diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,5558 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 13.280975609756098,
+  "global_step": 850,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 6.25e-06,
+      "loss": 2.9098,
+      "step": 1
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.25e-05,
+      "loss": 2.8818,
+      "step": 2
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.875e-05,
+      "loss": 2.9635,
+      "step": 3
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.5e-05,
+      "loss": 2.8696,
+      "step": 4
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 3.125e-05,
+      "loss": 2.8205,
+      "step": 5
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 3.75e-05,
+      "loss": 2.7737,
+      "step": 6
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.375e-05,
+      "loss": 2.751,
+      "step": 7
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 5e-05,
+      "loss": 2.7525,
+      "step": 8
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 5.6250000000000005e-05,
+      "loss": 2.7303,
+      "step": 9
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 6.25e-05,
+      "loss": 2.7412,
+      "step": 10
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 6.875e-05,
+      "loss": 2.4844,
+      "step": 11
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 7.5e-05,
+      "loss": 2.5187,
+      "step": 12
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 8.125000000000001e-05,
+      "loss": 2.4352,
+      "step": 13
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 8.75e-05,
+      "loss": 2.3852,
+      "step": 14
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.375e-05,
+      "loss": 2.388,
+      "step": 15
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0001,
+      "loss": 2.4145,
+      "step": 16
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00010625,
+      "loss": 2.3989,
+      "step": 17
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00011250000000000001,
+      "loss": 2.2944,
+      "step": 18
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00011875,
+      "loss": 2.2215,
+      "step": 19
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.000125,
+      "loss": 2.1633,
+      "step": 20
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00013125000000000002,
+      "loss": 2.2124,
+      "step": 21
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.0001375,
+      "loss": 2.1828,
+      "step": 22
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00014375,
+      "loss": 2.0791,
+      "step": 23
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00015,
+      "loss": 2.1757,
+      "step": 24
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00015625,
+      "loss": 2.1198,
+      "step": 25
+    },
+    {
+      "epoch": 0.39,
+      "eval_gen_len": 805.336,
+      "eval_loss": 1.8719514608383179,
+      "eval_rouge1": 29.4332,
+      "eval_rouge2": 7.3761,
+      "eval_rougeL": 17.0816,
+      "eval_rougeLsum": 25.065,
+      "eval_runtime": 886.4717,
+      "eval_samples_per_second": 0.282,
+      "eval_steps_per_second": 0.282,
+      "step": 25
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00016250000000000002,
+      "loss": 2.1116,
+      "step": 26
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00016875,
+      "loss": 2.0973,
+      "step": 27
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.000175,
+      "loss": 2.0698,
+      "step": 28
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00018125,
+      "loss": 2.1397,
+      "step": 29
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0001875,
+      "loss": 2.0961,
+      "step": 30
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00019375000000000002,
+      "loss": 2.0803,
+      "step": 31
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002,
+      "loss": 2.1034,
+      "step": 32
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00020625,
+      "loss": 2.0722,
+      "step": 33
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0002125,
+      "loss": 2.0459,
+      "step": 34
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00021875,
+      "loss": 2.0633,
+      "step": 35
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00022500000000000002,
+      "loss": 1.9343,
+      "step": 36
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00023125,
+      "loss": 1.9526,
+      "step": 37
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0002375,
+      "loss": 1.9989,
+      "step": 38
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00024375,
+      "loss": 1.9767,
+      "step": 39
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00025,
+      "loss": 1.9694,
+      "step": 40
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00025624999999999997,
+      "loss": 1.9294,
+      "step": 41
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00026250000000000004,
+      "loss": 1.9794,
+      "step": 42
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00026875,
+      "loss": 1.9808,
+      "step": 43
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.000275,
+      "loss": 1.9719,
+      "step": 44
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00028125000000000003,
+      "loss": 1.9447,
+      "step": 45
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0002875,
+      "loss": 1.9479,
+      "step": 46
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00029375,
+      "loss": 1.95,
+      "step": 47
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0003,
+      "loss": 1.9718,
+      "step": 48
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00030625000000000004,
+      "loss": 1.8785,
+      "step": 49
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0003125,
+      "loss": 1.8609,
+      "step": 50
+    },
+    {
+      "epoch": 0.78,
+      "eval_gen_len": 833.404,
+      "eval_loss": 1.7601044178009033,
+      "eval_rouge1": 35.3533,
+      "eval_rouge2": 10.6624,
+      "eval_rougeL": 18.643,
+      "eval_rougeLsum": 31.6979,
+      "eval_runtime": 912.2832,
+      "eval_samples_per_second": 0.274,
+      "eval_steps_per_second": 0.274,
+      "step": 50
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00031874999999999997,
+      "loss": 1.9477,
+      "step": 51
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00032500000000000004,
+      "loss": 1.8888,
+      "step": 52
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00033125,
+      "loss": 1.8034,
+      "step": 53
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0003375,
+      "loss": 1.9102,
+      "step": 54
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00034375,
+      "loss": 1.9455,
+      "step": 55
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00035,
+      "loss": 1.8725,
+      "step": 56
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00035625,
+      "loss": 1.8375,
+      "step": 57
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0003625,
+      "loss": 1.8795,
+      "step": 58
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00036875000000000005,
+      "loss": 1.7643,
+      "step": 59
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.000375,
+      "loss": 1.7923,
+      "step": 60
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00038124999999999997,
+      "loss": 1.869,
+      "step": 61
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00038750000000000004,
+      "loss": 1.8725,
+      "step": 62
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00039375,
+      "loss": 1.9809,
+      "step": 63
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0004,
+      "loss": 1.8203,
+      "step": 64
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00040625000000000004,
+      "loss": 1.9199,
+      "step": 65
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.0004125,
+      "loss": 1.8922,
+      "step": 66
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00041875,
+      "loss": 1.8536,
+      "step": 67
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.000425,
+      "loss": 1.8278,
+      "step": 68
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00043125000000000005,
+      "loss": 1.79,
+      "step": 69
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0004375,
+      "loss": 1.8595,
+      "step": 70
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00044374999999999997,
+      "loss": 1.7372,
+      "step": 71
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 1.7542,
+      "step": 72
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00045625,
+      "loss": 1.7291,
+      "step": 73
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0004625,
+      "loss": 1.8462,
+      "step": 74
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00046875,
+      "loss": 1.7805,
+      "step": 75
+    },
+    {
+      "epoch": 1.17,
+      "eval_gen_len": 866.356,
+      "eval_loss": 1.683250069618225,
+      "eval_rouge1": 36.5786,
+      "eval_rouge2": 11.1185,
+      "eval_rougeL": 20.0358,
+      "eval_rougeLsum": 33.2116,
+      "eval_runtime": 941.453,
+      "eval_samples_per_second": 0.266,
+      "eval_steps_per_second": 0.266,
+      "step": 75
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.000475,
+      "loss": 1.7477,
+      "step": 76
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00048125,
+      "loss": 1.784,
+      "step": 77
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0004875,
+      "loss": 1.8665,
+      "step": 78
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00049375,
+      "loss": 1.7594,
+      "step": 79
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0005,
+      "loss": 1.7236,
+      "step": 80
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0004999994660231515,
+      "loss": 1.8202,
+      "step": 81
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.0004999978640948868,
+      "loss": 1.708,
+      "step": 82
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.0004999951942220491,
+      "loss": 1.7961,
+      "step": 83
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.0004999914564160436,
+      "loss": 1.771,
+      "step": 84
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0004999866506928376,
+      "loss": 1.7643,
+      "step": 85
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.0004999807770729603,
+      "loss": 1.8169,
+      "step": 86
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0004999738355815026,
+      "loss": 1.7454,
+      "step": 87
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.0004999658262481172,
+      "loss": 1.8016,
+      "step": 88
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0004999567491070187,
+      "loss": 1.752,
+      "step": 89
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.0004999466041969828,
+      "loss": 1.7792,
+      "step": 90
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.0004999353915613467,
+      "loss": 1.7139,
+      "step": 91
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.0004999231112480087,
+      "loss": 1.7583,
+      "step": 92
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.000499909763309428,
+      "loss": 1.7325,
+      "step": 93
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.0004998953478026247,
+      "loss": 1.7276,
+      "step": 94
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.000499879864789179,
+      "loss": 1.7598,
+      "step": 95
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0004998633143352315,
+      "loss": 1.7816,
+      "step": 96
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.0004998456965114827,
+      "loss": 1.7535,
+      "step": 97
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.0004998270113931926,
+      "loss": 1.753,
+      "step": 98
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.0004998072590601808,
+      "loss": 1.7538,
+      "step": 99
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0004997864395968252,
+      "loss": 1.7352,
+      "step": 100
+    },
+    {
+      "epoch": 1.56,
+      "eval_gen_len": 822.348,
+      "eval_loss": 1.6523675918579102,
+      "eval_rouge1": 40.5489,
+      "eval_rouge2": 13.0695,
+      "eval_rougeL": 20.1256,
+      "eval_rougeLsum": 37.1369,
+      "eval_runtime": 897.0701,
+      "eval_samples_per_second": 0.279,
+      "eval_steps_per_second": 0.279,
+      "step": 100
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.0004997645530920631,
+      "loss": 1.7694,
+      "step": 101
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.0004997415996393894,
+      "loss": 1.7363,
+      "step": 102
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.000499717579336857,
+      "loss": 1.7284,
+      "step": 103
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.0004996924922870762,
+      "loss": 1.8031,
+      "step": 104
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.0004996663385972142,
+      "loss": 1.7258,
+      "step": 105
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.0004996391183789949,
+      "loss": 1.7247,
+      "step": 106
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0004996108317486977,
