{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9995964161756397,
  "eval_steps": 50,
  "global_step": 1548,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.006457341189765114,
      "grad_norm": 30.267567909012136,
      "learning_rate": 6.451612903225807e-07,
      "loss": 1.3322,
      "step": 10
    },
    {
      "epoch": 0.012914682379530229,
      "grad_norm": 12.2802238167104,
      "learning_rate": 1.2903225806451614e-06,
      "loss": 1.0135,
      "step": 20
    },
    {
      "epoch": 0.019372023569295342,
      "grad_norm": 7.960455059761512,
      "learning_rate": 1.935483870967742e-06,
      "loss": 0.7236,
      "step": 30
    },
    {
      "epoch": 0.025829364759060457,
      "grad_norm": 7.71044168621431,
      "learning_rate": 2.580645161290323e-06,
      "loss": 0.5902,
      "step": 40
    },
    {
      "epoch": 0.03228670594882557,
      "grad_norm": 6.449722223080085,
      "learning_rate": 3.225806451612903e-06,
      "loss": 0.5341,
      "step": 50
    },
    {
      "epoch": 0.03228670594882557,
      "eval_loss": 0.4792882204055786,
      "eval_runtime": 113.1816,
      "eval_samples_per_second": 35.748,
      "eval_steps_per_second": 2.235,
      "step": 50
    },
    {
      "epoch": 0.038744047138590684,
      "grad_norm": 6.367936953266739,
      "learning_rate": 3.870967741935484e-06,
      "loss": 0.5369,
      "step": 60
    },
    {
      "epoch": 0.0452013883283558,
      "grad_norm": 7.951835284314935,
      "learning_rate": 4.516129032258065e-06,
      "loss": 0.5446,
      "step": 70
    },
    {
      "epoch": 0.051658729518120915,
      "grad_norm": 4.994508946480814,
      "learning_rate": 5.161290322580646e-06,
      "loss": 0.5406,
      "step": 80
    },
    {
      "epoch": 0.058116070707886026,
      "grad_norm": 4.305696951552292,
      "learning_rate": 5.806451612903226e-06,
      "loss": 0.4874,
      "step": 90
    },
    {
      "epoch": 0.06457341189765115,
      "grad_norm": 4.732009014796636,
      "learning_rate": 6.451612903225806e-06,
      "loss": 0.5261,
      "step": 100
    },
    {
      "epoch": 0.06457341189765115,
      "eval_loss": 0.4858162999153137,
      "eval_runtime": 113.0613,
      "eval_samples_per_second": 35.786,
      "eval_steps_per_second": 2.238,
      "step": 100
    },
    {
      "epoch": 0.07103075308741626,
      "grad_norm": 5.374388570485032,
      "learning_rate": 7.096774193548388e-06,
      "loss": 0.5207,
      "step": 110
    },
    {
      "epoch": 0.07748809427718137,
      "grad_norm": 4.47458103398587,
      "learning_rate": 7.741935483870968e-06,
      "loss": 0.5235,
      "step": 120
    },
    {
      "epoch": 0.08394543546694648,
      "grad_norm": 4.283025312349808,
      "learning_rate": 8.387096774193549e-06,
      "loss": 0.55,
      "step": 130
    },
    {
      "epoch": 0.0904027766567116,
      "grad_norm": 4.249145432032488,
      "learning_rate": 9.03225806451613e-06,
      "loss": 0.545,
      "step": 140
    },
    {
      "epoch": 0.09686011784647672,
      "grad_norm": 3.90455367854534,
      "learning_rate": 9.67741935483871e-06,
      "loss": 0.5103,
      "step": 150
    },
    {
      "epoch": 0.09686011784647672,
      "eval_loss": 0.4932478964328766,
      "eval_runtime": 113.1385,
      "eval_samples_per_second": 35.761,
      "eval_steps_per_second": 2.236,
      "step": 150
    },
    {
      "epoch": 0.10331745903624183,
      "grad_norm": 4.051124309367156,
      "learning_rate": 9.999682112882774e-06,
      "loss": 0.5344,
      "step": 160
    },
    {
      "epoch": 0.10977480022600694,
      "grad_norm": 3.7940763824495796,
      "learning_rate": 9.997139258465151e-06,
      "loss": 0.5323,
      "step": 170
    },
    {
      "epoch": 0.11623214141577205,
      "grad_norm": 3.409341219142384,
      "learning_rate": 9.992054842933846e-06,
      "loss": 0.5393,
      "step": 180
    },
    {
      "epoch": 0.12268948260553716,
      "grad_norm": 3.148797921891643,
      "learning_rate": 9.984431452238968e-06,
      "loss": 0.5208,
      "step": 190
    },
    {
      "epoch": 0.1291468237953023,
      "grad_norm": 3.9252127996896284,
      "learning_rate": 9.974272963661561e-06,
      "loss": 0.5236,
