{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4982640434434256,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008902341315766046,
      "grad_norm": 5.759979724884033,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 2.1539,
      "step": 10
    },
    {
      "epoch": 0.0017804682631532092,
      "grad_norm": 3.376688241958618,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 2.0986,
      "step": 20
    },
    {
      "epoch": 0.002670702394729814,
      "grad_norm": 2.2482147216796875,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.8036,
      "step": 30
    },
    {
      "epoch": 0.0035609365263064184,
      "grad_norm": 1.0116345882415771,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.4411,
      "step": 40
    },
    {
      "epoch": 0.004451170657883023,
      "grad_norm": 0.845442533493042,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 1.2537,
      "step": 50
    },
    {
      "epoch": 0.005341404789459628,
      "grad_norm": 0.7026575207710266,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 1.1555,
      "step": 60
    },
    {
      "epoch": 0.006231638921036233,
      "grad_norm": 0.6083096861839294,
      "learning_rate": 7.814044943820226e-06,
      "loss": 1.1106,
      "step": 70
    },
    {
      "epoch": 0.007121873052612837,
      "grad_norm": 0.5970392227172852,
      "learning_rate": 8.932022471910112e-06,
      "loss": 1.0414,
      "step": 80
    },
    {
      "epoch": 0.008012107184189442,
      "grad_norm": 0.6251844763755798,
      "learning_rate": 1.005e-05,
      "loss": 1.0337,
      "step": 90
    },
    {
      "epoch": 0.008902341315766046,
      "grad_norm": 0.610266923904419,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.9821,
      "step": 100
    },
    {
      "epoch": 0.00979257544734265,
      "grad_norm": 0.573571503162384,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 1.0131,
      "step": 110
    },
    {
      "epoch": 0.010682809578919256,
      "grad_norm": 0.6081265807151794,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 1.0025,
      "step": 120
    },
    {
      "epoch": 0.01157304371049586,
      "grad_norm": 0.6050735712051392,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 1.0098,
      "step": 130
    },
    {
      "epoch": 0.012463277842072466,
      "grad_norm": 0.5551068782806396,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.9697,
      "step": 140
    },
    {
      "epoch": 0.01335351197364907,
      "grad_norm": 0.6075469255447388,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.97,
      "step": 150
    },
    {
      "epoch": 0.014243746105225673,
      "grad_norm": 0.5624942779541016,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.9884,
      "step": 160
    },
    {
      "epoch": 0.01513398023680228,
      "grad_norm": 0.5492900013923645,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.9975,
      "step": 170
    },
    {
      "epoch": 0.016024214368378885,
      "grad_norm": 0.5675668716430664,
      "learning_rate": 2e-05,
      "loss": 0.9819,
      "step": 180
    },
    {
      "epoch": 0.01691444849995549,
      "grad_norm": 0.5347182154655457,
      "learning_rate": 2e-05,
      "loss": 0.9741,
      "step": 190
    },
    {
      "epoch": 0.017804682631532093,
      "grad_norm": 0.5786206722259521,
      "learning_rate": 2e-05,
      "loss": 0.9597,
      "step": 200
    },
    {
      "epoch": 0.018694916763108697,
      "grad_norm": 0.5470443964004517,
      "learning_rate": 2e-05,
      "loss": 0.9601,
      "step": 210
    },
    {
      "epoch": 0.0195851508946853,
      "grad_norm": 0.5782967805862427,
      "learning_rate": 2e-05,
      "loss": 0.9516,
      "step": 220
    },
    {
      "epoch": 0.020475385026261908,
      "grad_norm": 0.530522882938385,
      "learning_rate": 2e-05,
      "loss": 0.9763,
      "step": 230
    },
    {
      "epoch": 0.021365619157838512,
      "grad_norm": 0.5024887323379517,
      "learning_rate": 2e-05,
      "loss": 0.9646,
      "step": 240
    },
    {
      "epoch": 0.022255853289415116,
      "grad_norm": 0.5411601662635803,
      "learning_rate": 2e-05,
      "loss": 0.9619,
      "step": 250
    },
    {
      "epoch": 0.02314608742099172,
      "grad_norm": 0.5052138566970825,
      "learning_rate": 2e-05,
      "loss": 0.9627,
      "step": 260
    },
    {
      "epoch": 0.024036321552568324,
      "grad_norm": 0.5513818860054016,
      "learning_rate": 2e-05,
      "loss": 0.9474,
      "step": 270
    },
    {
      "epoch": 0.02492655568414493,
      "grad_norm": 0.5436130166053772,
      "learning_rate": 2e-05,
      "loss": 0.9411,
      "step": 280
    },
    {
      "epoch": 0.025816789815721535,
      "grad_norm": 0.5385321974754333,
      "learning_rate": 2e-05,
      "loss": 0.9381,
      "step": 290
    },
    {
      "epoch": 0.02670702394729814,
      "grad_norm": 0.5083889961242676,
      "learning_rate": 2e-05,
      "loss": 0.9327,
      "step": 300
    },
    {
      "epoch": 0.027597258078874743,
      "grad_norm": 0.5057015419006348,
      "learning_rate": 2e-05,
      "loss": 0.9284,
      "step": 310
    },
    {
      "epoch": 0.028487492210451347,
      "grad_norm": 0.4919343888759613,
      "learning_rate": 2e-05,
      "loss": 0.9157,
      "step": 320
    },
    {
      "epoch": 0.029377726342027954,
      "grad_norm": 0.527897298336029,
      "learning_rate": 2e-05,
      "loss": 0.9314,
      "step": 330
    },
    {
      "epoch": 0.03026796047360456,
      "grad_norm": 0.53957200050354,
      "learning_rate": 2e-05,
      "loss": 0.9301,
      "step": 340
    },
    {
      "epoch": 0.031158194605181162,
      "grad_norm": 0.5134000182151794,
      "learning_rate": 2e-05,
      "loss": 0.9426,
      "step": 350
    },
    {
      "epoch": 0.03204842873675777,
      "grad_norm": 0.5129437446594238,
      "learning_rate": 2e-05,
      "loss": 0.9316,
      "step": 360
    },
    {
      "epoch": 0.03293866286833437,
      "grad_norm": 0.515938401222229,
      "learning_rate": 2e-05,
      "loss": 0.9266,
      "step": 370
    },
    {
      "epoch": 0.03382889699991098,
      "grad_norm": 0.5681295394897461,
      "learning_rate": 2e-05,
      "loss": 0.9527,
      "step": 380
    },
    {
      "epoch": 0.03471913113148758,
      "grad_norm": 0.5140536427497864,
      "learning_rate": 2e-05,
      "loss": 0.921,
      "step": 390
    },
    {
      "epoch": 0.035609365263064185,
      "grad_norm": 0.5116504430770874,
      "learning_rate": 2e-05,
      "loss": 0.9392,
      "step": 400
    },
    {
      "epoch": 0.03649959939464079,
      "grad_norm": 0.47694700956344604,
      "learning_rate": 2e-05,
      "loss": 0.9324,
      "step": 410
    },
    {
      "epoch": 0.03738983352621739,
      "grad_norm": 0.4879381060600281,
      "learning_rate": 2e-05,
      "loss": 0.9268,
      "step": 420
    },
    {
      "epoch": 0.038280067657794,
      "grad_norm": 0.5697586536407471,
      "learning_rate": 2e-05,
      "loss": 0.9202,
      "step": 430
    },
    {
      "epoch": 0.0391703017893706,
      "grad_norm": 0.5065985918045044,
      "learning_rate": 2e-05,
      "loss": 0.9341,
      "step": 440
    },
    {
      "epoch": 0.04006053592094721,
      "grad_norm": 0.48673391342163086,
      "learning_rate": 2e-05,
      "loss": 0.9084,
      "step": 450
    },
    {
      "epoch": 0.040950770052523816,
      "grad_norm": 0.4933507442474365,
      "learning_rate": 2e-05,
      "loss": 0.9295,
      "step": 460
    },
    {
      "epoch": 0.04184100418410042,
      "grad_norm": 0.4880451560020447,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 470
    },
    {
      "epoch": 0.042731238315677024,
      "grad_norm": 0.512821614742279,
      "learning_rate": 2e-05,
      "loss": 0.9055,
      "step": 480
    },
    {
      "epoch": 0.043621472447253624,
      "grad_norm": 0.4931286871433258,
      "learning_rate": 2e-05,
      "loss": 0.9123,
      "step": 490
    },
    {
      "epoch": 0.04451170657883023,
      "grad_norm": 0.48850977420806885,
      "learning_rate": 2e-05,
      "loss": 0.8839,
      "step": 500
    },
    {
      "epoch": 0.04540194071040684,
      "grad_norm": 0.46563583612442017,
      "learning_rate": 2e-05,
      "loss": 0.9047,
      "step": 510
    },
    {
      "epoch": 0.04629217484198344,
      "grad_norm": 0.47478628158569336,
      "learning_rate": 2e-05,
      "loss": 0.9261,
      "step": 520
    },
    {
      "epoch": 0.04718240897356005,
      "grad_norm": 0.47876861691474915,
      "learning_rate": 2e-05,
      "loss": 0.9205,
      "step": 530
    },
    {
      "epoch": 0.04807264310513665,
      "grad_norm": 0.5156196355819702,
      "learning_rate": 2e-05,
      "loss": 0.9178,
      "step": 540
    },
    {
      "epoch": 0.048962877236713255,
      "grad_norm": 0.5431042313575745,
      "learning_rate": 2e-05,
      "loss": 0.9339,
      "step": 550
    },
    {
      "epoch": 0.04985311136828986,
      "grad_norm": 0.48477429151535034,
      "learning_rate": 2e-05,
      "loss": 0.9344,
      "step": 560
    },
    {
      "epoch": 0.05074334549986646,
      "grad_norm": 0.5291976928710938,
      "learning_rate": 2e-05,
      "loss": 0.905,
      "step": 570
    },
    {
      "epoch": 0.05163357963144307,
      "grad_norm": 0.4579319953918457,
      "learning_rate": 2e-05,
      "loss": 0.9021,
      "step": 580
    },
    {
      "epoch": 0.05252381376301967,
      "grad_norm": 0.7584505677223206,
      "learning_rate": 2e-05,
      "loss": 0.9038,
      "step": 590
    },
    {
      "epoch": 0.05341404789459628,
      "grad_norm": 0.4732418954372406,
      "learning_rate": 2e-05,
      "loss": 0.8982,
      "step": 600
    },
    {
      "epoch": 0.054304282026172886,
      "grad_norm": 0.4613445997238159,
      "learning_rate": 2e-05,
      "loss": 0.8935,
      "step": 610
    },
    {
      "epoch": 0.055194516157749486,
      "grad_norm": 0.4943762421607971,
      "learning_rate": 2e-05,
      "loss": 0.9184,
      "step": 620
    },
    {
      "epoch": 0.056084750289326094,
      "grad_norm": 0.4789336025714874,
      "learning_rate": 2e-05,
      "loss": 0.9178,
      "step": 630
    },
    {
      "epoch": 0.056974984420902694,
      "grad_norm": 0.46869930624961853,
      "learning_rate": 2e-05,
      "loss": 0.9144,
      "step": 640
    },
    {
      "epoch": 0.0578652185524793,
      "grad_norm": 0.5243553519248962,
      "learning_rate": 2e-05,
      "loss": 0.8863,
      "step": 650
    },
    {
      "epoch": 0.05875545268405591,
      "grad_norm": 0.49072882533073425,
      "learning_rate": 2e-05,
      "loss": 0.9299,
      "step": 660
    },
    {
      "epoch": 0.05964568681563251,
      "grad_norm": 0.47901391983032227,
      "learning_rate": 2e-05,
      "loss": 0.9348,
      "step": 670
    },
    {
      "epoch": 0.06053592094720912,
      "grad_norm": 0.49131375551223755,
      "learning_rate": 2e-05,
      "loss": 0.9102,
      "step": 680
    },
    {
      "epoch": 0.06142615507878572,
      "grad_norm": 0.46955153346061707,
