loss": 0.6911,
      "step": 1880
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026548792270531397,
      "loss": 0.7253,
      "step": 1890
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002652946859903381,
      "loss": 0.672,
      "step": 1900
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002651014492753623,
      "loss": 0.7156,
      "step": 1910
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00026490821256038645,
      "loss": 0.6929,
      "step": 1920
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002647149758454106,
      "loss": 0.714,
      "step": 1930
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00026452173913043474,
      "loss": 0.6834,
      "step": 1940
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002643285024154589,
      "loss": 0.7413,
      "step": 1950
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002641352657004831,
      "loss": 0.6369,
      "step": 1960
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002639420289855072,
      "loss": 0.6581,
      "step": 1970
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00026374879227053136,
      "loss": 0.644,
      "step": 1980
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002635555555555555,
      "loss": 0.675,
      "step": 1990
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002633623188405797,
      "loss": 0.6935,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.6894997954368591,
      "eval_runtime": 384.4995,
      "eval_samples_per_second": 5.202,
      "eval_steps_per_second": 0.65,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00026316908212560384,
      "loss": 0.6739,
      "step": 2010
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.000262975845410628,
      "loss": 0.719,
      "step": 2020
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00026278260869565213,
      "loss": 0.6724,
      "step": 2030
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002625893719806763,
      "loss": 0.6736,
      "step": 2040
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026239613526570047,
      "loss": 0.675,
      "step": 2050
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002622028985507246,
      "loss": 0.7289,
      "step": 2060
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026200966183574875,
      "loss": 0.7081,
      "step": 2070
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002618164251207729,
      "loss": 0.6529,
      "step": 2080
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002616231884057971,
      "loss": 0.6659,
      "step": 2090
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00026142995169082124,
      "loss": 0.663,
      "step": 2100
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002612367149758454,
      "loss": 0.7021,
      "step": 2110
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002610434782608695,
      "loss": 0.6856,
      "step": 2120
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002608502415458937,
      "loss": 0.6762,
      "step": 2130
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00026065700483091786,
      "loss": 0.6754,
      "step": 2140
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.000260463768115942,
      "loss": 0.7019,
      "step": 2150
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00026027053140096615,
      "loss": 0.6847,
      "step": 2160
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002600772946859903,
      "loss": 0.6971,
      "step": 2170
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002598840579710145,
      "loss": 0.6819,
      "step": 2180
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002596908212560386,
      "loss": 0.6898,
      "step": 2190
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00025949758454106277,
      "loss": 0.698,
      "step": 2200
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.6893799304962158,
      "eval_runtime": 386.4657,
      "eval_samples_per_second": 5.175,
      "eval_steps_per_second": 0.647,
      "step": 2200
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002593043478260869,
      "loss": 0.6876,
      "step": 2210
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002591111111111111,
      "loss": 0.6798,
      "step": 2220
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00025891787439613525,
      "loss": 0.706,
      "step": 2230
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002587246376811594,
      "loss": 0.6495,
      "step": 2240
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00025853140096618354,
      "loss": 0.6646,
      "step": 2250
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002583381642512077,
      "loss": 0.6417,
      "step": 2260
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002581449275362319,
      "loss": 0.6681,
      "step": 2270
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.000257951690821256,
      "loss": 0.6689,
      "step": 2280
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00025775845410628016,
      "loss": 0.6837,
      "step": 2290
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002575652173913043,
      "loss": 0.7031,
      "step": 2300
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002573719806763285,
      "loss": 0.6746,
      "step": 2310
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00025717874396135264,
      "loss": 0.6951,
      "step": 2320
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002569855072463768,
      "loss": 0.6988,
      "step": 2330
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00025679227053140093,
      "loss": 0.6541,
      "step": 2340
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002565990338164251,
      "loss": 0.6366,
      "step": 2350
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00025640579710144927,
      "loss": 0.7011,
      "step": 2360
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002562125603864734,
      "loss": 0.6935,
      "step": 2370
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00025601932367149755,
      "loss": 0.6931,
      "step": 2380
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002558260869565217,
      "loss": 0.7004,
      "step": 2390
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002556328502415459,
      "loss": 0.6556,
      "step": 2400
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.6892030239105225,
      "eval_runtime": 391.4843,
      "eval_samples_per_second": 5.109,
      "eval_steps_per_second": 0.639,
      "step": 2400
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00025543961352657003,
      "loss": 0.664,
      "step": 2410
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002552463768115942,
      "loss": 0.7162,