      "step": 200
    },
    {
      "epoch": 0.1291468237953023,
      "eval_loss": 0.48204439878463745,
      "eval_runtime": 113.1055,
      "eval_samples_per_second": 35.772,
      "eval_steps_per_second": 2.237,
      "step": 200
    },
    {
      "epoch": 0.1356041649850674,
      "grad_norm": 3.3116374434919793,
      "learning_rate": 9.961584543841625e-06,
      "loss": 0.5369,
      "step": 210
    },
    {
      "epoch": 0.1420615061748325,
      "grad_norm": 3.1130423272518297,
      "learning_rate": 9.946372646150322e-06,
      "loss": 0.5244,
      "step": 220
    },
    {
      "epoch": 0.14851884736459764,
      "grad_norm": 3.690719240877882,
      "learning_rate": 9.928645007407791e-06,
      "loss": 0.5336,
      "step": 230
    },
    {
      "epoch": 0.15497618855436274,
      "grad_norm": 3.128668223174866,
      "learning_rate": 9.908410643948162e-06,
      "loss": 0.5015,
      "step": 240
    },
    {
      "epoch": 0.16143352974412786,
      "grad_norm": 4.538599998705712,
      "learning_rate": 9.885679847033832e-06,
      "loss": 0.524,
      "step": 250
    },
    {
      "epoch": 0.16143352974412786,
      "eval_loss": 0.462333619594574,
      "eval_runtime": 113.1028,
      "eval_samples_per_second": 35.773,
      "eval_steps_per_second": 2.237,
      "step": 250
    },
    {
      "epoch": 0.16789087093389296,
      "grad_norm": 3.5269249588508567,
      "learning_rate": 9.860464177621286e-06,
      "loss": 0.5043,
      "step": 260
    },
    {
      "epoch": 0.1743482121236581,
      "grad_norm": 2.877745150352999,
      "learning_rate": 9.832776460481162e-06,
      "loss": 0.4924,
      "step": 270
    },
    {
      "epoch": 0.1808055533134232,
      "grad_norm": 3.2647742800708204,
      "learning_rate": 9.802630777675528e-06,
      "loss": 0.5015,
      "step": 280
    },
    {
      "epoch": 0.1872628945031883,
      "grad_norm": 2.9015958802819957,
      "learning_rate": 9.770042461395705e-06,
      "loss": 0.5159,
      "step": 290
    },
    {
      "epoch": 0.19372023569295344,
      "grad_norm": 3.092413401049756,
      "learning_rate": 9.735028086164265e-06,
      "loss": 0.5107,
      "step": 300
    },
    {
      "epoch": 0.19372023569295344,
      "eval_loss": 0.44542330503463745,
      "eval_runtime": 113.164,
      "eval_samples_per_second": 35.753,
      "eval_steps_per_second": 2.236,
      "step": 300
    },
    {
      "epoch": 0.20017757688271853,
      "grad_norm": 3.03775644003821,
      "learning_rate": 9.697605460405162e-06,
      "loss": 0.4875,
      "step": 310
    },
    {
      "epoch": 0.20663491807248366,
      "grad_norm": 2.921380242765568,
      "learning_rate": 9.657793617386337e-06,
      "loss": 0.5043,
      "step": 320
    },
    {
      "epoch": 0.21309225926224876,
      "grad_norm": 4.56091451954007,
      "learning_rate": 9.615612805539305e-06,
      "loss": 0.5056,
      "step": 330
    },
    {
      "epoch": 0.21954960045201388,
      "grad_norm": 2.7567152985698105,
      "learning_rate": 9.571084478160763e-06,
      "loss": 0.4821,
      "step": 340
    },
    {
      "epoch": 0.226006941641779,
      "grad_norm": 3.03143158915137,
      "learning_rate": 9.524231282501362e-06,
      "loss": 0.4723,
      "step": 350
    },
    {
      "epoch": 0.226006941641779,
      "eval_loss": 0.43804243206977844,
      "eval_runtime": 113.0529,
      "eval_samples_per_second": 35.789,
      "eval_steps_per_second": 2.238,
      "step": 350
    },
    {
      "epoch": 0.2324642828315441,
      "grad_norm": 3.1198966349566826,
      "learning_rate": 9.475077048247243e-06,
      "loss": 0.5259,
      "step": 360
    },
    {
      "epoch": 0.23892162402130923,
      "grad_norm": 2.4780950251904987,
      "learning_rate": 9.423646775400186e-06,
      "loss": 0.4965,
      "step": 370
    },
    {
      "epoch": 0.24537896521107433,
      "grad_norm": 2.8932931613051345,
      "learning_rate": 9.369966621562525e-06,
      "loss": 0.4769,
      "step": 380
    },
    {
      "epoch": 0.2518363064008394,
      "grad_norm": 3.0372103770382775,
      "learning_rate": 9.314063888633312e-06,
      "loss": 0.4993,
      "step": 390
    },
    {
      "epoch": 0.2582936475906046,