      "learning_rate": 2e-05,
      "loss": 0.9182,
      "step": 690
    },
    {
      "epoch": 0.062316389210362325,
      "grad_norm": 0.489824116230011,
      "learning_rate": 2e-05,
      "loss": 0.8952,
      "step": 700
    },
    {
      "epoch": 0.06320662334193893,
      "grad_norm": 0.4554087221622467,
      "learning_rate": 2e-05,
      "loss": 0.9188,
      "step": 710
    },
    {
      "epoch": 0.06409685747351554,
      "grad_norm": 0.45714613795280457,
      "learning_rate": 2e-05,
      "loss": 0.9245,
      "step": 720
    },
    {
      "epoch": 0.06498709160509214,
      "grad_norm": 0.48824822902679443,
      "learning_rate": 2e-05,
      "loss": 0.9084,
      "step": 730
    },
    {
      "epoch": 0.06587732573666874,
      "grad_norm": 0.4541068971157074,
      "learning_rate": 2e-05,
      "loss": 0.8886,
      "step": 740
    },
    {
      "epoch": 0.06676755986824535,
      "grad_norm": 0.4842844009399414,
      "learning_rate": 2e-05,
      "loss": 0.8879,
      "step": 750
    },
    {
      "epoch": 0.06765779399982196,
      "grad_norm": 0.4912959039211273,
      "learning_rate": 2e-05,
      "loss": 0.9136,
      "step": 760
    },
    {
      "epoch": 0.06854802813139856,
      "grad_norm": 0.4518549144268036,
      "learning_rate": 2e-05,
      "loss": 0.9045,
      "step": 770
    },
    {
      "epoch": 0.06943826226297516,
      "grad_norm": 0.4674036502838135,
      "learning_rate": 2e-05,
      "loss": 0.897,
      "step": 780
    },
    {
      "epoch": 0.07032849639455177,
      "grad_norm": 0.4854924976825714,
      "learning_rate": 2e-05,
      "loss": 0.8903,
      "step": 790
    },
    {
      "epoch": 0.07121873052612837,
      "grad_norm": 0.450965940952301,
      "learning_rate": 2e-05,
      "loss": 0.891,
      "step": 800
    },
    {
      "epoch": 0.07210896465770497,
      "grad_norm": 0.4560759663581848,
      "learning_rate": 2e-05,
      "loss": 0.8832,
      "step": 810
    },
    {
      "epoch": 0.07299919878928159,
      "grad_norm": 0.4890328049659729,
      "learning_rate": 2e-05,
      "loss": 0.8758,
      "step": 820
    },
    {
      "epoch": 0.07388943292085819,
      "grad_norm": 0.4705803692340851,
      "learning_rate": 2e-05,
      "loss": 0.8714,
      "step": 830
    },
    {
      "epoch": 0.07477966705243479,
      "grad_norm": 0.4944480061531067,
      "learning_rate": 2e-05,
      "loss": 0.882,
      "step": 840
    },
    {
      "epoch": 0.0756699011840114,
      "grad_norm": 0.4771805703639984,
      "learning_rate": 2e-05,
      "loss": 0.8985,
      "step": 850
    },
    {
      "epoch": 0.076560135315588,
      "grad_norm": 0.4645552933216095,
      "learning_rate": 2e-05,
      "loss": 0.9023,
      "step": 860
    },
    {
      "epoch": 0.0774503694471646,
      "grad_norm": 0.47317224740982056,
      "learning_rate": 2e-05,
      "loss": 0.8878,
      "step": 870
    },
    {
      "epoch": 0.0783406035787412,
      "grad_norm": 0.4755535125732422,
      "learning_rate": 2e-05,
      "loss": 0.8746,
      "step": 880
    },
    {
      "epoch": 0.07923083771031782,
      "grad_norm": 0.44543999433517456,
      "learning_rate": 2e-05,
      "loss": 0.891,
      "step": 890
    },
    {
      "epoch": 0.08012107184189442,
      "grad_norm": 0.4649183452129364,
      "learning_rate": 2e-05,
      "loss": 0.9276,
      "step": 900
    },
    {
      "epoch": 0.08101130597347102,
      "grad_norm": 0.4410511553287506,
      "learning_rate": 2e-05,
      "loss": 0.8973,
      "step": 910
    },
    {
      "epoch": 0.08190154010504763,
      "grad_norm": 0.45230555534362793,
      "learning_rate": 2e-05,
      "loss": 0.8735,
      "step": 920
    },
    {
      "epoch": 0.08279177423662423,
      "grad_norm": 0.48251569271087646,
      "learning_rate": 2e-05,
      "loss": 0.8711,
      "step": 930
    },
    {
      "epoch": 0.08368200836820083,
      "grad_norm": 0.4867289066314697,
      "learning_rate": 2e-05,
      "loss": 0.9189,
      "step": 940
    },
    {
      "epoch": 0.08457224249977745,
      "grad_norm": 0.46645817160606384,
      "learning_rate": 2e-05,
      "loss": 0.8748,
      "step": 950
    },
    {
      "epoch": 0.08546247663135405,
      "grad_norm": 0.48964402079582214,
      "learning_rate": 2e-05,
      "loss": 0.8659,
      "step": 960
    },
    {
      "epoch": 0.08635271076293065,
      "grad_norm": 0.5039545893669128,
      "learning_rate": 2e-05,
      "loss": 0.8709,
      "step": 970
    },
    {
      "epoch": 0.08724294489450725,
      "grad_norm": 0.4509641230106354,
      "learning_rate": 2e-05,
      "loss": 0.8713,
      "step": 980
    },
    {
      "epoch": 0.08813317902608386,
      "grad_norm": 0.4961320459842682,
      "learning_rate": 2e-05,
      "loss": 0.8893,
      "step": 990
    },
    {
      "epoch": 0.08902341315766046,
      "grad_norm": 0.45977702736854553,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 1000
    },
    {
      "epoch": 0.08991364728923706,
      "grad_norm": 0.4814492166042328,
      "learning_rate": 2e-05,
      "loss": 0.8972,
      "step": 1010
    },
    {
      "epoch": 0.09080388142081368,
      "grad_norm": 0.4550889730453491,
      "learning_rate": 2e-05,
      "loss": 0.8592,
      "step": 1020
    },
    {
      "epoch": 0.09169411555239028,
      "grad_norm": 0.44844773411750793,
      "learning_rate": 2e-05,
      "loss": 0.8857,
      "step": 1030
    },
    {
      "epoch": 0.09258434968396688,
      "grad_norm": 0.5210088491439819,
      "learning_rate": 2e-05,
      "loss": 0.8765,
      "step": 1040
    },
    {
      "epoch": 0.0934745838155435,
      "grad_norm": 0.49490031599998474,
      "learning_rate": 2e-05,
      "loss": 0.8392,
      "step": 1050
    },
    {
      "epoch": 0.0943648179471201,
      "grad_norm": 0.4991597533226013,
      "learning_rate": 2e-05,
      "loss": 0.9008,
      "step": 1060
    },
    {
      "epoch": 0.0952550520786967,
      "grad_norm": 0.4496327340602875,
      "learning_rate": 2e-05,
      "loss": 0.9008,
      "step": 1070
    },
    {
      "epoch": 0.0961452862102733,
      "grad_norm": 0.4505268335342407,
      "learning_rate": 2e-05,
      "loss": 0.8817,
      "step": 1080
    },
    {
      "epoch": 0.09703552034184991,
      "grad_norm": 0.43803438544273376,
      "learning_rate": 2e-05,
      "loss": 0.8719,
      "step": 1090
    },
    {
      "epoch": 0.09792575447342651,
      "grad_norm": 0.48525097966194153,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 1100
    },
    {
      "epoch": 0.09881598860500311,
      "grad_norm": 0.45317840576171875,
      "learning_rate": 2e-05,
      "loss": 0.8877,
      "step": 1110
    },
    {
      "epoch": 0.09970622273657972,
      "grad_norm": 0.43088456988334656,
      "learning_rate": 2e-05,
      "loss": 0.9022,
      "step": 1120
    },
    {
      "epoch": 0.10059645686815633,
      "grad_norm": 0.45597684383392334,
      "learning_rate": 2e-05,
      "loss": 0.8725,
      "step": 1130
    },
    {
      "epoch": 0.10148669099973293,
      "grad_norm": 0.4579595923423767,
      "learning_rate": 2e-05,
      "loss": 0.8722,
      "step": 1140
    },
    {
      "epoch": 0.10237692513130954,
      "grad_norm": 0.47375500202178955,
      "learning_rate": 2e-05,
      "loss": 0.8606,
      "step": 1150
    },
    {
      "epoch": 0.10326715926288614,
      "grad_norm": 0.4688866436481476,
      "learning_rate": 2e-05,
      "loss": 0.8636,
      "step": 1160
    },
    {
      "epoch": 0.10415739339446274,
      "grad_norm": 0.4577106833457947,
      "learning_rate": 2e-05,
      "loss": 0.8822,
      "step": 1170
    },
    {
      "epoch": 0.10504762752603934,
      "grad_norm": 0.4428924024105072,
      "learning_rate": 2e-05,
      "loss": 0.8618,
      "step": 1180
    },
    {
      "epoch": 0.10593786165761596,
      "grad_norm": 0.42710617184638977,
      "learning_rate": 2e-05,
      "loss": 0.8567,
      "step": 1190
    },
    {
      "epoch": 0.10682809578919256,
      "grad_norm": 0.4547199606895447,
      "learning_rate": 2e-05,
      "loss": 0.8782,
      "step": 1200
    },
    {
      "epoch": 0.10771832992076916,
      "grad_norm": 0.47715529799461365,
      "learning_rate": 2e-05,
      "loss": 0.871,
      "step": 1210
    },
    {
      "epoch": 0.10860856405234577,
      "grad_norm": 0.4618558883666992,
      "learning_rate": 2e-05,
      "loss": 0.8708,
      "step": 1220
    },
    {
      "epoch": 0.10949879818392237,
      "grad_norm": 0.44639959931373596,
      "learning_rate": 2e-05,
      "loss": 0.8563,
      "step": 1230
    },
    {
      "epoch": 0.11038903231549897,
      "grad_norm": 0.5406988263130188,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 1240
    },
    {
      "epoch": 0.11127926644707559,
      "grad_norm": 0.4740561246871948,
      "learning_rate": 2e-05,
      "loss": 0.8941,
      "step": 1250
    },
    {
      "epoch": 0.11216950057865219,
      "grad_norm": 0.46081990003585815,
      "learning_rate": 2e-05,
      "loss": 0.8764,
      "step": 1260
    },
    {
      "epoch": 0.11305973471022879,
      "grad_norm": 0.46846848726272583,
      "learning_rate": 2e-05,
      "loss": 0.8657,
      "step": 1270
    },
    {
      "epoch": 0.11394996884180539,
      "grad_norm": 0.45560920238494873,
      "learning_rate": 2e-05,
      "loss": 0.8858,
      "step": 1280
    },
    {
      "epoch": 0.114840202973382,
      "grad_norm": 0.44569844007492065,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 1290
    },
    {
      "epoch": 0.1157304371049586,
      "grad_norm": 0.47736498713493347,
      "learning_rate": 2e-05,
      "loss": 0.8654,
      "step": 1300
    },
    {
      "epoch": 0.1166206712365352,
      "grad_norm": 0.42953744530677795,
      "learning_rate": 2e-05,
      "loss": 0.8612,
      "step": 1310
    },
    {
      "epoch": 0.11751090536811182,
      "grad_norm": 0.472917765378952,
      "learning_rate": 2e-05,
      "loss": 0.8824,
      "step": 1320
    },
    {
      "epoch": 0.11840113949968842,
      "grad_norm": 0.4567590653896332,
      "learning_rate": 2e-05,
      "loss": 0.8887,
      "step": 1330
    },
    {
      "epoch": 0.11929137363126502,
      "grad_norm": 0.44442838430404663,
      "learning_rate": 2e-05,
      "loss": 0.8823,
      "step": 1340
    },
    {
      "epoch": 0.12018160776284163,
      "grad_norm": 0.44202491641044617,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 1350
    },
    {
      "epoch": 0.12107184189441823,
      "grad_norm": 0.5220369100570679,
      "learning_rate": 2e-05,
      "loss": 0.8664,
      "step": 1360
    },
    {
      "epoch": 0.12196207602599483,
      "grad_norm": 0.4746244549751282,
      "learning_rate": 2e-05,
      "loss": 0.8763,
      "step": 1370
    },
    {
      "epoch": 0.12285231015757143,
      "grad_norm": 0.4417015612125397,
      "learning_rate": 2e-05,
      "loss": 0.8805,
      "step": 1380
    },
    {
      "epoch": 0.12374254428914805,
      "grad_norm": 0.5035746693611145,
      "learning_rate": 2e-05,
      "loss": 0.8845,
