{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 29.41176470588235,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 2.2869467735290527,
      "learning_rate": 1.9911764705882353e-05,
      "loss": 2.6394,
      "step": 10
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 1.5338459014892578,
      "learning_rate": 1.981372549019608e-05,
      "loss": 2.6312,
      "step": 20
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 2.0163090229034424,
      "learning_rate": 1.9715686274509805e-05,
      "loss": 2.6409,
      "step": 30
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 3.02846097946167,
      "learning_rate": 1.961764705882353e-05,
      "loss": 2.6167,
      "step": 40
    },
    {
      "epoch": 0.7352941176470589,
      "grad_norm": 4.507878303527832,
      "learning_rate": 1.9519607843137257e-05,
      "loss": 2.6231,
      "step": 50
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 2.433847188949585,
      "learning_rate": 1.9421568627450982e-05,
      "loss": 2.6153,
      "step": 60
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.579890012741089,
      "eval_runtime": 1.6326,
      "eval_samples_per_second": 82.691,
      "eval_steps_per_second": 10.413,
      "step": 68
    },
    {
      "epoch": 1.0294117647058822,
      "grad_norm": 3.4059886932373047,
      "learning_rate": 1.9323529411764706e-05,
      "loss": 2.5863,
      "step": 70
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.305633068084717,
      "learning_rate": 1.9225490196078434e-05,
      "loss": 2.5677,
      "step": 80
    },
    {
      "epoch": 1.3235294117647058,
      "grad_norm": 2.493675470352173,
      "learning_rate": 1.912745098039216e-05,
      "loss": 2.5061,
      "step": 90
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 2.6469180583953857,
      "learning_rate": 1.9029411764705883e-05,
      "loss": 2.4657,
      "step": 100
    },
    {
      "epoch": 1.6176470588235294,
      "grad_norm": 3.5626657009124756,
      "learning_rate": 1.8931372549019607e-05,
      "loss": 2.4594,
      "step": 110
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 2.9186179637908936,
      "learning_rate": 1.8833333333333335e-05,
      "loss": 2.4393,
      "step": 120
    },
    {
      "epoch": 1.9117647058823528,
      "grad_norm": 4.3771772384643555,
      "learning_rate": 1.873529411764706e-05,
      "loss": 2.3884,
      "step": 130
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.311964750289917,
      "eval_runtime": 1.6428,
      "eval_samples_per_second": 82.178,
      "eval_steps_per_second": 10.348,
      "step": 136
    },
    {
      "epoch": 2.0588235294117645,
      "grad_norm": 2.706939935684204,
      "learning_rate": 1.8637254901960787e-05,
      "loss": 2.3395,
      "step": 140
    },
    {
      "epoch": 2.2058823529411766,
      "grad_norm": 3.1243748664855957,
      "learning_rate": 1.853921568627451e-05,
      "loss": 2.2711,
      "step": 150
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 2.9195902347564697,
      "learning_rate": 1.844117647058824e-05,
      "loss": 2.2252,
      "step": 160
    },
    {
      "epoch": 2.5,
      "grad_norm": 5.713516712188721,
      "learning_rate": 1.834313725490196e-05,
      "loss": 2.2211,
      "step": 170
    },
    {
      "epoch": 2.6470588235294117,
      "grad_norm": 2.779804229736328,
      "learning_rate": 1.8245098039215688e-05,
      "loss": 2.1251,
      "step": 180
    },
    {
      "epoch": 2.7941176470588234,
      "grad_norm": 2.9840755462646484,
      "learning_rate": 1.8147058823529412e-05,
      "loss": 2.1162,
      "step": 190
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 3.6137640476226807,
      "learning_rate": 1.804901960784314e-05,
      "loss": 2.0859,
      "step": 200
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.982031226158142,
      "eval_runtime": 1.6147,
      "eval_samples_per_second": 83.606,
      "eval_steps_per_second": 10.528,
      "step": 204
    },
    {
      "epoch": 3.088235294117647,
      "grad_norm": 2.8977713584899902,
      "learning_rate": 1.7950980392156864e-05,
      "loss": 2.0281,
      "step": 210
    },
    {
      "epoch": 3.235294117647059,
      "grad_norm": 3.248961925506592,
      "learning_rate": 1.7852941176470592e-05,
      "loss": 1.9454,
      "step": 220
    },
    {
      "epoch": 3.3823529411764706,
      "grad_norm": 3.525045394897461,
      "learning_rate": 1.7754901960784313e-05,
      "loss": 1.8824,
      "step": 230
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 3.95782470703125,
      "learning_rate": 1.765686274509804e-05,
      "loss": 1.8538,
      "step": 240
    },
    {
      "epoch": 3.6764705882352944,
      "grad_norm": 4.282369613647461,
      "learning_rate": 1.7558823529411765e-05,
      "loss": 1.823,
      "step": 250
    },
    {
      "epoch": 3.8235294117647056,
      "grad_norm": 6.753427505493164,
      "learning_rate": 1.7460784313725493e-05,
      "loss": 1.8116,
      "step": 260
    },
    {
      "epoch": 3.9705882352941178,