      "grad_norm": 3.0478169324968243,
      "learning_rate": 9.255967008922475e-06,
      "loss": 0.4771,
      "step": 400
    },
    {
      "epoch": 0.2582936475906046,
      "eval_loss": 0.4323236048221588,
      "eval_runtime": 113.3031,
      "eval_samples_per_second": 35.71,
      "eval_steps_per_second": 2.233,
      "step": 400
    },
    {
      "epoch": 0.2647509887803697,
      "grad_norm": 2.6400318143361625,
      "learning_rate": 9.19570553069006e-06,
      "loss": 0.4848,
      "step": 410
    },
    {
      "epoch": 0.2712083299701348,
      "grad_norm": 2.7120828608055754,
      "learning_rate": 9.133310103117891e-06,
      "loss": 0.4608,
      "step": 420
    },
    {
      "epoch": 0.27766567115989993,
      "grad_norm": 2.8436594181157586,
      "learning_rate": 9.068812460721296e-06,
      "loss": 0.4828,
      "step": 430
    },
    {
      "epoch": 0.284123012349665,
      "grad_norm": 2.6044245067625194,
      "learning_rate": 9.002245407208831e-06,
      "loss": 0.4789,
      "step": 440
    },
    {
      "epoch": 0.2905803535394301,
      "grad_norm": 2.595066239026282,
      "learning_rate": 8.9336427987982e-06,
      "loss": 0.4835,
      "step": 450
    },
    {
      "epoch": 0.2905803535394301,
      "eval_loss": 0.4248904883861542,
      "eval_runtime": 113.1758,
      "eval_samples_per_second": 35.75,
      "eval_steps_per_second": 2.235,
      "step": 450
    },
    {
      "epoch": 0.2970376947291953,
      "grad_norm": 2.6150093306079016,
      "learning_rate": 8.863039526996891e-06,
      "loss": 0.4727,
      "step": 460
    },
    {
      "epoch": 0.3034950359189604,
      "grad_norm": 3.1956013155166176,
      "learning_rate": 8.790471500856229e-06,
      "loss": 0.4808,
      "step": 470
    },
    {
      "epoch": 0.3099523771087255,
      "grad_norm": 9.375771855867336,
      "learning_rate": 8.715975628707917e-06,
      "loss": 0.4775,
      "step": 480
    },
    {
      "epoch": 0.31640971829849057,
      "grad_norm": 2.849712917564074,
      "learning_rate": 8.63958979939235e-06,
      "loss": 0.4683,
      "step": 490
    },
    {
      "epoch": 0.3228670594882557,
      "grad_norm": 2.375130169309877,
      "learning_rate": 8.56135286298822e-06,
      "loss": 0.455,
      "step": 500
    },
    {
      "epoch": 0.3228670594882557,
      "eval_loss": 0.4204798936843872,
      "eval_runtime": 113.137,
      "eval_samples_per_second": 35.762,
      "eval_steps_per_second": 2.236,
      "step": 500
    },
    {
      "epoch": 0.3293244006780208,
      "grad_norm": 2.629529905003801,
      "learning_rate": 8.48130461105324e-06,
      "loss": 0.4523,
      "step": 510
    },
    {
      "epoch": 0.3357817418677859,
      "grad_norm": 2.470805475469757,
      "learning_rate": 8.399485756386038e-06,
      "loss": 0.4688,
      "step": 520
    },
    {
      "epoch": 0.3422390830575511,
      "grad_norm": 2.6526173274106792,
      "learning_rate": 8.315937912319493e-06,
      "loss": 0.4657,
      "step": 530
    },
    {
      "epoch": 0.3486964242473162,
      "grad_norm": 2.303760308318325,
      "learning_rate": 8.23070357155605e-06,
      "loss": 0.4456,
      "step": 540
    },
    {
      "epoch": 0.35515376543708127,
      "grad_norm": 2.507306087692525,
      "learning_rate": 8.143826084555809e-06,
      "loss": 0.4724,
      "step": 550
    },
    {
      "epoch": 0.35515376543708127,
      "eval_loss": 0.41447779536247253,
      "eval_runtime": 113.0047,
      "eval_samples_per_second": 35.804,
      "eval_steps_per_second": 2.239,
      "step": 550
    },
    {
      "epoch": 0.3616111066268464,
      "grad_norm": 2.769318157795312,
      "learning_rate": 8.055349637488336e-06,
      "loss": 0.4676,
      "step": 560
    },
    {
      "epoch": 0.3680684478166115,
      "grad_norm": 2.605597590172185,
      "learning_rate": 7.965319229759435e-06,
      "loss": 0.4615,
      "step": 570
    },
    {
      "epoch": 0.3745257890063766,
      "grad_norm": 2.620923577607227,
      "learning_rate": 7.873780651124302e-06,
      "loss": 0.4732,
      "step": 580
    },
    {
      "epoch": 0.3809831301961417,
      "grad_norm": 2.360832365799073,
      "learning_rate": 7.78078045839871e-06,
