{
  "best_metric": 0.44520583748817444,
  "best_model_checkpoint": "saves/Orca/lora/checkpoint-3000",
  "epoch": 0.9998413957176844,
  "eval_steps": 500,
  "global_step": 3152,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0031720856463124504,
      "grad_norm": 0.8383375406265259,
      "learning_rate": 5.000000000000001e-07,
      "loss": 2.0081,
      "step": 10
    },
    {
      "epoch": 0.006344171292624901,
      "grad_norm": 0.9924185276031494,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 2.0141,
      "step": 20
    },
    {
      "epoch": 0.00951625693893735,
      "grad_norm": 1.7123312950134277,
      "learning_rate": 1.5e-06,
      "loss": 2.0134,
      "step": 30
    },
    {
      "epoch": 0.012688342585249802,
      "grad_norm": 1.6002644300460815,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.9886,
      "step": 40
    },
    {
      "epoch": 0.01586042823156225,
      "grad_norm": 1.4284940958023071,
      "learning_rate": 2.5e-06,
      "loss": 1.9963,
      "step": 50
    },
    {
      "epoch": 0.0190325138778747,
      "grad_norm": 1.9293030500411987,
      "learning_rate": 3e-06,
      "loss": 1.9858,
      "step": 60
    },
    {
      "epoch": 0.022204599524187154,
      "grad_norm": 1.8281705379486084,
      "learning_rate": 3.5000000000000004e-06,
      "loss": 1.9608,
      "step": 70
    },
    {
      "epoch": 0.025376685170499604,
      "grad_norm": 1.8404144048690796,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.9317,
      "step": 80
    },
    {
      "epoch": 0.028548770816812053,
      "grad_norm": 1.8080644607543945,
      "learning_rate": 4.5e-06,
      "loss": 1.8928,
      "step": 90
    },
    {
      "epoch": 0.0317208564631245,
      "grad_norm": 1.4789849519729614,
      "learning_rate": 5e-06,
      "loss": 1.8453,
      "step": 100
    },
    {
      "epoch": 0.034892942109436956,
      "grad_norm": 2.3675060272216797,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.7683,
      "step": 110
    },
    {
      "epoch": 0.0380650277557494,
      "grad_norm": 6.9262309074401855,
      "learning_rate": 6e-06,
      "loss": 1.6648,
      "step": 120
    },
    {
      "epoch": 0.041237113402061855,
      "grad_norm": 1.5342968702316284,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.5507,
      "step": 130
    },
    {
      "epoch": 0.04440919904837431,
      "grad_norm": 1.7138350009918213,
      "learning_rate": 7.000000000000001e-06,
      "loss": 1.4388,
      "step": 140
    },
    {
      "epoch": 0.047581284694686754,
      "grad_norm": 0.9868146181106567,
      "learning_rate": 7.5e-06,
      "loss": 1.3484,
      "step": 150
    },
    {
      "epoch": 0.05075337034099921,
      "grad_norm": 0.649817705154419,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.2671,
      "step": 160
    },
    {
      "epoch": 0.05392545598731166,
      "grad_norm": 0.5127522945404053,
      "learning_rate": 8.500000000000002e-06,
      "loss": 1.19,
      "step": 170
    },
    {
      "epoch": 0.057097541633624106,
      "grad_norm": 0.43639031052589417,
      "learning_rate": 9e-06,
      "loss": 1.1479,
      "step": 180
    },
    {
      "epoch": 0.06026962727993656,
      "grad_norm": 0.31427648663520813,
      "learning_rate": 9.5e-06,
      "loss": 1.0992,
      "step": 190
    },
    {
      "epoch": 0.063441712926249,
      "grad_norm": 0.2897365689277649,
      "learning_rate": 1e-05,
      "loss": 1.0673,
      "step": 200
    },
    {
      "epoch": 0.06661379857256146,
      "grad_norm": 0.43495845794677734,
      "learning_rate": 1.05e-05,
      "loss": 1.0399,
      "step": 210
    },
    {
      "epoch": 0.06978588421887391,
      "grad_norm": 0.2536582946777344,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 1.0043,
      "step": 220
    },
    {
      "epoch": 0.07295796986518636,
      "grad_norm": 0.3480685353279114,
      "learning_rate": 1.1500000000000002e-05,
      "loss": 0.9747,
      "step": 230
    },
    {
      "epoch": 0.0761300555114988,
      "grad_norm": 0.25048044323921204,
      "learning_rate": 1.2e-05,
      "loss": 0.9477,
      "step": 240
    },
    {
      "epoch": 0.07930214115781126,
      "grad_norm": 0.31881338357925415,
      "learning_rate": 1.25e-05,
      "loss": 0.9099,
      "step": 250
    },
    {
      "epoch": 0.08247422680412371,
      "grad_norm": 0.25372961163520813,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.8661,
      "step": 260
    },
    {
      "epoch": 0.08564631245043616,
      "grad_norm": 0.2892467975616455,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 0.8306,
      "step": 270
    },
    {
      "epoch": 0.08881839809674862,
      "grad_norm": 0.265289306640625,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.79,
      "step": 280
    },
    {
      "epoch": 0.09199048374306107,
      "grad_norm": 0.2384192794561386,
      "learning_rate": 1.45e-05,
      "loss": 0.7414,
      "step": 290
    },
    {
      "epoch": 0.09516256938937351,
      "grad_norm": 0.2318211942911148,
      "learning_rate": 1.5e-05,
      "loss": 0.6994,
      "step": 300
    },
    {
      "epoch": 0.09833465503568596,
      "grad_norm": 0.22584764659404755,
      "learning_rate": 1.55e-05,
      "loss": 0.6663,
      "step": 310
    },
    {
      "epoch": 0.10150674068199841,
      "grad_norm": 0.2281956821680069,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.6366,
      "step": 320
    },
    {
      "epoch": 0.10467882632831087,
      "grad_norm": 0.23853954672813416,
      "learning_rate": 1.65e-05,
      "loss": 0.6105,
      "step": 330
    },
    {
      "epoch": 0.10785091197462332,
      "grad_norm": 0.2249988317489624,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.5923,
      "step": 340
    },
    {
      "epoch": 0.11102299762093576,
      "grad_norm": 0.22325117886066437,
      "learning_rate": 1.75e-05,
      "loss": 0.5782,
      "step": 350
    },
    {
      "epoch": 0.11419508326724821,
      "grad_norm": 0.26995840668678284,
      "learning_rate": 1.8e-05,
      "loss": 0.5588,
      "step": 360
    },
    {
      "epoch": 0.11736716891356067,
      "grad_norm": 0.3128603994846344,
      "learning_rate": 1.85e-05,
      "loss": 0.5539,
      "step": 370
    },
    {
      "epoch": 0.12053925455987312,
      "grad_norm": 0.2401484102010727,
      "learning_rate": 1.9e-05,
      "loss": 0.5463,
      "step": 380
    },
    {
      "epoch": 0.12371134020618557,
      "grad_norm": 0.252087265253067,
      "learning_rate": 1.9500000000000003e-05,
      "loss": 0.5396,
      "step": 390
    },
    {
      "epoch": 0.126883425852498,
      "grad_norm": 0.3551071286201477,
      "learning_rate": 2e-05,
      "loss": 0.5293,
      "step": 400
    },
    {
      "epoch": 0.13005551149881048,
      "grad_norm": 0.2852734625339508,
      "learning_rate": 2.05e-05,
      "loss": 0.5327,
      "step": 410
    },
    {
      "epoch": 0.13322759714512292,
      "grad_norm": 0.25599339604377747,
