{
  "best_metric": 0.971350908279419,
  "best_model_checkpoint": "./fine_tuned_models/health/Llama-3.1-8B-Instruct/checkpoint-2500",
  "epoch": 2.5669135802469136,
  "eval_steps": 100,
  "global_step": 2600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.019753086419753086,
      "grad_norm": 1.1160211563110352,
      "learning_rate": 1.9736842105263157e-06,
      "loss": 2.0627,
      "step": 20
    },
    {
      "epoch": 0.03950617283950617,
      "grad_norm": 1.0905706882476807,
      "learning_rate": 3.9473684210526315e-06,
      "loss": 2.0246,
      "step": 40
    },
    {
      "epoch": 0.05925925925925926,
      "grad_norm": 0.9707510471343994,
      "learning_rate": 5.921052631578948e-06,
      "loss": 1.9243,
      "step": 60
    },
    {
      "epoch": 0.07901234567901234,
      "grad_norm": 1.0528677701950073,
      "learning_rate": 7.894736842105263e-06,
      "loss": 1.6172,
      "step": 80
    },
    {
      "epoch": 0.09876543209876543,
      "grad_norm": 0.7791302800178528,
      "learning_rate": 9.868421052631579e-06,
      "loss": 1.3386,
      "step": 100
    },
    {
      "epoch": 0.09876543209876543,
      "eval_loss": 1.2289670705795288,
      "eval_runtime": 404.6369,
      "eval_samples_per_second": 4.448,
      "eval_steps_per_second": 1.112,
      "step": 100
    },
    {
      "epoch": 0.11851851851851852,
      "grad_norm": 0.6323535442352295,
      "learning_rate": 1.1842105263157895e-05,
      "loss": 1.1669,
      "step": 120
    },
    {
      "epoch": 0.1382716049382716,
      "grad_norm": 0.6384269595146179,
      "learning_rate": 1.3815789473684211e-05,
      "loss": 1.1764,
      "step": 140
    },
    {
      "epoch": 0.1580246913580247,
      "grad_norm": 0.6977917551994324,
      "learning_rate": 1.5789473684210526e-05,
      "loss": 1.1308,
      "step": 160
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 0.6905924081802368,
      "learning_rate": 1.7763157894736842e-05,
      "loss": 1.1127,
      "step": 180
    },
    {
      "epoch": 0.19753086419753085,
      "grad_norm": 0.7211576700210571,
      "learning_rate": 1.9736842105263158e-05,
      "loss": 1.1204,
      "step": 200
    },
    {
      "epoch": 0.19753086419753085,
      "eval_loss": 1.10204017162323,
      "eval_runtime": 403.0411,
      "eval_samples_per_second": 4.466,
      "eval_steps_per_second": 1.117,
      "step": 200
    },
    {
      "epoch": 0.21728395061728395,
      "grad_norm": 0.7831864953041077,
      "learning_rate": 2.1710526315789474e-05,
      "loss": 1.0875,
      "step": 220
    },
    {
      "epoch": 0.23703703703703705,
      "grad_norm": 0.8442113399505615,
      "learning_rate": 2.368421052631579e-05,
      "loss": 1.0855,
      "step": 240
    },
    {
      "epoch": 0.25679012345679014,
      "grad_norm": 0.8692957162857056,
      "learning_rate": 2.5657894736842107e-05,
      "loss": 1.0797,
      "step": 260
    },
    {
      "epoch": 0.2765432098765432,
      "grad_norm": 0.8785028457641602,
      "learning_rate": 2.7631578947368423e-05,
      "loss": 1.1142,
      "step": 280
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 0.8870799541473389,
      "learning_rate": 2.9605263157894735e-05,
      "loss": 1.0707,
      "step": 300
    },
    {
      "epoch": 0.2962962962962963,
      "eval_loss": 1.074904203414917,
      "eval_runtime": 403.0664,
      "eval_samples_per_second": 4.466,
      "eval_steps_per_second": 1.116,
      "step": 300
    },
    {
      "epoch": 0.3160493827160494,
      "grad_norm": 0.9721025228500366,
