{
  "best_metric": 0.986103355884552,
  "best_model_checkpoint": "runs/deepseek_lora_20240422-095359/checkpoint-2500",
  "epoch": 0.05788142411455891,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 11.34415054321289,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.2669,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 7.290144443511963,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.3878,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 5.349151134490967,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.1855,
      "step": 30
    },
    {
      "epoch": 0.0,
      "grad_norm": 14.503132820129395,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.0839,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.504097938537598,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.3778,
      "step": 50
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.759912490844727,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.2921,
      "step": 60
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.052595138549805,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 1.2835,
      "step": 70
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.416482925415039,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.1876,
      "step": 80
    },
    {
      "epoch": 0.0,
      "grad_norm": 8.467272758483887,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.0513,
      "step": 90
    },
    {
      "epoch": 0.0,
      "grad_norm": 7.2040510177612305,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1429,
      "step": 100
    },
    {
      "epoch": 0.0,
      "grad_norm": 22.72906494140625,
      "learning_rate": 4.4e-06,
      "loss": 1.344,
      "step": 110
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.6358766555786133,
      "learning_rate": 4.800000000000001e-06,
      "loss": 1.1027,
      "step": 120
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.7037551403045654,
      "learning_rate": 5.2e-06,
      "loss": 1.1483,
      "step": 130
    },
    {
      "epoch": 0.0,
      "grad_norm": 7.809167861938477,
      "learning_rate": 5.600000000000001e-06,
      "loss": 1.2556,
      "step": 140
    },
    {
      "epoch": 0.0,
      "grad_norm": 10.522435188293457,
      "learning_rate": 6e-06,
      "loss": 1.3272,
      "step": 150
    },
    {
      "epoch": 0.0,
      "grad_norm": 12.723409652709961,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 1.1342,
      "step": 160
    },
    {
      "epoch": 0.0,
      "grad_norm": 9.65882682800293,
      "learning_rate": 6.800000000000001e-06,
      "loss": 1.1509,
      "step": 170
    },
    {
      "epoch": 0.0,
      "grad_norm": 7.01952600479126,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 1.3069,
      "step": 180
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.8367834091186523,
      "learning_rate": 7.600000000000001e-06,
      "loss": 1.1551,
      "step": 190
    },
    {
      "epoch": 0.0,
      "grad_norm": 6.172680854797363,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4156,
      "step": 200
    },
    {
      "epoch": 0.0,
      "grad_norm": 13.904732704162598,
      "learning_rate": 8.400000000000001e-06,
      "loss": 1.2325,
      "step": 210
    },
    {
      "epoch": 0.01,
      "grad_norm": 7.113356590270996,
      "learning_rate": 8.8e-06,
      "loss": 1.142,
      "step": 220
    },
    {
      "epoch": 0.01,
      "grad_norm": 14.426301956176758,
      "learning_rate": 9.200000000000002e-06,
      "loss": 1.2648,
      "step": 230
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.00722885131836,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.3083,
      "step": 240
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.72845983505249,
      "learning_rate": 1e-05,
      "loss": 1.1742,
      "step": 250
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.7809460163116455,
      "learning_rate": 1.04e-05,
      "loss": 1.2009,
      "step": 260
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.464003562927246,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 1.1236,
      "step": 270
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.139829158782959,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 1.412,
      "step": 280
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.483891487121582,
      "learning_rate": 1.16e-05,
      "loss": 1.2858,
      "step": 290
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.432833671569824,
      "learning_rate": 1.2e-05,
      "loss": 1.2636,
      "step": 300
    },
    {
      "epoch": 0.01,
      "grad_norm": 8.610489845275879,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 1.3628,
      "step": 310
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.597244739532471,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 1.3585,
      "step": 320
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.667290687561035,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 1.4112,
      "step": 330