      "step": 1390
    },
    {
      "epoch": 0.12463277842072465,
      "grad_norm": 0.4350651502609253,
      "learning_rate": 2e-05,
      "loss": 0.8706,
      "step": 1400
    },
    {
      "epoch": 0.12552301255230125,
      "grad_norm": 0.44582507014274597,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 1410
    },
    {
      "epoch": 0.12641324668387785,
      "grad_norm": 0.45047810673713684,
      "learning_rate": 2e-05,
      "loss": 0.8803,
      "step": 1420
    },
    {
      "epoch": 0.12730348081545448,
      "grad_norm": 0.4596972167491913,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 1430
    },
    {
      "epoch": 0.12819371494703108,
      "grad_norm": 0.4250618517398834,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 1440
    },
    {
      "epoch": 0.12908394907860768,
      "grad_norm": 0.43938419222831726,
      "learning_rate": 2e-05,
      "loss": 0.863,
      "step": 1450
    },
    {
      "epoch": 0.12997418321018428,
      "grad_norm": 0.45980769395828247,
      "learning_rate": 2e-05,
      "loss": 0.8574,
      "step": 1460
    },
    {
      "epoch": 0.13086441734176088,
      "grad_norm": 0.4644986093044281,
      "learning_rate": 2e-05,
      "loss": 0.8749,
      "step": 1470
    },
    {
      "epoch": 0.13175465147333748,
      "grad_norm": 0.4807727038860321,
      "learning_rate": 2e-05,
      "loss": 0.8922,
      "step": 1480
    },
    {
      "epoch": 0.13264488560491408,
      "grad_norm": 0.421638548374176,
      "learning_rate": 2e-05,
      "loss": 0.8789,
      "step": 1490
    },
    {
      "epoch": 0.1335351197364907,
      "grad_norm": 0.44382473826408386,
      "learning_rate": 2e-05,
      "loss": 0.8874,
      "step": 1500
    },
    {
      "epoch": 0.1344253538680673,
      "grad_norm": 0.4820038974285126,
      "learning_rate": 2e-05,
      "loss": 0.8992,
      "step": 1510
    },
    {
      "epoch": 0.1353155879996439,
      "grad_norm": 0.505782961845398,
      "learning_rate": 2e-05,
      "loss": 0.8664,
      "step": 1520
    },
    {
      "epoch": 0.1362058221312205,
      "grad_norm": 0.44912824034690857,
      "learning_rate": 2e-05,
      "loss": 0.8596,
      "step": 1530
    },
    {
      "epoch": 0.1370960562627971,
      "grad_norm": 0.43143683671951294,
      "learning_rate": 2e-05,
      "loss": 0.8672,
      "step": 1540
    },
    {
      "epoch": 0.1379862903943737,
      "grad_norm": 0.44386616349220276,
      "learning_rate": 2e-05,
      "loss": 0.8771,
      "step": 1550
    },
    {
      "epoch": 0.1388765245259503,
      "grad_norm": 0.43272528052330017,
      "learning_rate": 2e-05,
      "loss": 0.8374,
      "step": 1560
    },
    {
      "epoch": 0.13976675865752694,
      "grad_norm": 0.47014743089675903,
      "learning_rate": 2e-05,
      "loss": 0.8627,
      "step": 1570
    },
    {
      "epoch": 0.14065699278910354,
      "grad_norm": 0.4546826183795929,
      "learning_rate": 2e-05,
      "loss": 0.8832,
      "step": 1580
    },
    {
      "epoch": 0.14154722692068014,
      "grad_norm": 0.43960079550743103,
      "learning_rate": 2e-05,
      "loss": 0.8884,
      "step": 1590
    },
    {
      "epoch": 0.14243746105225674,
      "grad_norm": 0.4659097194671631,
      "learning_rate": 2e-05,
      "loss": 0.8647,
      "step": 1600
    },
    {
      "epoch": 0.14332769518383334,
      "grad_norm": 0.456807941198349,
      "learning_rate": 2e-05,
      "loss": 0.8782,
      "step": 1610
    },
    {
      "epoch": 0.14421792931540994,
      "grad_norm": 0.4250633716583252,
      "learning_rate": 2e-05,
      "loss": 0.8507,
      "step": 1620
    },
    {
      "epoch": 0.14510816344698657,
      "grad_norm": 0.45779556035995483,
      "learning_rate": 2e-05,
      "loss": 0.8655,
      "step": 1630
    },
    {
      "epoch": 0.14599839757856317,
      "grad_norm": 0.43346431851387024,
      "learning_rate": 2e-05,
      "loss": 0.8509,
      "step": 1640
    },
    {
      "epoch": 0.14688863171013977,
      "grad_norm": 0.47265052795410156,
      "learning_rate": 2e-05,
      "loss": 0.8534,
      "step": 1650
    },
    {
      "epoch": 0.14777886584171637,
      "grad_norm": 0.464148610830307,
      "learning_rate": 2e-05,
      "loss": 0.8667,
      "step": 1660
    },
    {
      "epoch": 0.14866909997329297,
      "grad_norm": 0.45228493213653564,
      "learning_rate": 2e-05,
      "loss": 0.8638,
      "step": 1670
    },
    {
      "epoch": 0.14955933410486957,
      "grad_norm": 0.4178244173526764,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 1680
    },
    {
      "epoch": 0.15044956823644617,
      "grad_norm": 0.46974611282348633,
      "learning_rate": 2e-05,
      "loss": 0.8575,
      "step": 1690
    },
    {
      "epoch": 0.1513398023680228,
      "grad_norm": 0.4494037926197052,
      "learning_rate": 2e-05,
      "loss": 0.8655,
      "step": 1700
    },
    {
      "epoch": 0.1522300364995994,
      "grad_norm": 0.4633338153362274,
      "learning_rate": 2e-05,
      "loss": 0.8658,
      "step": 1710
    },
    {
      "epoch": 0.153120270631176,
      "grad_norm": 0.44515934586524963,
      "learning_rate": 2e-05,
      "loss": 0.8692,
      "step": 1720
    },
    {
      "epoch": 0.1540105047627526,
      "grad_norm": 0.4518592655658722,
      "learning_rate": 2e-05,
      "loss": 0.8566,
      "step": 1730
    },
    {
      "epoch": 0.1549007388943292,
      "grad_norm": 0.43090882897377014,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 1740
    },
    {
      "epoch": 0.1557909730259058,
      "grad_norm": 0.4627023935317993,
      "learning_rate": 2e-05,
      "loss": 0.8682,
      "step": 1750
    },
    {
      "epoch": 0.1566812071574824,
      "grad_norm": 0.428757905960083,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 1760
    },
    {
      "epoch": 0.15757144128905903,
      "grad_norm": 0.5828053951263428,
      "learning_rate": 2e-05,
      "loss": 0.8646,
      "step": 1770
    },
    {
      "epoch": 0.15846167542063563,
      "grad_norm": 0.44868114590644836,
      "learning_rate": 2e-05,
      "loss": 0.8737,
      "step": 1780
    },
    {
      "epoch": 0.15935190955221223,
      "grad_norm": 0.44050073623657227,
      "learning_rate": 2e-05,
      "loss": 0.8397,
      "step": 1790
    },
    {
      "epoch": 0.16024214368378883,
      "grad_norm": 0.4597872495651245,
      "learning_rate": 2e-05,
      "loss": 0.8264,
      "step": 1800
    },
    {
      "epoch": 0.16113237781536544,
      "grad_norm": 0.4526384472846985,
      "learning_rate": 2e-05,
      "loss": 0.8889,
      "step": 1810
    },
    {
      "epoch": 0.16202261194694204,
      "grad_norm": 0.4531833529472351,
      "learning_rate": 2e-05,
      "loss": 0.854,
      "step": 1820
    },
    {
      "epoch": 0.16291284607851866,
      "grad_norm": 0.4569271504878998,
      "learning_rate": 2e-05,
      "loss": 0.8702,
      "step": 1830
    },
    {
      "epoch": 0.16380308021009526,
      "grad_norm": 0.4483858346939087,
      "learning_rate": 2e-05,
      "loss": 0.8685,
      "step": 1840
    },
    {
      "epoch": 0.16469331434167186,
      "grad_norm": 0.4322422742843628,
      "learning_rate": 2e-05,
      "loss": 0.8549,
      "step": 1850
    },
    {
      "epoch": 0.16558354847324847,
      "grad_norm": 0.4667356610298157,
      "learning_rate": 2e-05,
      "loss": 0.8819,
      "step": 1860
    },
    {
      "epoch": 0.16647378260482507,
      "grad_norm": 0.44596439599990845,
      "learning_rate": 2e-05,
      "loss": 0.8665,
      "step": 1870
    },
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 0.44556376338005066,
      "learning_rate": 2e-05,
      "loss": 0.8674,
      "step": 1880
    },
    {
      "epoch": 0.16825425086797827,
      "grad_norm": 0.446918785572052,
      "learning_rate": 2e-05,
      "loss": 0.8712,
      "step": 1890
    },
    {
      "epoch": 0.1691444849995549,
      "grad_norm": 0.48224741220474243,
      "learning_rate": 2e-05,
      "loss": 0.8912,
      "step": 1900
    },
    {
      "epoch": 0.1700347191311315,
      "grad_norm": 0.42556098103523254,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 1910
    },
    {
      "epoch": 0.1709249532627081,
      "grad_norm": 0.4142284393310547,
      "learning_rate": 2e-05,
      "loss": 0.8283,
      "step": 1920
    },
    {
      "epoch": 0.1718151873942847,
      "grad_norm": 0.625287652015686,
      "learning_rate": 2e-05,
      "loss": 0.8809,
      "step": 1930
    },
    {
      "epoch": 0.1727054215258613,
      "grad_norm": 0.4345504641532898,
      "learning_rate": 2e-05,
      "loss": 0.8986,
      "step": 1940
    },
    {
      "epoch": 0.1735956556574379,
      "grad_norm": 0.43666818737983704,
      "learning_rate": 2e-05,
      "loss": 0.8623,
      "step": 1950
    },
    {
      "epoch": 0.1744858897890145,
      "grad_norm": 0.43533623218536377,
      "learning_rate": 2e-05,
      "loss": 0.8186,
      "step": 1960
    },
    {
      "epoch": 0.17537612392059113,
      "grad_norm": 0.4227614104747772,
      "learning_rate": 2e-05,
      "loss": 0.8761,
      "step": 1970
    },
    {
      "epoch": 0.17626635805216773,
      "grad_norm": 0.45506614446640015,
      "learning_rate": 2e-05,
      "loss": 0.8565,
      "step": 1980
    },
    {
      "epoch": 0.17715659218374433,
      "grad_norm": 0.44619250297546387,
      "learning_rate": 2e-05,
      "loss": 0.9029,
      "step": 1990
    },
    {
      "epoch": 0.17804682631532093,
      "grad_norm": 0.4685472846031189,
      "learning_rate": 2e-05,
      "loss": 0.8491,
      "step": 2000
    },
    {
      "epoch": 0.17893706044689753,
      "grad_norm": 0.43136295676231384,
      "learning_rate": 2e-05,
      "loss": 0.8497,
      "step": 2010
    },
    {
      "epoch": 0.17982729457847413,
      "grad_norm": 0.431625097990036,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 2020
    },
    {
      "epoch": 0.18071752871005076,
      "grad_norm": 0.4419959783554077,
      "learning_rate": 2e-05,
      "loss": 0.8614,
      "step": 2030
    },
    {
      "epoch": 0.18160776284162736,
      "grad_norm": 0.4239712059497833,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 2040
    },
    {
      "epoch": 0.18249799697320396,
      "grad_norm": 0.43728873133659363,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 2050
    },
    {
      "epoch": 0.18338823110478056,
      "grad_norm": 0.41318660974502563,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 2060
    },
    {
      "epoch": 0.18427846523635716,
      "grad_norm": 0.5283257961273193,
      "learning_rate": 2e-05,
      "loss": 0.8773,
      "step": 2070
    },
    {
      "epoch": 0.18516869936793376,
      "grad_norm": 0.45080527663230896,
      "learning_rate": 2e-05,
      "loss": 0.8651,
      "step": 2080
    },
    {
      "epoch": 0.18605893349951036,
      "grad_norm": 0.4390518367290497,
      "learning_rate": 2e-05,
      "loss": 0.8202,
      "step": 2090
    },
    {
      "epoch": 0.186949167631087,
      "grad_norm": 0.4196532070636749,