      "loss": 0.4663,
      "step": 590
    },
    {
      "epoch": 0.38744047138590687,
      "grad_norm": 2.382164893661759,
      "learning_rate": 7.686365951780071e-06,
      "loss": 0.4579,
      "step": 600
    },
    {
      "epoch": 0.38744047138590687,
      "eval_loss": 0.4005281925201416,
      "eval_runtime": 113.1546,
      "eval_samples_per_second": 35.756,
      "eval_steps_per_second": 2.236,
      "step": 600
    },
    {
      "epoch": 0.39389781257567197,
      "grad_norm": 2.365833998816792,
      "learning_rate": 7.590585150790388e-06,
      "loss": 0.4571,
      "step": 610
    },
    {
      "epoch": 0.40035515376543707,
      "grad_norm": 2.8970068498573407,
      "learning_rate": 7.493486769853383e-06,
      "loss": 0.4434,
      "step": 620
    },
    {
      "epoch": 0.4068124949552022,
      "grad_norm": 2.4583485406070964,
      "learning_rate": 7.39512019351818e-06,
      "loss": 0.4569,
      "step": 630
    },
    {
      "epoch": 0.4132698361449673,
      "grad_norm": 2.4315970598184267,
      "learning_rate": 7.295535451342166e-06,
      "loss": 0.436,
      "step": 640
    },
    {
      "epoch": 0.4197271773347324,
      "grad_norm": 2.28388365379662,
      "learning_rate": 7.194783192445795e-06,
      "loss": 0.4691,
      "step": 650
    },
    {
      "epoch": 0.4197271773347324,
      "eval_loss": 0.40489718317985535,
      "eval_runtime": 113.2039,
      "eval_samples_per_second": 35.741,
      "eval_steps_per_second": 2.235,
      "step": 650
    },
    {
      "epoch": 0.4261845185244975,
      "grad_norm": 2.3880709277885694,
      "learning_rate": 7.092914659752286e-06,
      "loss": 0.4739,
      "step": 660
    },
    {
      "epoch": 0.43264185971426267,
      "grad_norm": 2.489894784052563,
      "learning_rate": 6.989981663925301e-06,
      "loss": 0.4631,
      "step": 670
    },
    {
      "epoch": 0.43909920090402776,
      "grad_norm": 2.539339079118297,
      "learning_rate": 6.886036557017881e-06,
      "loss": 0.4451,
      "step": 680
    },
    {
      "epoch": 0.44555654209379286,
      "grad_norm": 2.478088709963733,
      "learning_rate": 6.781132205846019e-06,
      "loss": 0.4279,
      "step": 690
    },
    {
      "epoch": 0.452013883283558,
      "grad_norm": 2.4174050424059974,
      "learning_rate": 6.675321965100431e-06,
      "loss": 0.4405,
      "step": 700
    },
    {
      "epoch": 0.452013883283558,
      "eval_loss": 0.38827529549598694,
      "eval_runtime": 113.1619,
      "eval_samples_per_second": 35.754,
      "eval_steps_per_second": 2.236,
      "step": 700
    },
    {
      "epoch": 0.4584712244733231,
      "grad_norm": 2.014688109710611,
      "learning_rate": 6.568659650210184e-06,
      "loss": 0.443,
      "step": 710
    },
    {
      "epoch": 0.4649285656630882,
      "grad_norm": 2.2793036013581722,
      "learning_rate": 6.461199509972001e-06,
      "loss": 0.421,
      "step": 720
    },
    {
      "epoch": 0.47138590685285336,
      "grad_norm": 2.5589708675896072,
      "learning_rate": 6.35299619895914e-06,
      "loss": 0.4417,
      "step": 730
    },
    {
      "epoch": 0.47784324804261846,
      "grad_norm": 2.2481460698024396,
      "learning_rate": 6.244104749723916e-06,
      "loss": 0.427,
      "step": 740
    },
    {
      "epoch": 0.48430058923238356,
      "grad_norm": 2.3207297054640192,
      "learning_rate": 6.134580544807951e-06,
      "loss": 0.4443,
      "step": 750
    },
    {
      "epoch": 0.48430058923238356,
      "eval_loss": 0.38447996973991394,
      "eval_runtime": 112.9949,
      "eval_samples_per_second": 35.807,
      "eval_steps_per_second": 2.239,
      "step": 750
    },
    {
      "epoch": 0.49075793042214866,
      "grad_norm": 2.3996806799986814,
      "learning_rate": 6.024479288574448e-06,
      "loss": 0.4286,
      "step": 760
    },
    {
      "epoch": 0.4972152716119138,
      "grad_norm": 2.2155517916691534,
      "learning_rate": 5.9138569788767645e-06,
      "loss": 0.4301,
      "step": 770
    },
    {
      "epoch": 0.5036726128016789,
      "grad_norm": 2.328794137136517,
      "learning_rate": 5.802769878577729e-06,
      "loss": 0.4188,
      "step": 780
    },
    {