+      "loss": 1.6966,
+      "step": 107
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.0004995814788271582,
+      "loss": 1.7258,
+      "step": 108
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.0004995510597397663,
+      "loss": 1.8111,
+      "step": 109
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.0004995195746164671,
+      "loss": 1.7486,
+      "step": 110
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0004994870235917588,
+      "loss": 1.7584,
+      "step": 111
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.0004994534068046936,
+      "loss": 1.6991,
+      "step": 112
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.0004994187243988762,
+      "loss": 1.6487,
+      "step": 113
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.0004993829765224633,
+      "loss": 1.6318,
+      "step": 114
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.0004993461633281633,
+      "loss": 1.6386,
+      "step": 115
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.0004993082849732353,
+      "loss": 1.743,
+      "step": 116
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.0004992693416194886,
+      "loss": 1.6354,
+      "step": 117
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.000499229333433282,
+      "loss": 1.7054,
+      "step": 118
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.0004991882605855231,
+      "loss": 1.7264,
+      "step": 119
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.0004991461232516675,
+      "loss": 1.6596,
+      "step": 120
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.000499102921611718,
+      "loss": 1.6623,
+      "step": 121
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.0004990586558502241,
+      "loss": 1.698,
+      "step": 122
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.0004990133261562809,
+      "loss": 1.6981,
+      "step": 123
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.0004989669327235284,
+      "loss": 1.744,
+      "step": 124
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.000498919475750151,
+      "loss": 1.7371,
+      "step": 125
+    },
+    {
+      "epoch": 1.95,
+      "eval_gen_len": 765.6,
+      "eval_loss": 1.6293846368789673,
+      "eval_rouge1": 43.8594,
+      "eval_rouge2": 15.2962,
+      "eval_rougeL": 20.7807,
+      "eval_rougeLsum": 40.3461,
+      "eval_runtime": 841.3174,
+      "eval_samples_per_second": 0.297,
+      "eval_steps_per_second": 0.297,
+      "step": 125
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.0004988709554388757,
+      "loss": 1.7668,
+      "step": 126
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.0004988213719969726,
+      "loss": 1.7128,
+      "step": 127
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0004987707256362529,
+      "loss": 1.6625,
+      "step": 128
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.0004987190165730684,
+      "loss": 1.741,
+      "step": 129
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.0004986662450283106,
+      "loss": 1.6843,
+      "step": 130
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.0004986124112274099,
+      "loss": 1.5982,
+      "step": 131
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.0004985575154003344,
+      "loss": 1.5923,
+      "step": 132
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.0004985015577815888,
+      "loss": 1.7067,
+      "step": 133
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.0004984445386102135,
+      "loss": 1.6505,
+      "step": 134
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.0004983864581297841,
+      "loss": 1.6218,
+      "step": 135
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.0004983273165884096,
+      "loss": 1.6585,
+      "step": 136
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.0004982671142387316,
+      "loss": 1.6493,
+      "step": 137
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.0004982058513379235,
+      "loss": 1.6432,
+      "step": 138
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.000498143528147689,
+      "loss": 1.6534,
+      "step": 139
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.0004980801449342612,
+      "loss": 1.6082,
+      "step": 140
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.0004980157019684016,
+      "loss": 1.6816,
+      "step": 141
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.0004979501995253985,
+      "loss": 1.6422,
+      "step": 142
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.0004978836378850663,
+      "loss": 1.6053,
+      "step": 143
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 0.0004978160173317438,
+      "loss": 1.6641,
+      "step": 144
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.0004977473381542936,
+      "loss": 1.6122,
+      "step": 145
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0004976776006461005,
+      "loss": 1.7197,
+      "step": 146
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.0004976068051050702,
+      "loss": 1.5697,
+      "step": 147
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.000497534951833628,
+      "loss": 1.7142,
+      "step": 148
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.0004974620411387178,
+      "loss": 1.6504,
+      "step": 149
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.0004973880733318007,
+      "loss": 1.6428,
+      "step": 150
+    },
+    {
+      "epoch": 2.34,
+      "eval_gen_len": 844.184,
+      "eval_loss": 1.605539321899414,
+      "eval_rouge1": 44.5054,
+      "eval_rouge2": 15.731,
+      "eval_rougeL": 21.2582,
+      "eval_rougeLsum": 40.9775,
+      "eval_runtime": 922.4102,
+      "eval_samples_per_second": 0.271,
+      "eval_steps_per_second": 0.271,
+      "step": 150
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.0004973130487288534,
+      "loss": 1.6719,
+      "step": 151
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.0004972369676503671,
+      "loss": 1.6606,
+      "step": 152
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.0004971598304213461,
+      "loss": 1.6061,
+      "step": 153
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.0004970816373713064,
+      "loss": 1.666,
+      "step": 154
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.0004970023888342742,
+      "loss": 1.6483,
+      "step": 155
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.0004969220851487844,
+      "loss": 1.6359,
+      "step": 156
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.0004968407266578797,
+      "loss": 1.6336,
+      "step": 157
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.0004967583137091085,
+      "loss": 1.7383,
+      "step": 158
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.0004966748466545235,
+      "loss": 1.6486,
+      "step": 159
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.0004965903258506806,
+      "loss": 1.642,
+      "step": 160
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.000496504751658637,
+      "loss": 1.6086,
+      "step": 161
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.0004964181244439498,
+      "loss": 1.6132,
+      "step": 162
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.0004963304445766743,
+      "loss": 1.7406,
+      "step": 163
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.000496241712431363,
+      "loss": 1.5989,
+      "step": 164
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 0.0004961519283870628,
+      "loss": 1.6851,
+      "step": 165
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.0004960610928273146,
+      "loss": 1.7098,
+      "step": 166
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.0004959692061401512,
+      "loss": 1.6937,
+      "step": 167
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0004958762687180956,
+      "loss": 1.587,
+      "step": 168
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.000495782280958159,
+      "loss": 1.6361,
+      "step": 169
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.0004956872432618399,
+      "loss": 1.5235,
+      "step": 170
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.0004955911560351215,
+      "loss": 1.6453,
+      "step": 171
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.000495494019688471,
+      "loss": 1.695,
+      "step": 172
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0004953958346368365,
+      "loss": 1.6334,
+      "step": 173
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0004952966012996466,
+      "loss": 1.6352,
+      "step": 174
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.0004951963201008077,
+      "loss": 1.6567,
+      "step": 175
+    },
+    {
+      "epoch": 2.73,
+      "eval_gen_len": 857.236,
+      "eval_loss": 1.6031476259231567,
+      "eval_rouge1": 47.3641,
+      "eval_rouge2": 16.9664,
+      "eval_rougeL": 21.4998,
+      "eval_rougeLsum": 43.994,
+      "eval_runtime": 935.2302,
+      "eval_samples_per_second": 0.267,
+      "eval_steps_per_second": 0.267,
+      "step": 175
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.0004950949914687023,
+      "loss": 1.6236,
+      "step": 176
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.0004949926158361879,
+      "loss": 1.6126,
+      "step": 177
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.000494889193640594,
+      "loss": 1.5928,
+      "step": 178
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.0004947847253237212,
+      "loss": 1.5603,
+      "step": 179
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.0004946792113318387,
+      "loss": 1.595,
+      "step": 180
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.0004945726521156827,
+      "loss": 1.6604,
+      "step": 181
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0004944650481304545,
+      "loss": 1.7129,
+      "step": 182
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.0004943563998358185,
+      "loss": 1.6234,
+      "step": 183
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.0004942467076958999,
+      "loss": 1.6682,
+      "step": 184
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.0004941359721792832,
+      "loss": 1.5966,
+      "step": 185
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0004940241937590102,
+      "loss": 1.6909,
+      "step": 186
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0004939113729125775,
+      "loss": 1.6324,
+      "step": 187
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.000493797510121935,
+      "loss": 1.5672,
+      "step": 188
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.0004936826058734832,
+      "loss": 1.6071,
+      "step": 189
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 0.0004935666606580718,
+      "loss": 1.6646,
+      "step": 190
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 0.0004934496749709976,
+      "loss": 1.6769,
+      "step": 191
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0004933316493120015,
+      "loss": 1.6161,
+      "step": 192
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 0.0004932125841852673,
+      "loss": 1.6829,
+      "step": 193
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 0.0004930924800994192,
+      "loss": 1.6455,
+      "step": 194
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 0.0004929713375675195,
+      "loss": 1.5707,
+      "step": 195
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 0.0004928491571070669,
+      "loss": 1.6266,
+      "step": 196
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 0.0004927259392399936,
+      "loss": 1.6399,
+      "step": 197
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 0.0004926016844926634,
+      "loss": 1.6322,
+      "step": 198
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 0.0004924763933958695,
+      "loss": 1.5177,
+      "step": 199
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 0.0004923500664848326,
+      "loss": 1.5773,
+      "step": 200
+    },
+    {
+      "epoch": 3.12,
+      "eval_gen_len": 841.86,
+      "eval_loss": 1.5855237245559692,
+      "eval_rouge1": 47.2284,
+      "eval_rouge2": 17.3099,
+      "eval_rougeL": 21.6793,
+      "eval_rougeLsum": 43.9018,
+      "eval_runtime": 920.2336,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
+      "step": 200
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 0.0004922227042991976,
+      "loss": 1.5497,
+      "step": 201
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 0.0004920943073830322,
+      "loss": 1.6289,
+      "step": 202
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 0.0004919648762848243,
+      "loss": 1.5412,
+      "step": 203
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 0.0004918344115574796,
+      "loss": 1.608,
+      "step": 204
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 0.0004917029137583191,
+      "loss": 1.6501,
+      "step": 205
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 0.0004915703834490773,
+      "loss": 1.5859,
+      "step": 206
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 0.0004914368211958989,
+      "loss": 1.5605,