      "learning_rate": 2.9997461206762896e-05,
      "loss": 1.0471,
      "step": 320
    },
    {
      "epoch": 0.3358024691358025,
      "grad_norm": 0.9425962567329407,
      "learning_rate": 2.998714883211034e-05,
      "loss": 1.0614,
      "step": 340
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 1.0400207042694092,
      "learning_rate": 2.9968909651554313e-05,
      "loss": 1.0386,
      "step": 360
    },
    {
      "epoch": 0.37530864197530867,
      "grad_norm": 1.033016562461853,
      "learning_rate": 2.994275331192262e-05,
      "loss": 1.0591,
      "step": 380
    },
    {
      "epoch": 0.3950617283950617,
      "grad_norm": 1.202338695526123,
      "learning_rate": 2.9908693647482852e-05,
      "loss": 1.081,
      "step": 400
    },
    {
      "epoch": 0.3950617283950617,
      "eval_loss": 1.05765700340271,
      "eval_runtime": 402.7051,
      "eval_samples_per_second": 4.47,
      "eval_steps_per_second": 1.117,
      "step": 400
    },
    {
      "epoch": 0.4148148148148148,
      "grad_norm": 1.0287129878997803,
      "learning_rate": 2.9866748672625354e-05,
      "loss": 1.1013,
      "step": 420
    },
    {
      "epoch": 0.4345679012345679,
      "grad_norm": 1.114308476448059,
      "learning_rate": 2.9816940572335275e-05,
      "loss": 1.0684,
      "step": 440
    },
    {
      "epoch": 0.454320987654321,
      "grad_norm": 1.1122310161590576,
      "learning_rate": 2.9759295690458793e-05,
      "loss": 1.0675,
      "step": 460
    },
    {
      "epoch": 0.4740740740740741,
      "grad_norm": 1.0627741813659668,
      "learning_rate": 2.9693844515769658e-05,
      "loss": 1.0327,
      "step": 480
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 1.0726803541183472,
      "learning_rate": 2.9620621665843482e-05,
      "loss": 1.0413,
      "step": 500
    },
    {
      "epoch": 0.49382716049382713,
      "eval_loss": 1.0474056005477905,
      "eval_runtime": 403.0908,
      "eval_samples_per_second": 4.465,
      "eval_steps_per_second": 1.116,
      "step": 500
    },
    {
      "epoch": 0.5135802469135803,
      "grad_norm": 1.2839173078536987,
      "learning_rate": 2.953966586874827e-05,
      "loss": 1.0507,
      "step": 520
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 1.0455371141433716,
      "learning_rate": 2.9451019942560854e-05,
      "loss": 1.0484,
      "step": 540
    },
    {
      "epoch": 0.5530864197530864,
      "grad_norm": 1.0756592750549316,
      "learning_rate": 2.9354730772720152e-05,
      "loss": 1.0612,
      "step": 560
    },
    {
      "epoch": 0.5728395061728395,
      "grad_norm": 1.1546605825424194,
      "learning_rate": 2.9250849287229115e-05,
      "loss": 1.0338,
      "step": 580
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 1.062068223953247,
      "learning_rate": 2.9139430429718557e-05,
      "loss": 1.05,
      "step": 600
    },
    {
      "epoch": 0.5925925925925926,
      "eval_loss": 1.0370533466339111,
      "eval_runtime": 402.7813,
      "eval_samples_per_second": 4.469,
      "eval_steps_per_second": 1.117,
      "step": 600
    },
    {
      "epoch": 0.6123456790123457,
      "grad_norm": 1.0931472778320312,
      "learning_rate": 2.90205331303871e-05,
      "loss": 1.0282,
      "step": 620
    },
    {
      "epoch": 0.6320987654320988,
      "grad_norm": 1.1314401626586914,
      "learning_rate": 2.889422027483259e-05,
      "loss": 1.0185,
      "step": 640
    },
    {
      "epoch": 0.6518518518518519,
      "grad_norm": 1.1147228479385376,
