[
  {
    "loss": 0.4904,
    "grad_norm": 0.5958866477012634,
    "learning_rate": 1.9944444444444447e-05,
    "epoch": 0.03,
    "step": 50
  },
  {
    "loss": 0.3657,
    "grad_norm": 0.4602122902870178,
    "learning_rate": 1.988888888888889e-05,
    "epoch": 0.07,
    "step": 100
  },
  {
    "loss": 0.3182,
    "grad_norm": 0.6156640648841858,
    "learning_rate": 1.9833333333333335e-05,
    "epoch": 0.1,
    "step": 150
  },
  {
    "loss": 0.275,
    "grad_norm": 0.4630753695964813,
    "learning_rate": 1.977777777777778e-05,
    "epoch": 0.13,
    "step": 200
  },
  {
    "loss": 0.252,
    "grad_norm": 0.6395113468170166,
    "learning_rate": 1.9722222222222224e-05,
    "epoch": 0.17,
    "step": 250
  },
  {
    "loss": 0.2434,
    "grad_norm": 0.4501757025718689,
    "learning_rate": 1.9666666666666666e-05,
    "epoch": 0.2,
    "step": 300
  },
  {
    "loss": 0.2208,
    "grad_norm": 0.6610535383224487,
    "learning_rate": 1.961111111111111e-05,
    "epoch": 0.23,
    "step": 350
  },
  {
    "loss": 0.2192,
    "grad_norm": 0.40222012996673584,
    "learning_rate": 1.9555555555555557e-05,
    "epoch": 0.27,
    "step": 400
  },
  {
    "loss": 0.2138,
    "grad_norm": 0.7299392223358154,
    "learning_rate": 1.95e-05,
    "epoch": 0.3,
    "step": 450
  },
  {
    "loss": 0.2066,
    "grad_norm": 0.5643211007118225,
    "learning_rate": 1.9444444444444445e-05,
    "epoch": 0.33,
    "step": 500
  },
  {
    "loss": 0.2021,
    "grad_norm": 1.0369503498077393,
    "learning_rate": 1.938888888888889e-05,
    "epoch": 0.37,
    "step": 550
  },
  {
    "loss": 0.1908,
    "grad_norm": 0.37896671891212463,
    "learning_rate": 1.9333333333333333e-05,
    "epoch": 0.4,
    "step": 600
  },
  {
    "loss": 0.1831,
    "grad_norm": 0.3660513162612915,
    "learning_rate": 1.927777777777778e-05,
    "epoch": 0.43,
    "step": 650
  },
  {
    "loss": 0.1808,
    "grad_norm": 0.5333911776542664,
    "learning_rate": 1.9222222222222225e-05,
    "epoch": 0.47,
    "step": 700
  },
  {
    "loss": 0.1609,
    "grad_norm": 0.54070645570755,
    "learning_rate": 1.916666666666667e-05,
    "epoch": 0.5,
    "step": 750
  },
  {
    "loss": 0.1637,
    "grad_norm": 0.41102516651153564,
    "learning_rate": 1.9111111111111113e-05,
    "epoch": 0.53,
    "step": 800
  },
  {
    "loss": 0.1614,
    "grad_norm": 0.4941543936729431,
    "learning_rate": 1.9055555555555555e-05,
    "epoch": 0.57,
    "step": 850
  },
  {
    "loss": 0.1503,
    "grad_norm": 0.5441026091575623,
    "learning_rate": 1.9e-05,
    "epoch": 0.6,
    "step": 900
  },
  {
    "loss": 0.153,
    "grad_norm": 0.5017764568328857,
    "learning_rate": 1.8944444444444447e-05,
    "epoch": 0.63,
    "step": 950
  },
  {
    "loss": 0.1619,
    "grad_norm": 1.039005994796753,
    "learning_rate": 1.888888888888889e-05,
    "epoch": 0.67,
    "step": 1000
  },
  {
    "loss": 0.1571,
    "grad_norm": 0.401298463344574,
    "learning_rate": 1.8833333333333335e-05,
    "epoch": 0.7,
    "step": 1050
  },
  {
    "loss": 0.1549,
    "grad_norm": 0.5666447877883911,
    "learning_rate": 1.877777777777778e-05,
    "epoch": 0.73,
    "step": 1100
  },
  {
    "loss": 0.1458,
    "grad_norm": 0.6719347834587097,
    "learning_rate": 1.8722222222222223e-05,
    "epoch": 0.77,
    "step": 1150
  },
  {
    "loss": 0.1442,
    "grad_norm": 0.6617727875709534,
    "learning_rate": 1.866666666666667e-05,
    "epoch": 0.8,
    "step": 1200
  },
  {
    "loss": 0.1522,
    "grad_norm": 0.6959300637245178,
    "learning_rate": 1.8611111111111114e-05,
    "epoch": 0.83,
    "step": 1250
  },
  {
    "loss": 0.126,
    "grad_norm": 0.3736123740673065,
    "learning_rate": 1.8555555555555557e-05,
    "epoch": 0.87,
    "step": 1300
  },
  {
    "loss": 0.1373,
    "grad_norm": 0.20589721202850342,
    "learning_rate": 1.8500000000000002e-05,
    "epoch": 0.9,
    "step": 1350
  },
  {
    "loss": 0.138,
    "grad_norm": 0.9768441319465637,
    "learning_rate": 1.8444444444444448e-05,
    "epoch": 0.93,
    "step": 1400
  },
  {
    "loss": 0.1324,
    "grad_norm": 0.27331632375717163,
    "learning_rate": 1.838888888888889e-05,
    "epoch": 0.97,
    "step": 1450
  },
  {
    "loss": 0.1324,
    "grad_norm": 0.7121672034263611,
    "learning_rate": 1.8333333333333333e-05,
    "epoch": 1.0,
    "step": 1500
  },
  {
    "eval_loss": 0.12290950864553452,
    "eval_accuracy": 0.43533333333333335,
    "eval_f1": 0.8652792990142387,
    "eval_precision": 0.9544356230583362,
    "eval_recall": 0.7913566113337149,
    "eval_runtime": 403.441,
    "eval_samples_per_second": 3.718,
    "eval_steps_per_second": 0.466,
    "epoch": 1.0,
    "step": 1500
  },
  {
    "loss": 0.1187,
    "grad_norm": 0.5251692533493042,
    "learning_rate": 1.827777777777778e-05,
    "epoch": 1.03,
    "step": 1550
  },
  {
    "loss": 0.1219,
    "grad_norm": 0.7017450332641602,
    "learning_rate": 1.8222222222222224e-05,
    "epoch": 1.07,
    "step": 1600
  },
  {
    "loss": 0.1116,
    "grad_norm": 0.248159259557724,
    "learning_rate": 1.8166666666666667e-05,
    "epoch": 1.1,
    "step": 1650
  },
  {
    "loss": 0.127,
    "grad_norm": 0.9016281366348267,
    "learning_rate": 1.8111111111111112e-05,
    "epoch": 1.13,
    "step": 1700
  },
  {
    "loss": 0.1203,
    "grad_norm": 0.7193984389305115,
    "learning_rate": 1.8055555555555558e-05,
    "epoch": 1.17,
    "step": 1750
  },
  {
    "loss": 0.1195,
    "grad_norm": 0.7280067205429077,
    "learning_rate": 1.8e-05,
    "epoch": 1.2,
    "step": 1800
  },
  {
    "loss": 0.1094,
    "grad_norm": 0.906154453754425,
    "learning_rate": 1.7944444444444446e-05,
    "epoch": 1.23,
    "step": 1850
  },
  {
    "loss": 0.1155,
    "grad_norm": 0.4714917838573456,
    "learning_rate": 1.7888888888888892e-05,
    "epoch": 1.27,
    "step": 1900
  },
  {
    "loss": 0.1106,
    "grad_norm": 0.7589433789253235,
    "learning_rate": 1.7833333333333334e-05,
    "epoch": 1.3,
    "step": 1950
  },
  {
    "loss": 0.1052,
    "grad_norm": 0.5668336153030396,
    "learning_rate": 1.7777777777777777e-05,
    "epoch": 1.33,
    "step": 2000
  },
  {
    "loss": 0.1084,
    "grad_norm": 1.1015430688858032,
    "learning_rate": 1.7722222222222222e-05,
    "epoch": 1.37,
    "step": 2050
  },
  {
    "loss": 0.1017,
    "grad_norm": 0.23899418115615845,
    "learning_rate": 1.7666666666666668e-05,
    "epoch": 1.4,
    "step": 2100
  },
  {
    "loss": 0.1042,
    "grad_norm": 0.6596136689186096,
    "learning_rate": 1.7611111111111114e-05,
    "epoch": 1.43,
    "step": 2150
  },
  {
    "loss": 0.1054,
    "grad_norm": 0.5246695280075073,
    "learning_rate": 1.7555555555555556e-05,
    "epoch": 1.47,
    "step": 2200
  },
  {
    "loss": 0.103,
    "grad_norm": 0.7709386348724365,
    "learning_rate": 1.7500000000000002e-05,
    "epoch": 1.5,
    "step": 2250
  },
  {
    "loss": 0.0977,
    "grad_norm": 0.6520675420761108,
    "learning_rate": 1.7444444444444448e-05,
    "epoch": 1.53,
    "step": 2300
  },
  {
    "loss": 0.0931,
    "grad_norm": 0.44711801409721375,