+      "step": 207
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 0.0004913022275693372,
+      "loss": 1.5337,
+      "step": 208
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 0.0004911666031443512,
+      "loss": 1.5706,
+      "step": 209
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 0.0004910299485003034,
+      "loss": 1.5739,
+      "step": 210
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 0.0004908922642209571,
+      "loss": 1.6501,
+      "step": 211
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 0.000490753550894474,
+      "loss": 1.5612,
+      "step": 212
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 0.0004906138091134118,
+      "loss": 1.6209,
+      "step": 213
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 0.0004904730394747215,
+      "loss": 1.5751,
+      "step": 214
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 0.0004903312425797449,
+      "loss": 1.6414,
+      "step": 215
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 0.0004901884190342121,
+      "loss": 1.6426,
+      "step": 216
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 0.0004900445694482387,
+      "loss": 1.5086,
+      "step": 217
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 0.0004898996944363237,
+      "loss": 1.5461,
+      "step": 218
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 0.0004897537946173461,
+      "loss": 1.6256,
+      "step": 219
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 0.0004896068706145633,
+      "loss": 1.5871,
+      "step": 220
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 0.0004894589230556069,
+      "loss": 1.6357,
+      "step": 221
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 0.0004893099525724818,
+      "loss": 1.5664,
+      "step": 222
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 0.0004891599598015621,
+      "loss": 1.5824,
+      "step": 223
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 0.0004890089453835894,
+      "loss": 1.5546,
+      "step": 224
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 0.0004888569099636692,
+      "loss": 1.5614,
+      "step": 225
+    },
+    {
+      "epoch": 3.52,
+      "eval_gen_len": 832.8,
+      "eval_loss": 1.5882554054260254,
+      "eval_rouge1": 46.4612,
+      "eval_rouge2": 17.1368,
+      "eval_rougeL": 21.5931,
+      "eval_rougeLsum": 43.1184,
+      "eval_runtime": 912.3034,
+      "eval_samples_per_second": 0.274,
+      "eval_steps_per_second": 0.274,
+      "step": 225
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 0.0004887038541912687,
+      "loss": 1.5713,
+      "step": 226
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 0.0004885497787202137,
+      "loss": 1.5822,
+      "step": 227
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 0.0004883946842086861,
+      "loss": 1.571,
+      "step": 228
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 0.00048823857131922093,
+      "loss": 1.6026,
+      "step": 229
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 0.00048808144071870363,
+      "loss": 1.5299,
+      "step": 230
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 0.00048792329307836705,
+      "loss": 1.5651,
+      "step": 231
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 0.0004877641290737884,
+      "loss": 1.5702,
+      "step": 232
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 0.00048760394938488703,
+      "loss": 1.5424,
+      "step": 233
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 0.0004874427546959208,
+      "loss": 1.5763,
+      "step": 234
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 0.0004872805456954837,
+      "loss": 1.493,
+      "step": 235
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 0.00048711732307650236,
+      "loss": 1.5906,
+      "step": 236
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 0.00048695308753623367,
+      "loss": 1.5255,
+      "step": 237
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 0.0004867878397762615,
+      "loss": 1.4685,
+      "step": 238
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 0.0004866215805024935,
+      "loss": 1.5349,
+      "step": 239
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 0.00048645431042515866,
+      "loss": 1.6122,
+      "step": 240
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 0.0004862860302588037,
+      "loss": 1.5036,
+      "step": 241
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 0.0004861167407222904,
+      "loss": 1.5529,
+      "step": 242
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 0.0004859464425387922,
+      "loss": 1.6724,
+      "step": 243
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 0.0004857751364357913,
+      "loss": 1.6384,
+      "step": 244
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 0.00048560282314507594,
+      "loss": 1.5088,
+      "step": 245
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 0.0004854295034027364,
+      "loss": 1.5407,
+      "step": 246
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 0.00048525517794916254,
+      "loss": 1.6202,
+      "step": 247
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 0.00048507984752904035,
+      "loss": 1.5655,
+      "step": 248
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 0.000484903512891349,
+      "loss": 1.5927,
+      "step": 249
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 0.0004847261747893574,
+      "loss": 1.5328,
+      "step": 250
+    },
+    {
+      "epoch": 3.91,
+      "eval_gen_len": 790.056,
+      "eval_loss": 1.5730280876159668,
+      "eval_rouge1": 46.5685,
+      "eval_rouge2": 17.5423,
+      "eval_rougeL": 22.2082,
+      "eval_rougeLsum": 43.1811,
+      "eval_runtime": 869.0966,
+      "eval_samples_per_second": 0.288,
+      "eval_steps_per_second": 0.288,
+      "step": 250
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 0.0004845478339806211,
+      "loss": 1.613,
+      "step": 251
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 0.00048436849122697885,
+      "loss": 1.5404,
+      "step": 252
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 0.00048418814729454985,
+      "loss": 1.5256,
+      "step": 253
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 0.0004840068029537299,
+      "loss": 1.5344,
+      "step": 254
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 0.00048382445897918847,
+      "loss": 1.5377,
+      "step": 255
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0004836411161498652,
+      "loss": 1.5981,
+      "step": 256
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 0.0004834567752489668,
+      "loss": 1.645,
+      "step": 257
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 0.0004832714370639633,
+      "loss": 1.4417,
+      "step": 258
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 0.0004830851023865852,
+      "loss": 1.5535,
+      "step": 259
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 0.00048289777201281976,
+      "loss": 1.5704,
+      "step": 260
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 0.0004827094467429076,
+      "loss": 1.5286,
+      "step": 261
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 0.0004825201273813393,
+      "loss": 1.6414,
+      "step": 262
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 0.0004823298147368523,
+      "loss": 1.4781,
+      "step": 263
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 0.0004821385096224268,
+      "loss": 1.4318,
+      "step": 264
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 0.0004819462128552829,
+      "loss": 1.5628,
+      "step": 265
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 0.00048175292525687676,
+      "loss": 1.474,
+      "step": 266
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 0.00048155864765289717,
+      "loss": 1.4858,
+      "step": 267
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 0.00048136338087326213,
+      "loss": 1.5357,
+      "step": 268
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 0.0004811671257521151,
+      "loss": 1.5283,
+      "step": 269
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 0.0004809698831278217,
+      "loss": 1.4921,
+      "step": 270
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 0.0004807716538429658,
+      "loss": 1.5086,
+      "step": 271
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 0.0004805724387443462,
+      "loss": 1.5572,
+      "step": 272
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 0.0004803722386829729,
+      "loss": 1.4695,
+      "step": 273
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 0.0004801710545140635,
+      "loss": 1.534,
+      "step": 274
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 0.00047996888709703954,
+      "loss": 1.5194,
+      "step": 275
+    },
+    {
+      "epoch": 4.3,
+      "eval_gen_len": 825.868,
+      "eval_loss": 1.568995475769043,
+      "eval_rouge1": 47.6205,
+      "eval_rouge2": 18.377,
+      "eval_rougeL": 22.7639,
+      "eval_rougeLsum": 44.3701,
+      "eval_runtime": 903.1087,
+      "eval_samples_per_second": 0.277,
+      "eval_steps_per_second": 0.277,
+      "step": 275
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 0.00047976573729552276,
+      "loss": 1.5358,
+      "step": 276
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 0.0004795616059773315,
+      "loss": 1.5047,
+      "step": 277
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 0.0004793564940144769,
+      "loss": 1.5063,
+      "step": 278
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 0.00047915040228315933,
+      "loss": 1.6249,
+      "step": 279
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 0.00047894333166376434,
+      "loss": 1.5403,
+      "step": 280
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 0.0004787352830408595,
+      "loss": 1.5495,
+      "step": 281
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 0.0004785262573031899,
+      "loss": 1.5596,
+      "step": 282
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 0.0004783162553436747,
+      "loss": 1.5155,
+      "step": 283
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 0.0004781052780594034,
+      "loss": 1.4722,
+      "step": 284
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 0.00047789332635163204,
+      "loss": 1.5476,
+      "step": 285
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 0.0004776804011257788,
+      "loss": 1.5014,
+      "step": 286
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 0.00047746650329142103,
+      "loss": 1.4926,
+      "step": 287
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 0.00047725163376229063,
+      "loss": 1.5515,
+      "step": 288
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 0.00047703579345627036,
+      "loss": 1.4934,
+      "step": 289
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 0.00047681898329539,
+      "loss": 1.5362,
+      "step": 290
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 0.0004766012042058225,
+      "loss": 1.5657,
+      "step": 291
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 0.00047638245711787976,
+      "loss": 1.5209,
+      "step": 292
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 0.00047616274296600873,
+      "loss": 1.5339,
+      "step": 293
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.0004759420626887877,
+      "loss": 1.5366,
+      "step": 294
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 0.00047572041722892194,
+      "loss": 1.4674,
+      "step": 295
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 0.0004754978075332398,
+      "loss": 1.4913,
+      "step": 296
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 0.00047527423455268855,
+      "loss": 1.5963,
+      "step": 297
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 0.00047504969924233066,
+      "loss": 1.5394,
+      "step": 298
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 0.00047482420256133933,
+      "loss": 1.4853,
+      "step": 299
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 0.00047459774547299475,
+      "loss": 1.571,
+      "step": 300
+    },
+    {
+      "epoch": 4.69,
+      "eval_gen_len": 794.032,
+      "eval_loss": 1.5675740242004395,
+      "eval_rouge1": 49.2203,
+      "eval_rouge2": 19.1109,
+      "eval_rougeL": 22.8005,
+      "eval_rougeLsum": 46.0679,
+      "eval_runtime": 869.0821,
+      "eval_samples_per_second": 0.288,
+      "eval_steps_per_second": 0.288,
+      "step": 300
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 0.00047437032894467944,
+      "loss": 1.5875,
+      "step": 301
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 0.000474141953947875,
+      "loss": 1.4682,
+      "step": 302
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 0.00047391262145815674,
+      "loss": 1.5559,
+      "step": 303
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 0.0004736823324551909,
+      "loss": 1.5251,
+      "step": 304
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 0.00047345108792272927,
+      "loss": 1.5021,
+      "step": 305