      "learning_rate": 2.876055867079146e-05,
      "loss": 1.0552,
      "step": 660
    },
    {
      "epoch": 0.671604938271605,
      "grad_norm": 1.2712368965148926,
      "learning_rate": 2.8619619012803674e-05,
      "loss": 1.0241,
      "step": 680
    },
    {
      "epoch": 0.691358024691358,
      "grad_norm": 1.1172308921813965,
      "learning_rate": 2.847147584482188e-05,
      "loss": 1.018,
      "step": 700
    },
    {
      "epoch": 0.691358024691358,
      "eval_loss": 1.0282899141311646,
      "eval_runtime": 403.0768,
      "eval_samples_per_second": 4.466,
      "eval_steps_per_second": 1.116,
      "step": 700
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 1.1631219387054443,
      "learning_rate": 2.8316207520784622e-05,
      "loss": 1.037,
      "step": 720
    },
    {
      "epoch": 0.7308641975308642,
      "grad_norm": 1.0679621696472168,
      "learning_rate": 2.81538961631744e-05,
      "loss": 0.9998,
      "step": 740
    },
    {
      "epoch": 0.7506172839506173,
      "grad_norm": 1.098596215248108,
      "learning_rate": 2.7984627619582507e-05,
      "loss": 1.0164,
      "step": 760
    },
    {
      "epoch": 0.7703703703703704,
      "grad_norm": 1.1393965482711792,
      "learning_rate": 2.7808491417303672e-05,
      "loss": 1.0407,
      "step": 780
    },
    {
      "epoch": 0.7901234567901234,
      "grad_norm": 1.34330153465271,
      "learning_rate": 2.7625580715984452e-05,
      "loss": 1.0432,
      "step": 800
    },
    {
      "epoch": 0.7901234567901234,
      "eval_loss": 1.0187339782714844,
      "eval_runtime": 402.8401,
      "eval_samples_per_second": 4.468,
      "eval_steps_per_second": 1.117,
      "step": 800
    },
    {
      "epoch": 0.8098765432098766,
      "grad_norm": 1.1482850313186646,
      "learning_rate": 2.7435992258350444e-05,
      "loss": 1.0214,
      "step": 820
    },
    {
      "epoch": 0.8296296296296296,
      "grad_norm": 1.0984679460525513,
      "learning_rate": 2.7239826319038413e-05,
      "loss": 1.0104,
      "step": 840
    },
    {
      "epoch": 0.8493827160493828,
      "grad_norm": 1.1276164054870605,
      "learning_rate": 2.703718665156033e-05,
      "loss": 1.0151,
      "step": 860
    },
    {
      "epoch": 0.8691358024691358,
      "grad_norm": 1.1957815885543823,
      "learning_rate": 2.6828180433427437e-05,
      "loss": 1.0254,
      "step": 880
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.1888211965560913,
      "learning_rate": 2.6612918209463323e-05,
      "loss": 1.0056,
      "step": 900
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.0113353729248047,
      "eval_runtime": 403.2634,
      "eval_samples_per_second": 4.464,
      "eval_steps_per_second": 1.116,
      "step": 900
    },
    {
      "epoch": 0.908641975308642,
      "grad_norm": 1.1143324375152588,
      "learning_rate": 2.6391513833336013e-05,
      "loss": 1.038,
      "step": 920
    },
    {
      "epoch": 0.928395061728395,
      "grad_norm": 1.1294788122177124,
      "learning_rate": 2.616408440733997e-05,
      "loss": 1.0346,
      "step": 940
    },
    {
      "epoch": 0.9481481481481482,
      "grad_norm": 1.1774975061416626,
      "learning_rate": 2.5930750220459893e-05,
      "loss": 1.004,
      "step": 960
    },
    {
      "epoch": 0.9679012345679012,
      "grad_norm": 1.1493123769760132,
      "learning_rate": 2.569163468474906e-05,
      "loss": 0.9974,
      "step": 980
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 1.2377318143844604,
      "learning_rate": 2.5446864270055825e-05,