    "learning_rate": 1.738888888888889e-05,
    "epoch": 1.57,
    "step": 2350
  },
  {
    "loss": 0.0937,
    "grad_norm": 0.7022482752799988,
    "learning_rate": 1.7333333333333336e-05,
    "epoch": 1.6,
    "step": 2400
  },
  {
    "loss": 0.0961,
    "grad_norm": 0.9846331477165222,
    "learning_rate": 1.727777777777778e-05,
    "epoch": 1.63,
    "step": 2450
  },
  {
    "loss": 0.0974,
    "grad_norm": 0.5135018229484558,
    "learning_rate": 1.7222222222222224e-05,
    "epoch": 1.67,
    "step": 2500
  },
  {
    "loss": 0.0822,
    "grad_norm": 0.47495317459106445,
    "learning_rate": 1.7166666666666666e-05,
    "epoch": 1.7,
    "step": 2550
  },
  {
    "loss": 0.0883,
    "grad_norm": 0.6149684190750122,
    "learning_rate": 1.7111111111111112e-05,
    "epoch": 1.73,
    "step": 2600
  },
  {
    "loss": 0.0932,
    "grad_norm": 0.85074782371521,
    "learning_rate": 1.7055555555555558e-05,
    "epoch": 1.77,
    "step": 2650
  },
  {
    "loss": 0.083,
    "grad_norm": 0.5721635222434998,
    "learning_rate": 1.7e-05,
    "epoch": 1.8,
    "step": 2700
  },
  {
    "loss": 0.0829,
    "grad_norm": 0.31835615634918213,
    "learning_rate": 1.6944444444444446e-05,
    "epoch": 1.83,
    "step": 2750
  },
  {
    "loss": 0.079,
    "grad_norm": 0.616396963596344,
    "learning_rate": 1.688888888888889e-05,
    "epoch": 1.87,
    "step": 2800
  },
  {
    "loss": 0.0855,
    "grad_norm": 0.711447536945343,
    "learning_rate": 1.6833333333333334e-05,
    "epoch": 1.9,
    "step": 2850
  },
  {
    "loss": 0.0895,
    "grad_norm": 0.48665234446525574,
    "learning_rate": 1.677777777777778e-05,
    "epoch": 1.93,
    "step": 2900
  },
  {
    "loss": 0.0874,
    "grad_norm": 0.7903698086738586,
    "learning_rate": 1.6722222222222225e-05,
    "epoch": 1.97,
    "step": 2950
  },
  {
    "loss": 0.0838,
    "grad_norm": 0.3140181005001068,
    "learning_rate": 1.6666666666666667e-05,
    "epoch": 2.0,
    "step": 3000
  },
  {
    "eval_loss": 0.07791818678379059,
    "eval_accuracy": 0.604,
    "eval_f1": 0.925355272999252,
    "eval_precision": 0.9692886242557192,
    "eval_recall": 0.885231825987407,
    "eval_runtime": 404.0361,
    "eval_samples_per_second": 3.713,
    "eval_steps_per_second": 0.465,
    "epoch": 2.0,
    "step": 3000
  },
  {
    "loss": 0.0793,
    "grad_norm": 0.4299638271331787,
    "learning_rate": 1.6611111111111113e-05,
    "epoch": 2.03,
    "step": 3050
  },
  {
    "loss": 0.0719,
    "grad_norm": 0.5105370283126831,
    "learning_rate": 1.6555555555555556e-05,
    "epoch": 2.07,
    "step": 3100
  },
  {
    "loss": 0.0675,
    "grad_norm": 0.8778389096260071,
    "learning_rate": 1.65e-05,
    "epoch": 2.1,
    "step": 3150
  },
  {
    "loss": 0.071,
    "grad_norm": 0.4817022681236267,
    "learning_rate": 1.6444444444444444e-05,
    "epoch": 2.13,
    "step": 3200
  },
  {
    "loss": 0.0744,
    "grad_norm": 0.5327057242393494,
    "learning_rate": 1.638888888888889e-05,
    "epoch": 2.17,
    "step": 3250
  },
  {
    "loss": 0.0761,
    "grad_norm": 0.3640955090522766,
    "learning_rate": 1.6333333333333335e-05,
    "epoch": 2.2,
    "step": 3300
  },
  {
    "loss": 0.0788,
    "grad_norm": 0.3881916105747223,
    "learning_rate": 1.6277777777777777e-05,
    "epoch": 2.23,
    "step": 3350
  },
  {
    "loss": 0.0701,
    "grad_norm": 0.30048519372940063,
    "learning_rate": 1.6222222222222223e-05,
    "epoch": 2.27,
    "step": 3400
  },
  {
    "loss": 0.0716,
    "grad_norm": 0.6386255621910095,
    "learning_rate": 1.616666666666667e-05,
    "epoch": 2.3,
    "step": 3450
  },
  {
    "loss": 0.0665,
    "grad_norm": 0.950891375541687,
    "learning_rate": 1.6111111111111115e-05,
    "epoch": 2.33,
    "step": 3500
  },
  {
    "loss": 0.0659,
    "grad_norm": 0.5039156675338745,
    "learning_rate": 1.6055555555555557e-05,
    "epoch": 2.37,
    "step": 3550
  },
  {
    "loss": 0.0706,
    "grad_norm": 0.20529164373874664,
    "learning_rate": 1.6000000000000003e-05,
    "epoch": 2.4,
    "step": 3600
  },
  {
    "loss": 0.0653,
    "grad_norm": 0.4774051010608673,
    "learning_rate": 1.594444444444445e-05,
    "epoch": 2.43,
    "step": 3650
  },
  {
    "loss": 0.0674,
    "grad_norm": 0.4316297471523285,
    "learning_rate": 1.588888888888889e-05,
    "epoch": 2.47,
    "step": 3700
  },
  {
    "loss": 0.073,
    "grad_norm": 1.2835804224014282,
    "learning_rate": 1.5833333333333333e-05,
    "epoch": 2.5,
    "step": 3750
  },
  {
    "loss": 0.064,
    "grad_norm": 0.27332884073257446,
    "learning_rate": 1.577777777777778e-05,
    "epoch": 2.53,
    "step": 3800
  },
  {
    "loss": 0.0695,
    "grad_norm": 3.6316583156585693,
    "learning_rate": 1.5722222222222225e-05,
    "epoch": 2.57,
    "step": 3850
  },
  {
    "loss": 0.0626,
    "grad_norm": 0.5828558206558228,
    "learning_rate": 1.5666666666666667e-05,
    "epoch": 2.6,
    "step": 3900
  },
  {
    "loss": 0.0703,
    "grad_norm": 0.4646282494068146,
    "learning_rate": 1.5611111111111113e-05,
    "epoch": 2.63,
    "step": 3950
  },
  {
    "loss": 0.0654,
    "grad_norm": 0.7953578233718872,
    "learning_rate": 1.555555555555556e-05,
    "epoch": 2.67,
    "step": 4000
  },
  {
    "loss": 0.0647,
    "grad_norm": 1.1880340576171875,
    "learning_rate": 1.55e-05,
    "epoch": 2.7,
    "step": 4050
  },
  {
    "loss": 0.0637,
    "grad_norm": 0.5192042589187622,
    "learning_rate": 1.5444444444444446e-05,
    "epoch": 2.73,
    "step": 4100
  },
  {
    "loss": 0.0648,
    "grad_norm": 0.5041971802711487,
    "learning_rate": 1.5388888888888892e-05,
    "epoch": 2.77,
    "step": 4150
  },
  {
    "loss": 0.0651,
    "grad_norm": 0.5176190733909607,
    "learning_rate": 1.5333333333333334e-05,
    "epoch": 2.8,
    "step": 4200
  },
  {
    "loss": 0.0591,
    "grad_norm": 0.6175917983055115,
    "learning_rate": 1.5277777777777777e-05,
    "epoch": 2.83,
    "step": 4250
  },
  {
    "loss": 0.0681,
    "grad_norm": 0.9707493185997009,
    "learning_rate": 1.5222222222222223e-05,
    "epoch": 2.87,
    "step": 4300
  },
  {
    "loss": 0.0624,
    "grad_norm": 0.5937142372131348,
    "learning_rate": 1.5166666666666667e-05,
    "epoch": 2.9,
    "step": 4350
  },
  {
    "loss": 0.0632,
    "grad_norm": 0.45223355293273926,
    "learning_rate": 1.5111111111111112e-05,
    "epoch": 2.93,
    "step": 4400
  },
  {
    "loss": 0.0623,
    "grad_norm": 0.6646191477775574,
    "learning_rate": 1.5055555555555556e-05,
    "epoch": 2.97,
    "step": 4450
  },
  {
    "loss": 0.0589,
    "grad_norm": 0.29040953516960144,
    "learning_rate": 1.5000000000000002e-05,
    "epoch": 3.0,
    "step": 4500
  },
  {
    "eval_loss": 0.059341732412576675,
    "eval_accuracy": 0.7013333333333334,
    "eval_f1": 0.9475547687541491,
    "eval_precision": 0.9777743948850662,
    "eval_recall": 0.9191471093302804,
    "eval_runtime": 373.05,
    "eval_samples_per_second": 4.021,
    "eval_steps_per_second": 0.504,
    "epoch": 3.0,
    "step": 4500
  },
  {
    "loss": 0.0614,
    "grad_norm": 0.9815644025802612,
    "learning_rate": 1.4944444444444446e-05,