      "learning_rate": 2e-05,
      "loss": 0.8632,
      "step": 2100
    },
    {
      "epoch": 0.1878394017626636,
      "grad_norm": 0.41699790954589844,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 2110
    },
    {
      "epoch": 0.1887296358942402,
      "grad_norm": 0.4441491663455963,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 2120
    },
    {
      "epoch": 0.1896198700258168,
      "grad_norm": 0.4168669581413269,
      "learning_rate": 2e-05,
      "loss": 0.8455,
      "step": 2130
    },
    {
      "epoch": 0.1905101041573934,
      "grad_norm": 0.41449984908103943,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 2140
    },
    {
      "epoch": 0.19140033828897,
      "grad_norm": 0.4350728392601013,
      "learning_rate": 2e-05,
      "loss": 0.8658,
      "step": 2150
    },
    {
      "epoch": 0.1922905724205466,
      "grad_norm": 0.43918314576148987,
      "learning_rate": 2e-05,
      "loss": 0.8631,
      "step": 2160
    },
    {
      "epoch": 0.19318080655212322,
      "grad_norm": 0.5561551451683044,
      "learning_rate": 2e-05,
      "loss": 0.8762,
      "step": 2170
    },
    {
      "epoch": 0.19407104068369982,
      "grad_norm": 0.5069448351860046,
      "learning_rate": 2e-05,
      "loss": 0.8711,
      "step": 2180
    },
    {
      "epoch": 0.19496127481527642,
      "grad_norm": 0.41097161173820496,
      "learning_rate": 2e-05,
      "loss": 0.8717,
      "step": 2190
    },
    {
      "epoch": 0.19585150894685302,
      "grad_norm": 0.43706122040748596,
      "learning_rate": 2e-05,
      "loss": 0.8889,
      "step": 2200
    },
    {
      "epoch": 0.19674174307842962,
      "grad_norm": 0.41716432571411133,
      "learning_rate": 2e-05,
      "loss": 0.8548,
      "step": 2210
    },
    {
      "epoch": 0.19763197721000622,
      "grad_norm": 0.4192309081554413,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 2220
    },
    {
      "epoch": 0.19852221134158285,
      "grad_norm": 0.43411344289779663,
      "learning_rate": 2e-05,
      "loss": 0.8507,
      "step": 2230
    },
    {
      "epoch": 0.19941244547315945,
      "grad_norm": 0.42707616090774536,
      "learning_rate": 2e-05,
      "loss": 0.8532,
      "step": 2240
    },
    {
      "epoch": 0.20030267960473605,
      "grad_norm": 0.4212580621242523,
      "learning_rate": 2e-05,
      "loss": 0.8526,
      "step": 2250
    },
    {
      "epoch": 0.20119291373631265,
      "grad_norm": 0.396260142326355,
      "learning_rate": 2e-05,
      "loss": 0.8735,
      "step": 2260
    },
    {
      "epoch": 0.20208314786788925,
      "grad_norm": 0.445537269115448,
      "learning_rate": 2e-05,
      "loss": 0.8616,
      "step": 2270
    },
    {
      "epoch": 0.20297338199946585,
      "grad_norm": 0.4279395043849945,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 2280
    },
    {
      "epoch": 0.20386361613104245,
      "grad_norm": 0.46049898862838745,
      "learning_rate": 2e-05,
      "loss": 0.8404,
      "step": 2290
    },
    {
      "epoch": 0.20475385026261908,
      "grad_norm": 0.4414907395839691,
      "learning_rate": 2e-05,
      "loss": 0.8566,
      "step": 2300
    },
    {
      "epoch": 0.20564408439419568,
      "grad_norm": 0.415762722492218,
      "learning_rate": 2e-05,
      "loss": 0.8233,
      "step": 2310
    },
    {
      "epoch": 0.20653431852577228,
      "grad_norm": 0.4279174506664276,
      "learning_rate": 2e-05,
      "loss": 0.895,
      "step": 2320
    },
    {
      "epoch": 0.20742455265734888,
      "grad_norm": 0.44296035170555115,
      "learning_rate": 2e-05,
      "loss": 0.8615,
      "step": 2330
    },
    {
      "epoch": 0.20831478678892548,
      "grad_norm": 0.411670446395874,
      "learning_rate": 2e-05,
      "loss": 0.814,
      "step": 2340
    },
    {
      "epoch": 0.20920502092050208,
      "grad_norm": 0.42919275164604187,
      "learning_rate": 2e-05,
      "loss": 0.869,
      "step": 2350
    },
    {
      "epoch": 0.21009525505207868,
      "grad_norm": 0.42778828740119934,
      "learning_rate": 2e-05,
      "loss": 0.837,
      "step": 2360
    },
    {
      "epoch": 0.2109854891836553,
      "grad_norm": 0.4137372672557831,
      "learning_rate": 2e-05,
      "loss": 0.8727,
      "step": 2370
    },
    {
      "epoch": 0.2118757233152319,
      "grad_norm": 0.4157104194164276,
      "learning_rate": 2e-05,
      "loss": 0.8629,
      "step": 2380
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 0.42591139674186707,
      "learning_rate": 2e-05,
      "loss": 0.815,
      "step": 2390
    },
    {
      "epoch": 0.2136561915783851,
      "grad_norm": 0.46538254618644714,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 2400
    },
    {
      "epoch": 0.2145464257099617,
      "grad_norm": 0.4228883683681488,
      "learning_rate": 2e-05,
      "loss": 0.8553,
      "step": 2410
    },
    {
      "epoch": 0.2154366598415383,
      "grad_norm": 0.42215660214424133,
      "learning_rate": 2e-05,
      "loss": 0.841,
      "step": 2420
    },
    {
      "epoch": 0.21632689397311494,
      "grad_norm": 0.44334888458251953,
      "learning_rate": 2e-05,
      "loss": 0.8605,
      "step": 2430
    },
    {
      "epoch": 0.21721712810469154,
      "grad_norm": 0.4166254699230194,
      "learning_rate": 2e-05,
      "loss": 0.8485,
      "step": 2440
    },
    {
      "epoch": 0.21810736223626814,
      "grad_norm": 0.4625045359134674,
      "learning_rate": 2e-05,
      "loss": 0.8516,
      "step": 2450
    },
    {
      "epoch": 0.21899759636784474,
      "grad_norm": 0.44102203845977783,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 2460
    },
    {
      "epoch": 0.21988783049942134,
      "grad_norm": 0.5162444710731506,
      "learning_rate": 2e-05,
      "loss": 0.8652,
      "step": 2470
    },
    {
      "epoch": 0.22077806463099794,
      "grad_norm": 0.42829203605651855,
      "learning_rate": 2e-05,
      "loss": 0.8843,
      "step": 2480
    },
    {
      "epoch": 0.22166829876257454,
      "grad_norm": 0.48756518959999084,
      "learning_rate": 2e-05,
      "loss": 0.8504,
      "step": 2490
    },
    {
      "epoch": 0.22255853289415117,
      "grad_norm": 0.4166606366634369,
      "learning_rate": 2e-05,
      "loss": 0.8347,
      "step": 2500
    },
    {
      "epoch": 0.22344876702572777,
      "grad_norm": 0.4540061950683594,
      "learning_rate": 2e-05,
      "loss": 0.8424,
      "step": 2510
    },
    {
      "epoch": 0.22433900115730437,
      "grad_norm": 0.44452282786369324,
      "learning_rate": 2e-05,
      "loss": 0.856,
      "step": 2520
    },
    {
      "epoch": 0.22522923528888097,
      "grad_norm": 0.4142967164516449,
      "learning_rate": 2e-05,
      "loss": 0.8352,
      "step": 2530
    },
    {
      "epoch": 0.22611946942045758,
      "grad_norm": 0.4105556607246399,
      "learning_rate": 2e-05,
      "loss": 0.8166,
      "step": 2540
    },
    {
      "epoch": 0.22700970355203418,
      "grad_norm": 0.47593531012535095,
      "learning_rate": 2e-05,
      "loss": 0.8494,
      "step": 2550
    },
    {
      "epoch": 0.22789993768361078,
      "grad_norm": 0.42076337337493896,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 2560
    },
    {
      "epoch": 0.2287901718151874,
      "grad_norm": 0.4057317078113556,
      "learning_rate": 2e-05,
      "loss": 0.8373,
      "step": 2570
    },
    {
      "epoch": 0.229680405946764,
      "grad_norm": 0.8177383542060852,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 2580
    },
    {
      "epoch": 0.2305706400783406,
      "grad_norm": 0.44814127683639526,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 2590
    },
    {
      "epoch": 0.2314608742099172,
      "grad_norm": 0.4277145266532898,
      "learning_rate": 2e-05,
      "loss": 0.839,
      "step": 2600
    },
    {
      "epoch": 0.2323511083414938,
      "grad_norm": 0.6474114060401917,
      "learning_rate": 2e-05,
      "loss": 0.8188,
      "step": 2610
    },
    {
      "epoch": 0.2332413424730704,
      "grad_norm": 0.414085328578949,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 2620
    },
    {
      "epoch": 0.23413157660464703,
      "grad_norm": 0.45837730169296265,
      "learning_rate": 2e-05,
      "loss": 0.8541,
      "step": 2630
    },
    {
      "epoch": 0.23502181073622364,
      "grad_norm": 0.6366609334945679,
      "learning_rate": 2e-05,
      "loss": 0.8286,
      "step": 2640
    },
    {
      "epoch": 0.23591204486780024,
      "grad_norm": 0.43247151374816895,
      "learning_rate": 2e-05,
      "loss": 0.8307,
      "step": 2650
    },
    {
      "epoch": 0.23680227899937684,
      "grad_norm": 0.43700429797172546,
      "learning_rate": 2e-05,
      "loss": 0.8523,
      "step": 2660
    },
    {
      "epoch": 0.23769251313095344,
      "grad_norm": 0.43458279967308044,
      "learning_rate": 2e-05,
      "loss": 0.852,
      "step": 2670
    },
    {
      "epoch": 0.23858274726253004,
      "grad_norm": 0.4290573000907898,
      "learning_rate": 2e-05,
      "loss": 0.8692,
      "step": 2680
    },
    {
      "epoch": 0.23947298139410664,
      "grad_norm": 0.4276871979236603,
      "learning_rate": 2e-05,
      "loss": 0.8262,
      "step": 2690
    },
    {
      "epoch": 0.24036321552568327,
      "grad_norm": 0.47235074639320374,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 2700
    },
    {
      "epoch": 0.24125344965725987,
      "grad_norm": 0.4367382526397705,
      "learning_rate": 2e-05,
      "loss": 0.896,
      "step": 2710
    },
    {
      "epoch": 0.24214368378883647,
      "grad_norm": 0.5260432362556458,
      "learning_rate": 2e-05,
      "loss": 0.8189,
      "step": 2720
    },
    {
      "epoch": 0.24303391792041307,
      "grad_norm": 0.43498674035072327,
      "learning_rate": 2e-05,
      "loss": 0.8639,
      "step": 2730
    },
    {
      "epoch": 0.24392415205198967,
      "grad_norm": 0.4313834309577942,
      "learning_rate": 2e-05,
      "loss": 0.8802,
      "step": 2740
    },
    {
      "epoch": 0.24481438618356627,
      "grad_norm": 0.4172224700450897,
      "learning_rate": 2e-05,
      "loss": 0.8259,
      "step": 2750
    },
    {
      "epoch": 0.24570462031514287,
      "grad_norm": 0.4088374674320221,
      "learning_rate": 2e-05,
      "loss": 0.844,
      "step": 2760
    },
    {
      "epoch": 0.2465948544467195,
      "grad_norm": 0.5691619515419006,
      "learning_rate": 2e-05,
      "loss": 0.8679,
      "step": 2770
    },
    {
      "epoch": 0.2474850885782961,
      "grad_norm": 0.42780718207359314,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 2780
    },
    {
      "epoch": 0.2483753227098727,
      "grad_norm": 0.41950735449790955,
      "learning_rate": 2e-05,
      "loss": 0.8807,
      "step": 2790
    },
    {
      "epoch": 0.2492655568414493,
      "grad_norm": 0.4328187108039856,
      "learning_rate": 2e-05,
      "loss": 0.8315,
      "step": 2800
    },
    {
      "epoch": 0.2501557909730259,