      "loss": 0.9987,
      "step": 1000
    },
    {
      "epoch": 0.9876543209876543,
      "eval_loss": 1.0076097249984741,
      "eval_runtime": 402.948,
      "eval_samples_per_second": 4.467,
      "eval_steps_per_second": 1.117,
      "step": 1000
    },
    {
      "epoch": 1.0069135802469136,
      "grad_norm": 1.225941777229309,
      "learning_rate": 2.5196568437132855e-05,
      "loss": 0.9939,
      "step": 1020
    },
    {
      "epoch": 1.0266666666666666,
      "grad_norm": 1.240697979927063,
      "learning_rate": 2.4940879569164453e-05,
      "loss": 0.9806,
      "step": 1040
    },
    {
      "epoch": 1.0464197530864197,
      "grad_norm": 1.2726426124572754,
      "learning_rate": 2.467993290174819e-05,
      "loss": 0.9758,
      "step": 1060
    },
    {
      "epoch": 1.066172839506173,
      "grad_norm": 1.3213516473770142,
      "learning_rate": 2.4413866451367836e-05,
      "loss": 0.9822,
      "step": 1080
    },
    {
      "epoch": 1.085925925925926,
      "grad_norm": 1.2369683980941772,
      "learning_rate": 2.414282094239555e-05,
      "loss": 0.9467,
      "step": 1100
    },
    {
      "epoch": 1.085925925925926,
      "eval_loss": 1.0044056177139282,
      "eval_runtime": 402.972,
      "eval_samples_per_second": 4.467,
      "eval_steps_per_second": 1.117,
      "step": 1100
    },
    {
      "epoch": 1.105679012345679,
      "grad_norm": 1.1692577600479126,
      "learning_rate": 2.386693973266173e-05,
      "loss": 0.9586,
      "step": 1120
    },
    {
      "epoch": 1.125432098765432,
      "grad_norm": 1.1815599203109741,
      "learning_rate": 2.3586368737632124e-05,
      "loss": 0.9711,
      "step": 1140
    },
    {
      "epoch": 1.145185185185185,
      "grad_norm": 1.2383053302764893,
      "learning_rate": 2.3301256353232112e-05,
      "loss": 0.9668,
      "step": 1160
    },
    {
      "epoch": 1.1649382716049383,
      "grad_norm": 1.4748594760894775,
      "learning_rate": 2.3011753377359084e-05,
      "loss": 0.977,
      "step": 1180
    },
    {
      "epoch": 1.1846913580246914,
      "grad_norm": 1.3367607593536377,
      "learning_rate": 2.2718012930124407e-05,
      "loss": 0.9461,
      "step": 1200
    },
    {
      "epoch": 1.1846913580246914,
      "eval_loss": 0.9985090494155884,
      "eval_runtime": 402.6419,
      "eval_samples_per_second": 4.47,
      "eval_steps_per_second": 1.118,
      "step": 1200
    },
    {
      "epoch": 1.2044444444444444,
      "grad_norm": 1.3120101690292358,
      "learning_rate": 2.2420190372867144e-05,
      "loss": 0.9728,
      "step": 1220
    },
    {
      "epoch": 1.2241975308641975,
      "grad_norm": 1.4451560974121094,
      "learning_rate": 2.2118443225982365e-05,
      "loss": 0.962,
      "step": 1240
    },
    {
      "epoch": 1.2439506172839505,
      "grad_norm": 1.2705769538879395,
      "learning_rate": 2.1812931085607537e-05,
      "loss": 0.9425,
      "step": 1260
    },
    {
      "epoch": 1.2637037037037038,
      "grad_norm": 1.3269940614700317,
      "learning_rate": 2.1503815539211034e-05,
      "loss": 0.9519,
      "step": 1280
    },
    {
      "epoch": 1.2834567901234568,
      "grad_norm": 1.2251733541488647,
      "learning_rate": 2.119126008012741e-05,
      "loss": 0.9671,
      "step": 1300
    },
    {
      "epoch": 1.2834567901234568,
      "eval_loss": 0.9977313280105591,
      "eval_runtime": 402.6952,
      "eval_samples_per_second": 4.47,
      "eval_steps_per_second": 1.117,
      "step": 1300
    },
    {
      "epoch": 1.3032098765432099,