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 0.0004732188888486057,
+      "loss": 1.5081,
+      "step": 306
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 0.00047298573622473166,
+      "loss": 1.5025,
+      "step": 307
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 0.00047275163104709196,
+      "loss": 1.5625,
+      "step": 308
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 0.00047251657431574055,
+      "loss": 1.6046,
+      "step": 309
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 0.0004722805670347963,
+      "loss": 1.5373,
+      "step": 310
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 0.0004720436102124386,
+      "loss": 1.5182,
+      "step": 311
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 0.000471805704860903,
+      "loss": 1.5097,
+      "step": 312
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 0.00047156685199647714,
+      "loss": 1.5336,
+      "step": 313
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 0.0004713270526394963,
+      "loss": 1.5097,
+      "step": 314
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 0.00047108630781433886,
+      "loss": 1.5173,
+      "step": 315
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 0.0004708446185494222,
+      "loss": 1.5142,
+      "step": 316
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 0.00047060198587719793,
+      "loss": 1.466,
+      "step": 317
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 0.00047035841083414804,
+      "loss": 1.484,
+      "step": 318
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 0.0004701138944607799,
+      "loss": 1.5159,
+      "step": 319
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00046986843780162223,
+      "loss": 1.4444,
+      "step": 320
+    },
+    {
+      "epoch": 5.02,
+      "learning_rate": 0.0004696220419052205,
+      "loss": 1.5786,
+      "step": 321
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 0.00046937470782413215,
+      "loss": 1.5069,
+      "step": 322
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 0.0004691264366149227,
+      "loss": 1.4675,
+      "step": 323
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 0.00046887722933816074,
+      "loss": 1.4876,
+      "step": 324
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 0.0004686270870584136,
+      "loss": 1.4275,
+      "step": 325
+    },
+    {
+      "epoch": 5.08,
+      "eval_gen_len": 833.068,
+      "eval_loss": 1.5656464099884033,
+      "eval_rouge1": 50.6982,
+      "eval_rouge2": 20.0278,
+      "eval_rougeL": 23.5585,
+      "eval_rougeLsum": 47.5036,
+      "eval_runtime": 908.634,
+      "eval_samples_per_second": 0.275,
+      "eval_steps_per_second": 0.275,
+      "step": 325
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 0.00046837601084424277,
+      "loss": 1.5108,
+      "step": 326
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 0.0004681240017681993,
+      "loss": 1.4831,
+      "step": 327
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 0.0004678710609068193,
+      "loss": 1.421,
+      "step": 328
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 0.0004676171893406194,
+      "loss": 1.4381,
+      "step": 329
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 0.0004673623881540917,
+      "loss": 1.5484,
+      "step": 330
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 0.0004671066584356999,
+      "loss": 1.4882,
+      "step": 331
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 0.0004668500012778738,
+      "loss": 1.4782,
+      "step": 332
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 0.0004665924177770053,
+      "loss": 1.515,
+      "step": 333
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 0.00046633390903344355,
+      "loss": 1.5186,
+      "step": 334
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 0.00046607447615148984,
+      "loss": 1.5037,
+      "step": 335
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 0.0004658141202393935,
+      "loss": 1.5083,
+      "step": 336
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 0.0004655528424093467,
+      "loss": 1.3176,
+      "step": 337
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 0.00046529064377747993,
+      "loss": 1.491,
+      "step": 338
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 0.00046502752546385723,
+      "loss": 1.4108,
+      "step": 339
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 0.0004647634885924713,
+      "loss": 1.5044,
+      "step": 340
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 0.0004644985342912388,
+      "loss": 1.4611,
+      "step": 341
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 0.0004642326636919952,
+      "loss": 1.4847,
+      "step": 342
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 0.00046396587793049083,
+      "loss": 1.5091,
+      "step": 343
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 0.00046369817814638476,
+      "loss": 1.498,
+      "step": 344
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 0.000463429565483241,
+      "loss": 1.4504,
+      "step": 345
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 0.00046316004108852305,
+      "loss": 1.5032,
+      "step": 346
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 0.00046288960611358924,
+      "loss": 1.4502,
+      "step": 347
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 0.0004626182617136877,
+      "loss": 1.5371,
+      "step": 348
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 0.00046234600904795144,
+      "loss": 1.4603,
+      "step": 349
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 0.0004620728492793934,
+      "loss": 1.4912,
+      "step": 350
+    },
+    {
+      "epoch": 5.47,
+      "eval_gen_len": 793.068,
+      "eval_loss": 1.5625073909759521,
+      "eval_rouge1": 50.3371,
+      "eval_rouge2": 19.8639,
+      "eval_rougeL": 23.3666,
+      "eval_rougeLsum": 47.1898,
+      "eval_runtime": 869.2692,
+      "eval_samples_per_second": 0.288,
+      "eval_steps_per_second": 0.288,
+      "step": 350
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 0.00046179878357490155,
+      "loss": 1.4555,
+      "step": 351
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 0.00046152381310523384,
+      "loss": 1.5145,
+      "step": 352
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 0.00046124793904501316,
+      "loss": 1.4795,
+      "step": 353
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 0.0004609711625727224,
+      "loss": 1.4764,
+      "step": 354
+    },
+    {
+      "epoch": 5.55,
+      "learning_rate": 0.00046069348487069934,
+      "loss": 1.5356,
+      "step": 355
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 0.00046041490712513177,
+      "loss": 1.4876,
+      "step": 356
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 0.0004601354305260522,
+      "loss": 1.5048,
+      "step": 357
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 0.00045985505626733287,
+      "loss": 1.4918,
+      "step": 358
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 0.0004595737855466807,
+      "loss": 1.5214,
+      "step": 359
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 0.00045929161956563216,
+      "loss": 1.5063,
+      "step": 360
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 0.000459008559529548,
+      "loss": 1.4175,
+      "step": 361
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 0.00045872460664760827,
+      "loss": 1.4646,
+      "step": 362
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 0.00045843976213280716,
+      "loss": 1.469,
+      "step": 363
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 0.0004581540272019476,
+      "loss": 1.4715,
+      "step": 364
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 0.00045786740307563633,
+      "loss": 1.5284,
+      "step": 365
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 0.0004575798909782785,
+      "loss": 1.4902,
+      "step": 366
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 0.0004572914921380726,
+      "loss": 1.4994,
+      "step": 367
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 0.000457002207787005,
+      "loss": 1.4458,
+      "step": 368
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 0.00045671203916084494,
+      "loss": 1.4403,
+      "step": 369
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 0.00045642098749913896,
+      "loss": 1.4553,
+      "step": 370
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 0.00045612905404520586,
+      "loss": 1.506,
+      "step": 371
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 0.00045583624004613145,
+      "loss": 1.4695,
+      "step": 372
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 0.00045554254675276266,
+      "loss": 1.5557,
+      "step": 373
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 0.00045524797541970306,
+      "loss": 1.4717,
+      "step": 374
+    },
+    {
+      "epoch": 5.86,
+      "learning_rate": 0.0004549525273053067,
+      "loss": 1.4764,
+      "step": 375
+    },
+    {
+      "epoch": 5.86,
+      "eval_gen_len": 819.86,
+      "eval_loss": 1.5531580448150635,
+      "eval_rouge1": 50.9702,
+      "eval_rouge2": 20.7532,
+      "eval_rougeL": 23.8765,
+      "eval_rougeLsum": 47.9915,
+      "eval_runtime": 894.7339,
+      "eval_samples_per_second": 0.279,
+      "eval_steps_per_second": 0.279,
+      "step": 375
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 0.0004546562036716732,
+      "loss": 1.4498,
+      "step": 376
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 0.0004543590057846422,
+      "loss": 1.5125,
+      "step": 377
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 0.00045406093491378815,
+      "loss": 1.4536,
+      "step": 378
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 0.00045376199233241454,
+      "loss": 1.5585,
+      "step": 379
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 0.00045346217931754873,
+      "loss": 1.4651,
+      "step": 380
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 0.00045316149714993645,
+      "loss": 1.5225,
+      "step": 381
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 0.00045285994711403613,
+      "loss": 1.4312,
+      "step": 382
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 0.00045255753049801365,
+      "loss": 1.5308,
+      "step": 383
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0004522542485937369,
+      "loss": 1.4896,
+      "step": 384
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 0.0004519501026967699,
+      "loss": 1.5514,
+      "step": 385
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 0.0004516450941063677,
+      "loss": 1.4681,
+      "step": 386
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 0.00045133922412547034,
+      "loss": 1.3779,
+      "step": 387
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 0.0004510324940606979,
+      "loss": 1.4405,
+      "step": 388
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 0.00045072490522234435,
+      "loss": 1.4182,
+      "step": 389
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 0.0004504164589243721,
+      "loss": 1.4541,
+      "step": 390
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 0.00045010715648440686,
+      "loss": 1.4429,
+      "step": 391
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 0.0004497969992237312,
+      "loss": 1.3827,
+      "step": 392
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 0.0004494859884672795,
+      "loss": 1.4369,
+      "step": 393
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 0.00044917412554363223,
+      "loss": 1.4117,
+      "step": 394
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 0.00044886141178500984,
+      "loss": 1.368,
+      "step": 395
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 0.00044854784852726775,
+      "loss": 1.5125,
+      "step": 396
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 0.0004482334371098901,
+      "loss": 1.4048,
+      "step": 397
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 0.0004479181788759842,
+      "loss": 1.4849,
+      "step": 398
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 0.0004476020751722748,
+      "loss": 1.4988,
+      "step": 399
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 0.00044728512734909845,
+      "loss": 1.3972,
+      "step": 400
+    },
+    {
+      "epoch": 6.25,
+      "eval_gen_len": 770.78,
+      "eval_loss": 1.5564013719558716,
+      "eval_rouge1": 49.279,
+      "eval_rouge2": 19.4781,
+      "eval_rougeL": 23.1018,
+      "eval_rougeLsum": 46.1942,
+      "eval_runtime": 846.8994,
+      "eval_samples_per_second": 0.295,
+      "eval_steps_per_second": 0.295,
+      "step": 400
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 0.00044696733676039745,
+      "loss": 1.4684,
+      "step": 401
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 0.00044664870476371447,
+      "loss": 1.4135,
+      "step": 402
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 0.0004463292327201862,
+      "loss": 1.4849,
+      "step": 403