      "epoch": 0.510129953991444,
      "grad_norm": 2.38850296662767,
      "learning_rate": 5.691274486934166e-06,
      "loss": 0.4151,
      "step": 790
    },
    {
      "epoch": 0.5165872951812092,
      "grad_norm": 2.4014383644305917,
      "learning_rate": 5.579427510861194e-06,
      "loss": 0.4348,
      "step": 800
    },
    {
      "epoch": 0.5165872951812092,
      "eval_loss": 0.3787990212440491,
      "eval_runtime": 113.2045,
      "eval_samples_per_second": 35.741,
      "eval_steps_per_second": 2.235,
      "step": 800
    },
    {
      "epoch": 0.5230446363709742,
      "grad_norm": 2.4690922827935577,
      "learning_rate": 5.4672858360909e-06,
      "loss": 0.4273,
      "step": 810
    },
    {
      "epoch": 0.5295019775607394,
      "grad_norm": 2.279081423589604,
      "learning_rate": 5.35490649824008e-06,
      "loss": 0.4239,
      "step": 820
    },
    {
      "epoch": 0.5359593187505045,
      "grad_norm": 2.421246449319143,
      "learning_rate": 5.242346653801739e-06,
      "loss": 0.4189,
      "step": 830
    },
    {
      "epoch": 0.5424166599402696,
      "grad_norm": 2.2071350841498907,
      "learning_rate": 5.129663551075109e-06,
      "loss": 0.4227,
      "step": 840
    },
    {
      "epoch": 0.5488740011300347,
      "grad_norm": 2.4413807769668914,
      "learning_rate": 5.016914501048988e-06,
      "loss": 0.4153,
      "step": 850
    },
    {
      "epoch": 0.5488740011300347,
      "eval_loss": 0.36745426058769226,
      "eval_runtime": 113.0766,
      "eval_samples_per_second": 35.781,
      "eval_steps_per_second": 2.237,
      "step": 850
    },
    {
      "epoch": 0.5553313423197999,
      "grad_norm": 2.3371066482893568,
      "learning_rate": 4.904156848253184e-06,
      "loss": 0.4075,
      "step": 860
    },
    {
      "epoch": 0.5617886835095649,
      "grad_norm": 2.4368573269676745,
      "learning_rate": 4.791447941592896e-06,
      "loss": 0.4004,
      "step": 870
    },
    {
      "epoch": 0.56824602469933,
      "grad_norm": 2.4336186859151505,
      "learning_rate": 4.678845105180886e-06,
      "loss": 0.4114,
      "step": 880
    },
    {
      "epoch": 0.5747033658890952,
      "grad_norm": 2.3609720443049036,
      "learning_rate": 4.566405609182247e-06,
      "loss": 0.4128,
      "step": 890
    },
    {
      "epoch": 0.5811607070788603,
      "grad_norm": 2.4853090854239173,
      "learning_rate": 4.454186640686607e-06,
      "loss": 0.4123,
      "step": 900
    },
    {
      "epoch": 0.5811607070788603,
      "eval_loss": 0.36468505859375,
      "eval_runtime": 112.9737,
      "eval_samples_per_second": 35.814,
      "eval_steps_per_second": 2.239,
      "step": 900
    },
    {
      "epoch": 0.5876180482686254,
      "grad_norm": 2.766568871134695,
      "learning_rate": 4.3422452746226e-06,
      "loss": 0.3981,
      "step": 910
    },
    {
      "epoch": 0.5940753894583906,
      "grad_norm": 2.119770631202364,
      "learning_rate": 4.230638444729368e-06,
      "loss": 0.4198,
      "step": 920
    },
    {
      "epoch": 0.6005327306481556,
      "grad_norm": 2.599131418020882,
      "learning_rate": 4.1194229145998795e-06,
      "loss": 0.4138,
      "step": 930
    },
    {
      "epoch": 0.6069900718379208,
      "grad_norm": 2.401638444335127,
      "learning_rate": 4.008655248810787e-06,
      "loss": 0.4168,
      "step": 940
    },
    {
      "epoch": 0.6134474130276858,
      "grad_norm": 2.2202747583859224,
      "learning_rate": 3.898391784153494e-06,
      "loss": 0.3943,
      "step": 950
    },
    {
      "epoch": 0.6134474130276858,
      "eval_loss": 0.35901644825935364,
      "eval_runtime": 113.1417,
      "eval_samples_per_second": 35.76,
      "eval_steps_per_second": 2.236,
      "step": 950
    },
    {
      "epoch": 0.619904754217451,
      "grad_norm": 2.248321604412392,
      "learning_rate": 3.788688600981085e-06,
      "loss": 0.4115,
      "step": 960
    },
    {
      "epoch": 0.6263620954072161,
      "grad_norm": 2.211374108911103,
      "learning_rate": 3.679601494685679e-06,
      "loss": 0.3825,
      "step": 970
    },
    {
      "epoch": 0.6328194365969811,
      "grad_norm": 2.3000022394290154,