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.102590084075928,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 1.2255,
      "step": 340
    },
    {
      "epoch": 0.01,
      "grad_norm": 15.342700958251953,
      "learning_rate": 1.4e-05,
      "loss": 1.2915,
      "step": 350
    },
    {
      "epoch": 0.01,
      "grad_norm": 13.78890323638916,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 1.5371,
      "step": 360
    },
    {
      "epoch": 0.01,
      "grad_norm": 13.456836700439453,
      "learning_rate": 1.48e-05,
      "loss": 1.0586,
      "step": 370
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.98771071434021,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 1.3475,
      "step": 380
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.61796236038208,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 1.1562,
      "step": 390
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.871325492858887,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.3575,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.282336235046387,
      "learning_rate": 1.64e-05,
      "loss": 1.3261,
      "step": 410
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.444363594055176,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 1.0091,
      "step": 420
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7336666584014893,
      "learning_rate": 1.72e-05,
      "loss": 1.3492,
      "step": 430
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.654507637023926,
      "learning_rate": 1.76e-05,
      "loss": 1.2701,
      "step": 440
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.83292555809021,
      "learning_rate": 1.8e-05,
      "loss": 1.1073,
      "step": 450
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.379516124725342,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 1.2015,
      "step": 460
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.365577220916748,
      "learning_rate": 1.88e-05,
      "loss": 1.1174,
      "step": 470
    },
    {
      "epoch": 0.01,
      "grad_norm": 11.620206832885742,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 1.3091,
      "step": 480
    },
    {
      "epoch": 0.01,
      "grad_norm": 10.400999069213867,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.98,
      "step": 490
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.732550144195557,
      "learning_rate": 2e-05,
      "loss": 1.3638,
      "step": 500
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.0330229997634888,
      "eval_runtime": 66.9668,
      "eval_samples_per_second": 14.933,
      "eval_steps_per_second": 14.933,
      "step": 500
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.0168514251708984,
      "learning_rate": 1.9955555555555557e-05,
      "loss": 1.2762,
      "step": 510
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.4872841835021973,
      "learning_rate": 1.9911111111111112e-05,
      "loss": 1.3287,
      "step": 520
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.1740336418151855,
      "learning_rate": 1.9866666666666667e-05,
      "loss": 1.2577,
      "step": 530
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.3551251888275146,
      "learning_rate": 1.9822222222222226e-05,
      "loss": 1.1244,
      "step": 540
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.230109691619873,
      "learning_rate": 1.977777777777778e-05,
      "loss": 1.2021,
      "step": 550
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.284055709838867,
      "learning_rate": 1.9733333333333336e-05,
      "loss": 1.2919,
      "step": 560
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.5889816284179688,
      "learning_rate": 1.968888888888889e-05,
      "loss": 1.3648,
      "step": 570
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.799360752105713,
      "learning_rate": 1.9644444444444447e-05,
      "loss": 1.396,
      "step": 580
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.2582507133483887,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 1.2263,
      "step": 590
    },
    {
      "epoch": 0.01,
      "grad_norm": 9.519213676452637,
      "learning_rate": 1.9555555555555557e-05,
      "loss": 1.1475,
      "step": 600
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.0547032356262207,
      "learning_rate": 1.9511111111111113e-05,
      "loss": 1.4278,
      "step": 610
    },
    {
      "epoch": 0.01,
      "grad_norm": 9.448994636535645,
      "learning_rate": 1.9466666666666668e-05,
      "loss": 1.3394,
      "step": 620
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.21259069442749,
      "learning_rate": 1.9422222222222223e-05,
      "loss": 1.206,
      "step": 630
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.4361753463745117,
      "learning_rate": 1.9377777777777778e-05,
      "loss": 1.2472,
      "step": 640
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.0469722747802734,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 1.1283,