      "grad_norm": 0.4625028669834137,
      "learning_rate": 2e-05,
      "loss": 0.8474,
      "step": 2810
    },
    {
      "epoch": 0.2510460251046025,
      "grad_norm": 0.4164404571056366,
      "learning_rate": 2e-05,
      "loss": 0.8384,
      "step": 2820
    },
    {
      "epoch": 0.2519362592361791,
      "grad_norm": 0.4204826056957245,
      "learning_rate": 2e-05,
      "loss": 0.8354,
      "step": 2830
    },
    {
      "epoch": 0.2528264933677557,
      "grad_norm": 0.43664735555648804,
      "learning_rate": 2e-05,
      "loss": 0.8742,
      "step": 2840
    },
    {
      "epoch": 0.25371672749933233,
      "grad_norm": 0.41397643089294434,
      "learning_rate": 2e-05,
      "loss": 0.8335,
      "step": 2850
    },
    {
      "epoch": 0.25460696163090896,
      "grad_norm": 0.43303459882736206,
      "learning_rate": 2e-05,
      "loss": 0.8563,
      "step": 2860
    },
    {
      "epoch": 0.25549719576248553,
      "grad_norm": 0.42382875084877014,
      "learning_rate": 2e-05,
      "loss": 0.8662,
      "step": 2870
    },
    {
      "epoch": 0.25638742989406216,
      "grad_norm": 0.4423929452896118,
      "learning_rate": 2e-05,
      "loss": 0.8393,
      "step": 2880
    },
    {
      "epoch": 0.25727766402563873,
      "grad_norm": 0.456138014793396,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 2890
    },
    {
      "epoch": 0.25816789815721536,
      "grad_norm": 0.39736881852149963,
      "learning_rate": 2e-05,
      "loss": 0.8444,
      "step": 2900
    },
    {
      "epoch": 0.25905813228879193,
      "grad_norm": 0.43651801347732544,
      "learning_rate": 2e-05,
      "loss": 0.858,
      "step": 2910
    },
    {
      "epoch": 0.25994836642036856,
      "grad_norm": 0.4284229874610901,
      "learning_rate": 2e-05,
      "loss": 0.8435,
      "step": 2920
    },
    {
      "epoch": 0.2608386005519452,
      "grad_norm": 0.42209291458129883,
      "learning_rate": 2e-05,
      "loss": 0.8283,
      "step": 2930
    },
    {
      "epoch": 0.26172883468352176,
      "grad_norm": 0.4549107253551483,
      "learning_rate": 2e-05,
      "loss": 0.8425,
      "step": 2940
    },
    {
      "epoch": 0.2626190688150984,
      "grad_norm": 0.4218333065509796,
      "learning_rate": 2e-05,
      "loss": 0.8205,
      "step": 2950
    },
    {
      "epoch": 0.26350930294667496,
      "grad_norm": 0.3946894109249115,
      "learning_rate": 2e-05,
      "loss": 0.838,
      "step": 2960
    },
    {
      "epoch": 0.2643995370782516,
      "grad_norm": 0.4280807673931122,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 2970
    },
    {
      "epoch": 0.26528977120982816,
      "grad_norm": 0.4125746786594391,
      "learning_rate": 2e-05,
      "loss": 0.8336,
      "step": 2980
    },
    {
      "epoch": 0.2661800053414048,
      "grad_norm": 0.40464118123054504,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 2990
    },
    {
      "epoch": 0.2670702394729814,
      "grad_norm": 0.44472426176071167,
      "learning_rate": 2e-05,
      "loss": 0.8417,
      "step": 3000
    },
    {
      "epoch": 0.267960473604558,
      "grad_norm": 0.4557833671569824,
      "learning_rate": 2e-05,
      "loss": 0.8394,
      "step": 3010
    },
    {
      "epoch": 0.2688507077361346,
      "grad_norm": 0.42392170429229736,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 3020
    },
    {
      "epoch": 0.2697409418677112,
      "grad_norm": 0.43136587738990784,
      "learning_rate": 2e-05,
      "loss": 0.8434,
      "step": 3030
    },
    {
      "epoch": 0.2706311759992878,
      "grad_norm": 0.40608543157577515,
      "learning_rate": 2e-05,
      "loss": 0.8295,
      "step": 3040
    },
    {
      "epoch": 0.2715214101308644,
      "grad_norm": 0.421271413564682,
      "learning_rate": 2e-05,
      "loss": 0.821,
      "step": 3050
    },
    {
      "epoch": 0.272411644262441,
      "grad_norm": 0.4330582022666931,
      "learning_rate": 2e-05,
      "loss": 0.7876,
      "step": 3060
    },
    {
      "epoch": 0.27330187839401765,
      "grad_norm": 0.4295467138290405,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 3070
    },
    {
      "epoch": 0.2741921125255942,
      "grad_norm": 0.394844114780426,
      "learning_rate": 2e-05,
      "loss": 0.8487,
      "step": 3080
    },
    {
      "epoch": 0.27508234665717085,
      "grad_norm": 0.40360546112060547,
      "learning_rate": 2e-05,
      "loss": 0.853,
      "step": 3090
    },
    {
      "epoch": 0.2759725807887474,
      "grad_norm": 0.42461588978767395,
      "learning_rate": 2e-05,
      "loss": 0.854,
      "step": 3100
    },
    {
      "epoch": 0.27686281492032405,
      "grad_norm": 0.43159249424934387,
      "learning_rate": 2e-05,
      "loss": 0.8393,
      "step": 3110
    },
    {
      "epoch": 0.2777530490519006,
      "grad_norm": 0.4250316917896271,
      "learning_rate": 2e-05,
      "loss": 0.8478,
      "step": 3120
    },
    {
      "epoch": 0.27864328318347725,
      "grad_norm": 0.42414626479148865,
      "learning_rate": 2e-05,
      "loss": 0.8401,
      "step": 3130
    },
    {
      "epoch": 0.2795335173150539,
      "grad_norm": 0.4358814060688019,
      "learning_rate": 2e-05,
      "loss": 0.8455,
      "step": 3140
    },
    {
      "epoch": 0.28042375144663045,
      "grad_norm": 0.437195748090744,
      "learning_rate": 2e-05,
      "loss": 0.8527,
      "step": 3150
    },
    {
      "epoch": 0.2813139855782071,
      "grad_norm": 0.6838892698287964,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 3160
    },
    {
      "epoch": 0.28220421970978365,
      "grad_norm": 0.42534223198890686,
      "learning_rate": 2e-05,
      "loss": 0.866,
      "step": 3170
    },
    {
      "epoch": 0.2830944538413603,
      "grad_norm": 0.4798223376274109,
      "learning_rate": 2e-05,
      "loss": 0.828,
      "step": 3180
    },
    {
      "epoch": 0.28398468797293686,
      "grad_norm": 0.5947292447090149,
      "learning_rate": 2e-05,
      "loss": 0.8322,
      "step": 3190
    },
    {
      "epoch": 0.2848749221045135,
      "grad_norm": 0.4492744505405426,
      "learning_rate": 2e-05,
      "loss": 0.8138,
      "step": 3200
    },
    {
      "epoch": 0.2857651562360901,
      "grad_norm": 0.4417429268360138,
      "learning_rate": 2e-05,
      "loss": 0.8309,
      "step": 3210
    },
    {
      "epoch": 0.2866553903676667,
      "grad_norm": 0.3862607777118683,
      "learning_rate": 2e-05,
      "loss": 0.8358,
      "step": 3220
    },
    {
      "epoch": 0.2875456244992433,
      "grad_norm": 0.43765485286712646,
      "learning_rate": 2e-05,
      "loss": 0.8451,
      "step": 3230
    },
    {
      "epoch": 0.2884358586308199,
      "grad_norm": 0.46194157004356384,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 3240
    },
    {
      "epoch": 0.2893260927623965,
      "grad_norm": 0.430292010307312,
      "learning_rate": 2e-05,
      "loss": 0.8561,
      "step": 3250
    },
    {
      "epoch": 0.29021632689397314,
      "grad_norm": 0.4018983542919159,
      "learning_rate": 2e-05,
      "loss": 0.8233,
      "step": 3260
    },
    {
      "epoch": 0.2911065610255497,
      "grad_norm": 0.4397847354412079,
      "learning_rate": 2e-05,
      "loss": 0.8583,
      "step": 3270
    },
    {
      "epoch": 0.29199679515712634,
      "grad_norm": 0.40910622477531433,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 3280
    },
    {
      "epoch": 0.2928870292887029,
      "grad_norm": 0.42725783586502075,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 3290
    },
    {
      "epoch": 0.29377726342027954,
      "grad_norm": 0.4168187975883484,
      "learning_rate": 2e-05,
      "loss": 0.8129,
      "step": 3300
    },
    {
      "epoch": 0.2946674975518561,
      "grad_norm": 0.40158092975616455,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 3310
    },
    {
      "epoch": 0.29555773168343275,
      "grad_norm": 0.4245370924472809,
      "learning_rate": 2e-05,
      "loss": 0.8518,
      "step": 3320
    },
    {
      "epoch": 0.2964479658150094,
      "grad_norm": 0.3972856402397156,
      "learning_rate": 2e-05,
      "loss": 0.8296,
      "step": 3330
    },
    {
      "epoch": 0.29733819994658595,
      "grad_norm": 0.4222424328327179,
      "learning_rate": 2e-05,
      "loss": 0.8411,
      "step": 3340
    },
    {
      "epoch": 0.2982284340781626,
      "grad_norm": 0.41043180227279663,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 3350
    },
    {
      "epoch": 0.29911866820973915,
      "grad_norm": 0.45406341552734375,
      "learning_rate": 2e-05,
      "loss": 0.8463,
      "step": 3360
    },
    {
      "epoch": 0.3000089023413158,
      "grad_norm": 0.4208841025829315,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 3370
    },
    {
      "epoch": 0.30089913647289235,
      "grad_norm": 0.41306719183921814,
      "learning_rate": 2e-05,
      "loss": 0.8495,
      "step": 3380
    },
    {
      "epoch": 0.301789370604469,
      "grad_norm": 0.4270111322402954,
      "learning_rate": 2e-05,
      "loss": 0.8339,
      "step": 3390
    },
    {
      "epoch": 0.3026796047360456,
      "grad_norm": 0.4206894636154175,
      "learning_rate": 2e-05,
      "loss": 0.857,
      "step": 3400
    },
    {
      "epoch": 0.3035698388676222,
      "grad_norm": 0.42541080713272095,
      "learning_rate": 2e-05,
      "loss": 0.8095,
      "step": 3410
    },
    {
      "epoch": 0.3044600729991988,
      "grad_norm": 0.4107275605201721,
      "learning_rate": 2e-05,
      "loss": 0.8311,
      "step": 3420
    },
    {
      "epoch": 0.3053503071307754,
      "grad_norm": 0.4141565263271332,
      "learning_rate": 2e-05,
      "loss": 0.7865,
      "step": 3430
    },
    {
      "epoch": 0.306240541262352,
      "grad_norm": 0.4420231878757477,
      "learning_rate": 2e-05,
      "loss": 0.8488,
      "step": 3440
    },
    {
      "epoch": 0.3071307753939286,
      "grad_norm": 0.40426716208457947,
      "learning_rate": 2e-05,
      "loss": 0.8184,
      "step": 3450
    },
    {
      "epoch": 0.3080210095255052,
      "grad_norm": 0.42423561215400696,
      "learning_rate": 2e-05,
      "loss": 0.7847,
      "step": 3460
    },
    {
      "epoch": 0.30891124365708184,
      "grad_norm": 0.4481654763221741,
      "learning_rate": 2e-05,
      "loss": 0.823,
      "step": 3470
    },
    {
      "epoch": 0.3098014777886584,
      "grad_norm": 0.4493624269962311,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 3480
    },
    {
      "epoch": 0.31069171192023504,
      "grad_norm": 0.4318940341472626,
      "learning_rate": 2e-05,
      "loss": 0.8152,
      "step": 3490
    },
    {
      "epoch": 0.3115819460518116,
      "grad_norm": 0.41274625062942505,
      "learning_rate": 2e-05,
      "loss": 0.8531,
      "step": 3500
    },
    {
      "epoch": 0.31247218018338824,
      "grad_norm": 0.43591421842575073,
      "learning_rate": 2e-05,
      "loss": 0.8076,
      "step": 3510
    },
    {
      "epoch": 0.3133624143149648,