      "step": 2420
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002550531400966183,
      "loss": 0.646,
      "step": 2430
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002548599033816425,
      "loss": 0.6515,
      "step": 2440
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00025466666666666666,
      "loss": 0.6953,
      "step": 2450
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002544734299516908,
      "loss": 0.6887,
      "step": 2460
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00025428019323671494,
      "loss": 0.6739,
      "step": 2470
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002540869565217391,
      "loss": 0.6923,
      "step": 2480
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002538937198067633,
      "loss": 0.6877,
      "step": 2490
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002537004830917874,
      "loss": 0.6865,
      "step": 2500
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00025350724637681157,
      "loss": 0.6337,
      "step": 2510
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002533140096618357,
      "loss": 0.7073,
      "step": 2520
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002531207729468599,
      "loss": 0.6973,
      "step": 2530
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00025292753623188405,
      "loss": 0.6719,
      "step": 2540
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002527342995169082,
      "loss": 0.6674,
      "step": 2550
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00025254106280193234,
      "loss": 0.6745,
      "step": 2560
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002523478260869565,
      "loss": 0.6914,
      "step": 2570
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002521545893719807,
      "loss": 0.6382,
      "step": 2580
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002519613526570048,
      "loss": 0.6644,
      "step": 2590
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00025176811594202896,
      "loss": 0.6892,
      "step": 2600
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.6873727440834045,
      "eval_runtime": 390.6367,
      "eval_samples_per_second": 5.12,
      "eval_steps_per_second": 0.64,
      "step": 2600
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002515748792270531,
      "loss": 0.6592,
      "step": 2610
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002513816425120773,
      "loss": 0.6827,
      "step": 2620
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00025118840579710144,
      "loss": 0.6436,
      "step": 2630
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002509951690821256,
      "loss": 0.6969,
      "step": 2640
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002508019323671497,
      "loss": 0.6747,
      "step": 2650
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002506086956521739,
      "loss": 0.697,
      "step": 2660
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00025041545893719807,
      "loss": 0.7146,
      "step": 2670
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002502222222222222,
      "loss": 0.689,
      "step": 2680
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00025002898550724635,
      "loss": 0.6957,
      "step": 2690
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002498357487922705,
      "loss": 0.708,
      "step": 2700
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002496425120772947,
      "loss": 0.7306,
      "step": 2710
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00024944927536231883,
      "loss": 0.6418,
      "step": 2720
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.000249256038647343,
      "loss": 0.6888,
      "step": 2730
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002490628019323671,
      "loss": 0.6573,
      "step": 2740
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002488695652173913,
      "loss": 0.7008,
      "step": 2750
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00024867632850241546,
      "loss": 0.689,
      "step": 2760
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002484830917874396,
      "loss": 0.6739,
      "step": 2770
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00024828985507246374,
      "loss": 0.6867,
      "step": 2780
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002480966183574879,
      "loss": 0.6874,
      "step": 2790
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002479033816425121,
      "loss": 0.6858,
      "step": 2800
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.686144232749939,
      "eval_runtime": 396.2716,
      "eval_samples_per_second": 5.047,
      "eval_steps_per_second": 0.631,
      "step": 2800
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002477101449275362,
      "loss": 0.659,
      "step": 2810
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00024751690821256037,
      "loss": 0.6537,
      "step": 2820
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002473236714975845,
      "loss": 0.7331,
      "step": 2830
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002471304347826087,
      "loss": 0.6855,
      "step": 2840
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00024693719806763285,
      "loss": 0.7252,
      "step": 2850
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.000246743961352657,
      "loss": 0.7,
      "step": 2860
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00024655072463768113,
      "loss": 0.6917,
      "step": 2870
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002463574879227053,
      "loss": 0.6828,
      "step": 2880
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002461642512077295,
      "loss": 0.6747,
      "step": 2890
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002459710144927536,
      "loss": 0.696,
      "step": 2900
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00024577777777777776,
      "loss": 0.6573,
      "step": 2910
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002455845410628019,
      "loss": 0.6811,
      "step": 2920
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002453913043478261,
      "loss": 0.6992,
      "step": 2930
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00024519806763285024,
      "loss": 0.6972,
      "step": 2940
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002450048309178744,
      "loss": 0.6685,
      "step": 2950
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002448115942028985,
      "loss": 0.6531,
      "step": 2960
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002446183574879227,
      "loss": 0.6749,
      "step": 2970