      "step": 650
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.071639060974121,
      "learning_rate": 1.928888888888889e-05,
      "loss": 1.2191,
      "step": 660
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.428431272506714,
      "learning_rate": 1.9244444444444444e-05,
      "loss": 1.3863,
      "step": 670
    },
    {
      "epoch": 0.02,
      "grad_norm": 7.4005656242370605,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 1.0794,
      "step": 680
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.636923313140869,
      "learning_rate": 1.9155555555555558e-05,
      "loss": 1.3761,
      "step": 690
    },
    {
      "epoch": 0.02,
      "grad_norm": 5.060346603393555,
      "learning_rate": 1.9111111111111113e-05,
      "loss": 1.4087,
      "step": 700
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.77576208114624,
      "learning_rate": 1.9066666666666668e-05,
      "loss": 1.2442,
      "step": 710
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.3169615268707275,
      "learning_rate": 1.9022222222222223e-05,
      "loss": 1.2498,
      "step": 720
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.9390623569488525,
      "learning_rate": 1.897777777777778e-05,
      "loss": 1.4098,
      "step": 730
    },
    {
      "epoch": 0.02,
      "grad_norm": 5.928336143493652,
      "learning_rate": 1.8933333333333334e-05,
      "loss": 1.235,
      "step": 740
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.329615592956543,
      "learning_rate": 1.888888888888889e-05,
      "loss": 1.2166,
      "step": 750
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.106197357177734,
      "learning_rate": 1.8844444444444444e-05,
      "loss": 1.3728,
      "step": 760
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.729337215423584,
      "learning_rate": 1.88e-05,
      "loss": 1.2131,
      "step": 770
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.116116762161255,
      "learning_rate": 1.8755555555555558e-05,
      "loss": 1.4169,
      "step": 780
    },
    {
      "epoch": 0.02,
      "grad_norm": 8.869202613830566,
      "learning_rate": 1.8711111111111113e-05,
      "loss": 1.2377,
      "step": 790
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.858852863311768,
      "learning_rate": 1.866666666666667e-05,
      "loss": 1.3863,
      "step": 800
    },
    {
      "epoch": 0.02,
      "grad_norm": 10.197395324707031,
      "learning_rate": 1.8622222222222224e-05,
      "loss": 1.1053,
      "step": 810
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.7740931510925293,
      "learning_rate": 1.857777777777778e-05,
      "loss": 1.41,
      "step": 820
    },
    {
      "epoch": 0.02,
      "grad_norm": 8.306866645812988,
      "learning_rate": 1.8533333333333334e-05,
      "loss": 1.2661,
      "step": 830
    },
    {
      "epoch": 0.02,
      "grad_norm": 5.462616920471191,
      "learning_rate": 1.848888888888889e-05,
      "loss": 1.1217,
      "step": 840
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.6351518630981445,
      "learning_rate": 1.8444444444444448e-05,
      "loss": 1.2512,
      "step": 850
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.930731773376465,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 1.3136,
      "step": 860
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.907737731933594,
      "learning_rate": 1.835555555555556e-05,
      "loss": 1.1728,
      "step": 870
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.2834455966949463,
      "learning_rate": 1.8311111111111114e-05,
      "loss": 1.3566,
      "step": 880
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.938192367553711,
      "learning_rate": 1.826666666666667e-05,
      "loss": 1.2879,
      "step": 890
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.4376509189605713,
      "learning_rate": 1.8222222222222224e-05,
      "loss": 1.1971,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.437848091125488,
      "learning_rate": 1.817777777777778e-05,
      "loss": 1.3583,
      "step": 910
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.843893051147461,
      "learning_rate": 1.8133333333333335e-05,
      "loss": 1.2323,
      "step": 920
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.2527034282684326,
      "learning_rate": 1.808888888888889e-05,
      "loss": 1.2691,
      "step": 930
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.6528022289276123,
      "learning_rate": 1.8044444444444445e-05,
      "loss": 1.2525,
      "step": 940
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.8804101943969727,
      "learning_rate": 1.8e-05,
      "loss": 1.3387,
      "step": 950
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.754573822021484,
      "learning_rate": 1.7955555555555556e-05,
      "loss": 1.265,
      "step": 960
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.54309344291687,
      "learning_rate": 1.791111111111111e-05,