      "learning_rate": 2.1e-05,
      "loss": 0.5215,
      "step": 420
    },
    {
      "epoch": 0.13639968279143536,
      "grad_norm": 0.2308124303817749,
      "learning_rate": 2.15e-05,
      "loss": 0.5179,
      "step": 430
    },
    {
      "epoch": 0.13957176843774782,
      "grad_norm": 0.28417715430259705,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.5122,
      "step": 440
    },
    {
      "epoch": 0.14274385408406026,
      "grad_norm": 0.32094892859458923,
      "learning_rate": 2.25e-05,
      "loss": 0.5078,
      "step": 450
    },
    {
      "epoch": 0.14591593973037273,
      "grad_norm": 0.28217512369155884,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.5064,
      "step": 460
    },
    {
      "epoch": 0.14908802537668517,
      "grad_norm": 0.3319534659385681,
      "learning_rate": 2.35e-05,
      "loss": 0.5056,
      "step": 470
    },
    {
      "epoch": 0.1522601110229976,
      "grad_norm": 0.35069146752357483,
      "learning_rate": 2.4e-05,
      "loss": 0.5001,
      "step": 480
    },
    {
      "epoch": 0.15543219666931007,
      "grad_norm": 0.2778172492980957,
      "learning_rate": 2.45e-05,
      "loss": 0.4994,
      "step": 490
    },
    {
      "epoch": 0.1586042823156225,
      "grad_norm": 0.2950364053249359,
      "learning_rate": 2.5e-05,
      "loss": 0.5009,
      "step": 500
    },
    {
      "epoch": 0.1586042823156225,
      "eval_loss": 0.49637770652770996,
      "eval_runtime": 23627.0877,
      "eval_samples_per_second": 0.044,
      "eval_steps_per_second": 0.022,
      "step": 500
    },
    {
      "epoch": 0.16177636796193498,
      "grad_norm": 0.26569539308547974,
      "learning_rate": 2.5500000000000003e-05,
      "loss": 0.4978,
      "step": 510
    },
    {
      "epoch": 0.16494845360824742,
      "grad_norm": 0.26948851346969604,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.4926,
      "step": 520
    },
    {
      "epoch": 0.1681205392545599,
      "grad_norm": 0.29083552956581116,
      "learning_rate": 2.6500000000000004e-05,
      "loss": 0.4893,
      "step": 530
    },
    {
      "epoch": 0.17129262490087233,
      "grad_norm": 0.2939367890357971,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.491,
      "step": 540
    },
    {
      "epoch": 0.17446471054718476,
      "grad_norm": 0.28195658326148987,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.4865,
      "step": 550
    },
    {
      "epoch": 0.17763679619349723,
      "grad_norm": 0.2827642858028412,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.4851,
      "step": 560
    },
    {
      "epoch": 0.18080888183980967,
      "grad_norm": 0.2715272605419159,
      "learning_rate": 2.8499999999999998e-05,
      "loss": 0.4833,
      "step": 570
    },
    {
      "epoch": 0.18398096748612214,
      "grad_norm": 0.30428314208984375,
      "learning_rate": 2.9e-05,
      "loss": 0.4834,
      "step": 580
    },
    {
      "epoch": 0.18715305313243458,
      "grad_norm": 0.2857949435710907,
      "learning_rate": 2.95e-05,
      "loss": 0.4796,
      "step": 590
    },
    {
      "epoch": 0.19032513877874702,
      "grad_norm": 0.302212655544281,
      "learning_rate": 3e-05,
      "loss": 0.4826,
      "step": 600
    },
    {
      "epoch": 0.19349722442505948,
      "grad_norm": 0.34079357981681824,
      "learning_rate": 3.05e-05,
      "loss": 0.4826,
      "step": 610
    },
    {
      "epoch": 0.19666931007137192,
      "grad_norm": 0.3239024877548218,
      "learning_rate": 3.1e-05,
      "loss": 0.4804,
      "step": 620
    },
    {
      "epoch": 0.1998413957176844,
      "grad_norm": 0.36531686782836914,
      "learning_rate": 3.15e-05,
      "loss": 0.4782,
      "step": 630
    },
    {
      "epoch": 0.20301348136399683,
      "grad_norm": 0.3582271337509155,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.4784,
      "step": 640
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 0.3250937759876251,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.4716,
      "step": 650
    },
    {
      "epoch": 0.20935765265662173,
      "grad_norm": 0.28494277596473694,
      "learning_rate": 3.3e-05,
      "loss": 0.4764,
      "step": 660
    },
    {
      "epoch": 0.21252973830293417,
      "grad_norm": 0.3156846761703491,
      "learning_rate": 3.35e-05,
      "loss": 0.4742,
      "step": 670
    },
    {
      "epoch": 0.21570182394924664,
      "grad_norm": 0.27053794264793396,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.4765,
      "step": 680
    },
    {
      "epoch": 0.21887390959555908,
      "grad_norm": 0.31487756967544556,
      "learning_rate": 3.45e-05,
      "loss": 0.4749,
      "step": 690
    },
    {
      "epoch": 0.22204599524187152,
      "grad_norm": 0.2739400267601013,
      "learning_rate": 3.5e-05,
      "loss": 0.4693,
      "step": 700
    },
    {
      "epoch": 0.22521808088818399,
      "grad_norm": 0.2924807369709015,
      "learning_rate": 3.55e-05,
      "loss": 0.4697,
      "step": 710
    },
    {
      "epoch": 0.22839016653449642,
      "grad_norm": 0.27241086959838867,
      "learning_rate": 3.6e-05,
      "loss": 0.4711,
      "step": 720
    },
    {
      "epoch": 0.2315622521808089,
      "grad_norm": 0.249298095703125,
      "learning_rate": 3.65e-05,
      "loss": 0.4748,
      "step": 730
    },
    {
      "epoch": 0.23473433782712133,
      "grad_norm": 0.33761996030807495,
      "learning_rate": 3.7e-05,
      "loss": 0.4692,
      "step": 740
    },
    {
      "epoch": 0.23790642347343377,
      "grad_norm": 0.3454744517803192,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.4691,
      "step": 750
    },
    {
      "epoch": 0.24107850911974624,
      "grad_norm": 0.3940749168395996,
      "learning_rate": 3.8e-05,
      "loss": 0.4694,
      "step": 760
    },
    {
      "epoch": 0.24425059476605868,
      "grad_norm": 0.2833056151866913,
      "learning_rate": 3.85e-05,
      "loss": 0.4651,
      "step": 770
    },
    {
      "epoch": 0.24742268041237114,
      "grad_norm": 0.3439841866493225,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.4718,
      "step": 780
    },
    {
      "epoch": 0.2505947660586836,
      "grad_norm": 0.28585824370384216,
      "learning_rate": 3.9500000000000005e-05,
      "loss": 0.4678,
      "step": 790
    },
    {
      "epoch": 0.253766851704996,
      "grad_norm": 0.2761208415031433,
      "learning_rate": 4e-05,
      "loss": 0.4672,
      "step": 800
    },
    {
      "epoch": 0.25693893735130846,
      "grad_norm": 0.287431538105011,
      "learning_rate": 4.05e-05,
      "loss": 0.4654,
      "step": 810
    },
    {
      "epoch": 0.26011102299762096,
      "grad_norm": 0.2529178261756897,
      "learning_rate": 4.1e-05,
      "loss": 0.4658,
      "step": 820
    },
    {
      "epoch": 0.2632831086439334,
      "grad_norm": 0.22234879434108734,
      "learning_rate": 4.15e-05,
      "loss": 0.467,