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024442512077294686,
      "loss": 0.6811,
      "step": 2980
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.000244231884057971,
      "loss": 0.6904,
      "step": 2990
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00024403864734299515,
      "loss": 0.6819,
      "step": 3000
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.6858677864074707,
      "eval_runtime": 418.7569,
      "eval_samples_per_second": 4.776,
      "eval_steps_per_second": 0.597,
      "step": 3000
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002438454106280193,
      "loss": 0.6773,
      "step": 3010
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00024365217391304346,
      "loss": 0.7034,
      "step": 3020
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002434589371980676,
      "loss": 0.6804,
      "step": 3030
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00024326570048309177,
      "loss": 0.6912,
      "step": 3040
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00024307246376811592,
      "loss": 0.683,
      "step": 3050
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002428792270531401,
      "loss": 0.6767,
      "step": 3060
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00024268599033816423,
      "loss": 0.6683,
      "step": 3070
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024249275362318837,
      "loss": 0.6777,
      "step": 3080
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024229951690821254,
      "loss": 0.6813,
      "step": 3090
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024210628019323668,
      "loss": 0.6878,
      "step": 3100
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024191304347826085,
      "loss": 0.6885,
      "step": 3110
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.000241719806763285,
      "loss": 0.6373,
      "step": 3120
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024152657004830917,
      "loss": 0.6982,
      "step": 3130
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002413333333333333,
      "loss": 0.6485,
      "step": 3140
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024114009661835748,
      "loss": 0.6522,
      "step": 3150
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024094685990338162,
      "loss": 0.6092,
      "step": 3160
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024075362318840576,
      "loss": 0.6362,
      "step": 3170
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024056038647342993,
      "loss": 0.6725,
      "step": 3180
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024036714975845408,
      "loss": 0.6718,
      "step": 3190
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024017391304347825,
      "loss": 0.6738,
      "step": 3200
    },
    {
      "epoch": 1.02,
      "eval_loss": 0.6860418915748596,
      "eval_runtime": 407.1637,
      "eval_samples_per_second": 4.912,
      "eval_steps_per_second": 0.614,
      "step": 3200
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002399806763285024,
      "loss": 0.658,
      "step": 3210
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00023978743961352656,
      "loss": 0.6325,
      "step": 3220
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002395942028985507,
      "loss": 0.6372,
      "step": 3230
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00023940096618357487,
      "loss": 0.6075,
      "step": 3240
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.000239207729468599,
      "loss": 0.6212,
      "step": 3250
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00023901449275362315,
      "loss": 0.6373,
      "step": 3260
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023882125603864732,
      "loss": 0.6799,
      "step": 3270
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002386280193236715,
      "loss": 0.6769,
      "step": 3280
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023843478260869564,
      "loss": 0.6265,
      "step": 3290
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00023824154589371978,
      "loss": 0.6627,
      "step": 3300
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00023804830917874392,
      "loss": 0.6602,
      "step": 3310
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002378550724637681,
      "loss": 0.671,
      "step": 3320
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00023766183574879226,
      "loss": 0.6297,
      "step": 3330
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002374685990338164,
      "loss": 0.662,
      "step": 3340
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00023727536231884055,
      "loss": 0.6224,
      "step": 3350
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023708212560386472,
      "loss": 0.6527,
      "step": 3360
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023688888888888889,
      "loss": 0.6388,
      "step": 3370
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023669565217391303,
      "loss": 0.6626,
      "step": 3380
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023650241545893717,
      "loss": 0.6522,
      "step": 3390
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002363091787439613,
      "loss": 0.6514,
      "step": 3400
    },
    {
      "epoch": 1.09,
      "eval_loss": 0.6873291730880737,
      "eval_runtime": 392.2057,
      "eval_samples_per_second": 5.099,
      "eval_steps_per_second": 0.637,
      "step": 3400
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023611594202898548,
      "loss": 0.6147,
      "step": 3410
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023592270531400965,
      "loss": 0.6282,
      "step": 3420
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002357294685990338,
      "loss": 0.6168,
      "step": 3430
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00023553623188405794,
      "loss": 0.6858,
      "step": 3440
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00023534299516908208,
      "loss": 0.6219,
      "step": 3450
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023514975845410628,
      "loss": 0.6566,
      "step": 3460
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023495652173913042,
      "loss": 0.6633,
      "step": 3470
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023476328502415456,
      "loss": 0.6526,
      "step": 3480
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002345700483091787,
      "loss": 0.6544,
      "step": 3490
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023437681159420287,
      "loss": 0.6588,
      "step": 3500
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023418357487922704,
      "loss": 0.6477,
      "step": 3510