      "grad_norm": 4.552963733673096,
      "learning_rate": 1.7362745098039217e-05,
      "loss": 1.7387,
      "step": 270
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.6489075422286987,
      "eval_runtime": 1.6483,
      "eval_samples_per_second": 81.902,
      "eval_steps_per_second": 10.314,
      "step": 272
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 3.1702029705047607,
      "learning_rate": 1.7264705882352945e-05,
      "loss": 1.6522,
      "step": 280
    },
    {
      "epoch": 4.264705882352941,
      "grad_norm": 4.5589494705200195,
      "learning_rate": 1.7166666666666666e-05,
      "loss": 1.6078,
      "step": 290
    },
    {
      "epoch": 4.411764705882353,
      "grad_norm": 3.9019124507904053,
      "learning_rate": 1.7068627450980394e-05,
      "loss": 1.5599,
      "step": 300
    },
    {
      "epoch": 4.5588235294117645,
      "grad_norm": 4.917637348175049,
      "learning_rate": 1.6970588235294118e-05,
      "loss": 1.5237,
      "step": 310
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 3.638333320617676,
      "learning_rate": 1.6872549019607846e-05,
      "loss": 1.4842,
      "step": 320
    },
    {
      "epoch": 4.852941176470588,
      "grad_norm": 3.488672971725464,
      "learning_rate": 1.677450980392157e-05,
      "loss": 1.4391,
      "step": 330
    },
    {
      "epoch": 5.0,
      "grad_norm": 8.806844711303711,
      "learning_rate": 1.6676470588235295e-05,
      "loss": 1.4165,
      "step": 340
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.3353697061538696,
      "eval_runtime": 1.6321,
      "eval_samples_per_second": 82.718,
      "eval_steps_per_second": 10.416,
      "step": 340
    },
    {
      "epoch": 5.147058823529412,
      "grad_norm": 5.466882705688477,
      "learning_rate": 1.6578431372549022e-05,
      "loss": 1.3357,
      "step": 350
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 3.95070481300354,
      "learning_rate": 1.6480392156862747e-05,
      "loss": 1.3317,
      "step": 360
    },
    {
      "epoch": 5.4411764705882355,
      "grad_norm": 3.2761778831481934,
      "learning_rate": 1.638235294117647e-05,
      "loss": 1.1954,
      "step": 370
    },
    {
      "epoch": 5.588235294117647,
      "grad_norm": 2.8257029056549072,
      "learning_rate": 1.62843137254902e-05,
      "loss": 1.1713,
      "step": 380
    },
    {
      "epoch": 5.735294117647059,
      "grad_norm": 3.5745885372161865,
      "learning_rate": 1.6186274509803923e-05,
      "loss": 1.1767,
      "step": 390
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 3.625549077987671,
      "learning_rate": 1.6088235294117648e-05,
      "loss": 1.1028,
      "step": 400
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.0512332916259766,
      "eval_runtime": 1.6283,
      "eval_samples_per_second": 82.911,
      "eval_steps_per_second": 10.441,
      "step": 408
    },
    {
      "epoch": 6.029411764705882,
      "grad_norm": 3.7580277919769287,
      "learning_rate": 1.5990196078431375e-05,
      "loss": 1.0918,
      "step": 410
    },
    {
      "epoch": 6.176470588235294,
      "grad_norm": 5.0767292976379395,
      "learning_rate": 1.58921568627451e-05,
      "loss": 1.0799,
      "step": 420
    },
    {
      "epoch": 6.323529411764706,
      "grad_norm": 3.362245798110962,
      "learning_rate": 1.5794117647058824e-05,
      "loss": 1.0246,
      "step": 430
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 2.663010835647583,
      "learning_rate": 1.569607843137255e-05,
      "loss": 0.9335,
      "step": 440
    },
    {
      "epoch": 6.617647058823529,
      "grad_norm": 2.7811036109924316,
      "learning_rate": 1.5598039215686276e-05,
      "loss": 0.9084,
      "step": 450
    },
    {
      "epoch": 6.764705882352941,
      "grad_norm": 6.619537830352783,
      "learning_rate": 1.55e-05,
      "loss": 0.8652,
      "step": 460
    },
    {
      "epoch": 6.911764705882353,
      "grad_norm": 4.524352073669434,
      "learning_rate": 1.540196078431373e-05,
      "loss": 0.8706,
      "step": 470
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.837912380695343,
      "eval_runtime": 1.6194,
      "eval_samples_per_second": 83.366,
      "eval_steps_per_second": 10.498,
      "step": 476
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 4.127895832061768,
      "learning_rate": 1.5303921568627453e-05,
      "loss": 0.7721,
      "step": 480
    },
    {
      "epoch": 7.205882352941177,
      "grad_norm": 2.9023818969726562,
      "learning_rate": 1.5205882352941177e-05,
      "loss": 0.7268,
      "step": 490
    },
    {
      "epoch": 7.352941176470588,
      "grad_norm": 3.2444231510162354,
      "learning_rate": 1.5107843137254903e-05,
      "loss": 0.7142,
      "step": 500
    },
    {
      "epoch": 7.5,
      "grad_norm": 3.160628318786621,
      "learning_rate": 1.500980392156863e-05,
      "loss": 0.7019,
      "step": 510
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 2.913454294204712,