      "step": 830
    },
    {
      "epoch": 0.26645519429024583,
      "grad_norm": 0.2847014367580414,
      "learning_rate": 4.2e-05,
      "loss": 0.4643,
      "step": 840
    },
    {
      "epoch": 0.2696272799365583,
      "grad_norm": 0.27712419629096985,
      "learning_rate": 4.25e-05,
      "loss": 0.4604,
      "step": 850
    },
    {
      "epoch": 0.2727993655828707,
      "grad_norm": 0.24886064231395721,
      "learning_rate": 4.3e-05,
      "loss": 0.4657,
      "step": 860
    },
    {
      "epoch": 0.2759714512291832,
      "grad_norm": 0.3196752369403839,
      "learning_rate": 4.35e-05,
      "loss": 0.4652,
      "step": 870
    },
    {
      "epoch": 0.27914353687549565,
      "grad_norm": 0.27527421712875366,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.4634,
      "step": 880
    },
    {
      "epoch": 0.2823156225218081,
      "grad_norm": 0.26819467544555664,
      "learning_rate": 4.4500000000000004e-05,
      "loss": 0.4633,
      "step": 890
    },
    {
      "epoch": 0.2854877081681205,
      "grad_norm": 0.26196032762527466,
      "learning_rate": 4.5e-05,
      "loss": 0.4656,
      "step": 900
    },
    {
      "epoch": 0.28865979381443296,
      "grad_norm": 0.23337453603744507,
      "learning_rate": 4.55e-05,
      "loss": 0.4622,
      "step": 910
    },
    {
      "epoch": 0.29183187946074546,
      "grad_norm": 0.2958989143371582,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4627,
      "step": 920
    },
    {
      "epoch": 0.2950039651070579,
      "grad_norm": 0.23167885839939117,
      "learning_rate": 4.6500000000000005e-05,
      "loss": 0.4636,
      "step": 930
    },
    {
      "epoch": 0.29817605075337034,
      "grad_norm": 0.2608003318309784,
      "learning_rate": 4.7e-05,
      "loss": 0.4628,
      "step": 940
    },
    {
      "epoch": 0.3013481363996828,
      "grad_norm": 0.2806340456008911,
      "learning_rate": 4.75e-05,
      "loss": 0.4624,
      "step": 950
    },
    {
      "epoch": 0.3045202220459952,
      "grad_norm": 0.2206449806690216,
      "learning_rate": 4.8e-05,
      "loss": 0.4594,
      "step": 960
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.2883213460445404,
      "learning_rate": 4.85e-05,
      "loss": 0.4612,
      "step": 970
    },
    {
      "epoch": 0.31086439333862015,
      "grad_norm": 0.23365403711795807,
      "learning_rate": 4.9e-05,
      "loss": 0.463,
      "step": 980
    },
    {
      "epoch": 0.3140364789849326,
      "grad_norm": 0.28844788670539856,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 0.4612,
      "step": 990
    },
    {
      "epoch": 0.317208564631245,
      "grad_norm": 0.26974356174468994,
      "learning_rate": 5e-05,
      "loss": 0.4641,
      "step": 1000
    },
    {
      "epoch": 0.317208564631245,
      "eval_loss": 0.45910927653312683,
      "eval_runtime": 419.1696,
      "eval_samples_per_second": 2.457,
      "eval_steps_per_second": 1.229,
      "step": 1000
    },
    {
      "epoch": 0.3203806502775575,
      "grad_norm": 0.23146626353263855,
      "learning_rate": 4.9997336102492574e-05,
      "loss": 0.4583,
      "step": 1010
    },
    {
      "epoch": 0.32355273592386996,
      "grad_norm": 0.24651704728603363,
      "learning_rate": 4.9989344977678285e-05,
      "loss": 0.461,
      "step": 1020
    },
    {
      "epoch": 0.3267248215701824,
      "grad_norm": 0.2995850741863251,
      "learning_rate": 4.997602832856013e-05,
      "loss": 0.4583,
      "step": 1030
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 0.2886098325252533,
      "learning_rate": 4.995738899307319e-05,
      "loss": 0.4565,
      "step": 1040
    },
    {
      "epoch": 0.3330689928628073,
      "grad_norm": 0.2367076575756073,
      "learning_rate": 4.99334309434798e-05,
      "loss": 0.4598,
      "step": 1050
    },
    {
      "epoch": 0.3362410785091198,
      "grad_norm": 0.29130271077156067,
      "learning_rate": 4.990415928552305e-05,
      "loss": 0.4564,
      "step": 1060
    },
    {
      "epoch": 0.3394131641554322,
      "grad_norm": 0.23702391982078552,
      "learning_rate": 4.9869580257338685e-05,
      "loss": 0.4567,
      "step": 1070
    },
    {
      "epoch": 0.34258524980174465,
      "grad_norm": 0.24812078475952148,
      "learning_rate": 4.9829701228125655e-05,
      "loss": 0.4587,
      "step": 1080
    },
    {
      "epoch": 0.3457573354480571,
      "grad_norm": 0.2195880264043808,
      "learning_rate": 4.9784530696575684e-05,
      "loss": 0.4578,
      "step": 1090
    },
    {
      "epoch": 0.34892942109436953,
      "grad_norm": 0.2531008720397949,
      "learning_rate": 4.973407828906208e-05,
      "loss": 0.4551,
      "step": 1100
    },
    {
      "epoch": 0.352101506740682,
      "grad_norm": 0.2672252655029297,
      "learning_rate": 4.967835475758825e-05,
      "loss": 0.4596,
      "step": 1110
    },
    {
      "epoch": 0.35527359238699446,
      "grad_norm": 0.2659941613674164,
      "learning_rate": 4.961737197749633e-05,
      "loss": 0.4536,
      "step": 1120
    },
    {
      "epoch": 0.3584456780333069,
      "grad_norm": 0.2517182230949402,
      "learning_rate": 4.955114294493639e-05,
      "loss": 0.4579,
      "step": 1130
    },
    {
      "epoch": 0.36161776367961934,
      "grad_norm": 0.2654310166835785,
      "learning_rate": 4.947968177409681e-05,
      "loss": 0.4548,
      "step": 1140
    },
    {
      "epoch": 0.3647898493259318,
      "grad_norm": 0.36614441871643066,
      "learning_rate": 4.940300369419637e-05,
      "loss": 0.4575,
      "step": 1150
    },
    {
      "epoch": 0.3679619349722443,
      "grad_norm": 0.2583513855934143,
      "learning_rate": 4.9321125046238756e-05,
      "loss": 0.4598,
      "step": 1160
    },
    {
      "epoch": 0.3711340206185567,
      "grad_norm": 0.23981873691082,
      "learning_rate": 4.923406327953008e-05,
      "loss": 0.4546,
      "step": 1170
    },
    {
      "epoch": 0.37430610626486915,
      "grad_norm": 0.23798957467079163,
      "learning_rate": 4.9141836947960165e-05,
      "loss": 0.458,
      "step": 1180
    },
    {
      "epoch": 0.3774781919111816,
      "grad_norm": 0.2931083142757416,
      "learning_rate": 4.904446570604863e-05,
      "loss": 0.4543,
      "step": 1190
    },
    {
      "epoch": 0.38065027755749403,
      "grad_norm": 0.27312615513801575,
      "learning_rate": 4.8941970304756144e-05,
      "loss": 0.4552,
      "step": 1200
    },
    {
      "epoch": 0.3838223632038065,
      "grad_norm": 0.2659391164779663,
      "learning_rate": 4.883437258706224e-05,
      "loss": 0.4583,
      "step": 1210
    },
    {
      "epoch": 0.38699444885011897,
      "grad_norm": 0.23010990023612976,
      "learning_rate": 4.8721695483310275e-05,
      "loss": 0.4564,
      "step": 1220
    },
    {
      "epoch": 0.3901665344964314,
      "grad_norm": 0.2815942168235779,
      "learning_rate": 4.860396300632072e-05,
      "loss": 0.4548,