      "learning_rate": 3.571185947320712e-06,
      "loss": 0.3946,
      "step": 980
    },
    {
      "epoch": 0.6392767777867463,
      "grad_norm": 2.5678764569847976,
      "learning_rate": 3.4634970993825854e-06,
      "loss": 0.3959,
      "step": 990
    },
    {
      "epoch": 0.6457341189765115,
      "grad_norm": 2.102317352546757,
      "learning_rate": 3.356589721766034e-06,
      "loss": 0.4059,
      "step": 1000
    },
    {
      "epoch": 0.6457341189765115,
      "eval_loss": 0.34947654604911804,
      "eval_runtime": 113.1842,
      "eval_samples_per_second": 35.747,
      "eval_steps_per_second": 2.235,
      "step": 1000
    },
    {
      "epoch": 0.6521914601662765,
      "grad_norm": 2.136156581512006,
      "learning_rate": 3.250518187907469e-06,
      "loss": 0.3868,
      "step": 1010
    },
    {
      "epoch": 0.6586488013560416,
      "grad_norm": 2.2809574066745317,
      "learning_rate": 3.1453364461304795e-06,
      "loss": 0.3965,
      "step": 1020
    },
    {
      "epoch": 0.6651061425458068,
      "grad_norm": 2.245726944472294,
      "learning_rate": 3.0410979922075344e-06,
      "loss": 0.3896,
      "step": 1030
    },
    {
      "epoch": 0.6715634837355718,
      "grad_norm": 2.493934287857194,
      "learning_rate": 2.9378558421518645e-06,
      "loss": 0.3887,
      "step": 1040
    },
    {
      "epoch": 0.678020824925337,
      "grad_norm": 2.2918900021942465,
      "learning_rate": 2.835662505253344e-06,
      "loss": 0.3778,
      "step": 1050
    },
    {
      "epoch": 0.678020824925337,
      "eval_loss": 0.3436611592769623,
      "eval_runtime": 113.1253,
      "eval_samples_per_second": 35.766,
      "eval_steps_per_second": 2.236,
      "step": 1050
    },
    {
      "epoch": 0.6844781661151021,
      "grad_norm": 2.702991929059812,
      "learning_rate": 2.7345699573721014e-06,
      "loss": 0.3997,
      "step": 1060
    },
    {
      "epoch": 0.6909355073048672,
      "grad_norm": 2.239564635132244,
      "learning_rate": 2.634629614503428e-06,
      "loss": 0.3861,
      "step": 1070
    },
    {
      "epoch": 0.6973928484946323,
      "grad_norm": 2.0329067577466113,
      "learning_rate": 2.5358923066274354e-06,
      "loss": 0.3908,
      "step": 1080
    },
    {
      "epoch": 0.7038501896843975,
      "grad_norm": 2.37838080442691,
      "learning_rate": 2.4384082518567643e-06,
      "loss": 0.3982,
      "step": 1090
    },
    {
      "epoch": 0.7103075308741625,
      "grad_norm": 2.3024667656303706,
      "learning_rate": 2.3422270308954936e-06,
      "loss": 0.3734,
      "step": 1100
    },
    {
      "epoch": 0.7103075308741625,
      "eval_loss": 0.34295475482940674,
      "eval_runtime": 113.0131,
      "eval_samples_per_second": 35.801,
      "eval_steps_per_second": 2.239,
      "step": 1100
    },
    {
      "epoch": 0.7167648720639277,
      "grad_norm": 2.151969559566089,
      "learning_rate": 2.2473975618222304e-06,
      "loss": 0.3755,
      "step": 1110
    },
    {
      "epoch": 0.7232222132536928,
      "grad_norm": 2.1834116828550854,
      "learning_rate": 2.1539680752102217e-06,
      "loss": 0.3763,
      "step": 1120
    },
    {
      "epoch": 0.7296795544434579,
      "grad_norm": 2.067254081632634,
      "learning_rate": 2.0619860895971266e-06,
      "loss": 0.3613,
      "step": 1130
    },
    {
      "epoch": 0.736136895633223,
      "grad_norm": 2.523749347391733,
      "learning_rate": 1.9714983873169376e-06,
      "loss": 0.3959,
      "step": 1140
    },
    {
      "epoch": 0.7425942368229881,
      "grad_norm": 2.490370079710699,
      "learning_rate": 1.8825509907063328e-06,
      "loss": 0.3762,
      "step": 1150
    },
    {
      "epoch": 0.7425942368229881,
      "eval_loss": 0.33669936656951904,
      "eval_runtime": 113.1669,
      "eval_samples_per_second": 35.753,
      "eval_steps_per_second": 2.236,
      "step": 1150
    },
    {
      "epoch": 0.7490515780127532,
      "grad_norm": 2.4309754504331846,
      "learning_rate": 1.7951891386975684e-06,
      "loss": 0.3792,
      "step": 1160
    },
    {
      "epoch": 0.7555089192025184,
      "grad_norm": 2.195083204453119,
      "learning_rate": 1.7094572638098122e-06,