      "loss": 1.0998,
      "step": 970
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.926447868347168,
      "learning_rate": 1.7866666666666666e-05,
      "loss": 1.1927,
      "step": 980
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.4870805740356445,
      "learning_rate": 1.782222222222222e-05,
      "loss": 1.0565,
      "step": 990
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.2183382511138916,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 1.2572,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_loss": 1.019097089767456,
      "eval_runtime": 66.8787,
      "eval_samples_per_second": 14.952,
      "eval_steps_per_second": 14.952,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.61516284942627,
      "learning_rate": 1.7733333333333335e-05,
      "loss": 1.4306,
      "step": 1010
    },
    {
      "epoch": 0.02,
      "grad_norm": 10.755435943603516,
      "learning_rate": 1.768888888888889e-05,
      "loss": 1.3042,
      "step": 1020
    },
    {
      "epoch": 0.02,
      "grad_norm": 11.427221298217773,
      "learning_rate": 1.7644444444444446e-05,
      "loss": 1.3201,
      "step": 1030
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.656847953796387,
      "learning_rate": 1.76e-05,
      "loss": 1.2579,
      "step": 1040
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.030166149139404,
      "learning_rate": 1.7555555555555556e-05,
      "loss": 1.4462,
      "step": 1050
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.84343957901001,
      "learning_rate": 1.751111111111111e-05,
      "loss": 1.4213,
      "step": 1060
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.4579355716705322,
      "learning_rate": 1.7466666666666667e-05,
      "loss": 1.2673,
      "step": 1070
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.692587375640869,
      "learning_rate": 1.7422222222222222e-05,
      "loss": 1.1579,
      "step": 1080
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.579308032989502,
      "learning_rate": 1.737777777777778e-05,
      "loss": 1.3278,
      "step": 1090
    },
    {
      "epoch": 0.03,
      "grad_norm": 7.755845069885254,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 1.3471,
      "step": 1100
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.065462112426758,
      "learning_rate": 1.728888888888889e-05,
      "loss": 1.2811,
      "step": 1110
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.6691714525222778,
      "learning_rate": 1.7244444444444446e-05,
      "loss": 1.4433,
      "step": 1120
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.87007474899292,
      "learning_rate": 1.72e-05,
      "loss": 1.1916,
      "step": 1130
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.663476586341858,
      "learning_rate": 1.7155555555555557e-05,
      "loss": 1.3006,
      "step": 1140
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.493046522140503,
      "learning_rate": 1.7111111111111112e-05,
      "loss": 1.3099,
      "step": 1150
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.2659084796905518,
      "learning_rate": 1.706666666666667e-05,
      "loss": 1.1553,
      "step": 1160
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.07832670211792,
      "learning_rate": 1.7022222222222226e-05,
      "loss": 1.2604,
      "step": 1170
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.0068790912628174,
      "learning_rate": 1.697777777777778e-05,
      "loss": 0.9679,
      "step": 1180
    },
    {
      "epoch": 0.03,
      "grad_norm": 14.066789627075195,
      "learning_rate": 1.6933333333333336e-05,
      "loss": 1.2291,
      "step": 1190
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.058719158172607,
      "learning_rate": 1.688888888888889e-05,
      "loss": 1.3224,
      "step": 1200
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.309176445007324,
      "learning_rate": 1.6844444444444447e-05,
      "loss": 1.1184,
      "step": 1210
    },
    {
      "epoch": 0.03,
      "grad_norm": 9.05258846282959,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 1.345,
      "step": 1220
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.345346927642822,
      "learning_rate": 1.6755555555555557e-05,
      "loss": 1.2663,
      "step": 1230
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.119368076324463,
      "learning_rate": 1.6711111111111112e-05,
      "loss": 1.1441,
      "step": 1240
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.0422563552856445,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.2199,
      "step": 1250
    },
    {
      "epoch": 0.03,
      "grad_norm": 11.27535343170166,
      "learning_rate": 1.6622222222222223e-05,
      "loss": 1.1876,
      "step": 1260
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.767408847808838,
      "learning_rate": 1.6577777777777778e-05,
      "loss": 1.3928,
      "step": 1270
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.1706862449646,