      "step": 1230
    },
    {
      "epoch": 0.39333862014274384,
      "grad_norm": 0.36561158299446106,
      "learning_rate": 4.848120024627372e-05,
      "loss": 0.4528,
      "step": 1240
    },
    {
      "epoch": 0.3965107057890563,
      "grad_norm": 0.25015613436698914,
      "learning_rate": 4.835343336536212e-05,
      "loss": 0.4567,
      "step": 1250
    },
    {
      "epoch": 0.3996827914353688,
      "grad_norm": 0.2373395711183548,
      "learning_rate": 4.822068959221598e-05,
      "loss": 0.4544,
      "step": 1260
    },
    {
      "epoch": 0.4028548770816812,
      "grad_norm": 0.26331648230552673,
      "learning_rate": 4.80829972160998e-05,
      "loss": 0.4532,
      "step": 1270
    },
    {
      "epoch": 0.40602696272799366,
      "grad_norm": 0.2622736692428589,
      "learning_rate": 4.794038558088378e-05,
      "loss": 0.4535,
      "step": 1280
    },
    {
      "epoch": 0.4091990483743061,
      "grad_norm": 0.1999824345111847,
      "learning_rate": 4.779288507879031e-05,
      "loss": 0.4568,
      "step": 1290
    },
    {
      "epoch": 0.41237113402061853,
      "grad_norm": 0.24161754548549652,
      "learning_rate": 4.764052714391695e-05,
      "loss": 0.4534,
      "step": 1300
    },
    {
      "epoch": 0.41554321966693103,
      "grad_norm": 0.2368980497121811,
      "learning_rate": 4.7483344245537545e-05,
      "loss": 0.4558,
      "step": 1310
    },
    {
      "epoch": 0.41871530531324347,
      "grad_norm": 0.1948440819978714,
      "learning_rate": 4.7321369881182584e-05,
      "loss": 0.4564,
      "step": 1320
    },
    {
      "epoch": 0.4218873909595559,
      "grad_norm": 0.2509821355342865,
      "learning_rate": 4.715463856950053e-05,
      "loss": 0.4558,
      "step": 1330
    },
    {
      "epoch": 0.42505947660586835,
      "grad_norm": 0.23150520026683807,
      "learning_rate": 4.698318584290141e-05,
      "loss": 0.4538,
      "step": 1340
    },
    {
      "epoch": 0.4282315622521808,
      "grad_norm": 0.22577635943889618,
      "learning_rate": 4.680704823998452e-05,
      "loss": 0.4571,
      "step": 1350
    },
    {
      "epoch": 0.4314036478984933,
      "grad_norm": 0.22882512211799622,
      "learning_rate": 4.6626263297751546e-05,
      "loss": 0.453,
      "step": 1360
    },
    {
      "epoch": 0.4345757335448057,
      "grad_norm": 0.1979285478591919,
      "learning_rate": 4.644086954360708e-05,
      "loss": 0.4542,
      "step": 1370
    },
    {
      "epoch": 0.43774781919111816,
      "grad_norm": 0.2554062008857727,
      "learning_rate": 4.625090648714786e-05,
      "loss": 0.4561,
      "step": 1380
    },
    {
      "epoch": 0.4409199048374306,
      "grad_norm": 0.2330465316772461,
      "learning_rate": 4.60564146117429e-05,
      "loss": 0.4525,
      "step": 1390
    },
    {
      "epoch": 0.44409199048374304,
      "grad_norm": 0.22269243001937866,
      "learning_rate": 4.585743536590599e-05,
      "loss": 0.4544,
      "step": 1400
    },
    {
      "epoch": 0.44726407613005553,
      "grad_norm": 0.2291986495256424,
      "learning_rate": 4.565401115446246e-05,
      "loss": 0.4548,
      "step": 1410
    },
    {
      "epoch": 0.45043616177636797,
      "grad_norm": 0.21675272285938263,
      "learning_rate": 4.5446185329512314e-05,
      "loss": 0.4518,
      "step": 1420
    },
    {
      "epoch": 0.4536082474226804,
      "grad_norm": 0.21207275986671448,
      "learning_rate": 4.5234002181191304e-05,
      "loss": 0.4511,
      "step": 1430
    },
    {
      "epoch": 0.45678033306899285,
      "grad_norm": 0.22509776055812836,
      "learning_rate": 4.5017506928232247e-05,
      "loss": 0.4518,
      "step": 1440
    },
    {
      "epoch": 0.4599524187153053,
      "grad_norm": 0.23429358005523682,
      "learning_rate": 4.4796745708328294e-05,
      "loss": 0.4556,
      "step": 1450
    },
    {
      "epoch": 0.4631245043616178,
      "grad_norm": 0.18599826097488403,
      "learning_rate": 4.457176556830054e-05,
      "loss": 0.4538,
      "step": 1460
    },
    {
      "epoch": 0.4662965900079302,
      "grad_norm": 0.2500051259994507,
      "learning_rate": 4.434261445407172e-05,
      "loss": 0.4538,
      "step": 1470
    },
    {
      "epoch": 0.46946867565424266,
      "grad_norm": 0.213278666138649,
      "learning_rate": 4.410934120044838e-05,
      "loss": 0.4545,
      "step": 1480
    },
    {
      "epoch": 0.4726407613005551,
      "grad_norm": 0.2551578879356384,
      "learning_rate": 4.387199552071366e-05,
      "loss": 0.4512,
      "step": 1490
    },
    {
      "epoch": 0.47581284694686754,
      "grad_norm": 0.23356303572654724,
      "learning_rate": 4.3630627996032706e-05,
      "loss": 0.4514,
      "step": 1500
    },
    {
      "epoch": 0.47581284694686754,
      "eval_loss": 0.4515945613384247,
      "eval_runtime": 293.5383,
      "eval_samples_per_second": 3.509,
      "eval_steps_per_second": 1.754,
      "step": 1500
    },
    {
      "epoch": 0.47898493259318004,
      "grad_norm": 0.22167843580245972,
      "learning_rate": 4.3385290064673316e-05,
      "loss": 0.4529,
      "step": 1510
    },
    {
      "epoch": 0.4821570182394925,
      "grad_norm": 0.2573811709880829,
      "learning_rate": 4.31360340110438e-05,
      "loss": 0.4517,
      "step": 1520
    },
    {
      "epoch": 0.4853291038858049,
      "grad_norm": 0.20118848979473114,
      "learning_rate": 4.288291295455055e-05,
      "loss": 0.4538,
      "step": 1530
    },
    {
      "epoch": 0.48850118953211735,
      "grad_norm": 0.24297507107257843,
      "learning_rate": 4.262598083827769e-05,
      "loss": 0.4519,
      "step": 1540
    },
    {
      "epoch": 0.4916732751784298,
      "grad_norm": 0.18671482801437378,
      "learning_rate": 4.236529241749114e-05,
      "loss": 0.4515,
      "step": 1550
    },
    {
      "epoch": 0.4948453608247423,
      "grad_norm": 0.20851927995681763,
      "learning_rate": 4.2100903247969644e-05,
      "loss": 0.4526,
      "step": 1560
    },
    {
      "epoch": 0.4980174464710547,
      "grad_norm": 0.23694059252738953,
      "learning_rate": 4.1832869674165204e-05,
      "loss": 0.4519,
      "step": 1570
    },
    {
      "epoch": 0.5011895321173672,
      "grad_norm": 0.3004760146141052,
      "learning_rate": 4.156124881719533e-05,
      "loss": 0.4493,
      "step": 1580
    },
    {
      "epoch": 0.5043616177636796,
      "grad_norm": 0.20880846679210663,
      "learning_rate": 4.1286098562669925e-05,
      "loss": 0.4508,
      "step": 1590
    },
    {
      "epoch": 0.507533703409992,
      "grad_norm": 0.2033112347126007,
      "learning_rate": 4.1007477548355185e-05,
      "loss": 0.4522,
      "step": 1600
    },
    {
      "epoch": 0.5107057890563045,
      "grad_norm": 0.24385987222194672,
      "learning_rate": 4.072544515167714e-05,
      "loss": 0.45,
      "step": 1610
    },
    {
      "epoch": 0.5138778747026169,
      "grad_norm": 0.19505774974822998,