      "learning_rate": 1.4911764705882354e-05,
      "loss": 0.6911,
      "step": 520
    },
    {
      "epoch": 7.794117647058823,
      "grad_norm": 2.9466731548309326,
      "learning_rate": 1.481372549019608e-05,
      "loss": 0.6639,
      "step": 530
    },
    {
      "epoch": 7.9411764705882355,
      "grad_norm": 2.8029139041900635,
      "learning_rate": 1.4715686274509806e-05,
      "loss": 0.5554,
      "step": 540
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.5985532402992249,
      "eval_runtime": 1.6867,
      "eval_samples_per_second": 80.038,
      "eval_steps_per_second": 10.079,
      "step": 544
    },
    {
      "epoch": 8.088235294117647,
      "grad_norm": 2.267561912536621,
      "learning_rate": 1.461764705882353e-05,
      "loss": 0.5757,
      "step": 550
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 2.1232762336730957,
      "learning_rate": 1.4519607843137256e-05,
      "loss": 0.5091,
      "step": 560
    },
    {
      "epoch": 8.382352941176471,
      "grad_norm": 3.035518169403076,
      "learning_rate": 1.442156862745098e-05,
      "loss": 0.5158,
      "step": 570
    },
    {
      "epoch": 8.529411764705882,
      "grad_norm": 2.085606336593628,
      "learning_rate": 1.4323529411764707e-05,
      "loss": 0.4673,
      "step": 580
    },
    {
      "epoch": 8.676470588235293,
      "grad_norm": 1.8189260959625244,
      "learning_rate": 1.4225490196078433e-05,
      "loss": 0.4604,
      "step": 590
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 6.74431037902832,
      "learning_rate": 1.4127450980392159e-05,
      "loss": 0.4651,
      "step": 600
    },
    {
      "epoch": 8.970588235294118,
      "grad_norm": 1.56160306930542,
      "learning_rate": 1.4029411764705883e-05,
      "loss": 0.4017,
      "step": 610
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.40496423840522766,
      "eval_runtime": 1.6727,
      "eval_samples_per_second": 80.707,
      "eval_steps_per_second": 10.163,
      "step": 612
    },
    {
      "epoch": 9.117647058823529,
      "grad_norm": 1.9897135496139526,
      "learning_rate": 1.3931372549019608e-05,
      "loss": 0.3879,
      "step": 620
    },
    {
      "epoch": 9.264705882352942,
      "grad_norm": 1.6671721935272217,
      "learning_rate": 1.3833333333333334e-05,
      "loss": 0.3206,
      "step": 630
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.8617277145385742,
      "learning_rate": 1.373529411764706e-05,
      "loss": 0.3565,
      "step": 640
    },
    {
      "epoch": 9.558823529411764,
      "grad_norm": 1.64924955368042,
      "learning_rate": 1.3637254901960786e-05,
      "loss": 0.3457,
      "step": 650
    },
    {
      "epoch": 9.705882352941176,
      "grad_norm": 1.5722205638885498,
      "learning_rate": 1.3539215686274512e-05,
      "loss": 0.3123,
      "step": 660
    },
    {
      "epoch": 9.852941176470589,
      "grad_norm": 2.3067121505737305,
      "learning_rate": 1.3441176470588238e-05,
      "loss": 0.3137,
      "step": 670
    },
    {
      "epoch": 10.0,
      "grad_norm": 4.091474533081055,
      "learning_rate": 1.334313725490196e-05,
      "loss": 0.2663,
      "step": 680
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.29727470874786377,
      "eval_runtime": 1.6004,
      "eval_samples_per_second": 84.356,
      "eval_steps_per_second": 10.623,
      "step": 680
    },
    {
      "epoch": 10.147058823529411,
      "grad_norm": 1.217851161956787,
      "learning_rate": 1.3245098039215687e-05,
      "loss": 0.2133,
      "step": 690
    },
    {
      "epoch": 10.294117647058824,
      "grad_norm": 1.5889488458633423,
      "learning_rate": 1.3147058823529413e-05,
      "loss": 0.2299,
      "step": 700
    },
    {
      "epoch": 10.441176470588236,
      "grad_norm": 1.2852109670639038,
      "learning_rate": 1.3049019607843139e-05,
      "loss": 0.1991,
      "step": 710
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.1250885725021362,
      "learning_rate": 1.2950980392156865e-05,
      "loss": 0.2625,
      "step": 720
    },
    {
      "epoch": 10.735294117647058,
      "grad_norm": 0.8768565654754639,
      "learning_rate": 1.285294117647059e-05,
      "loss": 0.1657,
      "step": 730
    },
    {
      "epoch": 10.882352941176471,
      "grad_norm": 1.0275033712387085,
      "learning_rate": 1.2754901960784314e-05,
      "loss": 0.1814,
      "step": 740
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.22045107185840607,
      "eval_runtime": 1.6056,
      "eval_samples_per_second": 84.083,
      "eval_steps_per_second": 10.588,
      "step": 748
    },
    {
      "epoch": 11.029411764705882,
      "grad_norm": 1.1377185583114624,
      "learning_rate": 1.265686274509804e-05,
      "loss": 0.2123,
      "step": 750
    },
    {
      "epoch": 11.176470588235293,
      "grad_norm": 0.6125518679618835,
      "learning_rate": 1.2558823529411766e-05,
      "loss": 0.145,
      "step": 760
    },
    {
      "epoch": 11.323529411764707,
      "grad_norm": 3.132908344268799,