      "learning_rate": 1.6533333333333333e-05,
      "loss": 1.2268,
      "step": 1280
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.185644149780273,
      "learning_rate": 1.648888888888889e-05,
      "loss": 1.3086,
      "step": 1290
    },
    {
      "epoch": 0.03,
      "grad_norm": 9.57487678527832,
      "learning_rate": 1.6444444444444444e-05,
      "loss": 1.2079,
      "step": 1300
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.104099273681641,
      "learning_rate": 1.64e-05,
      "loss": 1.2012,
      "step": 1310
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.703191757202148,
      "learning_rate": 1.6355555555555557e-05,
      "loss": 1.2567,
      "step": 1320
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.194169521331787,
      "learning_rate": 1.6311111111111113e-05,
      "loss": 1.3687,
      "step": 1330
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.2333269119262695,
      "learning_rate": 1.6266666666666668e-05,
      "loss": 1.1547,
      "step": 1340
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.007706880569458,
      "learning_rate": 1.6222222222222223e-05,
      "loss": 1.2031,
      "step": 1350
    },
    {
      "epoch": 0.03,
      "grad_norm": 22.12256622314453,
      "learning_rate": 1.617777777777778e-05,
      "loss": 1.1117,
      "step": 1360
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.559662818908691,
      "learning_rate": 1.6133333333333334e-05,
      "loss": 1.0796,
      "step": 1370
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.618852615356445,
      "learning_rate": 1.608888888888889e-05,
      "loss": 1.3308,
      "step": 1380
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.6262142658233643,
      "learning_rate": 1.6044444444444444e-05,
      "loss": 1.1913,
      "step": 1390
    },
    {
      "epoch": 0.03,
      "grad_norm": 8.0167875289917,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.341,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.142825603485107,
      "learning_rate": 1.5955555555555558e-05,
      "loss": 1.03,
      "step": 1410
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.186448335647583,
      "learning_rate": 1.5911111111111113e-05,
      "loss": 1.2788,
      "step": 1420
    },
    {
      "epoch": 0.03,
      "grad_norm": 8.11291217803955,
      "learning_rate": 1.586666666666667e-05,
      "loss": 1.4419,
      "step": 1430
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.920192718505859,
      "learning_rate": 1.5822222222222224e-05,
      "loss": 1.3127,
      "step": 1440
    },
    {
      "epoch": 0.03,
      "grad_norm": 4.357940196990967,
      "learning_rate": 1.577777777777778e-05,
      "loss": 1.3184,
      "step": 1450
    },
    {
      "epoch": 0.03,
      "grad_norm": 8.83785343170166,
      "learning_rate": 1.5733333333333334e-05,
      "loss": 1.382,
      "step": 1460
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.565369129180908,
      "learning_rate": 1.5688888888888893e-05,
      "loss": 1.4148,
      "step": 1470
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.819150686264038,
      "learning_rate": 1.5644444444444448e-05,
      "loss": 1.4428,
      "step": 1480
    },
    {
      "epoch": 0.03,
      "grad_norm": 7.107534408569336,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 1.2188,
      "step": 1490
    },
    {
      "epoch": 0.03,
      "grad_norm": 9.566027641296387,
      "learning_rate": 1.555555555555556e-05,
      "loss": 1.2406,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "eval_loss": 1.0252078771591187,
      "eval_runtime": 66.9753,
      "eval_samples_per_second": 14.931,
      "eval_steps_per_second": 14.931,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "grad_norm": 8.054425239562988,
      "learning_rate": 1.5511111111111114e-05,
      "loss": 1.2767,
      "step": 1510
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.8494319915771484,
      "learning_rate": 1.546666666666667e-05,
      "loss": 1.1632,
      "step": 1520
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.9305810928344727,
      "learning_rate": 1.5422222222222224e-05,
      "loss": 1.2075,
      "step": 1530
    },
    {
      "epoch": 0.04,
      "grad_norm": 11.442655563354492,
      "learning_rate": 1.537777777777778e-05,
      "loss": 1.0231,
      "step": 1540
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.735372543334961,
      "learning_rate": 1.5333333333333334e-05,
      "loss": 1.282,
      "step": 1550
    },
    {
      "epoch": 0.04,
      "grad_norm": 11.126168251037598,
      "learning_rate": 1.528888888888889e-05,
      "loss": 1.2106,
      "step": 1560
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.569930076599121,
      "learning_rate": 1.5244444444444447e-05,
      "loss": 1.2016,
      "step": 1570
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.002272605895996,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 1.206,
      "step": 1580
    },
    {
      "epoch": 0.04,