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 0.00044600892199453824,
+      "loss": 1.4578,
+      "step": 404
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 0.00044568777395507863,
+      "loss": 1.4226,
+      "step": 405
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 0.0004453657899736923,
+      "loss": 1.4242,
+      "step": 406
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 0.00044504297142583505,
+      "loss": 1.4406,
+      "step": 407
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 0.00044471931969052817,
+      "loss": 1.5071,
+      "step": 408
+    },
+    {
+      "epoch": 6.39,
+      "learning_rate": 0.0004443948361503517,
+      "loss": 1.4148,
+      "step": 409
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 0.00044406952219143935,
+      "loss": 1.3703,
+      "step": 410
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 0.0004437433792034721,
+      "loss": 1.4795,
+      "step": 411
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 0.00044341640857967234,
+      "loss": 1.4414,
+      "step": 412
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 0.00044308861171679816,
+      "loss": 1.4081,
+      "step": 413
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 0.0004427599900151368,
+      "loss": 1.5113,
+      "step": 414
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 0.0004424305448784995,
+      "loss": 1.4275,
+      "step": 415
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 0.0004421002777142148,
+      "loss": 1.5111,
+      "step": 416
+    },
+    {
+      "epoch": 6.52,
+      "learning_rate": 0.00044176918993312284,
+      "loss": 1.467,
+      "step": 417
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 0.0004414372829495693,
+      "loss": 1.3737,
+      "step": 418
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 0.0004411045581813994,
+      "loss": 1.4639,
+      "step": 419
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 0.00044077101704995163,
+      "loss": 1.3785,
+      "step": 420
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 0.00044043666098005205,
+      "loss": 1.5022,
+      "step": 421
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 0.0004401014914000078,
+      "loss": 1.5083,
+      "step": 422
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 0.00043976550974160114,
+      "loss": 1.4377,
+      "step": 423
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 0.00043942871744008375,
+      "loss": 1.4162,
+      "step": 424
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 0.00043909111593416967,
+      "loss": 1.4479,
+      "step": 425
+    },
+    {
+      "epoch": 6.64,
+      "eval_gen_len": 806.244,
+      "eval_loss": 1.5529119968414307,
+      "eval_rouge1": 50.3317,
+      "eval_rouge2": 20.2888,
+      "eval_rougeL": 23.4454,
+      "eval_rougeLsum": 47.3491,
+      "eval_runtime": 882.7996,
+      "eval_samples_per_second": 0.283,
+      "eval_steps_per_second": 0.283,
+      "step": 425
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 0.00043875270666603015,
+      "loss": 1.4568,
+      "step": 426
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 0.00043841349108128704,
+      "loss": 1.4055,
+      "step": 427
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 0.00043807347062900623,
+      "loss": 1.4544,
+      "step": 428
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 0.00043773264676169225,
+      "loss": 1.4898,
+      "step": 429
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 0.00043739102093528153,
+      "loss": 1.4298,
+      "step": 430
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 0.00043704859460913634,
+      "loss": 1.4306,
+      "step": 431
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 0.0004367053692460385,
+      "loss": 1.4758,
+      "step": 432
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 0.00043636134631218316,
+      "loss": 1.4058,
+      "step": 433
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 0.00043601652727717267,
+      "loss": 1.4258,
+      "step": 434
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 0.00043567091361401,
+      "loss": 1.471,
+      "step": 435
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 0.00043532450679909277,
+      "loss": 1.4146,
+      "step": 436
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 0.0004349773083122066,
+      "loss": 1.5167,
+      "step": 437
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 0.0004346293196365193,
+      "loss": 1.4361,
+      "step": 438
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 0.0004342805422585739,
+      "loss": 1.5092,
+      "step": 439
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 0.00043393097766828293,
+      "loss": 1.5393,
+      "step": 440
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 0.0004335806273589214,
+      "loss": 1.4637,
+      "step": 441
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 0.000433229492827121,
+      "loss": 1.4411,
+      "step": 442
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 0.00043287757557286343,
+      "loss": 1.4045,
+      "step": 443
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 0.0004325248770994741,
+      "loss": 1.44,
+      "step": 444
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 0.0004321713989136153,
+      "loss": 1.371,
+      "step": 445
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 0.0004318171425252808,
+      "loss": 1.4431,
+      "step": 446
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 0.0004314621094477879,
+      "loss": 1.4203,
+      "step": 447
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.0004311063011977723,
+      "loss": 1.4698,
+      "step": 448
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 0.0004307497192951809,
+      "loss": 1.4772,
+      "step": 449
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 0.0004303923652632655,
+      "loss": 1.4567,
+      "step": 450
+    },
+    {
+      "epoch": 7.03,
+      "eval_gen_len": 787.48,
+      "eval_loss": 1.5590322017669678,
+      "eval_rouge1": 52.2209,
+      "eval_rouge2": 21.2868,
+      "eval_rougeL": 23.9284,
+      "eval_rougeLsum": 49.1691,
+      "eval_runtime": 864.0005,
+      "eval_samples_per_second": 0.289,
+      "eval_steps_per_second": 0.289,
+      "step": 450
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 0.00043003424062857656,
+      "loss": 1.4111,
+      "step": 451
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 0.000429675346920956,
+      "loss": 1.3935,
+      "step": 452
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 0.0004293156856735313,
+      "loss": 1.4402,
+      "step": 453
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 0.00042895525842270865,
+      "loss": 1.3481,
+      "step": 454
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 0.00042859406670816663,
+      "loss": 1.4009,
+      "step": 455
+    },
+    {
+      "epoch": 7.12,
+      "learning_rate": 0.0004282321120728493,
+      "loss": 1.3885,
+      "step": 456
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 0.00042786939606295976,
+      "loss": 1.3837,
+      "step": 457
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 0.0004275059202279537,
+      "loss": 1.3358,
+      "step": 458
+    },
+    {
+      "epoch": 7.17,
+      "learning_rate": 0.0004271416861205325,
+      "loss": 1.4314,
+      "step": 459
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 0.00042677669529663686,
+      "loss": 1.3707,
+      "step": 460
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 0.00042641094931544,
+      "loss": 1.421,
+      "step": 461
+    },
+    {
+      "epoch": 7.22,
+      "learning_rate": 0.00042604444973934097,
+      "loss": 1.3735,
+      "step": 462
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 0.0004256771981339581,
+      "loss": 1.4209,
+      "step": 463
+    },
+    {
+      "epoch": 7.25,
+      "learning_rate": 0.00042530919606812215,
+      "loss": 1.381,
+      "step": 464
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 0.0004249404451138699,
+      "loss": 1.4744,
+      "step": 465
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 0.0004245709468464371,
+      "loss": 1.4357,
+      "step": 466
+    },
+    {
+      "epoch": 7.3,
+      "learning_rate": 0.0004242007028442519,
+      "loss": 1.384,
+      "step": 467
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 0.00042382971468892807,
+      "loss": 1.437,
+      "step": 468
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 0.0004234579839652583,
+      "loss": 1.4199,
+      "step": 469
+    },
+    {
+      "epoch": 7.34,
+      "learning_rate": 0.0004230855122612074,
+      "loss": 1.4603,
+      "step": 470
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 0.0004227123011679056,
+      "loss": 1.3556,
+      "step": 471
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 0.00042233835227964146,
+      "loss": 1.3665,
+      "step": 472
+    },
+    {
+      "epoch": 7.39,
+      "learning_rate": 0.0004219636671938554,
+      "loss": 1.4353,
+      "step": 473
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 0.00042158824751113277,
+      "loss": 1.3791,
+      "step": 474
+    },
+    {
+      "epoch": 7.42,
+      "learning_rate": 0.0004212120948351968,
+      "loss": 1.3933,
+      "step": 475
+    },
+    {
+      "epoch": 7.42,
+      "eval_gen_len": 842.664,
+      "eval_loss": 1.556054949760437,
+      "eval_rouge1": 51.9578,
+      "eval_rouge2": 20.5806,
+      "eval_rougeL": 23.7177,
+      "eval_rougeLsum": 48.9121,
+      "eval_runtime": 918.7994,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
+      "step": 475
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 0.0004208352107729021,
+      "loss": 1.4203,
+      "step": 476
+    },
+    {
+      "epoch": 7.45,
+      "learning_rate": 0.0004204575969342277,
+      "loss": 1.3513,
+      "step": 477
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 0.0004200792549322698,
+      "loss": 1.3963,
+      "step": 478
+    },
+    {
+      "epoch": 7.48,
+      "learning_rate": 0.00041970018638323546,
+      "loss": 1.4117,
+      "step": 479
+    },
+    {
+      "epoch": 7.5,
+      "learning_rate": 0.0004193203929064353,
+      "loss": 1.4318,
+      "step": 480
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 0.00041893987612427665,
+      "loss": 1.4077,
+      "step": 481
+    },
+    {
+      "epoch": 7.53,
+      "learning_rate": 0.0004185586376622569,
+      "loss": 1.4389,
+      "step": 482
+    },
+    {
+      "epoch": 7.55,
+      "learning_rate": 0.0004181766791489559,
+      "loss": 1.4365,
+      "step": 483
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 0.0004177940022160299,
+      "loss": 1.3756,
+      "step": 484
+    },
+    {
+      "epoch": 7.58,
+      "learning_rate": 0.00041741060849820376,
+      "loss": 1.4333,
+      "step": 485
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 0.0004170264996332644,
+      "loss": 1.3782,
+      "step": 486
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 0.00041664167726205393,
+      "loss": 1.3374,
+      "step": 487
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 0.00041625614302846206,
+      "loss": 1.373,
+      "step": 488
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 0.0004158698985794197,
+      "loss": 1.4767,
+      "step": 489
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 0.00041548294556489163,
+      "loss": 1.421,
+      "step": 490
+    },
+    {
+      "epoch": 7.67,
+      "learning_rate": 0.00041509528563786946,
+      "loss": 1.4624,
+      "step": 491
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 0.00041470692045436446,
+      "loss": 1.4403,
+      "step": 492
+    },
+    {
+      "epoch": 7.7,
+      "learning_rate": 0.00041431785167340095,
+      "loss": 1.4075,
+      "step": 493
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 0.0004139280809570086,
+      "loss": 1.467,
+      "step": 494
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 0.0004135376099702158,
+      "loss": 1.4193,
+      "step": 495
+    },
+    {
+      "epoch": 7.75,
+      "learning_rate": 0.00041314644038104216,
+      "loss": 1.4129,
+      "step": 496
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 0.0004127545738604918,
+      "loss": 1.392,
+      "step": 497
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 0.0004123620120825459,
+      "loss": 1.3938,
+      "step": 498
+    },
+    {
+      "epoch": 7.8,
+      "learning_rate": 0.0004119687567241557,
+      "loss": 1.4033,
+      "step": 499
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 0.0004115748094652352,
+      "loss": 1.4245,
+      "step": 500
+    },
+    {
+      "epoch": 7.81,
+      "eval_gen_len": 813.772,
+      "eval_loss": 1.5419564247131348,
+      "eval_rouge1": 52.3725,
+      "eval_rouge2": 21.7787,
+      "eval_rougeL": 24.5209,
+      "eval_rougeLsum": 49.4003,
+      "eval_runtime": 890.9,
+      "eval_samples_per_second": 0.281,
+      "eval_steps_per_second": 0.281,
+      "step": 500
+    },
+    {
+      "epoch": 7.83,