      "grad_norm": 1.3956445455551147,
      "learning_rate": 2.0875430021084665e-05,
      "loss": 0.9742,
      "step": 1320
    },
    {
      "epoch": 1.322962962962963,
      "grad_norm": 1.2852596044540405,
      "learning_rate": 2.0556492406769176e-05,
      "loss": 0.9635,
      "step": 1340
    },
    {
      "epoch": 1.342716049382716,
      "grad_norm": 1.3903273344039917,
      "learning_rate": 2.0234615925474648e-05,
      "loss": 0.9622,
      "step": 1360
    },
    {
      "epoch": 1.3624691358024692,
      "grad_norm": 1.457078456878662,
      "learning_rate": 1.9909970819881708e-05,
      "loss": 0.9833,
      "step": 1380
    },
    {
      "epoch": 1.3822222222222222,
      "grad_norm": 1.2986372709274292,
      "learning_rate": 1.9582728797015385e-05,
      "loss": 0.9495,
      "step": 1400
    },
    {
      "epoch": 1.3822222222222222,
      "eval_loss": 0.9925122857093811,
      "eval_runtime": 402.5025,
      "eval_samples_per_second": 4.472,
      "eval_steps_per_second": 1.118,
      "step": 1400
    },
    {
      "epoch": 1.4019753086419753,
      "grad_norm": 1.342713475227356,
      "learning_rate": 1.9253062937428073e-05,
      "loss": 0.9609,
      "step": 1420
    },
    {
      "epoch": 1.4217283950617283,
      "grad_norm": 1.2822914123535156,
      "learning_rate": 1.892114760365605e-05,
      "loss": 0.9658,
      "step": 1440
    },
    {
      "epoch": 1.4414814814814814,
      "grad_norm": 1.331895351409912,
      "learning_rate": 1.8587158347997932e-05,
      "loss": 0.9769,
      "step": 1460
    },
    {
      "epoch": 1.4612345679012346,
      "grad_norm": 1.3801404237747192,
      "learning_rate": 1.8251271819663863e-05,
      "loss": 0.98,
      "step": 1480
    },
    {
      "epoch": 1.4809876543209877,
      "grad_norm": 1.337031364440918,
      "learning_rate": 1.7913665671344526e-05,
      "loss": 0.9716,
      "step": 1500
    },
    {
      "epoch": 1.4809876543209877,
      "eval_loss": 0.9894294738769531,
      "eval_runtime": 402.6051,
      "eval_samples_per_second": 4.471,
      "eval_steps_per_second": 1.118,
      "step": 1500
    },
    {
      "epoch": 1.5007407407407407,
      "grad_norm": 1.551279067993164,
      "learning_rate": 1.7574518465249413e-05,
      "loss": 0.9772,
      "step": 1520
    },
    {
      "epoch": 1.520493827160494,
      "grad_norm": 1.3463525772094727,
      "learning_rate": 1.7234009578664066e-05,
      "loss": 0.9477,
      "step": 1540
    },
    {
      "epoch": 1.5402469135802468,
      "grad_norm": 1.3253049850463867,
      "learning_rate": 1.6892319109076183e-05,
      "loss": 0.9641,
      "step": 1560
    },
    {
      "epoch": 1.56,
      "grad_norm": 1.4334008693695068,
      "learning_rate": 1.654962777892081e-05,
      "loss": 0.9584,
      "step": 1580
    },
    {
      "epoch": 1.579753086419753,
      "grad_norm": 1.412903904914856,
      "learning_rate": 1.6206116839995033e-05,
      "loss": 0.9779,
      "step": 1600
    },
    {
      "epoch": 1.579753086419753,
      "eval_loss": 0.9848657846450806,
      "eval_runtime": 402.5392,
      "eval_samples_per_second": 4.472,
      "eval_steps_per_second": 1.118,
      "step": 1600
    },
    {
      "epoch": 1.5995061728395061,
      "grad_norm": 1.251621961593628,
      "learning_rate": 1.5861967977592642e-05,
      "loss": 0.942,
      "step": 1620
    },
    {
      "epoch": 1.6192592592592594,
      "grad_norm": 1.3814697265625,
      "learning_rate": 1.5517363214409528e-05,
      "loss": 0.9469,
      "step": 1640
    },