      "grad_norm": 11.26425838470459,
      "learning_rate": 1.5155555555555557e-05,
      "loss": 1.3145,
      "step": 1590
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.265772819519043,
      "learning_rate": 1.5111111111111112e-05,
      "loss": 1.105,
      "step": 1600
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.139275550842285,
      "learning_rate": 1.5066666666666668e-05,
      "loss": 1.2016,
      "step": 1610
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.753066539764404,
      "learning_rate": 1.5022222222222223e-05,
      "loss": 1.3433,
      "step": 1620
    },
    {
      "epoch": 0.04,
      "grad_norm": 8.761942863464355,
      "learning_rate": 1.497777777777778e-05,
      "loss": 1.3153,
      "step": 1630
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.3448381423950195,
      "learning_rate": 1.4933333333333335e-05,
      "loss": 1.24,
      "step": 1640
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.818711519241333,
      "learning_rate": 1.488888888888889e-05,
      "loss": 1.2067,
      "step": 1650
    },
    {
      "epoch": 0.04,
      "grad_norm": 11.795276641845703,
      "learning_rate": 1.4844444444444445e-05,
      "loss": 1.3926,
      "step": 1660
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.47786283493042,
      "learning_rate": 1.48e-05,
      "loss": 1.5349,
      "step": 1670
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.560647010803223,
      "learning_rate": 1.4755555555555556e-05,
      "loss": 1.2318,
      "step": 1680
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.564818382263184,
      "learning_rate": 1.4711111111111111e-05,
      "loss": 1.2044,
      "step": 1690
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.4382556676864624,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 1.3367,
      "step": 1700
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.7976467609405518,
      "learning_rate": 1.4622222222222225e-05,
      "loss": 1.2304,
      "step": 1710
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.021933078765869,
      "learning_rate": 1.457777777777778e-05,
      "loss": 1.1339,
      "step": 1720
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.9026000499725342,
      "learning_rate": 1.4533333333333335e-05,
      "loss": 1.3733,
      "step": 1730
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.237542152404785,
      "learning_rate": 1.448888888888889e-05,
      "loss": 1.0956,
      "step": 1740
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.288679599761963,
      "learning_rate": 1.4444444444444446e-05,
      "loss": 1.0542,
      "step": 1750
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.956302642822266,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 1.326,
      "step": 1760
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.096738338470459,
      "learning_rate": 1.4355555555555556e-05,
      "loss": 1.1285,
      "step": 1770
    },
    {
      "epoch": 0.04,
      "grad_norm": 7.757137298583984,
      "learning_rate": 1.4311111111111111e-05,
      "loss": 1.3607,
      "step": 1780
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.415702819824219,
      "learning_rate": 1.4266666666666668e-05,
      "loss": 1.0,
      "step": 1790
    },
    {
      "epoch": 0.04,
      "grad_norm": 9.445133209228516,
      "learning_rate": 1.4222222222222224e-05,
      "loss": 1.2187,
      "step": 1800
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.406456470489502,
      "learning_rate": 1.4177777777777779e-05,
      "loss": 1.3459,
      "step": 1810
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.619770884513855,
      "learning_rate": 1.4133333333333334e-05,
      "loss": 1.2368,
      "step": 1820
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.307522296905518,
      "learning_rate": 1.408888888888889e-05,
      "loss": 1.0995,
      "step": 1830
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.3472185134887695,
      "learning_rate": 1.4044444444444445e-05,
      "loss": 1.2563,
      "step": 1840
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.04168701171875,
      "learning_rate": 1.4e-05,
      "loss": 1.2858,
      "step": 1850
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.180267810821533,
      "learning_rate": 1.3955555555555558e-05,
      "loss": 1.123,
      "step": 1860
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.560042381286621,
      "learning_rate": 1.3911111111111114e-05,
      "loss": 1.3962,
      "step": 1870
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.683982849121094,
      "learning_rate": 1.3866666666666669e-05,
      "loss": 1.0984,
      "step": 1880
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.1994190216064453,
      "learning_rate": 1.3822222222222224e-05,
      "loss": 1.1746,
      "step": 1890
    },
    {
      "epoch": 0.04,
      "grad_norm": 8.851926803588867,
      "learning_rate": 1.377777777777778e-05,