    "epoch": 3.03,
    "step": 4550
  },
  {
    "loss": 0.054,
    "grad_norm": 0.6002834439277649,
    "learning_rate": 1.488888888888889e-05,
    "epoch": 3.07,
    "step": 4600
  },
  {
    "loss": 0.0519,
    "grad_norm": 0.658513069152832,
    "learning_rate": 1.4833333333333336e-05,
    "epoch": 3.1,
    "step": 4650
  },
  {
    "loss": 0.0536,
    "grad_norm": 0.48609504103660583,
    "learning_rate": 1.477777777777778e-05,
    "epoch": 3.13,
    "step": 4700
  },
  {
    "loss": 0.0525,
    "grad_norm": 0.21742594242095947,
    "learning_rate": 1.4722222222222224e-05,
    "epoch": 3.17,
    "step": 4750
  },
  {
    "loss": 0.0584,
    "grad_norm": 0.845954418182373,
    "learning_rate": 1.4666666666666666e-05,
    "epoch": 3.2,
    "step": 4800
  },
  {
    "loss": 0.0494,
    "grad_norm": 0.9900146126747131,
    "learning_rate": 1.4611111111111112e-05,
    "epoch": 3.23,
    "step": 4850
  },
  {
    "loss": 0.0555,
    "grad_norm": 0.1688126176595688,
    "learning_rate": 1.4555555555555556e-05,
    "epoch": 3.27,
    "step": 4900
  },
  {
    "loss": 0.0518,
    "grad_norm": 0.2642045319080353,
    "learning_rate": 1.45e-05,
    "epoch": 3.3,
    "step": 4950
  },
  {
    "loss": 0.0436,
    "grad_norm": 0.3140838146209717,
    "learning_rate": 1.4444444444444446e-05,
    "epoch": 3.33,
    "step": 5000
  },
  {
    "loss": 0.0532,
    "grad_norm": 0.8053641319274902,
    "learning_rate": 1.438888888888889e-05,
    "epoch": 3.37,
    "step": 5050
  },
  {
    "loss": 0.0556,
    "grad_norm": 1.414271354675293,
    "learning_rate": 1.4333333333333334e-05,
    "epoch": 3.4,
    "step": 5100
  },
  {
    "loss": 0.0571,
    "grad_norm": 0.5971333384513855,
    "learning_rate": 1.427777777777778e-05,
    "epoch": 3.43,
    "step": 5150
  },
  {
    "loss": 0.053,
    "grad_norm": 0.4587002694606781,
    "learning_rate": 1.4222222222222224e-05,
    "epoch": 3.47,
    "step": 5200
  },
  {
    "loss": 0.0504,
    "grad_norm": 0.6501361131668091,
    "learning_rate": 1.416666666666667e-05,
    "epoch": 3.5,
    "step": 5250
  },
  {
    "loss": 0.0533,
    "grad_norm": 1.1308332681655884,
    "learning_rate": 1.4111111111111113e-05,
    "epoch": 3.53,
    "step": 5300
  },
  {
    "loss": 0.0508,
    "grad_norm": 1.671701192855835,
    "learning_rate": 1.4055555555555556e-05,
    "epoch": 3.57,
    "step": 5350
  },
  {
    "loss": 0.0526,
    "grad_norm": 0.7493703961372375,
    "learning_rate": 1.4e-05,
    "epoch": 3.6,
    "step": 5400
  },
  {
    "loss": 0.0523,
    "grad_norm": 0.1724095195531845,
    "learning_rate": 1.3944444444444446e-05,
    "epoch": 3.63,
    "step": 5450
  },
  {
    "loss": 0.0471,
    "grad_norm": 0.28638821840286255,
    "learning_rate": 1.388888888888889e-05,
    "epoch": 3.67,
    "step": 5500
  },
  {
    "loss": 0.0539,
    "grad_norm": 1.1178017854690552,
    "learning_rate": 1.3833333333333334e-05,
    "epoch": 3.7,
    "step": 5550
  },
  {
    "loss": 0.0454,
    "grad_norm": 0.3396628499031067,
    "learning_rate": 1.377777777777778e-05,
    "epoch": 3.73,
    "step": 5600
  },
  {
    "loss": 0.0458,
    "grad_norm": 0.6596197485923767,
    "learning_rate": 1.3722222222222223e-05,
    "epoch": 3.77,
    "step": 5650
  },
  {
    "loss": 0.0491,
    "grad_norm": 0.6554704308509827,
    "learning_rate": 1.3666666666666667e-05,
    "epoch": 3.8,
    "step": 5700
  },
  {
    "loss": 0.042,
    "grad_norm": 0.16569265723228455,
    "learning_rate": 1.3611111111111113e-05,
    "epoch": 3.83,
    "step": 5750
  },
  {
    "loss": 0.0489,
    "grad_norm": 0.4453265964984894,
    "learning_rate": 1.3555555555555557e-05,
    "epoch": 3.87,
    "step": 5800
  },
  {
    "loss": 0.0479,
    "grad_norm": 0.917383074760437,
    "learning_rate": 1.3500000000000001e-05,
    "epoch": 3.9,
    "step": 5850
  },
  {
    "loss": 0.0502,
    "grad_norm": 0.8061167597770691,
    "learning_rate": 1.3444444444444447e-05,
    "epoch": 3.93,
    "step": 5900
  },
  {
    "loss": 0.045,
    "grad_norm": 0.14958856999874115,
    "learning_rate": 1.338888888888889e-05,
    "epoch": 3.97,
    "step": 5950
  },
  {
    "loss": 0.0486,
    "grad_norm": 0.5399668216705322,
    "learning_rate": 1.3333333333333333e-05,
    "epoch": 4.0,
    "step": 6000
  },
  {
    "eval_loss": 0.05016530305147171,
    "eval_accuracy": 0.7253333333333334,
    "eval_f1": 0.952742677671463,
    "eval_precision": 0.9729971654483067,
    "eval_recall": 0.9333142530051517,
    "eval_runtime": 401.4594,
    "eval_samples_per_second": 3.736,
    "eval_steps_per_second": 0.468,
    "epoch": 4.0,
    "step": 6000
  },
  {
    "loss": 0.0403,
    "grad_norm": 1.0464340448379517,
    "learning_rate": 1.3277777777777779e-05,
    "epoch": 4.03,
    "step": 6050
  },
  {
    "loss": 0.0398,
    "grad_norm": 0.4619244635105133,
    "learning_rate": 1.3222222222222223e-05,
    "epoch": 4.07,
    "step": 6100
  },
  {
    "loss": 0.0446,
    "grad_norm": 0.8172645568847656,
    "learning_rate": 1.3166666666666667e-05,
    "epoch": 4.1,
    "step": 6150
  },
  {
    "loss": 0.0423,
    "grad_norm": 0.7000390291213989,
    "learning_rate": 1.3111111111111113e-05,
    "epoch": 4.13,
    "step": 6200
  },
  {
    "loss": 0.0373,
    "grad_norm": 1.2198138236999512,
    "learning_rate": 1.3055555555555557e-05,
    "epoch": 4.17,
    "step": 6250
  },
  {
    "loss": 0.0434,
    "grad_norm": 0.25625166296958923,
    "learning_rate": 1.3000000000000001e-05,
    "epoch": 4.2,
    "step": 6300
  },
  {
    "loss": 0.0434,
    "grad_norm": 0.8375250697135925,
    "learning_rate": 1.2944444444444447e-05,
    "epoch": 4.23,
    "step": 6350
  },
  {
    "loss": 0.0419,
    "grad_norm": 1.2568154335021973,
    "learning_rate": 1.288888888888889e-05,
    "epoch": 4.27,
    "step": 6400
  },
  {
    "loss": 0.04,
    "grad_norm": 0.08529137074947357,
    "learning_rate": 1.2833333333333335e-05,
    "epoch": 4.3,
    "step": 6450
  },
  {
    "loss": 0.0408,
    "grad_norm": 0.8144034147262573,
    "learning_rate": 1.2777777777777777e-05,
    "epoch": 4.33,
    "step": 6500
  },
  {
    "loss": 0.0421,
    "grad_norm": 0.41111770272254944,
    "learning_rate": 1.2722222222222223e-05,
    "epoch": 4.37,
    "step": 6550
  },
  {
    "loss": 0.0433,
    "grad_norm": 0.40817421674728394,
    "learning_rate": 1.2666666666666667e-05,
    "epoch": 4.4,
    "step": 6600
  },
  {
    "loss": 0.0393,
    "grad_norm": 0.6191626787185669,
    "learning_rate": 1.2611111111111113e-05,
    "epoch": 4.43,
    "step": 6650
  },
  {
    "loss": 0.0339,
    "grad_norm": 0.46081477403640747,
    "learning_rate": 1.2555555555555557e-05,
    "epoch": 4.47,
    "step": 6700
  },
  {
    "loss": 0.0422,
    "grad_norm": 0.453768789768219,
    "learning_rate": 1.25e-05,
    "epoch": 4.5,
    "step": 6750
  },
  {
    "loss": 0.0382,
    "grad_norm": 0.3311583399772644,
    "learning_rate": 1.2444444444444446e-05,
    "epoch": 4.53,
    "step": 6800
  },
  {
    "loss": 0.04,
    "grad_norm": 0.5008330345153809,
    "learning_rate": 1.238888888888889e-05,
    "epoch": 4.57,
    "step": 6850
  },
  {
    "loss": 0.0389,