      "learning_rate": 4.044006147706768e-05,
      "loss": 0.4528,
      "step": 1620
    },
    {
      "epoch": 0.5170499603489295,
      "grad_norm": 0.20980720221996307,
      "learning_rate": 4.0151387343155545e-05,
      "loss": 0.4512,
      "step": 1630
    },
    {
      "epoch": 0.5202220459952419,
      "grad_norm": 0.2582622766494751,
      "learning_rate": 3.985948426980521e-05,
      "loss": 0.452,
      "step": 1640
    },
    {
      "epoch": 0.5233941316415543,
      "grad_norm": 0.18902088701725006,
      "learning_rate": 3.956441446500624e-05,
      "loss": 0.4498,
      "step": 1650
    },
    {
      "epoch": 0.5265662172878668,
      "grad_norm": 0.21405866742134094,
      "learning_rate": 3.926624081161604e-05,
      "loss": 0.4462,
      "step": 1660
    },
    {
      "epoch": 0.5297383029341792,
      "grad_norm": 0.21340113878250122,
      "learning_rate": 3.896502685395876e-05,
      "loss": 0.449,
      "step": 1670
    },
    {
      "epoch": 0.5329103885804917,
      "grad_norm": 0.2296970933675766,
      "learning_rate": 3.8660836784283275e-05,
      "loss": 0.4513,
      "step": 1680
    },
    {
      "epoch": 0.5360824742268041,
      "grad_norm": 0.2143164724111557,
      "learning_rate": 3.835373542908308e-05,
      "loss": 0.4488,
      "step": 1690
    },
    {
      "epoch": 0.5392545598731165,
      "grad_norm": 0.1945250779390335,
      "learning_rate": 3.804378823528093e-05,
      "loss": 0.4529,
      "step": 1700
    },
    {
      "epoch": 0.542426645519429,
      "grad_norm": 0.20191755890846252,
      "learning_rate": 3.7731061256281394e-05,
      "loss": 0.4507,
      "step": 1710
    },
    {
      "epoch": 0.5455987311657414,
      "grad_norm": 0.21345514059066772,
      "learning_rate": 3.7415621137894056e-05,
      "loss": 0.4487,
      "step": 1720
    },
    {
      "epoch": 0.548770816812054,
      "grad_norm": 0.2062884271144867,
      "learning_rate": 3.709753510413052e-05,
      "loss": 0.4504,
      "step": 1730
    },
    {
      "epoch": 0.5519429024583664,
      "grad_norm": 0.20377105474472046,
      "learning_rate": 3.6776870942878196e-05,
      "loss": 0.4489,
      "step": 1740
    },
    {
      "epoch": 0.5551149881046789,
      "grad_norm": 0.19985787570476532,
      "learning_rate": 3.645369699145387e-05,
      "loss": 0.4497,
      "step": 1750
    },
    {
      "epoch": 0.5582870737509913,
      "grad_norm": 0.2308078110218048,
      "learning_rate": 3.6128082122040224e-05,
      "loss": 0.453,
      "step": 1760
    },
    {
      "epoch": 0.5614591593973037,
      "grad_norm": 0.19868697226047516,
      "learning_rate": 3.5800095727008395e-05,
      "loss": 0.4512,
      "step": 1770
    },
    {
      "epoch": 0.5646312450436162,
      "grad_norm": 0.20231026411056519,
      "learning_rate": 3.54698077041296e-05,
      "loss": 0.4482,
      "step": 1780
    },
    {
      "epoch": 0.5678033306899286,
      "grad_norm": 0.21103453636169434,
      "learning_rate": 3.51372884416791e-05,
      "loss": 0.452,
      "step": 1790
    },
    {
      "epoch": 0.570975416336241,
      "grad_norm": 0.21350406110286713,
      "learning_rate": 3.4802608803435646e-05,
      "loss": 0.4518,
      "step": 1800
    },
    {
      "epoch": 0.5741475019825535,
      "grad_norm": 0.2247006744146347,
      "learning_rate": 3.446584011357957e-05,
      "loss": 0.451,
      "step": 1810
    },
    {
      "epoch": 0.5773195876288659,
      "grad_norm": 0.20522333681583405,
      "learning_rate": 3.412705414149276e-05,
      "loss": 0.4491,
      "step": 1820
    },
    {
      "epoch": 0.5804916732751785,
      "grad_norm": 0.20249220728874207,
      "learning_rate": 3.3786323086463736e-05,
      "loss": 0.4508,
      "step": 1830
    },
    {
      "epoch": 0.5836637589214909,
      "grad_norm": 0.17756901681423187,
      "learning_rate": 3.3443719562301147e-05,
      "loss": 0.4493,
      "step": 1840
    },
    {
      "epoch": 0.5868358445678034,
      "grad_norm": 0.19246827065944672,
      "learning_rate": 3.309931658185892e-05,
      "loss": 0.4501,
      "step": 1850
    },
    {
      "epoch": 0.5900079302141158,
      "grad_norm": 0.20991584658622742,
      "learning_rate": 3.275318754147636e-05,
      "loss": 0.4478,
      "step": 1860
    },
    {
      "epoch": 0.5931800158604282,
      "grad_norm": 0.21498160064220428,
      "learning_rate": 3.240540620533649e-05,
      "loss": 0.4454,
      "step": 1870
    },
    {
      "epoch": 0.5963521015067407,
      "grad_norm": 0.199667289853096,
      "learning_rate": 3.205604668974607e-05,
      "loss": 0.4469,
      "step": 1880
    },
    {
      "epoch": 0.5995241871530531,
      "grad_norm": 0.21041718125343323,
      "learning_rate": 3.170518344734051e-05,
      "loss": 0.4517,
      "step": 1890
    },
    {
      "epoch": 0.6026962727993656,
      "grad_norm": 0.2041397988796234,
      "learning_rate": 3.135289125121718e-05,
      "loss": 0.4518,
      "step": 1900
    },
    {
      "epoch": 0.605868358445678,
      "grad_norm": 0.18860045075416565,
      "learning_rate": 3.0999245179000325e-05,
      "loss": 0.4511,
      "step": 1910
    },
    {
      "epoch": 0.6090404440919904,
      "grad_norm": 0.20879918336868286,
      "learning_rate": 3.064432059684117e-05,
      "loss": 0.4504,
      "step": 1920
    },
    {
      "epoch": 0.612212529738303,
      "grad_norm": 0.18158119916915894,
      "learning_rate": 3.0288193143356484e-05,
      "loss": 0.4501,
      "step": 1930
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.20405800640583038,
      "learning_rate": 2.9930938713509125e-05,
      "loss": 0.4478,
      "step": 1940
    },
    {
      "epoch": 0.6185567010309279,
      "grad_norm": 0.19368775188922882,
      "learning_rate": 2.9572633442433917e-05,
      "loss": 0.449,
      "step": 1950
    },
    {
      "epoch": 0.6217287866772403,
      "grad_norm": 0.19516830146312714,
      "learning_rate": 2.9213353689212337e-05,
      "loss": 0.4509,
      "step": 1960
    },
    {
      "epoch": 0.6249008723235527,
      "grad_norm": 0.2207994908094406,
      "learning_rate": 2.8853176020599504e-05,
      "loss": 0.4524,
      "step": 1970
    },
    {
      "epoch": 0.6280729579698652,
      "grad_norm": 0.20896515250205994,
      "learning_rate": 2.849217719470691e-05,
      "loss": 0.447,
      "step": 1980
    },
    {
      "epoch": 0.6312450436161776,
      "grad_norm": 0.19302915036678314,
      "learning_rate": 2.8130434144644364e-05,
      "loss": 0.4462,
      "step": 1990
    },
    {
      "epoch": 0.63441712926249,
      "grad_norm": 0.21044501662254333,
      "learning_rate": 2.776802396212461e-05,
      "loss": 0.4522,
      "step": 2000
    },
    {
      "epoch": 0.63441712926249,
      "eval_loss": 0.44816499948501587,
      "eval_runtime": 293.5621,
      "eval_samples_per_second": 3.509,
      "eval_steps_per_second": 1.754,
      "step": 2000
    },
    {
      "epoch": 0.6375892149088025,