      "grad_norm": 0.4300544857978821,
      "learning_rate": 2e-05,
      "loss": 0.8423,
      "step": 3520
    },
    {
      "epoch": 0.31425264844654144,
      "grad_norm": 0.43675532937049866,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 3530
    },
    {
      "epoch": 0.31514288257811807,
      "grad_norm": 0.47683587670326233,
      "learning_rate": 2e-05,
      "loss": 0.8102,
      "step": 3540
    },
    {
      "epoch": 0.31603311670969464,
      "grad_norm": 0.4293249249458313,
      "learning_rate": 2e-05,
      "loss": 0.8601,
      "step": 3550
    },
    {
      "epoch": 0.31692335084127127,
      "grad_norm": 0.41486814618110657,
      "learning_rate": 2e-05,
      "loss": 0.8412,
      "step": 3560
    },
    {
      "epoch": 0.31781358497284784,
      "grad_norm": 0.4220460057258606,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 3570
    },
    {
      "epoch": 0.31870381910442447,
      "grad_norm": 0.4201449751853943,
      "learning_rate": 2e-05,
      "loss": 0.8419,
      "step": 3580
    },
    {
      "epoch": 0.31959405323600104,
      "grad_norm": 0.43113377690315247,
      "learning_rate": 2e-05,
      "loss": 0.8153,
      "step": 3590
    },
    {
      "epoch": 0.32048428736757767,
      "grad_norm": 0.4121089279651642,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 3600
    },
    {
      "epoch": 0.3213745214991543,
      "grad_norm": 0.4429466724395752,
      "learning_rate": 2e-05,
      "loss": 0.8353,
      "step": 3610
    },
    {
      "epoch": 0.32226475563073087,
      "grad_norm": 0.41377443075180054,
      "learning_rate": 2e-05,
      "loss": 0.8239,
      "step": 3620
    },
    {
      "epoch": 0.3231549897623075,
      "grad_norm": 0.40802210569381714,
      "learning_rate": 2e-05,
      "loss": 0.8327,
      "step": 3630
    },
    {
      "epoch": 0.32404522389388407,
      "grad_norm": 0.4101453423500061,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 3640
    },
    {
      "epoch": 0.3249354580254607,
      "grad_norm": 0.45503178238868713,
      "learning_rate": 2e-05,
      "loss": 0.8468,
      "step": 3650
    },
    {
      "epoch": 0.3258256921570373,
      "grad_norm": 0.45985323190689087,
      "learning_rate": 2e-05,
      "loss": 0.8673,
      "step": 3660
    },
    {
      "epoch": 0.3267159262886139,
      "grad_norm": 0.43549156188964844,
      "learning_rate": 2e-05,
      "loss": 0.8377,
      "step": 3670
    },
    {
      "epoch": 0.32760616042019053,
      "grad_norm": 0.4208657741546631,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 3680
    },
    {
      "epoch": 0.3284963945517671,
      "grad_norm": 0.4263143539428711,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 3690
    },
    {
      "epoch": 0.32938662868334373,
      "grad_norm": 0.40754878520965576,
      "learning_rate": 2e-05,
      "loss": 0.8349,
      "step": 3700
    },
    {
      "epoch": 0.3302768628149203,
      "grad_norm": 0.43649813532829285,
      "learning_rate": 2e-05,
      "loss": 0.8513,
      "step": 3710
    },
    {
      "epoch": 0.33116709694649693,
      "grad_norm": 0.42877429723739624,
      "learning_rate": 2e-05,
      "loss": 0.8407,
      "step": 3720
    },
    {
      "epoch": 0.33205733107807356,
      "grad_norm": 0.4446612298488617,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 3730
    },
    {
      "epoch": 0.33294756520965013,
      "grad_norm": 0.40203750133514404,
      "learning_rate": 2e-05,
      "loss": 0.782,
      "step": 3740
    },
    {
      "epoch": 0.33383779934122676,
      "grad_norm": 0.42034125328063965,
      "learning_rate": 2e-05,
      "loss": 0.8237,
      "step": 3750
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 0.4304819405078888,
      "learning_rate": 2e-05,
      "loss": 0.8357,
      "step": 3760
    },
    {
      "epoch": 0.33561826760437996,
      "grad_norm": 0.45289233326911926,
      "learning_rate": 2e-05,
      "loss": 0.8474,
      "step": 3770
    },
    {
      "epoch": 0.33650850173595653,
      "grad_norm": 0.40869587659835815,
      "learning_rate": 2e-05,
      "loss": 0.8241,
      "step": 3780
    },
    {
      "epoch": 0.33739873586753316,
      "grad_norm": 0.41707906126976013,
      "learning_rate": 2e-05,
      "loss": 0.8145,
      "step": 3790
    },
    {
      "epoch": 0.3382889699991098,
      "grad_norm": 0.4553302526473999,
      "learning_rate": 2e-05,
      "loss": 0.8457,
      "step": 3800
    },
    {
      "epoch": 0.33917920413068636,
      "grad_norm": 0.41422930359840393,
      "learning_rate": 2e-05,
      "loss": 0.8064,
      "step": 3810
    },
    {
      "epoch": 0.340069438262263,
      "grad_norm": 0.40437325835227966,
      "learning_rate": 2e-05,
      "loss": 0.7979,
      "step": 3820
    },
    {
      "epoch": 0.34095967239383956,
      "grad_norm": 0.40993422269821167,
      "learning_rate": 2e-05,
      "loss": 0.7963,
      "step": 3830
    },
    {
      "epoch": 0.3418499065254162,
      "grad_norm": 0.4463188350200653,
      "learning_rate": 2e-05,
      "loss": 0.8383,
      "step": 3840
    },
    {
      "epoch": 0.34274014065699276,
      "grad_norm": 0.41199883818626404,
      "learning_rate": 2e-05,
      "loss": 0.8153,
      "step": 3850
    },
    {
      "epoch": 0.3436303747885694,
      "grad_norm": 0.39703360199928284,
      "learning_rate": 2e-05,
      "loss": 0.8322,
      "step": 3860
    },
    {
      "epoch": 0.344520608920146,
      "grad_norm": 0.4153502285480499,
      "learning_rate": 2e-05,
      "loss": 0.8482,
      "step": 3870
    },
    {
      "epoch": 0.3454108430517226,
      "grad_norm": 0.4348487854003906,
      "learning_rate": 2e-05,
      "loss": 0.8251,
      "step": 3880
    },
    {
      "epoch": 0.3463010771832992,
      "grad_norm": 0.3973188102245331,
      "learning_rate": 2e-05,
      "loss": 0.8155,
      "step": 3890
    },
    {
      "epoch": 0.3471913113148758,
      "grad_norm": 0.4261895716190338,
      "learning_rate": 2e-05,
      "loss": 0.8228,
      "step": 3900
    },
    {
      "epoch": 0.3480815454464524,
      "grad_norm": 0.4215255677700043,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 3910
    },
    {
      "epoch": 0.348971779578029,
      "grad_norm": 0.399711549282074,
      "learning_rate": 2e-05,
      "loss": 0.8168,
      "step": 3920
    },
    {
      "epoch": 0.3498620137096056,
      "grad_norm": 0.3966852128505707,
      "learning_rate": 2e-05,
      "loss": 0.8458,
      "step": 3930
    },
    {
      "epoch": 0.35075224784118225,
      "grad_norm": 0.40908384323120117,
      "learning_rate": 2e-05,
      "loss": 0.8446,
      "step": 3940
    },
    {
      "epoch": 0.3516424819727588,
      "grad_norm": 0.41330382227897644,
      "learning_rate": 2e-05,
      "loss": 0.8112,
      "step": 3950
    },
    {
      "epoch": 0.35253271610433545,
      "grad_norm": 0.41872796416282654,
      "learning_rate": 2e-05,
      "loss": 0.8362,
      "step": 3960
    },
    {
      "epoch": 0.353422950235912,
      "grad_norm": 0.5635706186294556,
      "learning_rate": 2e-05,
      "loss": 0.8535,
      "step": 3970
    },
    {
      "epoch": 0.35431318436748865,
      "grad_norm": 0.4385491609573364,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 3980
    },
    {
      "epoch": 0.3552034184990652,
      "grad_norm": 0.426220566034317,
      "learning_rate": 2e-05,
      "loss": 0.8331,
      "step": 3990
    },
    {
      "epoch": 0.35609365263064185,
      "grad_norm": 0.41327157616615295,
      "learning_rate": 2e-05,
      "loss": 0.8124,
      "step": 4000
    },
    {
      "epoch": 0.3569838867622185,
      "grad_norm": 0.4117506444454193,
      "learning_rate": 2e-05,
      "loss": 0.8212,
      "step": 4010
    },
    {
      "epoch": 0.35787412089379506,
      "grad_norm": 0.4391235113143921,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 4020
    },
    {
      "epoch": 0.3587643550253717,
      "grad_norm": 0.40042269229888916,
      "learning_rate": 2e-05,
      "loss": 0.807,
      "step": 4030
    },
    {
      "epoch": 0.35965458915694826,
      "grad_norm": 0.4029245376586914,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 4040
    },
    {
      "epoch": 0.3605448232885249,
      "grad_norm": 0.4109240174293518,
      "learning_rate": 2e-05,
      "loss": 0.8229,
      "step": 4050
    },
    {
      "epoch": 0.3614350574201015,
      "grad_norm": 0.4032326936721802,
      "learning_rate": 2e-05,
      "loss": 0.8035,
      "step": 4060
    },
    {
      "epoch": 0.3623252915516781,
      "grad_norm": 0.4191797077655792,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 4070
    },
    {
      "epoch": 0.3632155256832547,
      "grad_norm": 0.41351959109306335,
      "learning_rate": 2e-05,
      "loss": 0.8199,
      "step": 4080
    },
    {
      "epoch": 0.3641057598148313,
      "grad_norm": 0.40022891759872437,
      "learning_rate": 2e-05,
      "loss": 0.8199,
      "step": 4090
    },
    {
      "epoch": 0.3649959939464079,
      "grad_norm": 0.4206380844116211,
      "learning_rate": 2e-05,
      "loss": 0.8372,
      "step": 4100
    },
    {
      "epoch": 0.3658862280779845,
      "grad_norm": 0.40191662311553955,
      "learning_rate": 2e-05,
      "loss": 0.8434,
      "step": 4110
    },
    {
      "epoch": 0.3667764622095611,
      "grad_norm": 0.418744295835495,
      "learning_rate": 2e-05,
      "loss": 0.8345,
      "step": 4120
    },
    {
      "epoch": 0.36766669634113774,
      "grad_norm": 0.42262357473373413,
      "learning_rate": 2e-05,
      "loss": 0.824,
      "step": 4130
    },
    {
      "epoch": 0.3685569304727143,
      "grad_norm": 0.45413738489151,
      "learning_rate": 2e-05,
      "loss": 0.8102,
      "step": 4140
    },
    {
      "epoch": 0.36944716460429095,
      "grad_norm": 0.4292362928390503,
      "learning_rate": 2e-05,
      "loss": 0.8223,
      "step": 4150
    },
    {
      "epoch": 0.3703373987358675,
      "grad_norm": 0.3982831537723541,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 4160
    },
    {
      "epoch": 0.37122763286744415,
      "grad_norm": 0.4300854206085205,
      "learning_rate": 2e-05,
      "loss": 0.7877,
      "step": 4170
    },
    {
      "epoch": 0.3721178669990207,
      "grad_norm": 0.42021501064300537,
      "learning_rate": 2e-05,
      "loss": 0.8536,
      "step": 4180
    },
    {
      "epoch": 0.37300810113059735,
      "grad_norm": 0.42756009101867676,
      "learning_rate": 2e-05,
      "loss": 0.8555,
      "step": 4190
    },
    {
      "epoch": 0.373898335262174,
      "grad_norm": 0.42907434701919556,
      "learning_rate": 2e-05,
      "loss": 0.8145,
      "step": 4200
    },
    {
      "epoch": 0.37478856939375055,
      "grad_norm": 0.45126932859420776,
      "learning_rate": 2e-05,
      "loss": 0.8375,
      "step": 4210
    },
    {
      "epoch": 0.3756788035253272,
      "grad_norm": 0.38375094532966614,
      "learning_rate": 2e-05,
      "loss": 0.7933,
      "step": 4220
    },
    {
      "epoch": 0.37656903765690375,