    "grad_norm": 0.9684385061264038,
    "learning_rate": 1.2333333333333334e-05,
    "epoch": 4.6,
    "step": 6900
  },
  {
    "loss": 0.0393,
    "grad_norm": 0.38716915249824524,
    "learning_rate": 1.227777777777778e-05,
    "epoch": 4.63,
    "step": 6950
  },
  {
    "loss": 0.0428,
    "grad_norm": 0.7534592151641846,
    "learning_rate": 1.2222222222222224e-05,
    "epoch": 4.67,
    "step": 7000
  },
  {
    "loss": 0.0416,
    "grad_norm": 0.6481875777244568,
    "learning_rate": 1.2166666666666667e-05,
    "epoch": 4.7,
    "step": 7050
  },
  {
    "loss": 0.0377,
    "grad_norm": 0.93485027551651,
    "learning_rate": 1.211111111111111e-05,
    "epoch": 4.73,
    "step": 7100
  },
  {
    "loss": 0.0397,
    "grad_norm": 0.3216228485107422,
    "learning_rate": 1.2055555555555556e-05,
    "epoch": 4.77,
    "step": 7150
  },
  {
    "loss": 0.0382,
    "grad_norm": 0.6930127143859863,
    "learning_rate": 1.2e-05,
    "epoch": 4.8,
    "step": 7200
  },
  {
    "loss": 0.0378,
    "grad_norm": 0.2020283341407776,
    "learning_rate": 1.1944444444444444e-05,
    "epoch": 4.83,
    "step": 7250
  },
  {
    "loss": 0.0384,
    "grad_norm": 0.35124385356903076,
    "learning_rate": 1.188888888888889e-05,
    "epoch": 4.87,
    "step": 7300
  },
  {
    "loss": 0.038,
    "grad_norm": 0.5830740928649902,
    "learning_rate": 1.1833333333333334e-05,
    "epoch": 4.9,
    "step": 7350
  },
  {
    "loss": 0.04,
    "grad_norm": 0.8609558343887329,
    "learning_rate": 1.177777777777778e-05,
    "epoch": 4.93,
    "step": 7400
  },
  {
    "loss": 0.0446,
    "grad_norm": 0.8603160977363586,
    "learning_rate": 1.1722222222222224e-05,
    "epoch": 4.97,
    "step": 7450
  },
  {
    "loss": 0.0405,
    "grad_norm": 0.3362710475921631,
    "learning_rate": 1.1666666666666668e-05,
    "epoch": 5.0,
    "step": 7500
  },
  {
    "eval_loss": 0.04217784106731415,
    "eval_accuracy": 0.7713333333333333,
    "eval_f1": 0.9617064647641235,
    "eval_precision": 0.9788085358624777,
    "eval_recall": 0.9451917572982256,
    "eval_runtime": 421.9356,
    "eval_samples_per_second": 3.555,
    "eval_steps_per_second": 0.446,
    "epoch": 5.0,
    "step": 7500
  },
  {
    "loss": 0.031,
    "grad_norm": 0.6572943925857544,
    "learning_rate": 1.1611111111111114e-05,
    "epoch": 5.03,
    "step": 7550
  },
  {
    "loss": 0.0351,
    "grad_norm": 0.6608228087425232,
    "learning_rate": 1.1555555555555556e-05,
    "epoch": 5.07,
    "step": 7600
  },
  {
    "loss": 0.0335,
    "grad_norm": 0.34060031175613403,
    "learning_rate": 1.15e-05,
    "epoch": 5.1,
    "step": 7650
  },
  {
    "loss": 0.0358,
    "grad_norm": 0.8972667455673218,
    "learning_rate": 1.1444444444444444e-05,
    "epoch": 5.13,
    "step": 7700
  },
  {
    "loss": 0.0337,
    "grad_norm": 0.42261555790901184,
    "learning_rate": 1.138888888888889e-05,
    "epoch": 5.17,
    "step": 7750
  },
  {
    "loss": 0.0329,
    "grad_norm": 0.5377686619758606,
    "learning_rate": 1.1333333333333334e-05,
    "epoch": 5.2,
    "step": 7800
  },
  {
    "loss": 0.0359,
    "grad_norm": 0.5514285564422607,
    "learning_rate": 1.1277777777777778e-05,
    "epoch": 5.23,
    "step": 7850
  },
  {
    "loss": 0.035,
    "grad_norm": 0.3760692775249481,
    "learning_rate": 1.1222222222222224e-05,
    "epoch": 5.27,
    "step": 7900
  },
  {
    "loss": 0.0358,
    "grad_norm": 0.11783821135759354,
    "learning_rate": 1.1166666666666668e-05,
    "epoch": 5.3,
    "step": 7950
  },
  {
    "loss": 0.0351,
    "grad_norm": 1.0664758682250977,
    "learning_rate": 1.1111111111111113e-05,
    "epoch": 5.33,
    "step": 8000
  },
  {
    "loss": 0.0312,
    "grad_norm": 0.8289102911949158,
    "learning_rate": 1.1055555555555557e-05,
    "epoch": 5.37,
    "step": 8050
  },
  {
    "loss": 0.0325,
    "grad_norm": 0.4296257197856903,
    "learning_rate": 1.1000000000000001e-05,
    "epoch": 5.4,
    "step": 8100
  },
  {
    "loss": 0.0329,
    "grad_norm": 0.43858158588409424,
    "learning_rate": 1.0944444444444447e-05,
    "epoch": 5.43,
    "step": 8150
  },
  {
    "loss": 0.0313,
    "grad_norm": 0.7341222763061523,
    "learning_rate": 1.088888888888889e-05,
    "epoch": 5.47,
    "step": 8200
  },
  {
    "loss": 0.0318,
    "grad_norm": 0.0853128656744957,
    "learning_rate": 1.0833333333333334e-05,
    "epoch": 5.5,
    "step": 8250
  },
  {
    "loss": 0.0305,
    "grad_norm": 0.3742698132991791,
    "learning_rate": 1.0777777777777778e-05,
    "epoch": 5.53,
    "step": 8300
  },
  {
    "loss": 0.0347,
    "grad_norm": 1.1857047080993652,
    "learning_rate": 1.0722222222222223e-05,
    "epoch": 5.57,
    "step": 8350
  },
  {
    "loss": 0.0376,
    "grad_norm": 0.6907274127006531,
    "learning_rate": 1.0666666666666667e-05,
    "epoch": 5.6,
    "step": 8400
  },
  {
    "loss": 0.0362,
    "grad_norm": 1.3683446645736694,
    "learning_rate": 1.0611111111111111e-05,
    "epoch": 5.63,
    "step": 8450
  },
  {
    "loss": 0.028,
    "grad_norm": 0.4153459966182709,
    "learning_rate": 1.0555555555555557e-05,
    "epoch": 5.67,
    "step": 8500
  },
  {
    "loss": 0.029,
    "grad_norm": 0.3362661600112915,
    "learning_rate": 1.0500000000000001e-05,
    "epoch": 5.7,
    "step": 8550
  },
  {
    "loss": 0.0336,
    "grad_norm": 0.6201055645942688,
    "learning_rate": 1.0444444444444445e-05,
    "epoch": 5.73,
    "step": 8600
  },
  {
    "loss": 0.0299,
    "grad_norm": 0.4934876263141632,
    "learning_rate": 1.0388888888888891e-05,
    "epoch": 5.77,
    "step": 8650
  },
  {
    "loss": 0.0311,
    "grad_norm": 0.7425228953361511,
    "learning_rate": 1.0333333333333335e-05,
    "epoch": 5.8,
    "step": 8700
  },
  {
    "loss": 0.033,
    "grad_norm": 0.9107281565666199,
    "learning_rate": 1.0277777777777777e-05,
    "epoch": 5.83,
    "step": 8750
  },
  {
    "loss": 0.0284,
    "grad_norm": 0.7918747663497925,
    "learning_rate": 1.0222222222222223e-05,
    "epoch": 5.87,
    "step": 8800
  },
  {
    "loss": 0.0311,
    "grad_norm": 0.8067458868026733,
    "learning_rate": 1.0166666666666667e-05,
    "epoch": 5.9,
    "step": 8850
  },
  {
    "loss": 0.029,
    "grad_norm": 0.8999997973442078,
    "learning_rate": 1.0111111111111111e-05,
    "epoch": 5.93,
    "step": 8900
  },
  {
    "loss": 0.0288,
    "grad_norm": 0.5887193083763123,
    "learning_rate": 1.0055555555555557e-05,
    "epoch": 5.97,
    "step": 8950
  },
  {
    "loss": 0.029,
    "grad_norm": 0.12278777360916138,
    "learning_rate": 1e-05,
    "epoch": 6.0,
    "step": 9000
  },
  {
    "eval_loss": 0.038735076785087585,
    "eval_accuracy": 0.79,
    "eval_f1": 0.963403145155446,
    "eval_precision": 0.9759213037733079,
    "eval_recall": 0.9512020606754437,
    "eval_runtime": 420.923,
    "eval_samples_per_second": 3.564,
    "eval_steps_per_second": 0.447,
    "epoch": 6.0,
    "step": 9000
  },
  {
    "loss": 0.0311,
    "grad_norm": 0.15732979774475098,
    "learning_rate": 9.944444444444445e-06,
    "epoch": 6.03,
    "step": 9050
  },
  {
    "loss": 0.027,
    "grad_norm": 0.3718239367008209,