+      "learning_rate": 0.0004111801719886542,
+      "loss": 1.3977,
+      "step": 501
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 0.0004107848459802309,
+      "loss": 1.3682,
+      "step": 502
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 0.0004103888331287247,
+      "loss": 1.4175,
+      "step": 503
+    },
+    {
+      "epoch": 7.87,
+      "learning_rate": 0.0004099921351258292,
+      "loss": 1.4152,
+      "step": 504
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 0.00040959475366616474,
+      "loss": 1.4109,
+      "step": 505
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 0.0004091966904472715,
+      "loss": 1.4189,
+      "step": 506
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 0.0004087979471696016,
+      "loss": 1.3991,
+      "step": 507
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 0.0004083985255365126,
+      "loss": 1.4259,
+      "step": 508
+    },
+    {
+      "epoch": 7.95,
+      "learning_rate": 0.0004079984272542597,
+      "loss": 1.4068,
+      "step": 509
+    },
+    {
+      "epoch": 7.97,
+      "learning_rate": 0.00040759765403198877,
+      "loss": 1.3935,
+      "step": 510
+    },
+    {
+      "epoch": 7.98,
+      "learning_rate": 0.00040719620758172863,
+      "loss": 1.4305,
+      "step": 511
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00040679408961838426,
+      "loss": 1.4299,
+      "step": 512
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 0.00040639130185972903,
+      "loss": 1.4568,
+      "step": 513
+    },
+    {
+      "epoch": 8.03,
+      "learning_rate": 0.00040598784602639777,
+      "loss": 1.3649,
+      "step": 514
+    },
+    {
+      "epoch": 8.05,
+      "learning_rate": 0.000405583723841879,
+      "loss": 1.417,
+      "step": 515
+    },
+    {
+      "epoch": 8.06,
+      "learning_rate": 0.0004051789370325078,
+      "loss": 1.3732,
+      "step": 516
+    },
+    {
+      "epoch": 8.08,
+      "learning_rate": 0.00040477348732745853,
+      "loss": 1.346,
+      "step": 517
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 0.00040436737645873716,
+      "loss": 1.3935,
+      "step": 518
+    },
+    {
+      "epoch": 8.11,
+      "learning_rate": 0.00040396060616117423,
+      "loss": 1.3829,
+      "step": 519
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 0.000403553178172417,
+      "loss": 1.3617,
+      "step": 520
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 0.0004031450942329224,
+      "loss": 1.3344,
+      "step": 521
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 0.0004027363560859494,
+      "loss": 1.3731,
+      "step": 522
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 0.00040232696547755174,
+      "loss": 1.367,
+      "step": 523
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 0.0004019169241565703,
+      "loss": 1.4125,
+      "step": 524
+    },
+    {
+      "epoch": 8.2,
+      "learning_rate": 0.0004015062338746256,
+      "loss": 1.3033,
+      "step": 525
+    },
+    {
+      "epoch": 8.2,
+      "eval_gen_len": 824.66,
+      "eval_loss": 1.549867868423462,
+      "eval_rouge1": 52.7839,
+      "eval_rouge2": 21.589,
+      "eval_rougeL": 24.5617,
+      "eval_rougeLsum": 49.8609,
+      "eval_runtime": 900.6753,
+      "eval_samples_per_second": 0.278,
+      "eval_steps_per_second": 0.278,
+      "step": 525
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 0.0004010948963861104,
+      "loss": 1.3803,
+      "step": 526
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 0.0004006829134481824,
+      "loss": 1.3491,
+      "step": 527
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 0.00040027028682075626,
+      "loss": 1.4008,
+      "step": 528
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 0.00039985701826649665,
+      "loss": 1.3536,
+      "step": 529
+    },
+    {
+      "epoch": 8.28,
+      "learning_rate": 0.0003994431095508102,
+      "loss": 1.2883,
+      "step": 530
+    },
+    {
+      "epoch": 8.3,
+      "learning_rate": 0.0003990285624418384,
+      "loss": 1.356,
+      "step": 531
+    },
+    {
+      "epoch": 8.31,
+      "learning_rate": 0.00039861337871044953,
+      "loss": 1.3975,
+      "step": 532
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 0.0003981975601302317,
+      "loss": 1.3431,
+      "step": 533
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 0.00039778110847748485,
+      "loss": 1.3478,
+      "step": 534
+    },
+    {
+      "epoch": 8.36,
+      "learning_rate": 0.00039736402553121336,
+      "loss": 1.3866,
+      "step": 535
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 0.0003969463130731183,
+      "loss": 1.4443,
+      "step": 536
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 0.0003965279728875899,
+      "loss": 1.3758,
+      "step": 537
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 0.0003961090067617,
+      "loss": 1.3314,
+      "step": 538
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 0.00039568941648519416,
+      "loss": 1.3783,
+      "step": 539
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 0.0003952692038504846,
+      "loss": 1.4518,
+      "step": 540
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 0.00039484837065264155,
+      "loss": 1.3759,
+      "step": 541
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 0.00039442691868938674,
+      "loss": 1.3526,
+      "step": 542
+    },
+    {
+      "epoch": 8.48,
+      "learning_rate": 0.00039400484976108473,
+      "loss": 1.3463,
+      "step": 543
+    },
+    {
+      "epoch": 8.5,
+      "learning_rate": 0.0003935821656707359,
+      "loss": 1.3783,
+      "step": 544
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 0.0003931588682239684,
+      "loss": 1.4501,
+      "step": 545
+    },
+    {
+      "epoch": 8.53,
+      "learning_rate": 0.00039273495922903046,
+      "loss": 1.3807,
+      "step": 546
+    },
+    {
+      "epoch": 8.55,
+      "learning_rate": 0.00039231044049678287,
+      "loss": 1.3869,
+      "step": 547
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 0.0003918853138406909,
+      "loss": 1.3862,
+      "step": 548
+    },
+    {
+      "epoch": 8.58,
+      "learning_rate": 0.0003914595810768171,
+      "loss": 1.3999,
+      "step": 549
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 0.0003910332440238128,
+      "loss": 1.3673,
+      "step": 550
+    },
+    {
+      "epoch": 8.59,
+      "eval_gen_len": 807.348,
+      "eval_loss": 1.5529675483703613,
+      "eval_rouge1": 53.2339,
+      "eval_rouge2": 22.152,
+      "eval_rougeL": 24.7587,
+      "eval_rougeLsum": 50.2502,
+      "eval_runtime": 882.3041,
+      "eval_samples_per_second": 0.283,
+      "eval_steps_per_second": 0.283,
+      "step": 550
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 0.00039060630450291094,
+      "loss": 1.3473,
+      "step": 551
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 0.00039017876433791824,
+      "loss": 1.3673,
+      "step": 552
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 0.00038975062535520687,
+      "loss": 1.3757,
+      "step": 553
+    },
+    {
+      "epoch": 8.66,
+      "learning_rate": 0.00038932188938370745,
+      "loss": 1.3811,
+      "step": 554
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 0.00038889255825490053,
+      "loss": 1.4296,
+      "step": 555
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 0.00038846263380280934,
+      "loss": 1.3868,
+      "step": 556
+    },
+    {
+      "epoch": 8.7,
+      "learning_rate": 0.00038803211786399126,
+      "loss": 1.3756,
+      "step": 557
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 0.0003876010122775309,
+      "loss": 1.3312,
+      "step": 558
+    },
+    {
+      "epoch": 8.73,
+      "learning_rate": 0.00038716931888503126,
+      "loss": 1.3168,
+      "step": 559
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 0.00038673703953060677,
+      "loss": 1.3538,
+      "step": 560
+    },
+    {
+      "epoch": 8.76,
+      "learning_rate": 0.00038630417606087456,
+      "loss": 1.4301,
+      "step": 561
+    },
+    {
+      "epoch": 8.78,
+      "learning_rate": 0.00038587073032494734,
+      "loss": 1.4171,
+      "step": 562
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.000385436704174425,
+      "loss": 1.3681,
+      "step": 563
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 0.0003850020994633868,
+      "loss": 1.4067,
+      "step": 564
+    },
+    {
+      "epoch": 8.83,
+      "learning_rate": 0.0003845669180483836,
+      "loss": 1.3763,
+      "step": 565
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 0.0003841311617884299,
+      "loss": 1.3688,
+      "step": 566
+    },
+    {
+      "epoch": 8.86,
+      "learning_rate": 0.0003836948325449956,
+      "loss": 1.3775,
+      "step": 567
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 0.00038325793218199844,
+      "loss": 1.4386,
+      "step": 568
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 0.00038282046256579597,
+      "loss": 1.401,
+      "step": 569
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 0.00038238242556517724,
+      "loss": 1.3611,
+      "step": 570
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 0.00038194382305135525,
+      "loss": 1.3689,
+      "step": 571
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 0.00038150465689795857,
+      "loss": 1.3925,
+      "step": 572
+    },
+    {
+      "epoch": 8.95,
+      "learning_rate": 0.00038106492898102386,
+      "loss": 1.4082,
+      "step": 573
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 0.0003806246411789872,
+      "loss": 1.4461,
+      "step": 574
+    },
+    {
+      "epoch": 8.98,
+      "learning_rate": 0.00038018379537267666,
+      "loss": 1.3634,
+      "step": 575
+    },
+    {
+      "epoch": 8.98,
+      "eval_gen_len": 767.952,
+      "eval_loss": 1.5457568168640137,
+      "eval_rouge1": 53.0293,
+      "eval_rouge2": 22.3194,
+      "eval_rougeL": 25.174,
+      "eval_rougeLsum": 50.078,
+      "eval_runtime": 842.6599,
+      "eval_samples_per_second": 0.297,
+      "eval_steps_per_second": 0.297,
+      "step": 575
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 0.0003797423934453038,
+      "loss": 1.3482,
+      "step": 576
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 0.0003793004372824559,
+      "loss": 1.406,
+      "step": 577
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 0.0003788579287720878,
+      "loss": 1.3285,
+      "step": 578
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 0.000378414869804514,
+      "loss": 1.3259,
+      "step": 579
+    },
+    {
+      "epoch": 9.06,
+      "learning_rate": 0.00037797126227240025,
+      "loss": 1.2688,
+      "step": 580
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 0.00037752710807075585,
+      "loss": 1.3329,
+      "step": 581
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 0.00037708240909692516,
+      "loss": 1.337,
+      "step": 582
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 0.00037663716725058,
+      "loss": 1.3956,
+      "step": 583
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 0.000376191384433711,
+      "loss": 1.4836,
+      "step": 584
+    },
+    {
+      "epoch": 9.14,
+      "learning_rate": 0.00037574506255061966,
+      "loss": 1.3266,
+      "step": 585
+    },
+    {
+      "epoch": 9.16,
+      "learning_rate": 0.00037529820350791065,
+      "loss": 1.2573,
+      "step": 586
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 0.0003748508092144828,
+      "loss": 1.3676,
+      "step": 587
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 0.00037440288158152184,
+      "loss": 1.3651,
+      "step": 588
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 0.00037395442252249157,
+      "loss": 1.3189,
+      "step": 589
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 0.00037350543395312603,
+      "loss": 1.338,
+      "step": 590
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 0.0003730559177914212,
+      "loss": 1.3612,
+      "step": 591
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 0.00037260587595762705,
+      "loss": 1.3696,
+      "step": 592
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 0.0003721553103742388,
+      "loss": 1.3936,
+      "step": 593
+    },
+    {
+      "epoch": 9.28,
+      "learning_rate": 0.0003717042229659891,
+      "loss": 1.3243,
+      "step": 594
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 0.0003712526156598399,
+      "loss": 1.3273,
+      "step": 595
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 0.00037080049038497407,
+      "loss": 1.345,
+      "step": 596
+    },
+    {
+      "epoch": 9.33,
+      "learning_rate": 0.0003703478490727868,
+      "loss": 1.339,
+      "step": 597
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 0.00036989469365687815,
+      "loss": 1.4087,
+      "step": 598
+    },
+    {
+      "epoch": 9.36,
+      "learning_rate": 0.00036944102607304404,
+      "loss": 1.3427,
+      "step": 599
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 0.0003689868482592684,