      "loss": 0.3873,
      "step": 1170
    },
    {
      "epoch": 0.7619662603922834,
      "grad_norm": 2.21873276040462,
      "learning_rate": 1.6253989695506207e-06,
      "loss": 0.3986,
      "step": 1180
    },
    {
      "epoch": 0.7684236015820486,
      "grad_norm": 2.297794634380802,
      "learning_rate": 1.5430570082390545e-06,
      "loss": 0.3586,
      "step": 1190
    },
    {
      "epoch": 0.7748809427718137,
      "grad_norm": 2.258999946182705,
      "learning_rate": 1.462473259261713e-06,
      "loss": 0.3576,
      "step": 1200
    },
    {
      "epoch": 0.7748809427718137,
      "eval_loss": 0.33270442485809326,
      "eval_runtime": 113.2544,
      "eval_samples_per_second": 35.725,
      "eval_steps_per_second": 2.234,
      "step": 1200
    },
    {
      "epoch": 0.7813382839615788,
      "grad_norm": 2.3026248483307303,
      "learning_rate": 1.3836887077727424e-06,
      "loss": 0.3803,
      "step": 1210
    },
    {
      "epoch": 0.7877956251513439,
      "grad_norm": 2.3646358198138495,
      "learning_rate": 1.3067434238486527e-06,
      "loss": 0.364,
      "step": 1220
    },
    {
      "epoch": 0.7942529663411091,
      "grad_norm": 2.2738060011027503,
      "learning_rate": 1.2316765421085513e-06,
      "loss": 0.3836,
      "step": 1230
    },
    {
      "epoch": 0.8007103075308741,
      "grad_norm": 2.4056682587117995,
      "learning_rate": 1.1585262418101468e-06,
      "loss": 0.3761,
      "step": 1240
    },
    {
      "epoch": 0.8071676487206393,
      "grad_norm": 2.202675669752468,
      "learning_rate": 1.0873297274316568e-06,
      "loss": 0.3794,
      "step": 1250
    },
    {
      "epoch": 0.8071676487206393,
      "eval_loss": 0.3294529318809509,
      "eval_runtime": 113.0196,
      "eval_samples_per_second": 35.799,
      "eval_steps_per_second": 2.239,
      "step": 1250
    },
    {
      "epoch": 0.8136249899104044,
      "grad_norm": 2.2706575776707982,
      "learning_rate": 1.0181232097494904e-06,
      "loss": 0.3682,
      "step": 1260
    },
    {
      "epoch": 0.8200823311001695,
      "grad_norm": 2.318144415900623,
      "learning_rate": 9.509418874213316e-07,
      "loss": 0.3622,
      "step": 1270
    },
    {
      "epoch": 0.8265396722899346,
      "grad_norm": 2.1145575343867566,
      "learning_rate": 8.858199290839859e-07,
      "loss": 0.3503,
      "step": 1280
    },
    {
      "epoch": 0.8329970134796998,
      "grad_norm": 2.334481394095981,
      "learning_rate": 8.22790455975106e-07,
      "loss": 0.3633,
      "step": 1290
    },
    {
      "epoch": 0.8394543546694648,
      "grad_norm": 2.2990060578208658,
      "learning_rate": 7.61885525087619e-07,
      "loss": 0.3695,
      "step": 1300
    },
    {
      "epoch": 0.8394543546694648,
      "eval_loss": 0.3264792859554291,
      "eval_runtime": 113.0484,
      "eval_samples_per_second": 35.79,
      "eval_steps_per_second": 2.238,
      "step": 1300
    },
    {
      "epoch": 0.84591169585923,
      "grad_norm": 2.122385925316686,
      "learning_rate": 7.031361128654402e-07,
      "loss": 0.3604,
      "step": 1310
    },
    {
      "epoch": 0.852369037048995,
      "grad_norm": 2.1344509720833016,
      "learning_rate": 6.465720994487523e-07,
      "loss": 0.3632,
      "step": 1320
    },
    {
      "epoch": 0.8588263782387602,
      "grad_norm": 2.405834646760456,
      "learning_rate": 5.922222534768707e-07,
      "loss": 0.3782,
      "step": 1330
    },
    {
      "epoch": 0.8652837194285253,
      "grad_norm": 2.2658573865850267,
      "learning_rate": 5.401142174564195e-07,
      "loss": 0.3666,
      "step": 1340
    },
    {
      "epoch": 0.8717410606182904,
      "grad_norm": 2.1236117581013074,
      "learning_rate": 4.902744937022658e-07,
      "loss": 0.3571,
      "step": 1350
    },
    {
      "epoch": 0.8717410606182904,
      "eval_loss": 0.3233107328414917,
      "eval_runtime": 113.0912,
      "eval_samples_per_second": 35.776,
      "eval_steps_per_second": 2.237,
      "step": 1350
    },
    {
      "epoch": 0.8781984018080555,
      "grad_norm": 2.3825754101184367,
      "learning_rate": 4.4272843085835515e-07,
      "loss": 0.351,
      "step": 1360
    },