    "learning_rate": 9.88888888888889e-06,
    "epoch": 6.07,
    "step": 9100
  },
  {
    "loss": 0.0248,
    "grad_norm": 0.4709440767765045,
    "learning_rate": 9.833333333333333e-06,
    "epoch": 6.1,
    "step": 9150
  },
  {
    "loss": 0.0277,
    "grad_norm": 0.1499980241060257,
    "learning_rate": 9.777777777777779e-06,
    "epoch": 6.13,
    "step": 9200
  },
  {
    "loss": 0.0244,
    "grad_norm": 0.4170394837856293,
    "learning_rate": 9.722222222222223e-06,
    "epoch": 6.17,
    "step": 9250
  },
  {
    "loss": 0.0238,
    "grad_norm": 0.2653655409812927,
    "learning_rate": 9.666666666666667e-06,
    "epoch": 6.2,
    "step": 9300
  },
  {
    "loss": 0.0285,
    "grad_norm": 0.7988731265068054,
    "learning_rate": 9.611111111111112e-06,
    "epoch": 6.23,
    "step": 9350
  },
  {
    "loss": 0.0299,
    "grad_norm": 0.8069539666175842,
    "learning_rate": 9.555555555555556e-06,
    "epoch": 6.27,
    "step": 9400
  },
  {
    "loss": 0.0262,
    "grad_norm": 0.5015062093734741,
    "learning_rate": 9.5e-06,
    "epoch": 6.3,
    "step": 9450
  },
  {
    "loss": 0.0279,
    "grad_norm": 0.8432004451751709,
    "learning_rate": 9.444444444444445e-06,
    "epoch": 6.33,
    "step": 9500
  },
  {
    "loss": 0.0216,
    "grad_norm": 0.6179779171943665,
    "learning_rate": 9.38888888888889e-06,
    "epoch": 6.37,
    "step": 9550
  },
  {
    "loss": 0.0233,
    "grad_norm": 0.21573516726493835,
    "learning_rate": 9.333333333333334e-06,
    "epoch": 6.4,
    "step": 9600
  },
  {
    "loss": 0.026,
    "grad_norm": 0.1759645938873291,
    "learning_rate": 9.277777777777778e-06,
    "epoch": 6.43,
    "step": 9650
  },
  {
    "loss": 0.0274,
    "grad_norm": 0.6740387082099915,
    "learning_rate": 9.222222222222224e-06,
    "epoch": 6.47,
    "step": 9700
  },
  {
    "loss": 0.0323,
    "grad_norm": 0.7693589925765991,
    "learning_rate": 9.166666666666666e-06,
    "epoch": 6.5,
    "step": 9750
  },
  {
    "loss": 0.0278,
    "grad_norm": 0.5023288726806641,
    "learning_rate": 9.111111111111112e-06,
    "epoch": 6.53,
    "step": 9800
  },
  {
    "loss": 0.028,
    "grad_norm": 0.5492318272590637,
    "learning_rate": 9.055555555555556e-06,
    "epoch": 6.57,
    "step": 9850
  },
  {
    "loss": 0.0242,
    "grad_norm": 0.2837912440299988,
    "learning_rate": 9e-06,
    "epoch": 6.6,
    "step": 9900
  },
  {
    "loss": 0.0272,
    "grad_norm": 0.680008053779602,
    "learning_rate": 8.944444444444446e-06,
    "epoch": 6.63,
    "step": 9950
  },
  {
    "loss": 0.0289,
    "grad_norm": 1.9726448059082031,
    "learning_rate": 8.888888888888888e-06,
    "epoch": 6.67,
    "step": 10000
  },
  {
    "loss": 0.0271,
    "grad_norm": 1.6224364042282104,
    "learning_rate": 8.833333333333334e-06,
    "epoch": 6.7,
    "step": 10050
  },
  {
    "loss": 0.0275,
    "grad_norm": 0.19931891560554504,
    "learning_rate": 8.777777777777778e-06,
    "epoch": 6.73,
    "step": 10100
  },
  {
    "loss": 0.0281,
    "grad_norm": 0.5207138657569885,
    "learning_rate": 8.722222222222224e-06,
    "epoch": 6.77,
    "step": 10150
  },
  {
    "loss": 0.0265,
    "grad_norm": 0.29777082800865173,
    "learning_rate": 8.666666666666668e-06,
    "epoch": 6.8,
    "step": 10200
  },
  {
    "loss": 0.0292,
    "grad_norm": 0.8728300929069519,
    "learning_rate": 8.611111111111112e-06,
    "epoch": 6.83,
    "step": 10250
  },
  {
    "loss": 0.0226,
    "grad_norm": 0.7866277694702148,
    "learning_rate": 8.555555555555556e-06,
    "epoch": 6.87,
    "step": 10300
  },
  {
    "loss": 0.0273,
    "grad_norm": 0.3908791244029999,
    "learning_rate": 8.5e-06,
    "epoch": 6.9,
    "step": 10350
  },
  {
    "loss": 0.0253,
    "grad_norm": 0.9716600775718689,
    "learning_rate": 8.444444444444446e-06,
    "epoch": 6.93,
    "step": 10400
  },
  {
    "loss": 0.0272,
    "grad_norm": 0.49167874455451965,
    "learning_rate": 8.38888888888889e-06,
    "epoch": 6.97,
    "step": 10450
  },
  {
    "loss": 0.0264,
    "grad_norm": 0.7925304174423218,
    "learning_rate": 8.333333333333334e-06,
    "epoch": 7.0,
    "step": 10500
  },
  {
    "eval_loss": 0.036279063671827316,
    "eval_accuracy": 0.7973333333333333,
    "eval_f1": 0.965166908563135,
    "eval_precision": 0.9790930506478209,
    "eval_recall": 0.9516313680595306,
    "eval_runtime": 520.5295,
    "eval_samples_per_second": 2.882,
    "eval_steps_per_second": 0.361,
    "epoch": 7.0,
    "step": 10500
  },
  {
    "loss": 0.0222,
    "grad_norm": 0.8314028978347778,
    "learning_rate": 8.277777777777778e-06,
    "epoch": 7.03,
    "step": 10550
  },
  {
    "loss": 0.0246,
    "grad_norm": 0.16064128279685974,
    "learning_rate": 8.222222222222222e-06,
    "epoch": 7.07,
    "step": 10600
  },
  {
    "loss": 0.0236,
    "grad_norm": 0.6914618611335754,
    "learning_rate": 8.166666666666668e-06,
    "epoch": 7.1,
    "step": 10650
  },
  {
    "loss": 0.0197,
    "grad_norm": 0.2741406559944153,
    "learning_rate": 8.111111111111112e-06,
    "epoch": 7.13,
    "step": 10700
  },
  {
    "loss": 0.022,
    "grad_norm": 0.48848968744277954,
    "learning_rate": 8.055555555555557e-06,
    "epoch": 7.17,
    "step": 10750
  },
  {
    "loss": 0.0224,
    "grad_norm": 0.22712096571922302,
    "learning_rate": 8.000000000000001e-06,
    "epoch": 7.2,
    "step": 10800
  },
  {
    "loss": 0.021,
    "grad_norm": 0.4689682722091675,
    "learning_rate": 7.944444444444445e-06,
    "epoch": 7.23,
    "step": 10850
  },
  {
    "loss": 0.0224,
    "grad_norm": 0.8227229118347168,
    "learning_rate": 7.88888888888889e-06,
    "epoch": 7.27,
    "step": 10900
  },
  {
    "loss": 0.0225,
    "grad_norm": 0.08071909844875336,
    "learning_rate": 7.833333333333333e-06,
    "epoch": 7.3,
    "step": 10950
  },
  {
    "loss": 0.0251,
    "grad_norm": 0.674135684967041,
    "learning_rate": 7.77777777777778e-06,
    "epoch": 7.33,
    "step": 11000
  },
  {
    "loss": 0.024,
    "grad_norm": 0.5259861946105957,
    "learning_rate": 7.722222222222223e-06,
    "epoch": 7.37,
    "step": 11050
  },
  {
    "loss": 0.0228,
    "grad_norm": 0.3898846209049225,
    "learning_rate": 7.666666666666667e-06,
    "epoch": 7.4,
    "step": 11100
  },
  {
    "loss": 0.0232,
    "grad_norm": 0.6550074815750122,
    "learning_rate": 7.611111111111111e-06,
    "epoch": 7.43,
    "step": 11150
  },
  {
    "loss": 0.0221,
    "grad_norm": 0.41366636753082275,
    "learning_rate": 7.555555555555556e-06,
    "epoch": 7.47,
    "step": 11200
  },
  {
    "loss": 0.0237,
    "grad_norm": 0.37380266189575195,
    "learning_rate": 7.500000000000001e-06,
    "epoch": 7.5,
    "step": 11250
  },
  {
    "loss": 0.0235,
    "grad_norm": 0.4968361556529999,
    "learning_rate": 7.444444444444445e-06,
    "epoch": 7.53,
    "step": 11300
  },
  {
    "loss": 0.0198,
    "grad_norm": 0.8349938988685608,
    "learning_rate": 7.38888888888889e-06,
    "epoch": 7.57,
    "step": 11350
  },
  {
    "loss": 0.0242,
    "grad_norm": 0.8924107551574707,
    "learning_rate": 7.333333333333333e-06,