      "loss": 1.4551,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.670786142349243,
      "learning_rate": 1.3733333333333335e-05,
      "loss": 1.2446,
      "step": 1910
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.134795188903809,
      "learning_rate": 1.368888888888889e-05,
      "loss": 1.2128,
      "step": 1920
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.486630916595459,
      "learning_rate": 1.3644444444444445e-05,
      "loss": 1.1951,
      "step": 1930
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.025615215301514,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 1.3628,
      "step": 1940
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.232900619506836,
      "learning_rate": 1.3555555555555557e-05,
      "loss": 1.3367,
      "step": 1950
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.6469148397445679,
      "learning_rate": 1.3511111111111112e-05,
      "loss": 1.2144,
      "step": 1960
    },
    {
      "epoch": 0.05,
      "grad_norm": 8.560945510864258,
      "learning_rate": 1.3466666666666668e-05,
      "loss": 1.4125,
      "step": 1970
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.06697416305542,
      "learning_rate": 1.3422222222222223e-05,
      "loss": 1.1769,
      "step": 1980
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.721186399459839,
      "learning_rate": 1.3377777777777778e-05,
      "loss": 1.3266,
      "step": 1990
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.427524566650391,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 1.1697,
      "step": 2000
    },
    {
      "epoch": 0.05,
      "eval_loss": 1.042366623878479,
      "eval_runtime": 67.0143,
      "eval_samples_per_second": 14.922,
      "eval_steps_per_second": 14.922,
      "step": 2000
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.322222113609314,
      "learning_rate": 1.3288888888888889e-05,
      "loss": 1.3173,
      "step": 2010
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.731060028076172,
      "learning_rate": 1.3244444444444447e-05,
      "loss": 1.2578,
      "step": 2020
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8411041498184204,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 1.1796,
      "step": 2030
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.4170026779174805,
      "learning_rate": 1.3155555555555558e-05,
      "loss": 1.3608,
      "step": 2040
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.58616304397583,
      "learning_rate": 1.3111111111111113e-05,
      "loss": 1.2726,
      "step": 2050
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.936751365661621,
      "learning_rate": 1.3066666666666668e-05,
      "loss": 1.3327,
      "step": 2060
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.1074042320251465,
      "learning_rate": 1.3022222222222223e-05,
      "loss": 1.2446,
      "step": 2070
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.657953977584839,
      "learning_rate": 1.2977777777777779e-05,
      "loss": 0.9989,
      "step": 2080
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.181591987609863,
      "learning_rate": 1.2933333333333334e-05,
      "loss": 1.1732,
      "step": 2090
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.981390953063965,
      "learning_rate": 1.288888888888889e-05,
      "loss": 1.1463,
      "step": 2100
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.05696964263916,
      "learning_rate": 1.2844444444444446e-05,
      "loss": 1.2061,
      "step": 2110
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.697308540344238,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 1.3052,
      "step": 2120
    },
    {
      "epoch": 0.05,
      "grad_norm": 14.833478927612305,
      "learning_rate": 1.2755555555555556e-05,
      "loss": 1.1987,
      "step": 2130
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.479732990264893,
      "learning_rate": 1.2711111111111112e-05,
      "loss": 1.337,
      "step": 2140
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.746943235397339,
      "learning_rate": 1.2666666666666667e-05,
      "loss": 1.3082,
      "step": 2150
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.7201828956604,
      "learning_rate": 1.2622222222222222e-05,
      "loss": 1.1393,
      "step": 2160
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.082671642303467,
      "learning_rate": 1.257777777777778e-05,
      "loss": 1.3047,
      "step": 2170
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.5268049240112305,
      "learning_rate": 1.2533333333333336e-05,
      "loss": 1.1255,
      "step": 2180
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.8856544494628906,
      "learning_rate": 1.2488888888888891e-05,
      "loss": 1.4022,
      "step": 2190
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.552175521850586,
      "learning_rate": 1.2444444444444446e-05,
      "loss": 1.3444,
      "step": 2200
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.772660255432129,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 1.4274,