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002339903381642512,
      "loss": 0.6494,
      "step": 3520
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023379710144927533,
      "loss": 0.6219,
      "step": 3530
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023360386473429947,
      "loss": 0.6369,
      "step": 3540
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023341062801932367,
      "loss": 0.6647,
      "step": 3550
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002332173913043478,
      "loss": 0.6477,
      "step": 3560
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023302415458937195,
      "loss": 0.6721,
      "step": 3570
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002328309178743961,
      "loss": 0.6327,
      "step": 3580
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002326376811594203,
      "loss": 0.6668,
      "step": 3590
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00023244444444444444,
      "loss": 0.6422,
      "step": 3600
    },
    {
      "epoch": 1.15,
      "eval_loss": 0.6869779229164124,
      "eval_runtime": 396.5319,
      "eval_samples_per_second": 5.044,
      "eval_steps_per_second": 0.63,
      "step": 3600
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023225120772946858,
      "loss": 0.6774,
      "step": 3610
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023205797101449272,
      "loss": 0.6663,
      "step": 3620
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023186473429951686,
      "loss": 0.6903,
      "step": 3630
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023167149758454106,
      "loss": 0.6362,
      "step": 3640
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002314782608695652,
      "loss": 0.6187,
      "step": 3650
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00023128502415458935,
      "loss": 0.6256,
      "step": 3660
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002310917874396135,
      "loss": 0.6493,
      "step": 3670
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023089855072463768,
      "loss": 0.635,
      "step": 3680
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023070531400966183,
      "loss": 0.6973,
      "step": 3690
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023051207729468597,
      "loss": 0.6562,
      "step": 3700
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002303188405797101,
      "loss": 0.6327,
      "step": 3710
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00023012560386473425,
      "loss": 0.6736,
      "step": 3720
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00022993236714975845,
      "loss": 0.6178,
      "step": 3730
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002297391304347826,
      "loss": 0.6574,
      "step": 3740
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00022954589371980674,
      "loss": 0.6848,
      "step": 3750
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00022935265700483088,
      "loss": 0.6478,
      "step": 3760
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022915942028985508,
      "loss": 0.6144,
      "step": 3770
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022896618357487922,
      "loss": 0.6609,
      "step": 3780
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022877294685990336,
      "loss": 0.6744,
      "step": 3790
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002285797101449275,
      "loss": 0.636,
      "step": 3800
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.6865532398223877,
      "eval_runtime": 393.5432,
      "eval_samples_per_second": 5.082,
      "eval_steps_per_second": 0.635,
      "step": 3800
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00022838647342995165,
      "loss": 0.654,
      "step": 3810
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00022819323671497584,
      "loss": 0.6442,
      "step": 3820
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.6835,
      "step": 3830
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022780676328502413,
      "loss": 0.6339,
      "step": 3840
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022761352657004827,
      "loss": 0.6512,
      "step": 3850
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00022742028985507247,
      "loss": 0.622,
      "step": 3860
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002272270531400966,
      "loss": 0.6802,
      "step": 3870
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00022703381642512075,
      "loss": 0.6911,
      "step": 3880
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002268405797101449,
      "loss": 0.6815,
      "step": 3890
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002266473429951691,
      "loss": 0.6569,
      "step": 3900
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022645410628019323,
      "loss": 0.6363,
      "step": 3910
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022626086956521738,
      "loss": 0.6371,
      "step": 3920
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00022606763285024152,
      "loss": 0.6458,
      "step": 3930
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00022587439613526566,
      "loss": 0.668,
      "step": 3940
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00022568115942028986,
      "loss": 0.6676,
      "step": 3950
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.000225487922705314,
      "loss": 0.6692,
      "step": 3960
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022529468599033814,
      "loss": 0.6502,
      "step": 3970
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002251014492753623,
      "loss": 0.6387,
      "step": 3980
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00022490821256038646,
      "loss": 0.6475,
      "step": 3990
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00022471497584541063,
      "loss": 0.6627,
      "step": 4000
    },
    {
      "epoch": 1.28,
      "eval_loss": 0.6856961846351624,
      "eval_runtime": 387.1458,
      "eval_samples_per_second": 5.166,
      "eval_steps_per_second": 0.646,
      "step": 4000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00022452173913043477,
      "loss": 0.6346,
      "step": 4010
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002243285024154589,
      "loss": 0.6593,
      "step": 4020
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022413526570048305,
      "loss": 0.6397,
      "step": 4030
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022394202898550725,
      "loss": 0.641,
      "step": 4040
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002237487922705314,
      "loss": 0.6766,
      "step": 4050