    "epoch": 7.6,
    "step": 11400
  },
  {
    "loss": 0.0218,
    "grad_norm": 1.0067102909088135,
    "learning_rate": 7.277777777777778e-06,
    "epoch": 7.63,
    "step": 11450
  },
  {
    "loss": 0.0224,
    "grad_norm": 1.5542269945144653,
    "learning_rate": 7.222222222222223e-06,
    "epoch": 7.67,
    "step": 11500
  },
  {
    "loss": 0.023,
    "grad_norm": 0.7638463973999023,
    "learning_rate": 7.166666666666667e-06,
    "epoch": 7.7,
    "step": 11550
  },
  {
    "loss": 0.0211,
    "grad_norm": 0.43382686376571655,
    "learning_rate": 7.111111111111112e-06,
    "epoch": 7.73,
    "step": 11600
  },
  {
    "loss": 0.024,
    "grad_norm": 0.748521625995636,
    "learning_rate": 7.055555555555557e-06,
    "epoch": 7.77,
    "step": 11650
  },
  {
    "loss": 0.0223,
    "grad_norm": 0.3006242513656616,
    "learning_rate": 7e-06,
    "epoch": 7.8,
    "step": 11700
  },
  {
    "loss": 0.0233,
    "grad_norm": 0.5056409239768982,
    "learning_rate": 6.944444444444445e-06,
    "epoch": 7.83,
    "step": 11750
  },
  {
    "loss": 0.0204,
    "grad_norm": 0.2664984464645386,
    "learning_rate": 6.88888888888889e-06,
    "epoch": 7.87,
    "step": 11800
  },
  {
    "loss": 0.0218,
    "grad_norm": 0.18376298248767853,
    "learning_rate": 6.833333333333334e-06,
    "epoch": 7.9,
    "step": 11850
  },
  {
    "loss": 0.0244,
    "grad_norm": 0.9237328171730042,
    "learning_rate": 6.777777777777779e-06,
    "epoch": 7.93,
    "step": 11900
  },
  {
    "loss": 0.025,
    "grad_norm": 0.43559354543685913,
    "learning_rate": 6.7222222222222235e-06,
    "epoch": 7.97,
    "step": 11950
  },
  {
    "loss": 0.0239,
    "grad_norm": 0.6444123983383179,
    "learning_rate": 6.666666666666667e-06,
    "epoch": 8.0,
    "step": 12000
  },
  {
    "eval_loss": 0.03503166139125824,
    "eval_accuracy": 0.8213333333333334,
    "eval_f1": 0.9685324825986079,
    "eval_precision": 0.9816284538506761,
    "eval_recall": 0.9557813394390383,
    "eval_runtime": 487.6742,
    "eval_samples_per_second": 3.076,
    "eval_steps_per_second": 0.386,
    "epoch": 8.0,
    "step": 12000
  },
  {
    "loss": 0.019,
    "grad_norm": 0.10045385360717773,
    "learning_rate": 6.6111111111111115e-06,
    "epoch": 8.03,
    "step": 12050
  },
  {
    "loss": 0.0214,
    "grad_norm": 0.6069362759590149,
    "learning_rate": 6.555555555555556e-06,
    "epoch": 8.07,
    "step": 12100
  },
  {
    "loss": 0.0197,
    "grad_norm": 0.37218862771987915,
    "learning_rate": 6.5000000000000004e-06,
    "epoch": 8.1,
    "step": 12150
  },
  {
    "loss": 0.0203,
    "grad_norm": 0.2680938243865967,
    "learning_rate": 6.444444444444445e-06,
    "epoch": 8.13,
    "step": 12200
  },
  {
    "loss": 0.0202,
    "grad_norm": 0.6604072451591492,
    "learning_rate": 6.3888888888888885e-06,
    "epoch": 8.17,
    "step": 12250
  },
  {
    "loss": 0.0182,
    "grad_norm": 0.18383300304412842,
    "learning_rate": 6.333333333333333e-06,
    "epoch": 8.2,
    "step": 12300
  },
  {
    "loss": 0.02,
    "grad_norm": 0.6499873995780945,
    "learning_rate": 6.277777777777778e-06,
    "epoch": 8.23,
    "step": 12350
  },
  {
    "loss": 0.0191,
    "grad_norm": 0.6889489889144897,
    "learning_rate": 6.222222222222223e-06,
    "epoch": 8.27,
    "step": 12400
  },
  {
    "loss": 0.0202,
    "grad_norm": 0.33093497157096863,
    "learning_rate": 6.166666666666667e-06,
    "epoch": 8.3,
    "step": 12450
  },
  {
    "loss": 0.0212,
    "grad_norm": 0.531550407409668,
    "learning_rate": 6.111111111111112e-06,
    "epoch": 8.33,
    "step": 12500
  },
  {
    "loss": 0.0196,
    "grad_norm": 0.581438422203064,
    "learning_rate": 6.055555555555555e-06,
    "epoch": 8.37,
    "step": 12550
  },
  {
    "loss": 0.02,
    "grad_norm": 0.7175195217132568,
    "learning_rate": 6e-06,
    "epoch": 8.4,
    "step": 12600
  },
  {
    "loss": 0.02,
    "grad_norm": 0.354005366563797,
    "learning_rate": 5.944444444444445e-06,
    "epoch": 8.43,
    "step": 12650
  },
  {
    "loss": 0.0202,
    "grad_norm": 0.39726442098617554,
    "learning_rate": 5.88888888888889e-06,
    "epoch": 8.47,
    "step": 12700
  },
  {
    "loss": 0.0195,
    "grad_norm": 0.5678325891494751,
    "learning_rate": 5.833333333333334e-06,
    "epoch": 8.5,
    "step": 12750
  },
  {
    "loss": 0.0204,
    "grad_norm": 0.9092422723770142,
    "learning_rate": 5.777777777777778e-06,
    "epoch": 8.53,
    "step": 12800
  },
  {
    "loss": 0.0194,
    "grad_norm": 0.17264424264431,
    "learning_rate": 5.722222222222222e-06,
    "epoch": 8.57,
    "step": 12850
  },
  {
    "loss": 0.0185,
    "grad_norm": 0.5633407235145569,
    "learning_rate": 5.666666666666667e-06,
    "epoch": 8.6,
    "step": 12900
  },
  {
    "loss": 0.0163,
    "grad_norm": 0.38537177443504333,
    "learning_rate": 5.611111111111112e-06,
    "epoch": 8.63,
    "step": 12950
  },
  {
    "loss": 0.018,
    "grad_norm": 0.2082950472831726,
    "learning_rate": 5.555555555555557e-06,
    "epoch": 8.67,
    "step": 13000
  },
  {
    "loss": 0.0212,
    "grad_norm": 0.24411733448505402,
    "learning_rate": 5.500000000000001e-06,
    "epoch": 8.7,
    "step": 13050
  },
  {
    "loss": 0.02,
    "grad_norm": 0.236025869846344,
    "learning_rate": 5.444444444444445e-06,
    "epoch": 8.73,
    "step": 13100
  },
  {
    "loss": 0.0178,
    "grad_norm": 0.5068910121917725,
    "learning_rate": 5.388888888888889e-06,
    "epoch": 8.77,
    "step": 13150
  },
  {
    "loss": 0.0189,
    "grad_norm": 0.3922387957572937,
    "learning_rate": 5.333333333333334e-06,
    "epoch": 8.8,
    "step": 13200
  },
  {
    "loss": 0.0208,
    "grad_norm": 0.9244905710220337,
    "learning_rate": 5.2777777777777785e-06,
    "epoch": 8.83,
    "step": 13250
  },
  {
    "loss": 0.0199,
    "grad_norm": 0.8682542443275452,
    "learning_rate": 5.2222222222222226e-06,
    "epoch": 8.87,
    "step": 13300
  },
  {
    "loss": 0.019,
    "grad_norm": 1.3180102109909058,
    "learning_rate": 5.1666666666666675e-06,
    "epoch": 8.9,
    "step": 13350
  },
  {
    "loss": 0.021,
    "grad_norm": 0.4705464839935303,
    "learning_rate": 5.1111111111111115e-06,
    "epoch": 8.93,
    "step": 13400
  },
  {
    "loss": 0.0184,
    "grad_norm": 0.4992372989654541,
    "learning_rate": 5.0555555555555555e-06,
    "epoch": 8.97,
    "step": 13450
  },
  {
    "loss": 0.0191,
    "grad_norm": 1.026501178741455,
    "learning_rate": 5e-06,
    "epoch": 9.0,
    "step": 13500
  },
  {
    "eval_loss": 0.033837590366601944,
    "eval_accuracy": 0.814,
    "eval_f1": 0.9678493845039826,
    "eval_precision": 0.9796247434769862,
    "eval_recall": 0.9563537492844877,
    "eval_runtime": 520.4183,
    "eval_samples_per_second": 2.882,
    "eval_steps_per_second": 0.361,
    "epoch": 9.0,
    "step": 13500
  },
  {
    "loss": 0.0182,
    "grad_norm": 0.34491831064224243,
    "learning_rate": 4.944444444444445e-06,
    "epoch": 9.03,
    "step": 13550
  },
  {
    "loss": 0.0152,
    "grad_norm": 0.1780650019645691,
    "learning_rate": 4.888888888888889e-06,
    "epoch": 9.07,