      "grad_norm": 0.2075473517179489,
      "learning_rate": 2.7405023881034204e-05,
      "loss": 0.4487,
      "step": 2010
    },
    {
      "epoch": 0.640761300555115,
      "grad_norm": 0.1936168372631073,
      "learning_rate": 2.7041511260974028e-05,
      "loss": 0.4485,
      "step": 2020
    },
    {
      "epoch": 0.6439333862014275,
      "grad_norm": 0.2282039374113083,
      "learning_rate": 2.66775635707731e-05,
      "loss": 0.4478,
      "step": 2030
    },
    {
      "epoch": 0.6471054718477399,
      "grad_norm": 0.18652349710464478,
      "learning_rate": 2.6313258371978994e-05,
      "loss": 0.449,
      "step": 2040
    },
    {
      "epoch": 0.6502775574940524,
      "grad_norm": 0.20407763123512268,
      "learning_rate": 2.5948673302328587e-05,
      "loss": 0.4486,
      "step": 2050
    },
    {
      "epoch": 0.6534496431403648,
      "grad_norm": 0.22227470576763153,
      "learning_rate": 2.558388605920255e-05,
      "loss": 0.4506,
      "step": 2060
    },
    {
      "epoch": 0.6566217287866772,
      "grad_norm": 0.20888665318489075,
      "learning_rate": 2.5218974383067085e-05,
      "loss": 0.4487,
      "step": 2070
    },
    {
      "epoch": 0.6597938144329897,
      "grad_norm": 0.19797295331954956,
      "learning_rate": 2.4854016040906572e-05,
      "loss": 0.4521,
      "step": 2080
    },
    {
      "epoch": 0.6629659000793021,
      "grad_norm": 0.22293215990066528,
      "learning_rate": 2.4489088809650425e-05,
      "loss": 0.449,
      "step": 2090
    },
    {
      "epoch": 0.6661379857256146,
      "grad_norm": 0.2310066670179367,
      "learning_rate": 2.4124270459598007e-05,
      "loss": 0.4479,
      "step": 2100
    },
    {
      "epoch": 0.669310071371927,
      "grad_norm": 0.19995397329330444,
      "learning_rate": 2.375963873784478e-05,
      "loss": 0.4504,
      "step": 2110
    },
    {
      "epoch": 0.6724821570182395,
      "grad_norm": 0.18815317749977112,
      "learning_rate": 2.3395271351713515e-05,
      "loss": 0.4501,
      "step": 2120
    },
    {
      "epoch": 0.675654242664552,
      "grad_norm": 0.19861653447151184,
      "learning_rate": 2.303124595219395e-05,
      "loss": 0.4506,
      "step": 2130
    },
    {
      "epoch": 0.6788263283108644,
      "grad_norm": 0.17454959452152252,
      "learning_rate": 2.2667640117394442e-05,
      "loss": 0.4453,
      "step": 2140
    },
    {
      "epoch": 0.6819984139571769,
      "grad_norm": 0.19067569077014923,
      "learning_rate": 2.230453133600915e-05,
      "loss": 0.4466,
      "step": 2150
    },
    {
      "epoch": 0.6851704996034893,
      "grad_norm": 0.1979096531867981,
      "learning_rate": 2.1941996990804288e-05,
      "loss": 0.4479,
      "step": 2160
    },
    {
      "epoch": 0.6883425852498017,
      "grad_norm": 0.18306080996990204,
      "learning_rate": 2.1580114342126933e-05,
      "loss": 0.4482,
      "step": 2170
    },
    {
      "epoch": 0.6915146708961142,
      "grad_norm": 0.18473172187805176,
      "learning_rate": 2.1218960511439952e-05,
      "loss": 0.4468,
      "step": 2180
    },
    {
      "epoch": 0.6946867565424266,
      "grad_norm": 0.17530548572540283,
      "learning_rate": 2.0858612464886505e-05,
      "loss": 0.449,
      "step": 2190
    },
    {
      "epoch": 0.6978588421887391,
      "grad_norm": 0.1818351149559021,
      "learning_rate": 2.0499146996887618e-05,
      "loss": 0.4458,
      "step": 2200
    },
    {
      "epoch": 0.7010309278350515,
      "grad_norm": 0.19750525057315826,
      "learning_rate": 2.0140640713776438e-05,
      "loss": 0.4489,
      "step": 2210
    },
    {
      "epoch": 0.704203013481364,
      "grad_norm": 0.20060710608959198,
      "learning_rate": 1.978317001747248e-05,
      "loss": 0.4482,
      "step": 2220
    },
    {
      "epoch": 0.7073750991276765,
      "grad_norm": 0.19399641454219818,
      "learning_rate": 1.942681108919949e-05,
      "loss": 0.4494,
      "step": 2230
    },
    {
      "epoch": 0.7105471847739889,
      "grad_norm": 0.2124367654323578,
      "learning_rate": 1.9071639873250334e-05,
      "loss": 0.4491,
      "step": 2240
    },
    {
      "epoch": 0.7137192704203014,
      "grad_norm": 0.2064894288778305,
      "learning_rate": 1.871773206080236e-05,
      "loss": 0.4433,
      "step": 2250
    },
    {
      "epoch": 0.7168913560666138,
      "grad_norm": 0.20361967384815216,
      "learning_rate": 1.836516307378671e-05,
      "loss": 0.4482,
      "step": 2260
    },
    {
      "epoch": 0.7200634417129262,
      "grad_norm": 0.22753477096557617,
      "learning_rate": 1.801400804881507e-05,
      "loss": 0.4434,
      "step": 2270
    },
    {
      "epoch": 0.7232355273592387,
      "grad_norm": 0.2242439240217209,
      "learning_rate": 1.766434182116708e-05,
      "loss": 0.4487,
      "step": 2280
    },
    {
      "epoch": 0.7264076130055511,
      "grad_norm": 0.1731296181678772,
      "learning_rate": 1.7316238908842127e-05,
      "loss": 0.4456,
      "step": 2290
    },
    {
      "epoch": 0.7295796986518636,
      "grad_norm": 0.19372820854187012,
      "learning_rate": 1.6969773496678647e-05,
      "loss": 0.446,
      "step": 2300
    },
    {
      "epoch": 0.732751784298176,
      "grad_norm": 0.21722117066383362,
      "learning_rate": 1.662501942054447e-05,
      "loss": 0.4465,
      "step": 2310
    },
    {
      "epoch": 0.7359238699444886,
      "grad_norm": 0.24181324243545532,
      "learning_rate": 1.6282050151601518e-05,
      "loss": 0.4496,
      "step": 2320
    },
    {
      "epoch": 0.739095955590801,
      "grad_norm": 0.2091435343027115,
      "learning_rate": 1.594093878064825e-05,
      "loss": 0.4469,
      "step": 2330
    },
    {
      "epoch": 0.7422680412371134,
      "grad_norm": 0.19474650919437408,
      "learning_rate": 1.5601758002543137e-05,
      "loss": 0.4465,
      "step": 2340
    },
    {
      "epoch": 0.7454401268834259,
      "grad_norm": 0.21081089973449707,
      "learning_rate": 1.5264580100712507e-05,
      "loss": 0.4449,
      "step": 2350
    },
    {
      "epoch": 0.7486122125297383,
      "grad_norm": 0.21220536530017853,
      "learning_rate": 1.4929476931746167e-05,
      "loss": 0.4461,
      "step": 2360
    },
    {
      "epoch": 0.7517842981760507,
      "grad_norm": 0.19944216310977936,
      "learning_rate": 1.4596519910083825e-05,
      "loss": 0.4476,
      "step": 2370
    },
    {
      "epoch": 0.7549563838223632,
      "grad_norm": 0.19958476722240448,
      "learning_rate": 1.4265779992795893e-05,
      "loss": 0.4467,
      "step": 2380
    },
    {
      "epoch": 0.7581284694686756,
      "grad_norm": 0.22247722744941711,
      "learning_rate": 1.3937327664461672e-05,
      "loss": 0.4427,
      "step": 2390
    },
    {
      "epoch": 0.7613005551149881,
      "grad_norm": 0.19884702563285828,
      "learning_rate": 1.361123292214826e-05,
      "loss": 0.4498,
      "step": 2400
    },
    {