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00022355555555555554,
      "loss": 0.6931,
      "step": 4060
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00022336231884057968,
      "loss": 0.6618,
      "step": 4070
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022316908212560385,
      "loss": 0.6655,
      "step": 4080
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022297584541062802,
      "loss": 0.647,
      "step": 4090
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022278260869565216,
      "loss": 0.6337,
      "step": 4100
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002225893719806763,
      "loss": 0.6957,
      "step": 4110
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022239613526570044,
      "loss": 0.6435,
      "step": 4120
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022220289855072464,
      "loss": 0.6272,
      "step": 4130
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022200966183574878,
      "loss": 0.6502,
      "step": 4140
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022181642512077293,
      "loss": 0.6514,
      "step": 4150
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022162318840579707,
      "loss": 0.6625,
      "step": 4160
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022142995169082124,
      "loss": 0.6284,
      "step": 4170
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002212367149758454,
      "loss": 0.6503,
      "step": 4180
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00022104347826086955,
      "loss": 0.6425,
      "step": 4190
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002208502415458937,
      "loss": 0.6818,
      "step": 4200
    },
    {
      "epoch": 1.34,
      "eval_loss": 0.6859603524208069,
      "eval_runtime": 382.589,
      "eval_samples_per_second": 5.228,
      "eval_steps_per_second": 0.653,
      "step": 4200
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00022065700483091784,
      "loss": 0.6299,
      "step": 4210
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.000220463768115942,
      "loss": 0.6508,
      "step": 4220
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00022027053140096618,
      "loss": 0.6572,
      "step": 4230
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00022007729468599032,
      "loss": 0.6781,
      "step": 4240
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00021988405797101446,
      "loss": 0.6806,
      "step": 4250
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00021969082125603863,
      "loss": 0.632,
      "step": 4260
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002194975845410628,
      "loss": 0.6631,
      "step": 4270
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00021930434782608694,
      "loss": 0.6729,
      "step": 4280
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00021911111111111109,
      "loss": 0.6618,
      "step": 4290
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021891787439613525,
      "loss": 0.6132,
      "step": 4300
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002187246376811594,
      "loss": 0.6457,
      "step": 4310
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021853140096618357,
      "loss": 0.6443,
      "step": 4320
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002183381642512077,
      "loss": 0.653,
      "step": 4330
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021814492753623185,
      "loss": 0.6706,
      "step": 4340
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021795169082125602,
      "loss": 0.6684,
      "step": 4350
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021775845410628016,
      "loss": 0.6597,
      "step": 4360
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021756521739130433,
      "loss": 0.6478,
      "step": 4370
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021737198067632848,
      "loss": 0.6411,
      "step": 4380
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021717874396135265,
      "loss": 0.657,
      "step": 4390
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002169855072463768,
      "loss": 0.663,
      "step": 4400
    },
    {
      "epoch": 1.41,
      "eval_loss": 0.6874400973320007,
      "eval_runtime": 387.4088,
      "eval_samples_per_second": 5.163,
      "eval_steps_per_second": 0.645,
      "step": 4400
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021679227053140096,
      "loss": 0.6833,
      "step": 4410
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002165990338164251,
      "loss": 0.6827,
      "step": 4420
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021640579710144924,
      "loss": 0.6789,
      "step": 4430
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0002162125603864734,
      "loss": 0.6582,
      "step": 4440
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021601932367149756,
      "loss": 0.6222,
      "step": 4450
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021582608695652173,
      "loss": 0.6314,
      "step": 4460
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021563285024154587,
      "loss": 0.6466,
      "step": 4470
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021543961352657004,
      "loss": 0.6734,
      "step": 4480
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021524637681159418,
      "loss": 0.6552,
      "step": 4490
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021505314009661835,
      "loss": 0.7156,
      "step": 4500
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002148599033816425,
      "loss": 0.6548,
      "step": 4510
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00021466666666666664,
      "loss": 0.7265,
      "step": 4520
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002144734299516908,
      "loss": 0.6757,
      "step": 4530
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00021428019323671495,
      "loss": 0.6914,
      "step": 4540
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021408695652173912,
      "loss": 0.6746,
      "step": 4550
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021389371980676326,
      "loss": 0.7085,
      "step": 4560
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021370048309178743,
      "loss": 0.7058,
      "step": 4570
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021350724637681157,
      "loss": 0.6599,
      "step": 4580
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021331400966183571,
      "loss": 0.6653,
      "step": 4590
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021312077294685988,
      "loss": 0.6757,
      "step": 4600
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.6850671172142029,