      "grad_norm": 0.45565056800842285,
      "learning_rate": 2e-05,
      "loss": 0.8063,
      "step": 4230
    },
    {
      "epoch": 0.3774592717884804,
      "grad_norm": 0.4196164011955261,
      "learning_rate": 2e-05,
      "loss": 0.8233,
      "step": 4240
    },
    {
      "epoch": 0.37834950592005695,
      "grad_norm": 0.4495660960674286,
      "learning_rate": 2e-05,
      "loss": 0.8185,
      "step": 4250
    },
    {
      "epoch": 0.3792397400516336,
      "grad_norm": 0.41325727105140686,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 4260
    },
    {
      "epoch": 0.3801299741832102,
      "grad_norm": 0.40559035539627075,
      "learning_rate": 2e-05,
      "loss": 0.8262,
      "step": 4270
    },
    {
      "epoch": 0.3810202083147868,
      "grad_norm": 0.40767455101013184,
      "learning_rate": 2e-05,
      "loss": 0.8204,
      "step": 4280
    },
    {
      "epoch": 0.3819104424463634,
      "grad_norm": 0.42494624853134155,
      "learning_rate": 2e-05,
      "loss": 0.8241,
      "step": 4290
    },
    {
      "epoch": 0.38280067657794,
      "grad_norm": 0.40323498845100403,
      "learning_rate": 2e-05,
      "loss": 0.8006,
      "step": 4300
    },
    {
      "epoch": 0.3836909107095166,
      "grad_norm": 0.43134891986846924,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 4310
    },
    {
      "epoch": 0.3845811448410932,
      "grad_norm": 0.4199300706386566,
      "learning_rate": 2e-05,
      "loss": 0.8087,
      "step": 4320
    },
    {
      "epoch": 0.3854713789726698,
      "grad_norm": 0.4243045747280121,
      "learning_rate": 2e-05,
      "loss": 0.8268,
      "step": 4330
    },
    {
      "epoch": 0.38636161310424644,
      "grad_norm": 0.7792085409164429,
      "learning_rate": 2e-05,
      "loss": 0.8294,
      "step": 4340
    },
    {
      "epoch": 0.387251847235823,
      "grad_norm": 0.41471177339553833,
      "learning_rate": 2e-05,
      "loss": 0.8077,
      "step": 4350
    },
    {
      "epoch": 0.38814208136739964,
      "grad_norm": 0.42648547887802124,
      "learning_rate": 2e-05,
      "loss": 0.8305,
      "step": 4360
    },
    {
      "epoch": 0.3890323154989762,
      "grad_norm": 0.40763089060783386,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 4370
    },
    {
      "epoch": 0.38992254963055284,
      "grad_norm": 0.428901344537735,
      "learning_rate": 2e-05,
      "loss": 0.8117,
      "step": 4380
    },
    {
      "epoch": 0.39081278376212947,
      "grad_norm": 0.4178633391857147,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 4390
    },
    {
      "epoch": 0.39170301789370604,
      "grad_norm": 0.40298399329185486,
      "learning_rate": 2e-05,
      "loss": 0.8106,
      "step": 4400
    },
    {
      "epoch": 0.39259325202528267,
      "grad_norm": 0.42718732357025146,
      "learning_rate": 2e-05,
      "loss": 0.7935,
      "step": 4410
    },
    {
      "epoch": 0.39348348615685924,
      "grad_norm": 0.4392097592353821,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 4420
    },
    {
      "epoch": 0.39437372028843587,
      "grad_norm": 0.4516502022743225,
      "learning_rate": 2e-05,
      "loss": 0.8068,
      "step": 4430
    },
    {
      "epoch": 0.39526395442001244,
      "grad_norm": 0.42506465315818787,
      "learning_rate": 2e-05,
      "loss": 0.8022,
      "step": 4440
    },
    {
      "epoch": 0.39615418855158907,
      "grad_norm": 0.4037013053894043,
      "learning_rate": 2e-05,
      "loss": 0.8086,
      "step": 4450
    },
    {
      "epoch": 0.3970444226831657,
      "grad_norm": 0.4320470988750458,
      "learning_rate": 2e-05,
      "loss": 0.8309,
      "step": 4460
    },
    {
      "epoch": 0.39793465681474227,
      "grad_norm": 0.42536693811416626,
      "learning_rate": 2e-05,
      "loss": 0.7915,
      "step": 4470
    },
    {
      "epoch": 0.3988248909463189,
      "grad_norm": 0.41374078392982483,
      "learning_rate": 2e-05,
      "loss": 0.8017,
      "step": 4480
    },
    {
      "epoch": 0.3997151250778955,
      "grad_norm": 0.4253350794315338,
      "learning_rate": 2e-05,
      "loss": 0.8171,
      "step": 4490
    },
    {
      "epoch": 0.4006053592094721,
      "grad_norm": 0.4242878258228302,
      "learning_rate": 2e-05,
      "loss": 0.8544,
      "step": 4500
    },
    {
      "epoch": 0.4014955933410487,
      "grad_norm": 0.4610006511211395,
      "learning_rate": 2e-05,
      "loss": 0.7999,
      "step": 4510
    },
    {
      "epoch": 0.4023858274726253,
      "grad_norm": 0.3998091220855713,
      "learning_rate": 2e-05,
      "loss": 0.8245,
      "step": 4520
    },
    {
      "epoch": 0.40327606160420193,
      "grad_norm": 0.3994504511356354,
      "learning_rate": 2e-05,
      "loss": 0.7968,
      "step": 4530
    },
    {
      "epoch": 0.4041662957357785,
      "grad_norm": 0.432560533285141,
      "learning_rate": 2e-05,
      "loss": 0.8279,
      "step": 4540
    },
    {
      "epoch": 0.40505652986735513,
      "grad_norm": 0.4226333498954773,
      "learning_rate": 2e-05,
      "loss": 0.8073,
      "step": 4550
    },
    {
      "epoch": 0.4059467639989317,
      "grad_norm": 0.4384939968585968,
      "learning_rate": 2e-05,
      "loss": 0.8103,
      "step": 4560
    },
    {
      "epoch": 0.40683699813050833,
      "grad_norm": 0.41693630814552307,
      "learning_rate": 2e-05,
      "loss": 0.8009,
      "step": 4570
    },
    {
      "epoch": 0.4077272322620849,
      "grad_norm": 0.44103047251701355,
      "learning_rate": 2e-05,
      "loss": 0.843,
      "step": 4580
    },
    {
      "epoch": 0.40861746639366153,
      "grad_norm": 0.42429396510124207,
      "learning_rate": 2e-05,
      "loss": 0.7949,
      "step": 4590
    },
    {
      "epoch": 0.40950770052523816,
      "grad_norm": 0.4315495491027832,
      "learning_rate": 2e-05,
      "loss": 0.8995,
      "step": 4600
    },
    {
      "epoch": 0.41039793465681473,
      "grad_norm": 0.4785359799861908,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 4610
    },
    {
      "epoch": 0.41128816878839136,
      "grad_norm": 0.4082861542701721,
      "learning_rate": 2e-05,
      "loss": 0.8344,
      "step": 4620
    },
    {
      "epoch": 0.41217840291996793,
      "grad_norm": 0.40187034010887146,
      "learning_rate": 2e-05,
      "loss": 0.8117,
      "step": 4630
    },
    {
      "epoch": 0.41306863705154456,
      "grad_norm": 0.4089754819869995,
      "learning_rate": 2e-05,
      "loss": 0.8131,
      "step": 4640
    },
    {
      "epoch": 0.41395887118312114,
      "grad_norm": 0.40612879395484924,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 4650
    },
    {
      "epoch": 0.41484910531469776,
      "grad_norm": 0.4087541103363037,
      "learning_rate": 2e-05,
      "loss": 0.8059,
      "step": 4660
    },
    {
      "epoch": 0.4157393394462744,
      "grad_norm": 0.4274562895298004,
      "learning_rate": 2e-05,
      "loss": 0.8439,
      "step": 4670
    },
    {
      "epoch": 0.41662957357785096,
      "grad_norm": 0.4199943244457245,
      "learning_rate": 2e-05,
      "loss": 0.8202,
      "step": 4680
    },
    {
      "epoch": 0.4175198077094276,
      "grad_norm": 0.40392178297042847,
      "learning_rate": 2e-05,
      "loss": 0.8036,
      "step": 4690
    },
    {
      "epoch": 0.41841004184100417,
      "grad_norm": 0.4199431538581848,
      "learning_rate": 2e-05,
      "loss": 0.8089,
      "step": 4700
    },
    {
      "epoch": 0.4193002759725808,
      "grad_norm": 0.448820024728775,
      "learning_rate": 2e-05,
      "loss": 0.807,
      "step": 4710
    },
    {
      "epoch": 0.42019051010415737,
      "grad_norm": 0.3838268518447876,
      "learning_rate": 2e-05,
      "loss": 0.8141,
      "step": 4720
    },
    {
      "epoch": 0.421080744235734,
      "grad_norm": 0.3966931402683258,
      "learning_rate": 2e-05,
      "loss": 0.8114,
      "step": 4730
    },
    {
      "epoch": 0.4219709783673106,
      "grad_norm": 0.42101746797561646,
      "learning_rate": 2e-05,
      "loss": 0.8111,
      "step": 4740
    },
    {
      "epoch": 0.4228612124988872,
      "grad_norm": 0.41088178753852844,
      "learning_rate": 2e-05,
      "loss": 0.7957,
      "step": 4750
    },
    {
      "epoch": 0.4237514466304638,
      "grad_norm": 0.40913447737693787,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 4760
    },
    {
      "epoch": 0.4246416807620404,
      "grad_norm": 0.3723960816860199,
      "learning_rate": 2e-05,
      "loss": 0.7886,
      "step": 4770
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 0.4006163775920868,
      "learning_rate": 2e-05,
      "loss": 0.8081,
      "step": 4780
    },
    {
      "epoch": 0.42642214902519365,
      "grad_norm": 0.429553747177124,
      "learning_rate": 2e-05,
      "loss": 0.8333,
      "step": 4790
    },
    {
      "epoch": 0.4273123831567702,
      "grad_norm": 0.4222668707370758,
      "learning_rate": 2e-05,
      "loss": 0.8345,
      "step": 4800
    },
    {
      "epoch": 0.42820261728834685,
      "grad_norm": 0.3898997902870178,
      "learning_rate": 2e-05,
      "loss": 0.8193,
      "step": 4810
    },
    {
      "epoch": 0.4290928514199234,
      "grad_norm": 0.40183025598526,
      "learning_rate": 2e-05,
      "loss": 0.8114,
      "step": 4820
    },
    {
      "epoch": 0.42998308555150005,
      "grad_norm": 0.5734966397285461,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 4830
    },
    {
      "epoch": 0.4308733196830766,
      "grad_norm": 0.4249509572982788,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 4840
    },
    {
      "epoch": 0.43176355381465326,
      "grad_norm": 0.4371035397052765,
      "learning_rate": 2e-05,
      "loss": 0.8346,
      "step": 4850
    },
    {
      "epoch": 0.4326537879462299,
      "grad_norm": 0.7718930244445801,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 4860
    },
    {
      "epoch": 0.43354402207780646,
      "grad_norm": 0.42943865060806274,
      "learning_rate": 2e-05,
      "loss": 0.8259,
      "step": 4870
    },
    {
      "epoch": 0.4344342562093831,
      "grad_norm": 0.5938506722450256,
      "learning_rate": 2e-05,
      "loss": 0.8095,
      "step": 4880
    },
    {
      "epoch": 0.43532449034095966,
      "grad_norm": 0.4294741153717041,
      "learning_rate": 2e-05,
      "loss": 0.801,
      "step": 4890
    },
    {
      "epoch": 0.4362147244725363,
      "grad_norm": 0.38467851281166077,
      "learning_rate": 2e-05,
      "loss": 0.7963,
      "step": 4900
    },
    {
      "epoch": 0.43710495860411286,
      "grad_norm": 0.39362895488739014,
      "learning_rate": 2e-05,
      "loss": 0.789,
      "step": 4910
    },
    {
      "epoch": 0.4379951927356895,
      "grad_norm": 0.4231758415699005,
      "learning_rate": 2e-05,
      "loss": 0.8095,
      "step": 4920
    },
    {
      "epoch": 0.4388854268672661,
      "grad_norm": 0.41444429755210876,
      "learning_rate": 2e-05,
      "loss": 0.7958,
      "step": 4930
    },
    {