      "learning_rate": 1.2460784313725492e-05,
      "loss": 0.1409,
      "step": 770
    },
    {
      "epoch": 11.470588235294118,
      "grad_norm": 7.12354850769043,
      "learning_rate": 1.2362745098039218e-05,
      "loss": 0.171,
      "step": 780
    },
    {
      "epoch": 11.617647058823529,
      "grad_norm": 0.9431573152542114,
      "learning_rate": 1.2264705882352944e-05,
      "loss": 0.1561,
      "step": 790
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.5725237131118774,
      "learning_rate": 1.2166666666666667e-05,
      "loss": 0.1012,
      "step": 800
    },
    {
      "epoch": 11.911764705882353,
      "grad_norm": 0.6124898791313171,
      "learning_rate": 1.2068627450980393e-05,
      "loss": 0.1072,
      "step": 810
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.20114047825336456,
      "eval_runtime": 1.61,
      "eval_samples_per_second": 83.85,
      "eval_steps_per_second": 10.559,
      "step": 816
    },
    {
      "epoch": 12.058823529411764,
      "grad_norm": 0.6034418344497681,
      "learning_rate": 1.1970588235294119e-05,
      "loss": 0.1093,
      "step": 820
    },
    {
      "epoch": 12.205882352941176,
      "grad_norm": 0.6741713881492615,
      "learning_rate": 1.1872549019607845e-05,
      "loss": 0.0901,
      "step": 830
    },
    {
      "epoch": 12.352941176470589,
      "grad_norm": 0.7810522317886353,
      "learning_rate": 1.177450980392157e-05,
      "loss": 0.089,
      "step": 840
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.6442411541938782,
      "learning_rate": 1.1676470588235295e-05,
      "loss": 0.1219,
      "step": 850
    },
    {
      "epoch": 12.647058823529411,
      "grad_norm": 3.0149757862091064,
      "learning_rate": 1.1578431372549021e-05,
      "loss": 0.0814,
      "step": 860
    },
    {
      "epoch": 12.794117647058824,
      "grad_norm": 0.46087804436683655,
      "learning_rate": 1.1480392156862746e-05,
      "loss": 0.1077,
      "step": 870
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.56253981590271,
      "learning_rate": 1.1382352941176472e-05,
      "loss": 0.068,
      "step": 880
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.1641794890165329,
      "eval_runtime": 1.6752,
      "eval_samples_per_second": 80.589,
      "eval_steps_per_second": 10.148,
      "step": 884
    },
    {
      "epoch": 13.088235294117647,
      "grad_norm": 0.6027768850326538,
      "learning_rate": 1.1284313725490198e-05,
      "loss": 0.0689,
      "step": 890
    },
    {
      "epoch": 13.235294117647058,
      "grad_norm": 0.369243860244751,
      "learning_rate": 1.1186274509803922e-05,
      "loss": 0.0613,
      "step": 900
    },
    {
      "epoch": 13.382352941176471,
      "grad_norm": 0.3756796419620514,
      "learning_rate": 1.1088235294117648e-05,
      "loss": 0.0599,
      "step": 910
    },
    {
      "epoch": 13.529411764705882,
      "grad_norm": 0.42388486862182617,
      "learning_rate": 1.0990196078431374e-05,
      "loss": 0.0561,
      "step": 920
    },
    {
      "epoch": 13.676470588235293,
      "grad_norm": 0.40439847111701965,
      "learning_rate": 1.0892156862745099e-05,
      "loss": 0.0542,
      "step": 930
    },
    {
      "epoch": 13.823529411764707,
      "grad_norm": 0.39990735054016113,
      "learning_rate": 1.0794117647058825e-05,
      "loss": 0.0672,
      "step": 940
    },
    {
      "epoch": 13.970588235294118,
      "grad_norm": 0.3704688251018524,
      "learning_rate": 1.0696078431372549e-05,
      "loss": 0.1051,
      "step": 950
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.14415012300014496,
      "eval_runtime": 1.6471,
      "eval_samples_per_second": 81.961,
      "eval_steps_per_second": 10.321,
      "step": 952
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.33351826667785645,
      "learning_rate": 1.0598039215686275e-05,
      "loss": 0.0492,
      "step": 960
    },
    {
      "epoch": 14.264705882352942,
      "grad_norm": 0.3827894926071167,
      "learning_rate": 1.0500000000000001e-05,
      "loss": 0.0484,
      "step": 970
    },
    {
      "epoch": 14.411764705882353,
      "grad_norm": 0.4303022623062134,
      "learning_rate": 1.0401960784313727e-05,
      "loss": 0.0461,
      "step": 980
    },
    {
      "epoch": 14.558823529411764,
      "grad_norm": 0.3589096665382385,
      "learning_rate": 1.0303921568627452e-05,
      "loss": 0.0468,
      "step": 990
    },
    {
      "epoch": 14.705882352941176,
      "grad_norm": 0.30145418643951416,
      "learning_rate": 1.0205882352941176e-05,
      "loss": 0.0942,
      "step": 1000
    },
    {
      "epoch": 14.852941176470589,
      "grad_norm": 0.3331129550933838,
      "learning_rate": 1.0107843137254902e-05,
      "loss": 0.0465,
      "step": 1010
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.790107011795044,
      "learning_rate": 1.0009803921568628e-05,
      "loss": 0.0431,
      "step": 1020
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.1515953093767166,