    {
      "epoch": 0.8846557429978207,
      "grad_norm": 2.2027132795705455,
      "learning_rate": 3.9750021100531e-07,
      "loss": 0.357,
      "step": 1370
    },
    {
      "epoch": 0.8911130841875857,
      "grad_norm": 2.1229245134033956,
      "learning_rate": 3.546128373613472e-07,
      "loss": 0.3717,
      "step": 1380
    },
    {
      "epoch": 0.8975704253773509,
      "grad_norm": 2.0965567994973737,
      "learning_rate": 3.1408812258276575e-07,
      "loss": 0.3673,
      "step": 1390
    },
    {
      "epoch": 0.904027766567116,
      "grad_norm": 2.1314794438017888,
      "learning_rate": 2.7594667766995933e-07,
      "loss": 0.3655,
      "step": 1400
    },
    {
      "epoch": 0.904027766567116,
      "eval_loss": 0.32247301936149597,
      "eval_runtime": 113.1802,
      "eval_samples_per_second": 35.748,
      "eval_steps_per_second": 2.235,
      "step": 1400
    },
    {
      "epoch": 0.9104851077568811,
      "grad_norm": 2.0850201873294663,
      "learning_rate": 2.402079014845943e-07,
      "loss": 0.3675,
      "step": 1410
    },
    {
      "epoch": 0.9169424489466462,
      "grad_norm": 2.2935876803662234,
      "learning_rate": 2.068899708832911e-07,
      "loss": 0.3633,
      "step": 1420
    },
    {
      "epoch": 0.9233997901364114,
      "grad_norm": 2.181391325175196,
      "learning_rate": 1.7600983147280982e-07,
      "loss": 0.3581,
      "step": 1430
    },
    {
      "epoch": 0.9298571313261764,
      "grad_norm": 2.293868408013747,
      "learning_rate": 1.475831889914642e-07,
      "loss": 0.3462,
      "step": 1440
    },
    {
      "epoch": 0.9363144725159416,
      "grad_norm": 2.1846606725364253,
      "learning_rate": 1.2162450132113202e-07,
      "loss": 0.3801,
      "step": 1450
    },
    {
      "epoch": 0.9363144725159416,
      "eval_loss": 0.32113760709762573,
      "eval_runtime": 113.1982,
      "eval_samples_per_second": 35.743,
      "eval_steps_per_second": 2.235,
      "step": 1450
    },
    {
      "epoch": 0.9427718137057067,
      "grad_norm": 2.110156214362176,
      "learning_rate": 9.814697113392835e-08,
      "loss": 0.3451,
      "step": 1460
    },
    {
      "epoch": 0.9492291548954718,
      "grad_norm": 2.219030549712315,
      "learning_rate": 7.716253917728622e-08,
      "loss": 0.3579,
      "step": 1470
    },
    {
      "epoch": 0.9556864960852369,
      "grad_norm": 2.3426611489045035,
      "learning_rate": 5.8681878200850805e-08,
      "loss": 0.3655,
      "step": 1480
    },
    {
      "epoch": 0.962143837275002,
      "grad_norm": 2.209095014866135,
      "learning_rate": 4.27143875282876e-08,
      "loss": 0.3712,
      "step": 1490
    },
    {
      "epoch": 0.9686011784647671,
      "grad_norm": 2.165974141942845,
      "learning_rate": 2.9268188276757035e-08,
      "loss": 0.3704,
      "step": 1500
    },
    {
      "epoch": 0.9686011784647671,
      "eval_loss": 0.3208717405796051,
      "eval_runtime": 113.2653,
      "eval_samples_per_second": 35.721,
      "eval_steps_per_second": 2.234,
      "step": 1500
    },
    {
      "epoch": 0.9750585196545323,
      "grad_norm": 1.9902452057394107,
      "learning_rate": 1.8350119226483442e-08,
      "loss": 0.3645,
      "step": 1510
    },
    {
      "epoch": 0.9815158608442973,
      "grad_norm": 2.1757750886297016,
      "learning_rate": 9.965733342532925e-09,
      "loss": 0.3769,
      "step": 1520
    },
    {
      "epoch": 0.9879732020340625,
      "grad_norm": 2.2474809831164557,
      "learning_rate": 4.119294950558072e-09,
      "loss": 0.3462,
      "step": 1530
    },
    {
      "epoch": 0.9944305432238276,
      "grad_norm": 2.1927619111460768,
      "learning_rate": 8.137775679456505e-10,
      "loss": 0.3603,
      "step": 1540
    },
    {
      "epoch": 0.9995964161756397,
      "step": 1548,
      "total_flos": 836372543832064.0,
      "train_loss": 0.44547561819855247,
      "train_runtime": 23313.1683,
      "train_samples_per_second": 8.502,
      "train_steps_per_second": 0.066
    }
  ],
  "logging_steps": 10,
  "max_steps": 1548,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 836372543832064.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}