      "step": 2210
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.9137773513793945,
      "learning_rate": 1.2355555555555557e-05,
      "loss": 1.1356,
      "step": 2220
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.5068724155426025,
      "learning_rate": 1.2311111111111112e-05,
      "loss": 1.25,
      "step": 2230
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8587338924407959,
      "learning_rate": 1.2266666666666667e-05,
      "loss": 0.9182,
      "step": 2240
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.356514930725098,
      "learning_rate": 1.2222222222222224e-05,
      "loss": 1.2882,
      "step": 2250
    },
    {
      "epoch": 0.05,
      "grad_norm": 8.728981018066406,
      "learning_rate": 1.217777777777778e-05,
      "loss": 1.2974,
      "step": 2260
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.5229454040527344,
      "learning_rate": 1.2133333333333335e-05,
      "loss": 1.2368,
      "step": 2270
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.760233163833618,
      "learning_rate": 1.208888888888889e-05,
      "loss": 1.0504,
      "step": 2280
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.7355753183364868,
      "learning_rate": 1.2044444444444445e-05,
      "loss": 1.3547,
      "step": 2290
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.2375967502593994,
      "learning_rate": 1.2e-05,
      "loss": 1.3058,
      "step": 2300
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.5727386474609375,
      "learning_rate": 1.1955555555555556e-05,
      "loss": 1.3097,
      "step": 2310
    },
    {
      "epoch": 0.05,
      "grad_norm": 7.387458801269531,
      "learning_rate": 1.191111111111111e-05,
      "loss": 1.2058,
      "step": 2320
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.593174934387207,
      "learning_rate": 1.186666666666667e-05,
      "loss": 1.1711,
      "step": 2330
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.3077690601348877,
      "learning_rate": 1.1822222222222225e-05,
      "loss": 1.1175,
      "step": 2340
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.785140514373779,
      "learning_rate": 1.177777777777778e-05,
      "loss": 1.2618,
      "step": 2350
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.8957252502441406,
      "learning_rate": 1.1733333333333335e-05,
      "loss": 1.1641,
      "step": 2360
    },
    {
      "epoch": 0.05,
      "grad_norm": 7.033701419830322,
      "learning_rate": 1.168888888888889e-05,
      "loss": 1.3563,
      "step": 2370
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.6566953659057617,
      "learning_rate": 1.1644444444444446e-05,
      "loss": 1.3673,
      "step": 2380
    },
    {
      "epoch": 0.06,
      "grad_norm": 10.329083442687988,
      "learning_rate": 1.16e-05,
      "loss": 1.2953,
      "step": 2390
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.5499980449676514,
      "learning_rate": 1.1555555555555556e-05,
      "loss": 1.3136,
      "step": 2400
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.051661968231201,
      "learning_rate": 1.1511111111111113e-05,
      "loss": 1.3235,
      "step": 2410
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.9961462020874023,
      "learning_rate": 1.1466666666666668e-05,
      "loss": 1.1637,
      "step": 2420
    },
    {
      "epoch": 0.06,
      "grad_norm": 8.53834056854248,
      "learning_rate": 1.1422222222222223e-05,
      "loss": 1.1066,
      "step": 2430
    },
    {
      "epoch": 0.06,
      "grad_norm": 5.684910297393799,
      "learning_rate": 1.1377777777777779e-05,
      "loss": 1.4476,
      "step": 2440
    },
    {
      "epoch": 0.06,
      "grad_norm": 16.79145622253418,
      "learning_rate": 1.1333333333333334e-05,
      "loss": 1.3978,
      "step": 2450
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.3043346405029297,
      "learning_rate": 1.1288888888888889e-05,
      "loss": 1.1598,
      "step": 2460
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.7516391277313232,
      "learning_rate": 1.1244444444444444e-05,
      "loss": 1.3534,
      "step": 2470
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.0643539428710938,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 1.2924,
      "step": 2480
    },
    {
      "epoch": 0.06,
      "grad_norm": 4.207937717437744,
      "learning_rate": 1.1155555555555556e-05,
      "loss": 1.2699,
      "step": 2490
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.3195173740386963,
      "learning_rate": 1.1111111111111113e-05,
      "loss": 1.379,
      "step": 2500
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.986103355884552,
      "eval_runtime": 67.0933,
      "eval_samples_per_second": 14.905,
      "eval_steps_per_second": 14.905,
      "step": 2500
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2500,
  "total_flos": 4.025531498496e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}