      "eval_runtime": 410.1023,
      "eval_samples_per_second": 4.877,
      "eval_steps_per_second": 0.61,
      "step": 4600
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00021292753623188405,
      "loss": 0.6644,
      "step": 4610
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002127342995169082,
      "loss": 0.6444,
      "step": 4620
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00021254106280193234,
      "loss": 0.6548,
      "step": 4630
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002123478260869565,
      "loss": 0.6361,
      "step": 4640
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021215458937198065,
      "loss": 0.6726,
      "step": 4650
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021196135265700482,
      "loss": 0.6451,
      "step": 4660
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021176811594202896,
      "loss": 0.6939,
      "step": 4670
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0002115748792270531,
      "loss": 0.6569,
      "step": 4680
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021138164251207728,
      "loss": 0.6405,
      "step": 4690
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021118840579710145,
      "loss": 0.6261,
      "step": 4700
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002109951690821256,
      "loss": 0.6544,
      "step": 4710
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00021080193236714973,
      "loss": 0.6367,
      "step": 4720
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002106086956521739,
      "loss": 0.6691,
      "step": 4730
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021041545893719804,
      "loss": 0.6228,
      "step": 4740
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0002102222222222222,
      "loss": 0.6628,
      "step": 4750
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021002898550724635,
      "loss": 0.6678,
      "step": 4760
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002098357487922705,
      "loss": 0.6699,
      "step": 4770
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00020964251207729467,
      "loss": 0.683,
      "step": 4780
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00020944927536231884,
      "loss": 0.688,
      "step": 4790
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00020925603864734298,
      "loss": 0.6661,
      "step": 4800
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.6855675578117371,
      "eval_runtime": 385.7747,
      "eval_samples_per_second": 5.184,
      "eval_steps_per_second": 0.648,
      "step": 4800
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00020906280193236712,
      "loss": 0.6605,
      "step": 4810
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00020886956521739126,
      "loss": 0.7119,
      "step": 4820
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00020867632850241543,
      "loss": 0.6181,
      "step": 4830
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002084830917874396,
      "loss": 0.6484,
      "step": 4840
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00020828985507246375,
      "loss": 0.6747,
      "step": 4850
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002080966183574879,
      "loss": 0.6455,
      "step": 4860
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020790338164251206,
      "loss": 0.6591,
      "step": 4870
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020771014492753623,
      "loss": 0.6898,
      "step": 4880
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020751690821256037,
      "loss": 0.6491,
      "step": 4890
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002073236714975845,
      "loss": 0.66,
      "step": 4900
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00020713043478260866,
      "loss": 0.6423,
      "step": 4910
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00020693719806763285,
      "loss": 0.6102,
      "step": 4920
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.000206743961352657,
      "loss": 0.6709,
      "step": 4930
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020655072463768114,
      "loss": 0.6432,
      "step": 4940
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020635748792270528,
      "loss": 0.6388,
      "step": 4950
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020616425120772942,
      "loss": 0.6574,
      "step": 4960
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020597101449275362,
      "loss": 0.6687,
      "step": 4970
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020577777777777776,
      "loss": 0.674,
      "step": 4980
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002055845410628019,
      "loss": 0.618,
      "step": 4990
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00020539130434782605,
      "loss": 0.689,
      "step": 5000
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.6850703358650208,
      "eval_runtime": 372.4881,
      "eval_samples_per_second": 5.369,
      "eval_steps_per_second": 0.671,
      "step": 5000
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00020519806763285024,
      "loss": 0.6865,
      "step": 5010
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002050048309178744,
      "loss": 0.6529,
      "step": 5020
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00020481159420289853,
      "loss": 0.6495,
      "step": 5030
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00020461835748792267,
      "loss": 0.6517,
      "step": 5040
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020442512077294681,
      "loss": 0.6684,
      "step": 5050
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.000204231884057971,
      "loss": 0.6412,
      "step": 5060
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020403864734299515,
      "loss": 0.6463,
      "step": 5070
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002038454106280193,
      "loss": 0.633,
      "step": 5080
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020365217391304344,
      "loss": 0.6473,
      "step": 5090
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020345893719806764,
      "loss": 0.6102,
      "step": 5100
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00020326570048309178,
      "loss": 0.6816,
      "step": 5110
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00020307246376811592,
      "loss": 0.6743,
      "step": 5120
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00020287922705314006,
      "loss": 0.6745,
      "step": 5130
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002026859903381642,
      "loss": 0.6435,
      "step": 5140