      "eval_runtime": 1.5961,
      "eval_samples_per_second": 84.583,
      "eval_steps_per_second": 10.651,
      "step": 1020
    },
    {
      "epoch": 15.147058823529411,
      "grad_norm": 0.4401082992553711,
      "learning_rate": 9.911764705882354e-06,
      "loss": 0.0511,
      "step": 1030
    },
    {
      "epoch": 15.294117647058824,
      "grad_norm": 0.3193589150905609,
      "learning_rate": 9.813725490196078e-06,
      "loss": 0.0369,
      "step": 1040
    },
    {
      "epoch": 15.441176470588236,
      "grad_norm": 0.33625075221061707,
      "learning_rate": 9.715686274509805e-06,
      "loss": 0.0363,
      "step": 1050
    },
    {
      "epoch": 15.588235294117647,
      "grad_norm": 0.18927061557769775,
      "learning_rate": 9.61764705882353e-06,
      "loss": 0.0349,
      "step": 1060
    },
    {
      "epoch": 15.735294117647058,
      "grad_norm": 0.28063905239105225,
      "learning_rate": 9.519607843137255e-06,
      "loss": 0.0355,
      "step": 1070
    },
    {
      "epoch": 15.882352941176471,
      "grad_norm": 0.31923410296440125,
      "learning_rate": 9.421568627450981e-06,
      "loss": 0.0949,
      "step": 1080
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.16671152412891388,
      "eval_runtime": 1.6889,
      "eval_samples_per_second": 79.936,
      "eval_steps_per_second": 10.066,
      "step": 1088
    },
    {
      "epoch": 16.029411764705884,
      "grad_norm": 0.25026965141296387,
      "learning_rate": 9.323529411764707e-06,
      "loss": 0.0357,
      "step": 1090
    },
    {
      "epoch": 16.176470588235293,
      "grad_norm": 0.39554500579833984,
      "learning_rate": 9.225490196078433e-06,
      "loss": 0.0946,
      "step": 1100
    },
    {
      "epoch": 16.323529411764707,
      "grad_norm": 0.30123069882392883,
      "learning_rate": 9.127450980392158e-06,
      "loss": 0.0322,
      "step": 1110
    },
    {
      "epoch": 16.470588235294116,
      "grad_norm": 0.25656214356422424,
      "learning_rate": 9.029411764705884e-06,
      "loss": 0.0291,
      "step": 1120
    },
    {
      "epoch": 16.61764705882353,
      "grad_norm": 0.21169237792491913,
      "learning_rate": 8.93137254901961e-06,
      "loss": 0.029,
      "step": 1130
    },
    {
      "epoch": 16.764705882352942,
      "grad_norm": 0.22442790865898132,
      "learning_rate": 8.833333333333334e-06,
      "loss": 0.0316,
      "step": 1140
    },
    {
      "epoch": 16.91176470588235,
      "grad_norm": 0.20120474696159363,
      "learning_rate": 8.73529411764706e-06,
      "loss": 0.0287,
      "step": 1150
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.13797470927238464,
      "eval_runtime": 1.6743,
      "eval_samples_per_second": 80.63,
      "eval_steps_per_second": 10.153,
      "step": 1156
    },
    {
      "epoch": 17.058823529411764,
      "grad_norm": 0.20405986905097961,
      "learning_rate": 8.637254901960786e-06,
      "loss": 0.0321,
      "step": 1160
    },
    {
      "epoch": 17.205882352941178,
      "grad_norm": 0.22055025398731232,
      "learning_rate": 8.53921568627451e-06,
      "loss": 0.027,
      "step": 1170
    },
    {
      "epoch": 17.352941176470587,
      "grad_norm": 0.18829534947872162,
      "learning_rate": 8.441176470588237e-06,
      "loss": 0.0769,
      "step": 1180
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.7278969883918762,
      "learning_rate": 8.343137254901961e-06,
      "loss": 0.0276,
      "step": 1190
    },
    {
      "epoch": 17.647058823529413,
      "grad_norm": 0.22244372963905334,
      "learning_rate": 8.245098039215687e-06,
      "loss": 0.0263,
      "step": 1200
    },
    {
      "epoch": 17.794117647058822,
      "grad_norm": 0.1744055598974228,
      "learning_rate": 8.147058823529413e-06,
      "loss": 0.0241,
      "step": 1210
    },
    {
      "epoch": 17.941176470588236,
      "grad_norm": 0.16366326808929443,
      "learning_rate": 8.049019607843137e-06,
      "loss": 0.0246,
      "step": 1220
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.15256382524967194,
      "eval_runtime": 1.6045,
      "eval_samples_per_second": 84.139,
      "eval_steps_per_second": 10.595,
      "step": 1224
    },
    {
      "epoch": 18.08823529411765,
      "grad_norm": 0.15204133093357086,
      "learning_rate": 7.950980392156864e-06,
      "loss": 0.0254,
      "step": 1230
    },
    {
      "epoch": 18.235294117647058,
      "grad_norm": 0.16595087945461273,
      "learning_rate": 7.85294117647059e-06,
      "loss": 0.0243,
      "step": 1240
    },
    {
      "epoch": 18.38235294117647,
      "grad_norm": 0.2022247016429901,
      "learning_rate": 7.754901960784314e-06,
      "loss": 0.0242,
      "step": 1250
    },
    {
      "epoch": 18.529411764705884,
      "grad_norm": 0.17330515384674072,
      "learning_rate": 7.65686274509804e-06,
      "loss": 0.024,
      "step": 1260
    },
    {
      "epoch": 18.676470588235293,
      "grad_norm": 0.20826362073421478,
      "learning_rate": 7.558823529411765e-06,