      "epoch": 0.7644726407613005,
      "grad_norm": 0.17867441475391388,
      "learning_rate": 1.3287565260493357e-05,
      "loss": 0.4484,
      "step": 2410
    },
    {
      "epoch": 0.767644726407613,
      "grad_norm": 0.17917264997959137,
      "learning_rate": 1.2966393656895134e-05,
      "loss": 0.4439,
      "step": 2420
    },
    {
      "epoch": 0.7708168120539255,
      "grad_norm": 0.24791833758354187,
      "learning_rate": 1.2647786556812332e-05,
      "loss": 0.4463,
      "step": 2430
    },
    {
      "epoch": 0.7739888977002379,
      "grad_norm": 0.1971379816532135,
      "learning_rate": 1.2331811859177722e-05,
      "loss": 0.4475,
      "step": 2440
    },
    {
      "epoch": 0.7771609833465504,
      "grad_norm": 0.19460198283195496,
      "learning_rate": 1.2018536901928079e-05,
      "loss": 0.4468,
      "step": 2450
    },
    {
      "epoch": 0.7803330689928628,
      "grad_norm": 0.21018314361572266,
      "learning_rate": 1.1708028447653613e-05,
      "loss": 0.4489,
      "step": 2460
    },
    {
      "epoch": 0.7835051546391752,
      "grad_norm": 0.18419425189495087,
      "learning_rate": 1.1400352669370115e-05,
      "loss": 0.4469,
      "step": 2470
    },
    {
      "epoch": 0.7866772402854877,
      "grad_norm": 0.192842036485672,
      "learning_rate": 1.1095575136416695e-05,
      "loss": 0.4457,
      "step": 2480
    },
    {
      "epoch": 0.7898493259318001,
      "grad_norm": 0.21617653965950012,
      "learning_rate": 1.0793760800482179e-05,
      "loss": 0.4511,
      "step": 2490
    },
    {
      "epoch": 0.7930214115781126,
      "grad_norm": 0.1891343593597412,
      "learning_rate": 1.0494973981763146e-05,
      "loss": 0.4436,
      "step": 2500
    },
    {
      "epoch": 0.7930214115781126,
      "eval_loss": 0.44588717818260193,
      "eval_runtime": 293.564,
      "eval_samples_per_second": 3.509,
      "eval_steps_per_second": 1.754,
      "step": 2500
    },
    {
      "epoch": 0.796193497224425,
      "grad_norm": 0.193357452750206,
      "learning_rate": 1.0199278355256522e-05,
      "loss": 0.4436,
      "step": 2510
    },
    {
      "epoch": 0.7993655828707376,
      "grad_norm": 0.19498059153556824,
      "learning_rate": 9.906736937189697e-06,
      "loss": 0.4457,
      "step": 2520
    },
    {
      "epoch": 0.80253766851705,
      "grad_norm": 0.2086293250322342,
      "learning_rate": 9.61741207159104e-06,
      "loss": 0.4488,
      "step": 2530
    },
    {
      "epoch": 0.8057097541633624,
      "grad_norm": 0.1872304081916809,
      "learning_rate": 9.331365417003601e-06,
      "loss": 0.4436,
      "step": 2540
    },
    {
      "epoch": 0.8088818398096749,
      "grad_norm": 0.196652352809906,
      "learning_rate": 9.048657933345e-06,
      "loss": 0.4448,
      "step": 2550
    },
    {
      "epoch": 0.8120539254559873,
      "grad_norm": 0.21112217009067535,
      "learning_rate": 8.769349868916118e-06,
      "loss": 0.4481,
      "step": 2560
    },
    {
      "epoch": 0.8152260111022998,
      "grad_norm": 0.19768798351287842,
      "learning_rate": 8.493500747561488e-06,
      "loss": 0.4493,
      "step": 2570
    },
    {
      "epoch": 0.8183980967486122,
      "grad_norm": 0.19622166454792023,
      "learning_rate": 8.221169355984052e-06,
      "loss": 0.4448,
      "step": 2580
    },
    {
      "epoch": 0.8215701823949246,
      "grad_norm": 0.2044159471988678,
      "learning_rate": 7.952413731217025e-06,
      "loss": 0.4472,
      "step": 2590
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 0.206452876329422,
      "learning_rate": 7.687291148255527e-06,
      "loss": 0.4452,
      "step": 2600
    },
    {
      "epoch": 0.8279143536875495,
      "grad_norm": 0.20858535170555115,
      "learning_rate": 7.425858107850578e-06,
      "loss": 0.4419,
      "step": 2610
    },
    {
      "epoch": 0.8310864393338621,
      "grad_norm": 0.18986332416534424,
      "learning_rate": 7.168170324468171e-06,
      "loss": 0.449,
      "step": 2620
    },
    {
      "epoch": 0.8342585249801745,
      "grad_norm": 0.21051757037639618,
      "learning_rate": 6.9142827144158066e-06,
      "loss": 0.4468,
      "step": 2630
    },
    {
      "epoch": 0.8374306106264869,
      "grad_norm": 0.20453451573848724,
      "learning_rate": 6.66424938413921e-06,
      "loss": 0.4471,
      "step": 2640
    },
    {
      "epoch": 0.8406026962727994,
      "grad_norm": 0.18574683368206024,
      "learning_rate": 6.418123618691607e-06,
      "loss": 0.4473,
      "step": 2650
    },
    {
      "epoch": 0.8437747819191118,
      "grad_norm": 0.2309403419494629,
      "learning_rate": 6.175957870378043e-06,
      "loss": 0.4477,
      "step": 2660
    },
    {
      "epoch": 0.8469468675654243,
      "grad_norm": 0.19485293328762054,
      "learning_rate": 5.937803747577186e-06,
      "loss": 0.4443,
      "step": 2670
    },
    {
      "epoch": 0.8501189532117367,
      "grad_norm": 0.17677082121372223,
      "learning_rate": 5.7037120037429645e-06,
      "loss": 0.4501,
      "step": 2680
    },
    {
      "epoch": 0.8532910388580491,
      "grad_norm": 0.18835794925689697,
      "learning_rate": 5.473732526588407e-06,
      "loss": 0.4451,
      "step": 2690
    },
    {
      "epoch": 0.8564631245043616,
      "grad_norm": 0.2117832899093628,
      "learning_rate": 5.2479143274539954e-06,
      "loss": 0.4464,
      "step": 2700
    },
    {
      "epoch": 0.8596352101506741,
      "grad_norm": 0.18720006942749023,
      "learning_rate": 5.026305530862749e-06,
      "loss": 0.4484,
      "step": 2710
    },
    {
      "epoch": 0.8628072957969866,
      "grad_norm": 0.17910288274288177,
      "learning_rate": 4.80895336426434e-06,
      "loss": 0.4457,
      "step": 2720
    },
    {
      "epoch": 0.865979381443299,
      "grad_norm": 0.1965160220861435,
      "learning_rate": 4.595904147970356e-06,
      "loss": 0.4455,
      "step": 2730
    },
    {
      "epoch": 0.8691514670896114,
      "grad_norm": 0.16718564927577972,
      "learning_rate": 4.3872032852828955e-06,
      "loss": 0.4455,
      "step": 2740
    },
    {
      "epoch": 0.8723235527359239,
      "grad_norm": 0.19451048970222473,
      "learning_rate": 4.182895252818589e-06,
      "loss": 0.4454,
      "step": 2750
    },
    {
      "epoch": 0.8754956383822363,
      "grad_norm": 0.2160135954618454,
      "learning_rate": 3.983023591030113e-06,
      "loss": 0.4462,
      "step": 2760
    },
    {
      "epoch": 0.8786677240285488,
      "grad_norm": 0.1994054913520813,
      "learning_rate": 3.7876308949271995e-06,
      "loss": 0.4455,
      "step": 2770
    },
    {
      "epoch": 0.8818398096748612,
      "grad_norm": 0.20801877975463867,
      "learning_rate": 3.5967588049991313e-06,
      "loss": 0.4465,
      "step": 2780
    },
    {
      "epoch": 0.8850118953211736,
      "grad_norm": 0.20264749228954315,
      "learning_rate": 3.410447998340688e-06,
      "loss": 0.4451,