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002024927536231884,
      "loss": 0.6675,
      "step": 5150
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00020229951690821255,
      "loss": 0.6547,
      "step": 5160
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002021062801932367,
      "loss": 0.6464,
      "step": 5170
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020191304347826083,
      "loss": 0.6894,
      "step": 5180
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020171980676328503,
      "loss": 0.6613,
      "step": 5190
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020152657004830917,
      "loss": 0.6362,
      "step": 5200
    },
    {
      "epoch": 1.66,
      "eval_loss": 0.6850407719612122,
      "eval_runtime": 377.0156,
      "eval_samples_per_second": 5.305,
      "eval_steps_per_second": 0.663,
      "step": 5200
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002013333333333333,
      "loss": 0.636,
      "step": 5210
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020114009661835745,
      "loss": 0.6731,
      "step": 5220
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020094685990338165,
      "loss": 0.6429,
      "step": 5230
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002007536231884058,
      "loss": 0.6808,
      "step": 5240
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00020056038647342994,
      "loss": 0.6728,
      "step": 5250
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00020036714975845408,
      "loss": 0.6377,
      "step": 5260
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00020017391304347822,
      "loss": 0.6616,
      "step": 5270
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00019998067632850242,
      "loss": 0.6571,
      "step": 5280
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00019978743961352656,
      "loss": 0.6765,
      "step": 5290
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001995942028985507,
      "loss": 0.6636,
      "step": 5300
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00019940096618357485,
      "loss": 0.6621,
      "step": 5310
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00019920772946859904,
      "loss": 0.6788,
      "step": 5320
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019901449275362319,
      "loss": 0.6491,
      "step": 5330
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019882125603864733,
      "loss": 0.6413,
      "step": 5340
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019862801932367147,
      "loss": 0.6287,
      "step": 5350
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001984347826086956,
      "loss": 0.6624,
      "step": 5360
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001982415458937198,
      "loss": 0.6657,
      "step": 5370
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019804830917874395,
      "loss": 0.6617,
      "step": 5380
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0001978550724637681,
      "loss": 0.6791,
      "step": 5390
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00019766183574879224,
      "loss": 0.6537,
      "step": 5400
    },
    {
      "epoch": 1.73,
      "eval_loss": 0.683772623538971,
      "eval_runtime": 380.6447,
      "eval_samples_per_second": 5.254,
      "eval_steps_per_second": 0.657,
      "step": 5400
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00019746859903381643,
      "loss": 0.5705,
      "step": 5410
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00019727536231884058,
      "loss": 0.634,
      "step": 5420
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019708212560386472,
      "loss": 0.6558,
      "step": 5430
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019688888888888886,
      "loss": 0.657,
      "step": 5440
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.000196695652173913,
      "loss": 0.6615,
      "step": 5450
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001965024154589372,
      "loss": 0.6656,
      "step": 5460
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00019630917874396134,
      "loss": 0.6286,
      "step": 5470
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001961159420289855,
      "loss": 0.657,
      "step": 5480
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019592270531400963,
      "loss": 0.6891,
      "step": 5490
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001957294685990338,
      "loss": 0.6539,
      "step": 5500
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019553623188405797,
      "loss": 0.6335,
      "step": 5510
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001953429951690821,
      "loss": 0.6625,
      "step": 5520
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019514975845410625,
      "loss": 0.6597,
      "step": 5530
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019495652173913042,
      "loss": 0.6753,
      "step": 5540
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001947632850241546,
      "loss": 0.6644,
      "step": 5550
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00019457004830917874,
      "loss": 0.6434,
      "step": 5560
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00019437681159420288,
      "loss": 0.6604,
      "step": 5570
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019418357487922702,
      "loss": 0.6538,
      "step": 5580
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001939903381642512,
      "loss": 0.639,
      "step": 5590
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019379710144927536,
      "loss": 0.6668,
      "step": 5600
    },
    {
      "epoch": 1.79,
      "eval_loss": 0.6843588352203369,
      "eval_runtime": 378.4765,
      "eval_samples_per_second": 5.284,
      "eval_steps_per_second": 0.661,
      "step": 5600
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001936038647342995,
      "loss": 0.6375,
      "step": 5610
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019341062801932364,
      "loss": 0.6455,
      "step": 5620
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019321739130434781,
      "loss": 0.6856,
      "step": 5630
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019302415458937198,
      "loss": 0.648,
      "step": 5640
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019283091787439613,
      "loss": 0.6005,
      "step": 5650
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019263768115942027,
      "loss": 0.6719,
      "step": 5660
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001924444444444444,
      "loss": 0.6588,
      "step": 5670
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00019225120772946858,
      "loss": 0.6517,