      "loss": 0.0801,
      "step": 1270
    },
    {
      "epoch": 18.823529411764707,
      "grad_norm": 0.2036994844675064,
      "learning_rate": 7.460784313725491e-06,
      "loss": 0.0232,
      "step": 1280
    },
    {
      "epoch": 18.970588235294116,
      "grad_norm": 0.1467086523771286,
      "learning_rate": 7.3627450980392165e-06,
      "loss": 0.0217,
      "step": 1290
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.14522351324558258,
      "eval_runtime": 1.6184,
      "eval_samples_per_second": 83.417,
      "eval_steps_per_second": 10.504,
      "step": 1292
    },
    {
      "epoch": 19.11764705882353,
      "grad_norm": 0.21091414988040924,
      "learning_rate": 7.264705882352942e-06,
      "loss": 0.0217,
      "step": 1300
    },
    {
      "epoch": 19.264705882352942,
      "grad_norm": 0.16935530304908752,
      "learning_rate": 7.166666666666667e-06,
      "loss": 0.0742,
      "step": 1310
    },
    {
      "epoch": 19.41176470588235,
      "grad_norm": 0.19531255960464478,
      "learning_rate": 7.068627450980393e-06,
      "loss": 0.0203,
      "step": 1320
    },
    {
      "epoch": 19.558823529411764,
      "grad_norm": 0.18234974145889282,
      "learning_rate": 6.970588235294118e-06,
      "loss": 0.0539,
      "step": 1330
    },
    {
      "epoch": 19.705882352941178,
      "grad_norm": 0.1523897647857666,
      "learning_rate": 6.8725490196078434e-06,
      "loss": 0.0207,
      "step": 1340
    },
    {
      "epoch": 19.852941176470587,
      "grad_norm": 0.1887982189655304,
      "learning_rate": 6.7745098039215695e-06,
      "loss": 0.0205,
      "step": 1350
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.2818535566329956,
      "learning_rate": 6.676470588235294e-06,
      "loss": 0.0216,
      "step": 1360
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.15050244331359863,
      "eval_runtime": 1.6089,
      "eval_samples_per_second": 83.907,
      "eval_steps_per_second": 10.566,
      "step": 1360
    },
    {
      "epoch": 20.147058823529413,
      "grad_norm": 0.1478956639766693,
      "learning_rate": 6.57843137254902e-06,
      "loss": 0.0181,
      "step": 1370
    },
    {
      "epoch": 20.294117647058822,
      "grad_norm": 0.18312789499759674,
      "learning_rate": 6.480392156862746e-06,
      "loss": 0.1025,
      "step": 1380
    },
    {
      "epoch": 20.441176470588236,
      "grad_norm": 0.15513214468955994,
      "learning_rate": 6.38235294117647e-06,
      "loss": 0.0209,
      "step": 1390
    },
    {
      "epoch": 20.58823529411765,
      "grad_norm": 0.13837294280529022,
      "learning_rate": 6.2843137254901964e-06,
      "loss": 0.0199,
      "step": 1400
    },
    {
      "epoch": 20.735294117647058,
      "grad_norm": 0.1465756893157959,
      "learning_rate": 6.1862745098039225e-06,
      "loss": 0.0192,
      "step": 1410
    },
    {
      "epoch": 20.88235294117647,
      "grad_norm": 0.16790196299552917,
      "learning_rate": 6.088235294117647e-06,
      "loss": 0.0195,
      "step": 1420
    },
    {
      "epoch": 21.0,
      "eval_loss": 0.1510893851518631,
      "eval_runtime": 1.6277,
      "eval_samples_per_second": 82.937,
      "eval_steps_per_second": 10.444,
      "step": 1428
    },
    {
      "epoch": 21.029411764705884,
      "grad_norm": 0.29194769263267517,
      "learning_rate": 5.990196078431373e-06,
      "loss": 0.0181,
      "step": 1430
    },
    {
      "epoch": 21.176470588235293,
      "grad_norm": 0.16755153238773346,
      "learning_rate": 5.892156862745099e-06,
      "loss": 0.0186,
      "step": 1440
    },
    {
      "epoch": 21.323529411764707,
      "grad_norm": 0.19387616217136383,
      "learning_rate": 5.794117647058824e-06,
      "loss": 0.0184,
      "step": 1450
    },
    {
      "epoch": 21.470588235294116,
      "grad_norm": 0.15613146126270294,
      "learning_rate": 5.6960784313725494e-06,
      "loss": 0.0196,
      "step": 1460
    },
    {
      "epoch": 21.61764705882353,
      "grad_norm": 0.12751713395118713,
      "learning_rate": 5.598039215686275e-06,
      "loss": 0.0178,
      "step": 1470
    },
    {
      "epoch": 21.764705882352942,
      "grad_norm": 0.15414807200431824,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.0177,
      "step": 1480
    },
    {
      "epoch": 21.91176470588235,
      "grad_norm": 0.12436957657337189,
      "learning_rate": 5.401960784313726e-06,
      "loss": 0.0749,
      "step": 1490
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.14262542128562927,
      "eval_runtime": 1.6896,
      "eval_samples_per_second": 79.901,
      "eval_steps_per_second": 10.062,
      "step": 1496
    },
    {
      "epoch": 22.058823529411764,
      "grad_norm": 0.1797289252281189,
      "learning_rate": 5.303921568627451e-06,
      "loss": 0.0163,
      "step": 1500
    },
    {
      "epoch": 22.205882352941178,
      "grad_norm": 0.15396994352340698,
      "learning_rate": 5.205882352941177e-06,
      "loss": 0.0164,
      "step": 1510