      "step": 2790
    },
    {
      "epoch": 0.8881839809674861,
      "grad_norm": 0.173824280500412,
      "learning_rate": 3.2287381799833428e-06,
      "loss": 0.4453,
      "step": 2800
    },
    {
      "epoch": 0.8913560666137986,
      "grad_norm": 0.18437042832374573,
      "learning_rate": 3.051668074433667e-06,
      "loss": 0.449,
      "step": 2810
    },
    {
      "epoch": 0.8945281522601111,
      "grad_norm": 0.17420920729637146,
      "learning_rate": 2.8792754174206905e-06,
      "loss": 0.4453,
      "step": 2820
    },
    {
      "epoch": 0.8977002379064235,
      "grad_norm": 0.18584947288036346,
      "learning_rate": 2.7115969478539562e-06,
      "loss": 0.4446,
      "step": 2830
    },
    {
      "epoch": 0.9008723235527359,
      "grad_norm": 0.18006384372711182,
      "learning_rate": 2.5486683999940335e-06,
      "loss": 0.4449,
      "step": 2840
    },
    {
      "epoch": 0.9040444091990484,
      "grad_norm": 0.19664862751960754,
      "learning_rate": 2.3905244958371263e-06,
      "loss": 0.4463,
      "step": 2850
    },
    {
      "epoch": 0.9072164948453608,
      "grad_norm": 0.20159141719341278,
      "learning_rate": 2.2371989377154013e-06,
      "loss": 0.444,
      "step": 2860
    },
    {
      "epoch": 0.9103885804916733,
      "grad_norm": 0.23178640007972717,
      "learning_rate": 2.088724401114625e-06,
      "loss": 0.4472,
      "step": 2870
    },
    {
      "epoch": 0.9135606661379857,
      "grad_norm": 0.18047955632209778,
      "learning_rate": 1.9451325277106413e-06,
      "loss": 0.4489,
      "step": 2880
    },
    {
      "epoch": 0.9167327517842981,
      "grad_norm": 0.18607738614082336,
      "learning_rate": 1.8064539186261387e-06,
      "loss": 0.4489,
      "step": 2890
    },
    {
      "epoch": 0.9199048374306106,
      "grad_norm": 0.19285809993743896,
      "learning_rate": 1.6727181279092036e-06,
      "loss": 0.4448,
      "step": 2900
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.1946035474538803,
      "learning_rate": 1.5439536562349976e-06,
      "loss": 0.4474,
      "step": 2910
    },
    {
      "epoch": 0.9262490087232356,
      "grad_norm": 0.1947924941778183,
      "learning_rate": 1.4201879448319355e-06,
      "loss": 0.4465,
      "step": 2920
    },
    {
      "epoch": 0.929421094369548,
      "grad_norm": 0.17740875482559204,
      "learning_rate": 1.301447369633621e-06,
      "loss": 0.4475,
      "step": 2930
    },
    {
      "epoch": 0.9325931800158604,
      "grad_norm": 0.18701794743537903,
      "learning_rate": 1.187757235657841e-06,
      "loss": 0.4488,
      "step": 2940
    },
    {
      "epoch": 0.9357652656621729,
      "grad_norm": 0.19531536102294922,
      "learning_rate": 1.0791417716137565e-06,
      "loss": 0.4458,
      "step": 2950
    },
    {
      "epoch": 0.9389373513084853,
      "grad_norm": 0.17441792786121368,
      "learning_rate": 9.756241247384807e-07,
      "loss": 0.4426,
      "step": 2960
    },
    {
      "epoch": 0.9421094369547978,
      "grad_norm": 0.20463122427463531,
      "learning_rate": 8.772263558641386e-07,
      "loss": 0.4485,
      "step": 2970
    },
    {
      "epoch": 0.9452815226011102,
      "grad_norm": 0.18601520359516144,
      "learning_rate": 7.839694347164223e-07,
      "loss": 0.4421,
      "step": 2980
    },
    {
      "epoch": 0.9484536082474226,
      "grad_norm": 0.18209311366081238,
      "learning_rate": 6.958732354457209e-07,
      "loss": 0.4499,
      "step": 2990
    },
    {
      "epoch": 0.9516256938937351,
      "grad_norm": 0.1675465852022171,
      "learning_rate": 6.129565323916813e-07,
      "loss": 0.4463,
      "step": 3000
    },
    {
      "epoch": 0.9516256938937351,
      "eval_loss": 0.44520583748817444,
      "eval_runtime": 293.5596,
      "eval_samples_per_second": 3.509,
      "eval_steps_per_second": 1.754,
      "step": 3000
    },
    {
      "epoch": 0.9547977795400476,
      "grad_norm": 0.21267291903495789,
      "learning_rate": 5.352369960821946e-07,
      "loss": 0.4428,
      "step": 3010
    },
    {
      "epoch": 0.9579698651863601,
      "grad_norm": 0.19138512015342712,
      "learning_rate": 4.627311894675856e-07,
      "loss": 0.446,
      "step": 3020
    },
    {
      "epoch": 0.9611419508326725,
      "grad_norm": 0.19043755531311035,
      "learning_rate": 3.954545643908514e-07,
      "loss": 0.4446,
      "step": 3030
    },
    {
      "epoch": 0.964314036478985,
      "grad_norm": 0.1850534826517105,
      "learning_rate": 3.3342145829469983e-07,
      "loss": 0.4435,
      "step": 3040
    },
    {
      "epoch": 0.9674861221252974,
      "grad_norm": 0.21564887464046478,
      "learning_rate": 2.7664509116607506e-07,
      "loss": 0.4451,
      "step": 3050
    },
    {
      "epoch": 0.9706582077716098,
      "grad_norm": 0.1970747411251068,
      "learning_rate": 2.251375627187996e-07,
      "loss": 0.4465,
      "step": 3060
    },
    {
      "epoch": 0.9738302934179223,
      "grad_norm": 0.1968993842601776,
      "learning_rate": 1.789098498150066e-07,
      "loss": 0.4479,
      "step": 3070
    },
    {
      "epoch": 0.9770023790642347,
      "grad_norm": 0.18383654952049255,
      "learning_rate": 1.3797180412583322e-07,
      "loss": 0.4472,
      "step": 3080
    },
    {
      "epoch": 0.9801744647105471,
      "grad_norm": 0.18183210492134094,
      "learning_rate": 1.0233215003190577e-07,
      "loss": 0.4484,
      "step": 3090
    },
    {
      "epoch": 0.9833465503568596,
      "grad_norm": 0.18428853154182434,
      "learning_rate": 7.199848276408238e-08,
      "loss": 0.4445,
      "step": 3100
    },
    {
      "epoch": 0.9865186360031721,
      "grad_norm": 0.21674080193042755,
      "learning_rate": 4.6977266784811736e-08,
      "loss": 0.4465,
      "step": 3110
    },
    {
      "epoch": 0.9896907216494846,
      "grad_norm": 0.2023499608039856,
      "learning_rate": 2.7273834410485034e-08,
      "loss": 0.4448,
      "step": 3120
    },
    {
      "epoch": 0.992862807295797,
      "grad_norm": 0.19151511788368225,
      "learning_rate": 1.2892384675056158e-08,
      "loss": 0.4461,
      "step": 3130
    },
    {
      "epoch": 0.9960348929421095,
      "grad_norm": 0.16607537865638733,
      "learning_rate": 3.8359824351685836e-09,
      "loss": 0.4453,
      "step": 3140
    },
    {
      "epoch": 0.9992069785884219,
      "grad_norm": 0.21372340619564056,
      "learning_rate": 1.0655771701395534e-10,
      "loss": 0.448,
      "step": 3150
    },
    {
      "epoch": 0.9998413957176844,
      "step": 3152,
      "total_flos": 2.0499162760961065e+18,
      "train_loss": 0.5496831483344742,
      "train_runtime": 71646.9274,
      "train_samples_per_second": 0.704,
      "train_steps_per_second": 0.044
    }
  ],
  "logging_steps": 10,
  "max_steps": 3152,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2.0499162760961065e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}