    "step": 13600
  },
  {
    "loss": 0.016,
    "grad_norm": 0.49815917015075684,
    "learning_rate": 4.833333333333333e-06,
    "epoch": 9.1,
    "step": 13650
  },
  {
    "loss": 0.0178,
    "grad_norm": 0.21634286642074585,
    "learning_rate": 4.777777777777778e-06,
    "epoch": 9.13,
    "step": 13700
  },
  {
    "loss": 0.0174,
    "grad_norm": 0.512596845626831,
    "learning_rate": 4.722222222222222e-06,
    "epoch": 9.17,
    "step": 13750
  },
  {
    "loss": 0.0169,
    "grad_norm": 0.480957567691803,
    "learning_rate": 4.666666666666667e-06,
    "epoch": 9.2,
    "step": 13800
  },
  {
    "loss": 0.0187,
    "grad_norm": 0.4561571180820465,
    "learning_rate": 4.611111111111112e-06,
    "epoch": 9.23,
    "step": 13850
  },
  {
    "loss": 0.0192,
    "grad_norm": 0.31520533561706543,
    "learning_rate": 4.555555555555556e-06,
    "epoch": 9.27,
    "step": 13900
  },
  {
    "loss": 0.0186,
    "grad_norm": 0.33180102705955505,
    "learning_rate": 4.5e-06,
    "epoch": 9.3,
    "step": 13950
  },
  {
    "loss": 0.0199,
    "grad_norm": 0.8074162006378174,
    "learning_rate": 4.444444444444444e-06,
    "epoch": 9.33,
    "step": 14000
  },
  {
    "loss": 0.0187,
    "grad_norm": 0.07085882127285004,
    "learning_rate": 4.388888888888889e-06,
    "epoch": 9.37,
    "step": 14050
  },
  {
    "loss": 0.0156,
    "grad_norm": 0.2821616232395172,
    "learning_rate": 4.333333333333334e-06,
    "epoch": 9.4,
    "step": 14100
  },
  {
    "loss": 0.0163,
    "grad_norm": 0.7558738589286804,
    "learning_rate": 4.277777777777778e-06,
    "epoch": 9.43,
    "step": 14150
  },
  {
    "loss": 0.0191,
    "grad_norm": 0.23577666282653809,
    "learning_rate": 4.222222222222223e-06,
    "epoch": 9.47,
    "step": 14200
  },
  {
    "loss": 0.0141,
    "grad_norm": 0.6420838832855225,
    "learning_rate": 4.166666666666667e-06,
    "epoch": 9.5,
    "step": 14250
  },
  {
    "loss": 0.0175,
    "grad_norm": 0.4799270033836365,
    "learning_rate": 4.111111111111111e-06,
    "epoch": 9.53,
    "step": 14300
  },
  {
    "loss": 0.0176,
    "grad_norm": 0.15877492725849152,
    "learning_rate": 4.055555555555556e-06,
    "epoch": 9.57,
    "step": 14350
  },
  {
    "loss": 0.0174,
    "grad_norm": 0.2862108647823334,
    "learning_rate": 4.000000000000001e-06,
    "epoch": 9.6,
    "step": 14400
  },
  {
    "loss": 0.0181,
    "grad_norm": 0.6650111079216003,
    "learning_rate": 3.944444444444445e-06,
    "epoch": 9.63,
    "step": 14450
  },
  {
    "loss": 0.0164,
    "grad_norm": 0.170445516705513,
    "learning_rate": 3.88888888888889e-06,
    "epoch": 9.67,
    "step": 14500
  },
  {
    "loss": 0.0164,
    "grad_norm": 1.0318118333816528,
    "learning_rate": 3.833333333333334e-06,
    "epoch": 9.7,
    "step": 14550
  },
  {
    "loss": 0.0172,
    "grad_norm": 0.1030106320977211,
    "learning_rate": 3.777777777777778e-06,
    "epoch": 9.73,
    "step": 14600
  },
  {
    "loss": 0.0175,
    "grad_norm": 0.5192915797233582,
    "learning_rate": 3.7222222222222225e-06,
    "epoch": 9.77,
    "step": 14650
  },
  {
    "loss": 0.0192,
    "grad_norm": 0.5223665833473206,
    "learning_rate": 3.6666666666666666e-06,
    "epoch": 9.8,
    "step": 14700
  },
  {
    "loss": 0.0173,
    "grad_norm": 0.21479204297065735,
    "learning_rate": 3.6111111111111115e-06,
    "epoch": 9.83,
    "step": 14750
  },
  {
    "loss": 0.0177,
    "grad_norm": 0.18222138285636902,
    "learning_rate": 3.555555555555556e-06,
    "epoch": 9.87,
    "step": 14800
  },
  {
    "loss": 0.0161,
    "grad_norm": 0.5151605010032654,
    "learning_rate": 3.5e-06,
    "epoch": 9.9,
    "step": 14850
  },
  {
    "loss": 0.0154,
    "grad_norm": 0.41523924469947815,
    "learning_rate": 3.444444444444445e-06,
    "epoch": 9.93,
    "step": 14900
  },
  {
    "loss": 0.0155,
    "grad_norm": 0.5363922715187073,
    "learning_rate": 3.3888888888888893e-06,
    "epoch": 9.97,
    "step": 14950
  },
  {
    "loss": 0.0172,
    "grad_norm": 0.5515249967575073,
    "learning_rate": 3.3333333333333333e-06,
    "epoch": 10.0,
    "step": 15000
  },
  {
    "eval_loss": 0.03265129774808884,
    "eval_accuracy": 0.816,
    "eval_f1": 0.9686844579446011,
    "eval_precision": 0.9792367305161573,
    "eval_recall": 0.9583571837435604,
    "eval_runtime": 522.0948,
    "eval_samples_per_second": 2.873,
    "eval_steps_per_second": 0.36,
    "epoch": 10.0,
    "step": 15000
  },
  {
    "loss": 0.014,
    "grad_norm": 0.4168848693370819,
    "learning_rate": 3.277777777777778e-06,
    "epoch": 10.03,
    "step": 15050
  },
  {
    "loss": 0.0156,
    "grad_norm": 0.22728918492794037,
    "learning_rate": 3.2222222222222227e-06,
    "epoch": 10.07,
    "step": 15100
  },
  {
    "loss": 0.0164,
    "grad_norm": 0.11018675565719604,
    "learning_rate": 3.1666666666666667e-06,
    "epoch": 10.1,
    "step": 15150
  },
  {
    "loss": 0.0153,
    "grad_norm": 0.2938558757305145,
    "learning_rate": 3.1111111111111116e-06,
    "epoch": 10.13,
    "step": 15200
  },
  {
    "loss": 0.0174,
    "grad_norm": 0.48106732964515686,
    "learning_rate": 3.055555555555556e-06,
    "epoch": 10.17,
    "step": 15250
  },
  {
    "loss": 0.0157,
    "grad_norm": 0.6213731169700623,
    "learning_rate": 3e-06,
    "epoch": 10.2,
    "step": 15300
  },
  {
    "loss": 0.0136,
    "grad_norm": 0.12498883903026581,
    "learning_rate": 2.944444444444445e-06,
    "epoch": 10.23,
    "step": 15350
  },
  {
    "loss": 0.0136,
    "grad_norm": 0.4984402656555176,
    "learning_rate": 2.888888888888889e-06,
    "epoch": 10.27,
    "step": 15400
  },
  {
    "loss": 0.0143,
    "grad_norm": 0.1306380331516266,
    "learning_rate": 2.8333333333333335e-06,
    "epoch": 10.3,
    "step": 15450
  },
  {
    "loss": 0.0177,
    "grad_norm": 0.33903223276138306,
    "learning_rate": 2.7777777777777783e-06,
    "epoch": 10.33,
    "step": 15500
  },
  {
    "loss": 0.0168,
    "grad_norm": 0.4361119866371155,
    "learning_rate": 2.7222222222222224e-06,
    "epoch": 10.37,
    "step": 15550
  },
  {
    "loss": 0.0155,
    "grad_norm": 0.6033740043640137,
    "learning_rate": 2.666666666666667e-06,
    "epoch": 10.4,
    "step": 15600
  },
  {
    "loss": 0.0155,
    "grad_norm": 0.37458139657974243,
    "learning_rate": 2.6111111111111113e-06,
    "epoch": 10.43,
    "step": 15650
  },
  {
    "loss": 0.0191,
    "grad_norm": 0.6604194045066833,
    "learning_rate": 2.5555555555555557e-06,
    "epoch": 10.47,
    "step": 15700
  },
  {
    "loss": 0.0147,
    "grad_norm": 0.560930073261261,
    "learning_rate": 2.5e-06,
    "epoch": 10.5,
    "step": 15750
  },
  {
    "loss": 0.0196,
    "grad_norm": 0.24454829096794128,
    "learning_rate": 2.4444444444444447e-06,
    "epoch": 10.53,
    "step": 15800
  },
  {
    "loss": 0.0173,
    "grad_norm": 0.9778093099594116,
    "learning_rate": 2.388888888888889e-06,
    "epoch": 10.57,
    "step": 15850
  },
  {
    "loss": 0.0151,
    "grad_norm": 0.1762203425168991,
    "learning_rate": 2.3333333333333336e-06,
    "epoch": 10.6,
    "step": 15900
  },
  {