    },
    {
      "epoch": 22.352941176470587,
      "grad_norm": 0.43706995248794556,
      "learning_rate": 5.107843137254902e-06,
      "loss": 0.0173,
      "step": 1520
    },
    {
      "epoch": 22.5,
      "grad_norm": 0.13553638756275177,
      "learning_rate": 5.009803921568628e-06,
      "loss": 0.0701,
      "step": 1530
    },
    {
      "epoch": 22.647058823529413,
      "grad_norm": 0.14506298303604126,
      "learning_rate": 4.911764705882353e-06,
      "loss": 0.0158,
      "step": 1540
    },
    {
      "epoch": 22.794117647058822,
      "grad_norm": 0.25496020913124084,
      "learning_rate": 4.813725490196079e-06,
      "loss": 0.0152,
      "step": 1550
    },
    {
      "epoch": 22.941176470588236,
      "grad_norm": 0.16556541621685028,
      "learning_rate": 4.715686274509804e-06,
      "loss": 0.0157,
      "step": 1560
    },
    {
      "epoch": 23.0,
      "eval_loss": 0.1430669128894806,
      "eval_runtime": 1.6521,
      "eval_samples_per_second": 81.714,
      "eval_steps_per_second": 10.29,
      "step": 1564
    },
    {
      "epoch": 23.08823529411765,
      "grad_norm": 0.12428230047225952,
      "learning_rate": 4.61764705882353e-06,
      "loss": 0.0159,
      "step": 1570
    },
    {
      "epoch": 23.235294117647058,
      "grad_norm": 0.11920207738876343,
      "learning_rate": 4.519607843137255e-06,
      "loss": 0.0145,
      "step": 1580
    },
    {
      "epoch": 23.38235294117647,
      "grad_norm": 0.1940208524465561,
      "learning_rate": 4.421568627450981e-06,
      "loss": 0.0162,
      "step": 1590
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 0.1417732536792755,
      "learning_rate": 4.323529411764707e-06,
      "loss": 0.0149,
      "step": 1600
    },
    {
      "epoch": 23.676470588235293,
      "grad_norm": 0.13574868440628052,
      "learning_rate": 4.225490196078432e-06,
      "loss": 0.0152,
      "step": 1610
    },
    {
      "epoch": 23.823529411764707,
      "grad_norm": 0.1438971906900406,
      "learning_rate": 4.127450980392157e-06,
      "loss": 0.0677,
      "step": 1620
    },
    {
      "epoch": 23.970588235294116,
      "grad_norm": 0.12435536831617355,
      "learning_rate": 4.029411764705883e-06,
      "loss": 0.0152,
      "step": 1630
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.14459367096424103,
      "eval_runtime": 1.6409,
      "eval_samples_per_second": 82.271,
      "eval_steps_per_second": 10.36,
      "step": 1632
    },
    {
      "epoch": 24.11764705882353,
      "grad_norm": 0.15185730159282684,
      "learning_rate": 3.931372549019608e-06,
      "loss": 0.0146,
      "step": 1640
    },
    {
      "epoch": 24.264705882352942,
      "grad_norm": 0.14219020307064056,
      "learning_rate": 3.833333333333334e-06,
      "loss": 0.0141,
      "step": 1650
    },
    {
      "epoch": 24.41176470588235,
      "grad_norm": 0.1183333620429039,
      "learning_rate": 3.7352941176470593e-06,
      "loss": 0.0142,
      "step": 1660
    },
    {
      "epoch": 24.558823529411764,
      "grad_norm": 0.11505335569381714,
      "learning_rate": 3.6372549019607845e-06,
      "loss": 0.0601,
      "step": 1670
    },
    {
      "epoch": 24.705882352941178,
      "grad_norm": 0.1279689520597458,
      "learning_rate": 3.53921568627451e-06,
      "loss": 0.0149,
      "step": 1680
    },
    {
      "epoch": 24.852941176470587,
      "grad_norm": 0.09878543764352798,
      "learning_rate": 3.4411764705882358e-06,
      "loss": 0.0145,
      "step": 1690
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.2872982919216156,
      "learning_rate": 3.343137254901961e-06,
      "loss": 0.0142,
      "step": 1700
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.1434563398361206,
      "eval_runtime": 1.5906,
      "eval_samples_per_second": 84.875,
      "eval_steps_per_second": 10.688,
      "step": 1700
    },
    {
      "epoch": 25.147058823529413,
      "grad_norm": 13.370408058166504,
      "learning_rate": 3.2450980392156866e-06,
      "loss": 0.0515,
      "step": 1710
    },
    {
      "epoch": 25.294117647058822,
      "grad_norm": 0.11658412218093872,
      "learning_rate": 3.147058823529412e-06,
      "loss": 0.0136,
      "step": 1720
    },
    {
      "epoch": 25.441176470588236,
      "grad_norm": 0.082937091588974,
      "learning_rate": 3.049019607843138e-06,
      "loss": 0.0145,
      "step": 1730
    },
    {
      "epoch": 25.58823529411765,
      "grad_norm": 0.09096076339483261,
      "learning_rate": 2.950980392156863e-06,
      "loss": 0.0132,
      "step": 1740
    },
    {
      "epoch": 25.735294117647058,
      "grad_norm": 0.12208091467618942,
      "learning_rate": 2.8529411764705883e-06,
      "loss": 0.0144,
      "step": 1750
    },
    {
      "epoch": 25.88235294117647,
      "grad_norm": 0.11157109588384628,
      "learning_rate": 2.754901960784314e-06,
      "loss": 0.0134,
      "step": 1760
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.14469991624355316,
      "eval_runtime": 1.6847,