+      "loss": 1.3095,
+      "step": 600
+    },
+    {
+      "epoch": 9.37,
+      "eval_gen_len": 856.252,
+      "eval_loss": 1.5412006378173828,
+      "eval_rouge1": 53.7658,
+      "eval_rouge2": 22.5229,
+      "eval_rougeL": 25.0448,
+      "eval_rougeLsum": 50.708,
+      "eval_runtime": 931.7127,
+      "eval_samples_per_second": 0.268,
+      "eval_steps_per_second": 0.268,
+      "step": 600
+    },
+    {
+      "epoch": 9.39,
+      "learning_rate": 0.0003685321621557147,
+      "loss": 1.3381,
+      "step": 601
+    },
+    {
+      "epoch": 9.41,
+      "learning_rate": 0.00036807696970471784,
+      "loss": 1.3336,
+      "step": 602
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 0.00036762127285077544,
+      "loss": 1.403,
+      "step": 603
+    },
+    {
+      "epoch": 9.44,
+      "learning_rate": 0.0003671650735405404,
+      "loss": 1.3572,
+      "step": 604
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 0.00036670837372281136,
+      "loss": 1.3604,
+      "step": 605
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 0.0003662511753485256,
+      "loss": 1.3897,
+      "step": 606
+    },
+    {
+      "epoch": 9.48,
+      "learning_rate": 0.0003657934803707497,
+      "loss": 1.3179,
+      "step": 607
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 0.000365335290744672,
+      "loss": 1.3702,
+      "step": 608
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 0.00036487660842759355,
+      "loss": 1.3762,
+      "step": 609
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 0.00036441743537892043,
+      "loss": 1.3296,
+      "step": 610
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 0.0003639577735601548,
+      "loss": 1.3171,
+      "step": 611
+    },
+    {
+      "epoch": 9.56,
+      "learning_rate": 0.00036349762493488667,
+      "loss": 1.3241,
+      "step": 612
+    },
+    {
+      "epoch": 9.58,
+      "learning_rate": 0.000363036991468786,
+      "loss": 1.3366,
+      "step": 613
+    },
+    {
+      "epoch": 9.59,
+      "learning_rate": 0.00036257587512959343,
+      "loss": 1.3323,
+      "step": 614
+    },
+    {
+      "epoch": 9.61,
+      "learning_rate": 0.0003621142778871127,
+      "loss": 1.3814,
+      "step": 615
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 0.00036165220171320166,
+      "loss": 1.3978,
+      "step": 616
+    },
+    {
+      "epoch": 9.64,
+      "learning_rate": 0.00036118964858176413,
+      "loss": 1.339,
+      "step": 617
+    },
+    {
+      "epoch": 9.66,
+      "learning_rate": 0.00036072662046874145,
+      "loss": 1.3834,
+      "step": 618
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 0.00036026311935210393,
+      "loss": 1.3568,
+      "step": 619
+    },
+    {
+      "epoch": 9.69,
+      "learning_rate": 0.0003597991472118426,
+      "loss": 1.4211,
+      "step": 620
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 0.00035933470602996035,
+      "loss": 1.3621,
+      "step": 621
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 0.000358869797790464,
+      "loss": 1.3238,
+      "step": 622
+    },
+    {
+      "epoch": 9.73,
+      "learning_rate": 0.0003584044244793555,
+      "loss": 1.3429,
+      "step": 623
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 0.0003579385880846232,
+      "loss": 1.3871,
+      "step": 624
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 0.0003574722905962342,
+      "loss": 1.3492,
+      "step": 625
+    },
+    {
+      "epoch": 9.76,
+      "eval_gen_len": 826.064,
+      "eval_loss": 1.5389072895050049,
+      "eval_rouge1": 51.8662,
+      "eval_rouge2": 21.6229,
+      "eval_rougeL": 24.6819,
+      "eval_rougeLsum": 48.8648,
+      "eval_runtime": 900.6261,
+      "eval_samples_per_second": 0.278,
+      "eval_steps_per_second": 0.278,
+      "step": 625
+    },
+    {
+      "epoch": 9.78,
+      "learning_rate": 0.0003570055340061248,
+      "loss": 1.3172,
+      "step": 626
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 0.0003565383203081928,
+      "loss": 1.3828,
+      "step": 627
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 0.0003560706514982884,
+      "loss": 1.4252,
+      "step": 628
+    },
+    {
+      "epoch": 9.83,
+      "learning_rate": 0.0003556025295742065,
+      "loss": 1.3182,
+      "step": 629
+    },
+    {
+      "epoch": 9.84,
+      "learning_rate": 0.0003551339565356769,
+      "loss": 1.3041,
+      "step": 630
+    },
+    {
+      "epoch": 9.86,
+      "learning_rate": 0.00035466493438435703,
+      "loss": 1.359,
+      "step": 631
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 0.0003541954651238226,
+      "loss": 1.365,
+      "step": 632
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 0.00035372555075955937,
+      "loss": 1.2863,
+      "step": 633
+    },
+    {
+      "epoch": 9.91,
+      "learning_rate": 0.0003532551932989544,
+      "loss": 1.3145,
+      "step": 634
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 0.0003527843947512878,
+      "loss": 1.4008,
+      "step": 635
+    },
+    {
+      "epoch": 9.94,
+      "learning_rate": 0.00035231315712772347,
+      "loss": 1.2704,
+      "step": 636
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 0.0003518414824413017,
+      "loss": 1.3213,
+      "step": 637
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 0.00035136937270692893,
+      "loss": 1.3692,
+      "step": 638
+    },
+    {
+      "epoch": 9.98,
+      "learning_rate": 0.0003508968299413708,
+      "loss": 1.3404,
+      "step": 639
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 0.0003504238561632424,
+      "loss": 1.3378,
+      "step": 640
+    },
+    {
+      "epoch": 10.02,
+      "learning_rate": 0.0003499504533930001,
+      "loss": 1.3609,
+      "step": 641
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 0.0003494766236529329,
+      "loss": 1.3558,
+      "step": 642
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 0.00034900236896715355,
+      "loss": 1.3094,
+      "step": 643
+    },
+    {
+      "epoch": 10.06,
+      "learning_rate": 0.0003485276913615905,
+      "loss": 1.3134,
+      "step": 644
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 0.00034805259286397824,
+      "loss": 1.3176,
+      "step": 645
+    },
+    {
+      "epoch": 10.09,
+      "learning_rate": 0.00034757707550384976,
+      "loss": 1.3581,
+      "step": 646
+    },
+    {
+      "epoch": 10.11,
+      "learning_rate": 0.0003471011413125271,
+      "loss": 1.3142,
+      "step": 647
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 0.00034662479232311306,
+      "loss": 1.2926,
+      "step": 648
+    },
+    {
+      "epoch": 10.14,
+      "learning_rate": 0.0003461480305704822,
+      "loss": 1.2874,
+      "step": 649
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 0.0003456708580912725,
+      "loss": 1.3007,
+      "step": 650
+    },
+    {
+      "epoch": 10.16,
+      "eval_gen_len": 843.544,
+      "eval_loss": 1.5403586626052856,
+      "eval_rouge1": 53.6692,
+      "eval_rouge2": 22.154,
+      "eval_rougeL": 24.6218,
+      "eval_rougeLsum": 50.6864,
+      "eval_runtime": 918.5032,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
+      "step": 650
+    },
+    {
+      "epoch": 10.17,
+      "learning_rate": 0.0003451932769238763,
+      "loss": 1.311,
+      "step": 651
+    },
+    {
+      "epoch": 10.19,
+      "learning_rate": 0.0003447152891084319,
+      "loss": 1.3283,
+      "step": 652
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 0.0003442368966868149,
+      "loss": 1.3141,
+      "step": 653
+    },
+    {
+      "epoch": 10.22,
+      "learning_rate": 0.0003437581017026289,
+      "loss": 1.3405,
+      "step": 654
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 0.00034327890620119755,
+      "loss": 1.3586,
+      "step": 655
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 0.00034279931222955517,
+      "loss": 1.2894,
+      "step": 656
+    },
+    {
+      "epoch": 10.27,
+      "learning_rate": 0.0003423193218364385,
+      "loss": 1.2816,
+      "step": 657
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 0.0003418389370722775,
+      "loss": 1.2884,
+      "step": 658
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 0.00034135815998918693,
+      "loss": 1.331,
+      "step": 659
+    },
+    {
+      "epoch": 10.31,
+      "learning_rate": 0.00034087699264095743,
+      "loss": 1.3353,
+      "step": 660
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 0.00034039543708304696,
+      "loss": 1.3334,
+      "step": 661
+    },
+    {
+      "epoch": 10.34,
+      "learning_rate": 0.00033991349537257137,
+      "loss": 1.318,
+      "step": 662
+    },
+    {
+      "epoch": 10.36,
+      "learning_rate": 0.00033943116956829664,
+      "loss": 1.2815,
+      "step": 663
+    },
+    {
+      "epoch": 10.37,
+      "learning_rate": 0.00033894846173062915,
+      "loss": 1.3458,
+      "step": 664
+    },
+    {
+      "epoch": 10.39,
+      "learning_rate": 0.0003384653739216074,
+      "loss": 1.3625,
+      "step": 665
+    },
+    {
+      "epoch": 10.41,
+      "learning_rate": 0.00033798190820489313,
+      "loss": 1.3976,
+      "step": 666
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 0.0003374980666457621,
+      "loss": 1.3768,
+      "step": 667
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 0.00033701385131109616,
+      "loss": 1.2797,
+      "step": 668
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 0.0003365292642693733,
+      "loss": 1.3,
+      "step": 669
+    },
+    {
+      "epoch": 10.47,
+      "learning_rate": 0.00033604430759065967,
+      "loss": 1.331,
+      "step": 670
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 0.0003355589833466004,
+      "loss": 1.3494,
+      "step": 671
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 0.0003350732936104108,
+      "loss": 1.2914,
+      "step": 672
+    },
+    {
+      "epoch": 10.52,
+      "learning_rate": 0.0003345872404568674,
+      "loss": 1.3664,
+      "step": 673
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 0.0003341008259622993,
+      "loss": 1.3086,
+      "step": 674
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 0.00033361405220457913,
+      "loss": 1.2729,
+      "step": 675
+    },
+    {
+      "epoch": 10.55,
+      "eval_gen_len": 808.764,
+      "eval_loss": 1.5428452491760254,
+      "eval_rouge1": 54.6479,
+      "eval_rouge2": 23.3029,
+      "eval_rougeL": 25.5647,
+      "eval_rougeLsum": 51.6394,
+      "eval_runtime": 883.7152,
+      "eval_samples_per_second": 0.283,
+      "eval_steps_per_second": 0.283,
+      "step": 675
+    },
+    {
+      "epoch": 10.56,
+      "learning_rate": 0.00033312692126311424,
+      "loss": 1.2774,
+      "step": 676
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 0.00033263943521883766,
+      "loss": 1.3757,
+      "step": 677
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 0.0003321515961541997,
+      "loss": 1.3389,
+      "step": 678
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 0.00033166340615315836,
+      "loss": 1.3292,
+      "step": 679
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 0.00033117486730117093,
+      "loss": 1.3003,
+      "step": 680
+    },
+    {
+      "epoch": 10.64,
+      "learning_rate": 0.00033068598168518485,
+      "loss": 1.2605,
+      "step": 681
+    },
+    {
+      "epoch": 10.66,
+      "learning_rate": 0.000330196751393629,
+      "loss": 1.2984,
+      "step": 682
+    },
+    {
+      "epoch": 10.67,
+      "learning_rate": 0.0003297071785164045,
+      "loss": 1.3517,
+      "step": 683
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 0.0003292172651448761,
+      "loss": 1.3402,
+      "step": 684
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 0.00032872701337186293,
+      "loss": 1.2718,
+      "step": 685
+    },
+    {
+      "epoch": 10.72,
+      "learning_rate": 0.0003282364252916298,
+      "loss": 1.3146,
+      "step": 686
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 0.0003277455029998781,
+      "loss": 1.2862,
+      "step": 687
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 0.00032725424859373687,
+      "loss": 1.3264,
+      "step": 688
+    },
+    {
+      "epoch": 10.76,
+      "learning_rate": 0.0003267626641717541,
+      "loss": 1.3557,
+      "step": 689
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 0.0003262707518338872,
+      "loss": 1.3663,
+      "step": 690
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 0.0003257785136814948,
+      "loss": 1.3282,
+      "step": 691
+    },
+    {
+      "epoch": 10.81,
+      "learning_rate": 0.0003252859518173269,
+      "loss": 1.3846,
+      "step": 692
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 0.00032479306834551667,
+      "loss": 1.2832,
+      "step": 693
+    },
+    {
+      "epoch": 10.84,
+      "learning_rate": 0.00032429986537157096,
+      "loss": 1.2998,
+      "step": 694
+    },
+    {
+      "epoch": 10.86,
+      "learning_rate": 0.0003238063450023617,
+      "loss": 1.3911,
+      "step": 695