    {
      "epoch": 1.6390123456790122,
      "grad_norm": 1.388342022895813,
      "learning_rate": 1.5172484814270635e-05,
      "loss": 0.9716,
      "step": 1660
    },
    {
      "epoch": 1.6587654320987655,
      "grad_norm": 1.318664312362671,
      "learning_rate": 1.4827515185729368e-05,
      "loss": 0.98,
      "step": 1680
    },
    {
      "epoch": 1.6785185185185185,
      "grad_norm": 1.364259958267212,
      "learning_rate": 1.4482636785590474e-05,
      "loss": 0.9628,
      "step": 1700
    },
    {
      "epoch": 1.6785185185185185,
      "eval_loss": 0.9820418357849121,
      "eval_runtime": 402.744,
      "eval_samples_per_second": 4.469,
      "eval_steps_per_second": 1.117,
      "step": 1700
    },
    {
      "epoch": 1.6982716049382716,
      "grad_norm": 1.346475601196289,
      "learning_rate": 1.413803202240736e-05,
      "loss": 0.995,
      "step": 1720
    },
    {
      "epoch": 1.7180246913580248,
      "grad_norm": 1.350071907043457,
      "learning_rate": 1.3793883160004971e-05,
      "loss": 0.9792,
      "step": 1740
    },
    {
      "epoch": 1.7377777777777776,
      "grad_norm": 1.3962644338607788,
      "learning_rate": 1.3450372221079193e-05,
      "loss": 0.9885,
      "step": 1760
    },
    {
      "epoch": 1.757530864197531,
      "grad_norm": 1.4300681352615356,
      "learning_rate": 1.3107680890923821e-05,
      "loss": 0.9433,
      "step": 1780
    },
    {
      "epoch": 1.777283950617284,
      "grad_norm": 1.604736566543579,
      "learning_rate": 1.2765990421335937e-05,
      "loss": 0.966,
      "step": 1800
    },
    {
      "epoch": 1.777283950617284,
      "eval_loss": 0.9798233509063721,
      "eval_runtime": 402.6875,
      "eval_samples_per_second": 4.47,
      "eval_steps_per_second": 1.117,
      "step": 1800
    },
    {
      "epoch": 1.797037037037037,
      "grad_norm": 1.3228657245635986,
      "learning_rate": 1.2425481534750591e-05,
      "loss": 1.0006,
      "step": 1820
    },
    {
      "epoch": 1.8167901234567903,
      "grad_norm": 1.5004771947860718,
      "learning_rate": 1.2086334328655478e-05,
      "loss": 0.9622,
      "step": 1840
    },
    {
      "epoch": 1.836543209876543,
      "grad_norm": 1.412369966506958,
      "learning_rate": 1.1748728180336137e-05,
      "loss": 0.9764,
      "step": 1860
    },
    {
      "epoch": 1.8562962962962963,
      "grad_norm": 1.4217263460159302,
      "learning_rate": 1.141284165200207e-05,
      "loss": 0.9396,
      "step": 1880
    },
    {
      "epoch": 1.8760493827160494,
      "grad_norm": 1.3855938911437988,
      "learning_rate": 1.1078852396343955e-05,
      "loss": 0.9449,
      "step": 1900
    },
    {
      "epoch": 1.8760493827160494,
      "eval_loss": 0.9764742255210876,
      "eval_runtime": 402.9468,
      "eval_samples_per_second": 4.467,
      "eval_steps_per_second": 1.117,
      "step": 1900
    },
    {
      "epoch": 1.8958024691358024,
      "grad_norm": 1.3269942998886108,
      "learning_rate": 1.0746937062571928e-05,
      "loss": 0.9488,
      "step": 1920
    },
    {
      "epoch": 1.9155555555555557,
      "grad_norm": 1.3707290887832642,
      "learning_rate": 1.0417271202984614e-05,
      "loss": 0.9477,
      "step": 1940
    },
    {
      "epoch": 1.9353086419753085,
      "grad_norm": 1.4749308824539185,
      "learning_rate": 1.0090029180118293e-05,
      "loss": 0.9367,
      "step": 1960
    },
    {
      "epoch": 1.9550617283950618,
      "grad_norm": 1.537101149559021,
      "learning_rate": 9.765384074525358e-06,