      "eval_samples_per_second": 80.133,
      "eval_steps_per_second": 10.091,
      "step": 1768
    },
    {
      "epoch": 26.029411764705884,
      "grad_norm": 0.13175475597381592,
      "learning_rate": 2.6568627450980396e-06,
      "loss": 0.0139,
      "step": 1770
    },
    {
      "epoch": 26.176470588235293,
      "grad_norm": 0.12328551709651947,
      "learning_rate": 2.558823529411765e-06,
      "loss": 0.0127,
      "step": 1780
    },
    {
      "epoch": 26.323529411764707,
      "grad_norm": 0.08687178790569305,
      "learning_rate": 2.4607843137254905e-06,
      "loss": 0.0731,
      "step": 1790
    },
    {
      "epoch": 26.470588235294116,
      "grad_norm": 0.11597796529531479,
      "learning_rate": 2.3627450980392157e-06,
      "loss": 0.0135,
      "step": 1800
    },
    {
      "epoch": 26.61764705882353,
      "grad_norm": 0.09727944433689117,
      "learning_rate": 2.2647058823529413e-06,
      "loss": 0.014,
      "step": 1810
    },
    {
      "epoch": 26.764705882352942,
      "grad_norm": 0.1099948063492775,
      "learning_rate": 2.166666666666667e-06,
      "loss": 0.0139,
      "step": 1820
    },
    {
      "epoch": 26.91176470588235,
      "grad_norm": 0.1616813987493515,
      "learning_rate": 2.068627450980392e-06,
      "loss": 0.0137,
      "step": 1830
    },
    {
      "epoch": 27.0,
      "eval_loss": 0.15348146855831146,
      "eval_runtime": 1.6268,
      "eval_samples_per_second": 82.982,
      "eval_steps_per_second": 10.45,
      "step": 1836
    },
    {
      "epoch": 27.058823529411764,
      "grad_norm": 0.09702517092227936,
      "learning_rate": 1.970588235294118e-06,
      "loss": 0.013,
      "step": 1840
    },
    {
      "epoch": 27.205882352941178,
      "grad_norm": 0.4694797992706299,
      "learning_rate": 1.8725490196078432e-06,
      "loss": 0.014,
      "step": 1850
    },
    {
      "epoch": 27.352941176470587,
      "grad_norm": 0.11010562628507614,
      "learning_rate": 1.7745098039215689e-06,
      "loss": 0.0126,
      "step": 1860
    },
    {
      "epoch": 27.5,
      "grad_norm": 0.11507220566272736,
      "learning_rate": 1.676470588235294e-06,
      "loss": 0.0134,
      "step": 1870
    },
    {
      "epoch": 27.647058823529413,
      "grad_norm": 0.10667148232460022,
      "learning_rate": 1.5784313725490197e-06,
      "loss": 0.0703,
      "step": 1880
    },
    {
      "epoch": 27.794117647058822,
      "grad_norm": 0.09035525470972061,
      "learning_rate": 1.4803921568627452e-06,
      "loss": 0.0127,
      "step": 1890
    },
    {
      "epoch": 27.941176470588236,
      "grad_norm": 0.13801293075084686,
      "learning_rate": 1.3823529411764708e-06,
      "loss": 0.0127,
      "step": 1900
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.151557058095932,
      "eval_runtime": 1.6327,
      "eval_samples_per_second": 82.687,
      "eval_steps_per_second": 10.412,
      "step": 1904
    },
    {
      "epoch": 28.08823529411765,
      "grad_norm": 0.12456930428743362,
      "learning_rate": 1.284313725490196e-06,
      "loss": 0.0121,
      "step": 1910
    },
    {
      "epoch": 28.235294117647058,
      "grad_norm": 0.1714206337928772,
      "learning_rate": 1.1862745098039217e-06,
      "loss": 0.0143,
      "step": 1920
    },
    {
      "epoch": 28.38235294117647,
      "grad_norm": 0.09358106553554535,
      "learning_rate": 1.088235294117647e-06,
      "loss": 0.0124,
      "step": 1930
    },
    {
      "epoch": 28.529411764705884,
      "grad_norm": 0.12065358459949493,
      "learning_rate": 9.901960784313725e-07,
      "loss": 0.013,
      "step": 1940
    },
    {
      "epoch": 28.676470588235293,
      "grad_norm": 0.10697101801633835,
      "learning_rate": 8.921568627450982e-07,
      "loss": 0.0128,
      "step": 1950
    },
    {
      "epoch": 28.823529411764707,
      "grad_norm": 0.11727521568536758,
      "learning_rate": 7.941176470588236e-07,
      "loss": 0.0121,
      "step": 1960
    },
    {
      "epoch": 28.970588235294116,
      "grad_norm": 0.10474637895822525,
      "learning_rate": 6.960784313725491e-07,
      "loss": 0.0548,
      "step": 1970
    },
    {
      "epoch": 29.0,
      "eval_loss": 0.15194356441497803,
      "eval_runtime": 1.6187,
      "eval_samples_per_second": 83.399,
      "eval_steps_per_second": 10.502,
      "step": 1972
    },
    {
      "epoch": 29.11764705882353,
      "grad_norm": 0.11856939643621445,
      "learning_rate": 5.980392156862745e-07,
      "loss": 0.0121,
      "step": 1980
    },
    {
      "epoch": 29.264705882352942,
      "grad_norm": 0.12045503407716751,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.0128,
      "step": 1990
    },
    {
      "epoch": 29.41176470588235,
      "grad_norm": 0.11332525312900543,
      "learning_rate": 4.0196078431372556e-07,
      "loss": 0.0119,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 2040,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4156813161449472.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}