    "loss": 0.0152,
    "grad_norm": 0.49755942821502686,
    "learning_rate": 2.277777777777778e-06,
    "epoch": 10.63,
    "step": 15950
  },
  {
    "loss": 0.016,
    "grad_norm": 0.4183555245399475,
    "learning_rate": 2.222222222222222e-06,
    "epoch": 10.67,
    "step": 16000
  },
  {
    "loss": 0.0163,
    "grad_norm": 0.2946385145187378,
    "learning_rate": 2.166666666666667e-06,
    "epoch": 10.7,
    "step": 16050
  },
  {
    "loss": 0.0155,
    "grad_norm": 0.27264583110809326,
    "learning_rate": 2.1111111111111114e-06,
    "epoch": 10.73,
    "step": 16100
  },
  {
    "loss": 0.0137,
    "grad_norm": 0.4318552017211914,
    "learning_rate": 2.0555555555555555e-06,
    "epoch": 10.77,
    "step": 16150
  },
  {
    "loss": 0.0147,
    "grad_norm": 0.2526138722896576,
    "learning_rate": 2.0000000000000003e-06,
    "epoch": 10.8,
    "step": 16200
  },
  {
    "loss": 0.0142,
    "grad_norm": 0.12450098246335983,
    "learning_rate": 1.944444444444445e-06,
    "epoch": 10.83,
    "step": 16250
  },
  {
    "loss": 0.015,
    "grad_norm": 0.5571430325508118,
    "learning_rate": 1.888888888888889e-06,
    "epoch": 10.87,
    "step": 16300
  },
  {
    "loss": 0.016,
    "grad_norm": 0.48039206862449646,
    "learning_rate": 1.8333333333333333e-06,
    "epoch": 10.9,
    "step": 16350
  },
  {
    "loss": 0.0175,
    "grad_norm": 0.13909302651882172,
    "learning_rate": 1.777777777777778e-06,
    "epoch": 10.93,
    "step": 16400
  },
  {
    "loss": 0.0142,
    "grad_norm": 0.754486083984375,
    "learning_rate": 1.7222222222222224e-06,
    "epoch": 10.97,
    "step": 16450
  },
  {
    "loss": 0.0147,
    "grad_norm": 0.6141240000724792,
    "learning_rate": 1.6666666666666667e-06,
    "epoch": 11.0,
    "step": 16500
  },
  {
    "eval_loss": 0.03264497593045235,
    "eval_accuracy": 0.822,
    "eval_f1": 0.9696138578130639,
    "eval_precision": 0.9781564001747488,
    "eval_recall": 0.9612192329708071,
    "eval_runtime": 521.02,
    "eval_samples_per_second": 2.879,
    "eval_steps_per_second": 0.361,
    "epoch": 11.0,
    "step": 16500
  },
  {
    "loss": 0.0148,
    "grad_norm": 0.6877202391624451,
    "learning_rate": 1.6111111111111113e-06,
    "epoch": 11.03,
    "step": 16550
  },
  {
    "loss": 0.0155,
    "grad_norm": 0.5596755743026733,
    "learning_rate": 1.5555555555555558e-06,
    "epoch": 11.07,
    "step": 16600
  },
  {
    "loss": 0.0165,
    "grad_norm": 0.268638551235199,
    "learning_rate": 1.5e-06,
    "epoch": 11.1,
    "step": 16650
  },
  {
    "loss": 0.0146,
    "grad_norm": 0.09044534713029861,
    "learning_rate": 1.4444444444444445e-06,
    "epoch": 11.13,
    "step": 16700
  },
  {
    "loss": 0.0153,
    "grad_norm": 0.29204678535461426,
    "learning_rate": 1.3888888888888892e-06,
    "epoch": 11.17,
    "step": 16750
  },
  {
    "loss": 0.0152,
    "grad_norm": 0.3674470782279968,
    "learning_rate": 1.3333333333333334e-06,
    "epoch": 11.2,
    "step": 16800
  },
  {
    "loss": 0.0148,
    "grad_norm": 0.07375849038362503,
    "learning_rate": 1.2777777777777779e-06,
    "epoch": 11.23,
    "step": 16850
  },
  {
    "loss": 0.014,
    "grad_norm": 0.23876281082630157,
    "learning_rate": 1.2222222222222223e-06,
    "epoch": 11.27,
    "step": 16900
  },
  {
    "loss": 0.0139,
    "grad_norm": 0.5587507486343384,
    "learning_rate": 1.1666666666666668e-06,
    "epoch": 11.3,
    "step": 16950
  },
  {
    "loss": 0.0153,
    "grad_norm": 0.19995635747909546,
    "learning_rate": 1.111111111111111e-06,
    "epoch": 11.33,
    "step": 17000
  },
  {
    "loss": 0.0142,
    "grad_norm": 0.1458493173122406,
    "learning_rate": 1.0555555555555557e-06,
    "epoch": 11.37,
    "step": 17050
  },
  {
    "loss": 0.0146,
    "grad_norm": 0.5687617063522339,
    "learning_rate": 1.0000000000000002e-06,
    "epoch": 11.4,
    "step": 17100
  },
  {
    "loss": 0.0133,
    "grad_norm": 0.5970475077629089,
    "learning_rate": 9.444444444444445e-07,
    "epoch": 11.43,
    "step": 17150
  },
  {
    "loss": 0.0142,
    "grad_norm": 0.3932097852230072,
    "learning_rate": 8.88888888888889e-07,
    "epoch": 11.47,
    "step": 17200
  },
  {
    "loss": 0.0149,
    "grad_norm": 0.1657884120941162,
    "learning_rate": 8.333333333333333e-07,
    "epoch": 11.5,
    "step": 17250
  },
  {
    "loss": 0.0131,
    "grad_norm": 0.10400531440973282,
    "learning_rate": 7.777777777777779e-07,
    "epoch": 11.53,
    "step": 17300
  },
  {
    "loss": 0.0143,
    "grad_norm": 0.27298837900161743,
    "learning_rate": 7.222222222222222e-07,
    "epoch": 11.57,
    "step": 17350
  },
  {
    "loss": 0.0157,
    "grad_norm": 0.5544312596321106,
    "learning_rate": 6.666666666666667e-07,
    "epoch": 11.6,
    "step": 17400
  },
  {
    "loss": 0.0134,
    "grad_norm": 0.027582811191678047,
    "learning_rate": 6.111111111111112e-07,
    "epoch": 11.63,
    "step": 17450
  },
  {
    "loss": 0.0141,
    "grad_norm": 0.5370105504989624,
    "learning_rate": 5.555555555555555e-07,
    "epoch": 11.67,
    "step": 17500
  },
  {
    "loss": 0.0159,
    "grad_norm": 0.7681547999382019,
    "learning_rate": 5.000000000000001e-07,
    "epoch": 11.7,
    "step": 17550
  },
  {
    "loss": 0.0152,
    "grad_norm": 0.5727664232254028,
    "learning_rate": 4.444444444444445e-07,
    "epoch": 11.73,
    "step": 17600
  },
  {
    "loss": 0.0139,
    "grad_norm": 0.324990838766098,
    "learning_rate": 3.8888888888888895e-07,
    "epoch": 11.77,
    "step": 17650
  },
  {
    "loss": 0.0139,
    "grad_norm": 0.16624578833580017,
    "learning_rate": 3.3333333333333335e-07,
    "epoch": 11.8,
    "step": 17700
  },
  {
    "loss": 0.0151,
    "grad_norm": 0.5462536215782166,
    "learning_rate": 2.7777777777777776e-07,
    "epoch": 11.83,
    "step": 17750
  },
  {
    "loss": 0.0157,
    "grad_norm": 0.3523215055465698,
    "learning_rate": 2.2222222222222224e-07,
    "epoch": 11.87,
    "step": 17800
  },
  {
    "loss": 0.0135,
    "grad_norm": 0.5089453458786011,
    "learning_rate": 1.6666666666666668e-07,
    "epoch": 11.9,
    "step": 17850
  },
  {
    "loss": 0.0151,
    "grad_norm": 0.45039597153663635,
    "learning_rate": 1.1111111111111112e-07,
    "epoch": 11.93,
    "step": 17900
  },
  {
    "loss": 0.0154,
    "grad_norm": 0.5319696068763733,
    "learning_rate": 5.555555555555556e-08,
    "epoch": 11.97,
    "step": 17950
  },
  {
    "loss": 0.0146,
    "grad_norm": 0.4072882831096649,
    "learning_rate": 0.0,
    "epoch": 12.0,
    "step": 18000
  },
  {
    "eval_loss": 0.03210272639989853,
    "eval_accuracy": 0.8293333333333334,
    "eval_f1": 0.9706901530464915,
    "eval_precision": 0.979458041958042,
    "eval_recall": 0.9620778477389811,
    "eval_runtime": 365.9115,
    "eval_samples_per_second": 4.099,
    "eval_steps_per_second": 0.514,
    "epoch": 12.0,
    "step": 18000
  },
  {
    "train_runtime": 94832.1488,
    "train_samples_per_second": 1.518,
    "train_steps_per_second": 0.19,
    "total_flos": 3.7897517039616e+16,
    "train_loss": 0.05056379745403926,
    "epoch": 12.0,
    "step": 18000
  }
]