      "loss": 0.919,
      "step": 1980
    },
    {
      "epoch": 1.9748148148148148,
      "grad_norm": 1.251558780670166,
      "learning_rate": 9.443507593230829e-06,
      "loss": 0.9454,
      "step": 2000
    },
    {
      "epoch": 1.9748148148148148,
      "eval_loss": 0.9748017191886902,
      "eval_runtime": 403.0506,
      "eval_samples_per_second": 4.466,
      "eval_steps_per_second": 1.116,
      "step": 2000
    },
    {
      "epoch": 1.9945679012345678,
      "grad_norm": 1.4743434190750122,
      "learning_rate": 9.124569978915336e-06,
      "loss": 0.9237,
      "step": 2020
    },
    {
      "epoch": 2.013827160493827,
      "grad_norm": 1.2874201536178589,
      "learning_rate": 8.808739919872588e-06,
      "loss": 0.8969,
      "step": 2040
    },
    {
      "epoch": 2.0335802469135804,
      "grad_norm": 1.4540894031524658,
      "learning_rate": 8.496184460788969e-06,
      "loss": 0.9179,
      "step": 2060
    },
    {
      "epoch": 2.0533333333333332,
      "grad_norm": 1.5025060176849365,
      "learning_rate": 8.187068914392464e-06,
      "loss": 0.9091,
      "step": 2080
    },
    {
      "epoch": 2.0730864197530865,
      "grad_norm": 1.4199432134628296,
      "learning_rate": 7.881556774017635e-06,
      "loss": 0.9209,
      "step": 2100
    },
    {
      "epoch": 2.0730864197530865,
      "eval_loss": 0.9763918519020081,
      "eval_runtime": 402.9594,
      "eval_samples_per_second": 4.467,
      "eval_steps_per_second": 1.117,
      "step": 2100
    },
    {
      "epoch": 2.0928395061728393,
      "grad_norm": 1.5283966064453125,
      "learning_rate": 7.579809627132857e-06,
      "loss": 0.9488,
      "step": 2120
    },
    {
      "epoch": 2.1125925925925926,
      "grad_norm": 1.4568904638290405,
      "learning_rate": 7.281987069875591e-06,
      "loss": 0.9094,
      "step": 2140
    },
    {
      "epoch": 2.132345679012346,
      "grad_norm": 1.4240208864212036,
      "learning_rate": 6.988246622640921e-06,
      "loss": 0.9166,
      "step": 2160
    },
    {
      "epoch": 2.1520987654320987,
      "grad_norm": 1.5284944772720337,
      "learning_rate": 6.698743646767891e-06,
      "loss": 0.9205,
      "step": 2180
    },
    {
      "epoch": 2.171851851851852,
      "grad_norm": 1.4134441614151,
      "learning_rate": 6.413631262367882e-06,
      "loss": 0.9121,
      "step": 2200
    },
    {
      "epoch": 2.171851851851852,
      "eval_loss": 0.9747745990753174,
      "eval_runtime": 402.9155,
      "eval_samples_per_second": 4.467,
      "eval_steps_per_second": 1.117,
      "step": 2200
    },
    {
      "epoch": 2.1916049382716047,
      "grad_norm": 1.4537668228149414,
      "learning_rate": 6.1330602673382725e-06,
      "loss": 0.8968,
      "step": 2220
    },
    {
      "epoch": 2.211358024691358,
      "grad_norm": 1.4549959897994995,
      "learning_rate": 5.857179057604451e-06,
      "loss": 0.8984,
      "step": 2240
    },
    {
      "epoch": 2.2311111111111113,
      "grad_norm": 1.5467361211776733,
      "learning_rate": 5.586133548632161e-06,
      "loss": 0.9044,
      "step": 2260
    },
    {
      "epoch": 2.250864197530864,
      "grad_norm": 1.5347260236740112,
      "learning_rate": 5.320067098251815e-06,
      "loss": 0.9008,
      "step": 2280
    },
    {
      "epoch": 2.2706172839506173,
      "grad_norm": 1.5415095090866089,
      "learning_rate": 5.0591204308355465e-06,
      "loss": 0.92,
      "step": 2300
    },
    {
      "epoch": 2.2706172839506173,
      "eval_loss": 0.9742743372917175,