      "step": 5680
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00019205797101449275,
      "loss": 0.6489,
      "step": 5690
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001918647342995169,
      "loss": 0.6283,
      "step": 5700
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00019167149758454104,
      "loss": 0.6523,
      "step": 5710
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001914782608695652,
      "loss": 0.5914,
      "step": 5720
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00019128502415458935,
      "loss": 0.6403,
      "step": 5730
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00019109178743961352,
      "loss": 0.6426,
      "step": 5740
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00019089855072463766,
      "loss": 0.6609,
      "step": 5750
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001907053140096618,
      "loss": 0.6394,
      "step": 5760
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019051207729468597,
      "loss": 0.6546,
      "step": 5770
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019031884057971014,
      "loss": 0.6879,
      "step": 5780
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019012560386473429,
      "loss": 0.6116,
      "step": 5790
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00018993236714975843,
      "loss": 0.6477,
      "step": 5800
    },
    {
      "epoch": 1.86,
      "eval_loss": 0.6840969920158386,
      "eval_runtime": 385.7288,
      "eval_samples_per_second": 5.185,
      "eval_steps_per_second": 0.648,
      "step": 5800
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001897391304347826,
      "loss": 0.6657,
      "step": 5810
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00018954589371980674,
      "loss": 0.6399,
      "step": 5820
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001893526570048309,
      "loss": 0.6587,
      "step": 5830
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00018915942028985505,
      "loss": 0.6333,
      "step": 5840
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00018896618357487922,
      "loss": 0.6327,
      "step": 5850
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018877294685990336,
      "loss": 0.6788,
      "step": 5860
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001885797101449275,
      "loss": 0.6765,
      "step": 5870
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018838647342995168,
      "loss": 0.6742,
      "step": 5880
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018819323671497582,
      "loss": 0.6462,
      "step": 5890
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.000188,
      "loss": 0.6907,
      "step": 5900
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00018780676328502413,
      "loss": 0.6645,
      "step": 5910
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001876135265700483,
      "loss": 0.6719,
      "step": 5920
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00018742028985507244,
      "loss": 0.6417,
      "step": 5930
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001872270531400966,
      "loss": 0.6383,
      "step": 5940
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00018703381642512076,
      "loss": 0.6342,
      "step": 5950
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001868405797101449,
      "loss": 0.657,
      "step": 5960
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00018664734299516907,
      "loss": 0.6832,
      "step": 5970
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001864541062801932,
      "loss": 0.6658,
      "step": 5980
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018626086956521738,
      "loss": 0.6614,
      "step": 5990
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018606763285024152,
      "loss": 0.6878,
      "step": 6000
    },
    {
      "epoch": 1.92,
      "eval_loss": 0.6828343272209167,
      "eval_runtime": 392.2472,
      "eval_samples_per_second": 5.099,
      "eval_steps_per_second": 0.637,
      "step": 6000
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001858743961352657,
      "loss": 0.6408,
      "step": 6010
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00018568115942028984,
      "loss": 0.6498,
      "step": 6020
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.000185487922705314,
      "loss": 0.6598,
      "step": 6030
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00018529468599033815,
      "loss": 0.6231,
      "step": 6040
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001851014492753623,
      "loss": 0.6906,
      "step": 6050
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00018490821256038646,
      "loss": 0.6467,
      "step": 6060
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001847149758454106,
      "loss": 0.6355,
      "step": 6070
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018452173913043477,
      "loss": 0.6794,
      "step": 6080
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018432850241545891,
      "loss": 0.6475,
      "step": 6090
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018413526570048306,
      "loss": 0.6165,
      "step": 6100
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018394202898550723,
      "loss": 0.6281,
      "step": 6110
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001837487922705314,
      "loss": 0.6473,
      "step": 6120
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018355555555555554,
      "loss": 0.6415,
      "step": 6130
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018336231884057968,
      "loss": 0.659,
      "step": 6140
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018316908212560385,
      "loss": 0.6821,
      "step": 6150
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018297584541062802,
      "loss": 0.6631,
      "step": 6160
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018278260869565216,
      "loss": 0.6332,
      "step": 6170
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001825893719806763,
      "loss": 0.6561,
      "step": 6180
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00018239613526570045,
      "loss": 0.6654,
      "step": 6190
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00018220289855072462,
      "loss": 0.6656,
      "step": 6200
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.6827040314674377,
      "eval_runtime": 411.107,
      "eval_samples_per_second": 4.865,
      "eval_steps_per_second": 0.608,
      "step": 6200
    }
  ],
  "logging_steps": 10,
  "max_steps": 15625,
  "num_train_epochs": 5,
  "save_steps": 200,
  "total_flos": 9.076709518992998e+17,
  "trial_name": null,
  "trial_params": null
}