      "epoch": 0.4397756609988427,
      "grad_norm": 0.4043222963809967,
      "learning_rate": 2e-05,
      "loss": 0.8357,
      "step": 4940
    },
    {
      "epoch": 0.4406658951304193,
      "grad_norm": 0.424271821975708,
      "learning_rate": 2e-05,
      "loss": 0.8152,
      "step": 4950
    },
    {
      "epoch": 0.4415561292619959,
      "grad_norm": 0.40251779556274414,
      "learning_rate": 2e-05,
      "loss": 0.8167,
      "step": 4960
    },
    {
      "epoch": 0.4424463633935725,
      "grad_norm": 0.4041527509689331,
      "learning_rate": 2e-05,
      "loss": 0.8093,
      "step": 4970
    },
    {
      "epoch": 0.4433365975251491,
      "grad_norm": 0.43732142448425293,
      "learning_rate": 2e-05,
      "loss": 0.8325,
      "step": 4980
    },
    {
      "epoch": 0.4442268316567257,
      "grad_norm": 0.3886256217956543,
      "learning_rate": 2e-05,
      "loss": 0.7962,
      "step": 4990
    },
    {
      "epoch": 0.44511706578830235,
      "grad_norm": 0.4678460359573364,
      "learning_rate": 2e-05,
      "loss": 0.8153,
      "step": 5000
    },
    {
      "epoch": 0.4460072999198789,
      "grad_norm": 0.4146559536457062,
      "learning_rate": 2e-05,
      "loss": 0.8186,
      "step": 5010
    },
    {
      "epoch": 0.44689753405145555,
      "grad_norm": 0.3865691125392914,
      "learning_rate": 2e-05,
      "loss": 0.8149,
      "step": 5020
    },
    {
      "epoch": 0.4477877681830321,
      "grad_norm": 0.4151422083377838,
      "learning_rate": 2e-05,
      "loss": 0.8363,
      "step": 5030
    },
    {
      "epoch": 0.44867800231460875,
      "grad_norm": 0.38880807161331177,
      "learning_rate": 2e-05,
      "loss": 0.8135,
      "step": 5040
    },
    {
      "epoch": 0.4495682364461853,
      "grad_norm": 0.38919609785079956,
      "learning_rate": 2e-05,
      "loss": 0.8178,
      "step": 5050
    },
    {
      "epoch": 0.45045847057776195,
      "grad_norm": 0.40866583585739136,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 5060
    },
    {
      "epoch": 0.4513487047093386,
      "grad_norm": 0.4290905296802521,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 5070
    },
    {
      "epoch": 0.45223893884091515,
      "grad_norm": 0.415767639875412,
      "learning_rate": 2e-05,
      "loss": 0.7904,
      "step": 5080
    },
    {
      "epoch": 0.4531291729724918,
      "grad_norm": 0.39671918749809265,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.7758,
      "step": 5090
    },
    {
      "epoch": 0.45401940710406835,
      "grad_norm": 0.4186277985572815,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.8025,
      "step": 5100
    },
    {
      "epoch": 0.454909641235645,
      "grad_norm": 0.39761802554130554,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.7829,
      "step": 5110
    },
    {
      "epoch": 0.45579987536722155,
      "grad_norm": 0.3970823884010315,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.8108,
      "step": 5120
    },
    {
      "epoch": 0.4566901094987982,
      "grad_norm": 0.40173348784446716,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.805,
      "step": 5130
    },
    {
      "epoch": 0.4575803436303748,
      "grad_norm": 0.42634904384613037,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.8138,
      "step": 5140
    },
    {
      "epoch": 0.4584705777619514,
      "grad_norm": 0.44743019342422485,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.8162,
      "step": 5150
    },
    {
      "epoch": 0.459360811893528,
      "grad_norm": 0.4315200448036194,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.8116,
      "step": 5160
    },
    {
      "epoch": 0.4602510460251046,
      "grad_norm": 0.4362252354621887,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.8204,
      "step": 5170
    },
    {
      "epoch": 0.4611412801566812,
      "grad_norm": 0.39729076623916626,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.7876,
      "step": 5180
    },
    {
      "epoch": 0.46203151428825784,
      "grad_norm": 0.3996797800064087,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.8213,
      "step": 5190
    },
    {
      "epoch": 0.4629217484198344,
      "grad_norm": 0.41905173659324646,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.8086,
      "step": 5200
    },
    {
      "epoch": 0.46381198255141104,
      "grad_norm": 0.4100542962551117,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.8181,
      "step": 5210
    },
    {
      "epoch": 0.4647022166829876,
      "grad_norm": 0.4531281590461731,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.8487,
      "step": 5220
    },
    {
      "epoch": 0.46559245081456424,
      "grad_norm": 0.401101678609848,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.7935,
      "step": 5230
    },
    {
      "epoch": 0.4664826849461408,
      "grad_norm": 0.38295483589172363,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.7905,
      "step": 5240
    },
    {
      "epoch": 0.46737291907771744,
      "grad_norm": 0.39416199922561646,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.7816,
      "step": 5250
    },
    {
      "epoch": 0.46826315320929407,
      "grad_norm": 0.43410876393318176,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.8149,
      "step": 5260
    },
    {
      "epoch": 0.46915338734087064,
      "grad_norm": 0.4208872616291046,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.8048,
      "step": 5270
    },
    {
      "epoch": 0.47004362147244727,
      "grad_norm": 0.4065808653831482,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.8189,
      "step": 5280
    },
    {
      "epoch": 0.47093385560402384,
      "grad_norm": 0.3961937129497528,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.7763,
      "step": 5290
    },
    {
      "epoch": 0.47182408973560047,
      "grad_norm": 0.38339245319366455,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.7954,
      "step": 5300
    },
    {
      "epoch": 0.47271432386717704,
      "grad_norm": 0.41130658984184265,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.7957,
      "step": 5310
    },
    {
      "epoch": 0.4736045579987537,
      "grad_norm": 0.4106111228466034,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.8074,
      "step": 5320
    },
    {
      "epoch": 0.4744947921303303,
      "grad_norm": 0.379512220621109,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.7987,
      "step": 5330
    },
    {
      "epoch": 0.4753850262619069,
      "grad_norm": 0.4216710329055786,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.8162,
      "step": 5340
    },
    {
      "epoch": 0.4762752603934835,
      "grad_norm": 0.3908363878726959,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.816,
      "step": 5350
    },
    {
      "epoch": 0.4771654945250601,
      "grad_norm": 0.39039817452430725,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.7842,
      "step": 5360
    },
    {
      "epoch": 0.4780557286566367,
      "grad_norm": 0.39243635535240173,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.8085,
      "step": 5370
    },
    {
      "epoch": 0.4789459627882133,
      "grad_norm": 0.428039014339447,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.7982,
      "step": 5380
    },
    {
      "epoch": 0.4798361969197899,
      "grad_norm": 0.38504689931869507,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.7726,
      "step": 5390
    },
    {
      "epoch": 0.48072643105136653,
      "grad_norm": 0.5742550492286682,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.804,
      "step": 5400
    },
    {
      "epoch": 0.4816166651829431,
      "grad_norm": 0.4175790250301361,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.7661,
      "step": 5410
    },
    {
      "epoch": 0.48250689931451973,
      "grad_norm": 0.39529699087142944,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.7932,
      "step": 5420
    },
    {
      "epoch": 0.4833971334460963,
      "grad_norm": 0.39059698581695557,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.8144,
      "step": 5430
    },
    {
      "epoch": 0.48428736757767293,
      "grad_norm": 0.4001123011112213,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.7875,
      "step": 5440
    },
    {
      "epoch": 0.4851776017092495,
      "grad_norm": 0.38896510004997253,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.7737,
      "step": 5450
    },
    {
      "epoch": 0.48606783584082613,
      "grad_norm": 0.40694886445999146,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.797,
      "step": 5460
    },
    {
      "epoch": 0.48695806997240276,
      "grad_norm": 0.3884669542312622,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.7948,
      "step": 5470
    },
    {
      "epoch": 0.48784830410397934,
      "grad_norm": 0.4052155911922455,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.8013,
      "step": 5480
    },
    {
      "epoch": 0.48873853823555596,
      "grad_norm": 0.41877853870391846,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.7843,
      "step": 5490
    },
    {
      "epoch": 0.48962877236713254,
      "grad_norm": 0.43613216280937195,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.7806,
      "step": 5500
    },
    {
      "epoch": 0.49051900649870916,
      "grad_norm": 0.3943822979927063,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.7622,
      "step": 5510
    },
    {
      "epoch": 0.49140924063028574,
      "grad_norm": 0.40710416436195374,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.7804,
      "step": 5520
    },
    {
      "epoch": 0.49229947476186237,
      "grad_norm": 0.40323033928871155,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.8216,
      "step": 5530
    },
    {
      "epoch": 0.493189708893439,
      "grad_norm": 0.37536442279815674,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.7813,
      "step": 5540
    },
    {
      "epoch": 0.49407994302501557,
      "grad_norm": 0.39955657720565796,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.779,
      "step": 5550
    },
    {
      "epoch": 0.4949701771565922,
      "grad_norm": 0.42673394083976746,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.8108,
      "step": 5560
    },
    {
      "epoch": 0.49586041128816877,
      "grad_norm": 0.3725492060184479,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.7745,
      "step": 5570
    },
    {
      "epoch": 0.4967506454197454,
      "grad_norm": 0.4250732660293579,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.7984,
      "step": 5580
    },
    {
      "epoch": 0.497640879551322,
      "grad_norm": 0.3803497552871704,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.7935,
      "step": 5590
    },
    {
      "epoch": 0.4982640434434256,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.026943307658146592,
      "train_runtime": 1761.6953,
      "train_samples_per_second": 38.125,
      "train_steps_per_second": 3.177
    }
  ],
  "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
}