      "eval_runtime": 402.5831,
      "eval_samples_per_second": 4.471,
      "eval_steps_per_second": 1.118,
      "step": 2300
    },
    {
      "epoch": 2.29037037037037,
      "grad_norm": 1.4806352853775024,
      "learning_rate": 4.803431562867145e-06,
      "loss": 0.9074,
      "step": 2320
    },
    {
      "epoch": 2.3101234567901234,
      "grad_norm": 1.5135067701339722,
      "learning_rate": 4.5531357299441774e-06,
      "loss": 0.9279,
      "step": 2340
    },
    {
      "epoch": 2.3298765432098767,
      "grad_norm": 1.5284662246704102,
      "learning_rate": 4.3083653152509434e-06,
      "loss": 0.918,
      "step": 2360
    },
    {
      "epoch": 2.3496296296296295,
      "grad_norm": 1.5032851696014404,
      "learning_rate": 4.069249779540108e-06,
      "loss": 0.8999,
      "step": 2380
    },
    {
      "epoch": 2.3693827160493828,
      "grad_norm": 1.510326862335205,
      "learning_rate": 3.835915592660032e-06,
      "loss": 0.8842,
      "step": 2400
    },
    {
      "epoch": 2.3693827160493828,
      "eval_loss": 0.9732682108879089,
      "eval_runtime": 402.6779,
      "eval_samples_per_second": 4.47,
      "eval_steps_per_second": 1.118,
      "step": 2400
    },
    {
      "epoch": 2.389135802469136,
      "grad_norm": 1.5141196250915527,
      "learning_rate": 3.6084861666639856e-06,
      "loss": 0.8984,
      "step": 2420
    },
    {
      "epoch": 2.408888888888889,
      "grad_norm": 1.5734765529632568,
      "learning_rate": 3.387081790536677e-06,
      "loss": 0.9365,
      "step": 2440
    },
    {
      "epoch": 2.428641975308642,
      "grad_norm": 1.4879003763198853,
      "learning_rate": 3.171819566572563e-06,
      "loss": 0.8829,
      "step": 2460
    },
    {
      "epoch": 2.448395061728395,
      "grad_norm": 1.4807547330856323,
      "learning_rate": 2.962813348439669e-06,
      "loss": 0.9212,
      "step": 2480
    },
    {
      "epoch": 2.468148148148148,
      "grad_norm": 1.492896556854248,
      "learning_rate": 2.7601736809615883e-06,
      "loss": 0.9079,
      "step": 2500
    },
    {
      "epoch": 2.468148148148148,
      "eval_loss": 0.971350908279419,
      "eval_runtime": 402.7891,
      "eval_samples_per_second": 4.469,
      "eval_steps_per_second": 1.117,
      "step": 2500
    },
    {
      "epoch": 2.487901234567901,
      "grad_norm": 1.49791419506073,
      "learning_rate": 2.564007741649559e-06,
      "loss": 0.9271,
      "step": 2520
    },
    {
      "epoch": 2.5076543209876543,
      "grad_norm": 1.5255753993988037,
      "learning_rate": 2.374419284015552e-06,
      "loss": 0.9105,
      "step": 2540
    },
    {
      "epoch": 2.5274074074074075,
      "grad_norm": 1.4556204080581665,
      "learning_rate": 2.191508582696329e-06,
      "loss": 0.9169,
      "step": 2560
    },
    {
      "epoch": 2.5471604938271604,
      "grad_norm": 1.5181186199188232,
      "learning_rate": 2.0153723804174945e-06,
      "loss": 0.8952,
      "step": 2580
    },
    {
      "epoch": 2.5669135802469136,
      "grad_norm": 1.5350087881088257,
      "learning_rate": 1.8461038368256027e-06,
      "loss": 0.909,
      "step": 2600
    },
    {
      "epoch": 2.5669135802469136,
      "eval_loss": 0.9715363383293152,
      "eval_runtime": 402.7586,
      "eval_samples_per_second": 4.469,
      "eval_steps_per_second": 1.117,
      "step": 2600
    }
  ],
  "logging_steps": 20,
  "max_steps": 3036,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.8563300999472087e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}