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 0.0003233125093461162,
+      "loss": 1.3407,
+      "step": 696
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 0.0003228183605124092,
+      "loss": 1.358,
+      "step": 697
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 0.0003223239006121528,
+      "loss": 1.3951,
+      "step": 698
+    },
+    {
+      "epoch": 10.92,
+      "learning_rate": 0.0003218291317575882,
+      "loss": 1.3062,
+      "step": 699
+    },
+    {
+      "epoch": 10.94,
+      "learning_rate": 0.00032133405606227635,
+      "loss": 1.3758,
+      "step": 700
+    },
+    {
+      "epoch": 10.94,
+      "eval_gen_len": 800.152,
+      "eval_loss": 1.5403379201889038,
+      "eval_rouge1": 54.9418,
+      "eval_rouge2": 23.3323,
+      "eval_rougeL": 25.6087,
+      "eval_rougeLsum": 51.9256,
+      "eval_runtime": 874.0945,
+      "eval_samples_per_second": 0.286,
+      "eval_steps_per_second": 0.286,
+      "step": 700
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 0.00032083867564108886,
+      "loss": 1.3445,
+      "step": 701
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 0.0003203429926101991,
+      "loss": 1.3103,
+      "step": 702
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 0.00031984700908707333,
+      "loss": 1.2862,
+      "step": 703
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 0.00031935072719046115,
+      "loss": 1.3072,
+      "step": 704
+    },
+    {
+      "epoch": 11.02,
+      "learning_rate": 0.000318854149040387,
+      "loss": 1.3737,
+      "step": 705
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 0.0003183572767581406,
+      "loss": 1.2964,
+      "step": 706
+    },
+    {
+      "epoch": 11.05,
+      "learning_rate": 0.00031786011246626855,
+      "loss": 1.3451,
+      "step": 707
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 0.00031736265828856446,
+      "loss": 1.2713,
+      "step": 708
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 0.00031686491635006054,
+      "loss": 1.3316,
+      "step": 709
+    },
+    {
+      "epoch": 11.09,
+      "learning_rate": 0.0003163668887770181,
+      "loss": 1.2551,
+      "step": 710
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 0.0003158685776969187,
+      "loss": 1.2511,
+      "step": 711
+    },
+    {
+      "epoch": 11.12,
+      "learning_rate": 0.00031536998523845497,
+      "loss": 1.2933,
+      "step": 712
+    },
+    {
+      "epoch": 11.14,
+      "learning_rate": 0.0003148711135315215,
+      "loss": 1.2744,
+      "step": 713
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 0.00031437196470720597,
+      "loss": 1.3624,
+      "step": 714
+    },
+    {
+      "epoch": 11.17,
+      "learning_rate": 0.00031387254089777967,
+      "loss": 1.232,
+      "step": 715
+    },
+    {
+      "epoch": 11.19,
+      "learning_rate": 0.0003133728442366885,
+      "loss": 1.2982,
+      "step": 716
+    },
+    {
+      "epoch": 11.2,
+      "learning_rate": 0.00031287287685854417,
+      "loss": 1.3225,
+      "step": 717
+    },
+    {
+      "epoch": 11.22,
+      "learning_rate": 0.00031237264089911465,
+      "loss": 1.3021,
+      "step": 718
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 0.0003118721384953153,
+      "loss": 1.3114,
+      "step": 719
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 0.0003113713717851998,
+      "loss": 1.2986,
+      "step": 720
+    },
+    {
+      "epoch": 11.27,
+      "learning_rate": 0.0003108703429079506,
+      "loss": 1.2522,
+      "step": 721
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 0.0003103690540038705,
+      "loss": 1.3075,
+      "step": 722
+    },
+    {
+      "epoch": 11.3,
+      "learning_rate": 0.0003098675072143727,
+      "loss": 1.3449,
+      "step": 723
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 0.00030936570468197215,
+      "loss": 1.2293,
+      "step": 724
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 0.00030886364855027646,
+      "loss": 1.3357,
+      "step": 725
+    },
+    {
+      "epoch": 11.33,
+      "eval_gen_len": 814.496,
+      "eval_loss": 1.5454789400100708,
+      "eval_rouge1": 55.2511,
+      "eval_rouge2": 23.5606,
+      "eval_rougeL": 25.8237,
+      "eval_rougeLsum": 52.3183,
+      "eval_runtime": 889.3375,
+      "eval_samples_per_second": 0.281,
+      "eval_steps_per_second": 0.281,
+      "step": 725
+    },
+    {
+      "epoch": 11.34,
+      "learning_rate": 0.0003083613409639764,
+      "loss": 1.3662,
+      "step": 726
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 0.0003078587840688368,
+      "loss": 1.3794,
+      "step": 727
+    },
+    {
+      "epoch": 11.37,
+      "learning_rate": 0.0003073559800116879,
+      "loss": 1.2878,
+      "step": 728
+    },
+    {
+      "epoch": 11.39,
+      "learning_rate": 0.00030685293094041534,
+      "loss": 1.2361,
+      "step": 729
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 0.00030634963900395156,
+      "loss": 1.3076,
+      "step": 730
+    },
+    {
+      "epoch": 11.42,
+      "learning_rate": 0.00030584610635226654,
+      "loss": 1.3512,
+      "step": 731
+    },
+    {
+      "epoch": 11.44,
+      "learning_rate": 0.0003053423351363586,
+      "loss": 1.2882,
+      "step": 732
+    },
+    {
+      "epoch": 11.45,
+      "learning_rate": 0.00030483832750824495,
+      "loss": 1.3675,
+      "step": 733
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 0.0003043340856209529,
+      "loss": 1.3085,
+      "step": 734
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 0.0003038296116285103,
+      "loss": 1.2838,
+      "step": 735
+    },
+    {
+      "epoch": 11.5,
+      "learning_rate": 0.0003033249076859367,
+      "loss": 1.3298,
+      "step": 736
+    },
+    {
+      "epoch": 11.52,
+      "learning_rate": 0.0003028199759492339,
+      "loss": 1.3187,
+      "step": 737
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 0.0003023148185753766,
+      "loss": 1.3395,
+      "step": 738
+    },
+    {
+      "epoch": 11.55,
+      "learning_rate": 0.0003018094377223037,
+      "loss": 1.3484,
+      "step": 739
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 0.00030130383554890854,
+      "loss": 1.2969,
+      "step": 740
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 0.00030079801421503005,
+      "loss": 1.2698,
+      "step": 741
+    },
+    {
+      "epoch": 11.59,
+      "learning_rate": 0.0003002919758814431,
+      "loss": 1.3265,
+      "step": 742
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 0.00029978572270984987,
+      "loss": 1.3148,
+      "step": 743
+    },
+    {
+      "epoch": 11.62,
+      "learning_rate": 0.00029927925686287006,
+      "loss": 1.2497,
+      "step": 744
+    },
+    {
+      "epoch": 11.64,
+      "learning_rate": 0.0002987725805040321,
+      "loss": 1.3015,
+      "step": 745
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 0.0002982656957977634,
+      "loss": 1.3065,
+      "step": 746
+    },
+    {
+      "epoch": 11.67,
+      "learning_rate": 0.0002977586049093816,
+      "loss": 1.2443,
+      "step": 747
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 0.0002972513100050851,
+      "loss": 1.303,
+      "step": 748
+    },
+    {
+      "epoch": 11.7,
+      "learning_rate": 0.00029674381325194373,
+      "loss": 1.3074,
+      "step": 749
+    },
+    {
+      "epoch": 11.72,
+      "learning_rate": 0.00029623611681788963,
+      "loss": 1.2817,
+      "step": 750
+    },
+    {
+      "epoch": 11.72,
+      "eval_gen_len": 811.144,
+      "eval_loss": 1.5412460565567017,
+      "eval_rouge1": 55.2847,
+      "eval_rouge2": 23.6632,
+      "eval_rougeL": 25.9341,
+      "eval_rougeLsum": 52.3146,
+      "eval_runtime": 886.0701,
+      "eval_samples_per_second": 0.282,
+      "eval_steps_per_second": 0.282,
+      "step": 750
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 0.00029572822287170794,
+      "loss": 1.2607,
+      "step": 751
+    },
+    {
+      "epoch": 11.75,
+      "learning_rate": 0.0002952201335830275,
+      "loss": 1.2668,
+      "step": 752
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 0.0002947118511223117,
+      "loss": 1.3079,
+      "step": 753
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 0.00029420337766084903,
+      "loss": 1.2802,
+      "step": 754
+    },
+    {
+      "epoch": 11.8,
+      "learning_rate": 0.0002936947153707439,
+      "loss": 1.3151,
+      "step": 755
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 0.00029318586642490763,
+      "loss": 1.2845,
+      "step": 756
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 0.0002926768329970484,
+      "loss": 1.2877,
+      "step": 757
+    },
+    {
+      "epoch": 11.84,
+      "learning_rate": 0.00029216761726166287,
+      "loss": 1.2546,
+      "step": 758
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 0.0002916582213940263,
+      "loss": 1.253,
+      "step": 759
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 0.0002911486475701835,
+      "loss": 1.3328,
+      "step": 760
+    },
+    {
+      "epoch": 11.89,
+      "learning_rate": 0.00029063889796693954,
+      "loss": 1.309,
+      "step": 761
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 0.0002901289747618502,
+      "loss": 1.3281,
+      "step": 762
+    },
+    {
+      "epoch": 11.92,
+      "learning_rate": 0.000289618880133213,
+      "loss": 1.287,
+      "step": 763
+    },
+    {
+      "epoch": 11.94,
+      "learning_rate": 0.00028910861626005774,
+      "loss": 1.3711,
+      "step": 764
+    },
+    {
+      "epoch": 11.95,
+      "learning_rate": 0.00028859818532213724,
+      "loss": 1.2575,
+      "step": 765
+    },
+    {
+      "epoch": 11.97,
+      "learning_rate": 0.00028808758949991776,
+      "loss": 1.2802,
+      "step": 766
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 0.00028757683097457035,
+      "loss": 1.3022,
+      "step": 767
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 0.0002870659119279605,
+      "loss": 1.2998,
+      "step": 768
+    },
+    {
+      "epoch": 12.02,
+      "learning_rate": 0.00028655483454264,
+      "loss": 1.3663,
+      "step": 769
+    },
+    {
+      "epoch": 12.03,
+      "learning_rate": 0.00028604360100183666,
+      "loss": 1.3068,
+      "step": 770
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 0.00028553221348944565,
+      "loss": 1.2841,
+      "step": 771
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 0.0002850206741900195,
+      "loss": 1.315,
+      "step": 772
+    },
+    {
+      "epoch": 12.08,
+      "learning_rate": 0.0002845089852887595,
+      "loss": 1.267,
+      "step": 773
+    },
+    {
+      "epoch": 12.09,
+      "learning_rate": 0.0002839971489715057,
+      "loss": 1.2197,
+      "step": 774
+    },
+    {
+      "epoch": 12.11,
+      "learning_rate": 0.0002834851674247282,
+      "loss": 1.2771,
+      "step": 775
+    },
+    {
+      "epoch": 12.11,
+      "eval_gen_len": 852.704,
+      "eval_loss": 1.5449961423873901,
+      "eval_rouge1": 55.1956,
+      "eval_rouge2": 23.5545,
+      "eval_rougeL": 25.677,
+      "eval_rougeLsum": 52.1841,
+      "eval_runtime": 928.7433,
+      "eval_samples_per_second": 0.269,
+      "eval_steps_per_second": 0.269,
+      "step": 775
+    },
+    {
+      "epoch": 12.12,
+      "learning_rate": 0.00028297304283551725,
+      "loss": 1.2442,
+      "step": 776
+    },
+    {
+      "epoch": 12.14,
+      "learning_rate": 0.00028246077739157435,
+      "loss": 1.316,
+      "step": 777
+    },
+    {
+      "epoch": 12.16,
+      "learning_rate": 0.0002819483732812025,
+      "loss": 1.2588,
+      "step": 778
+    },
+    {
+      "epoch": 12.17,
+      "learning_rate": 0.0002814358326932972,
+      "loss": 1.2439,
+      "step": 779
+    },
+    {
+      "epoch": 12.19,
+      "learning_rate": 0.000280923157817337,
+      "loss": 1.223,
+      "step": 780
+    },
+    {
+      "epoch": 12.2,
+      "learning_rate": 0.0002804103508433738,
+      "loss": 1.3348,
+      "step": 781
+    },
+    {
+      "epoch": 12.22,
+      "learning_rate": 0.0002798974139620242,
+      "loss": 1.2156,
+      "step": 782
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 0.00027938434936445943,
+      "loss": 1.3101,
+      "step": 783
+    },
+    {
+      "epoch": 12.25,
+      "learning_rate": 0.0002788711592423966,
+      "loss": 1.2715,
+      "step": 784
+    },
+    {
+      "epoch": 12.27,
+      "learning_rate": 0.00027835784578808867,
+      "loss": 1.373,
+      "step": 785
+    },
+    {
+      "epoch": 12.28,
+      "learning_rate": 0.00027784441119431576,
+      "loss": 1.2366,
+      "step": 786
+    },
+    {
+      "epoch": 12.3,
+      "learning_rate": 0.00027733085765437526,
+      "loss": 1.2454,
+      "step": 787
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 0.00027681718736207296,
+      "loss": 1.3599,
+      "step": 788
+    },
+    {
+      "epoch": 12.33,
+      "learning_rate": 0.00027630340251171297,
+      "loss": 1.2374,
+      "step": 789
+    },
+    {
+      "epoch": 12.34,
+      "learning_rate": 0.00027578950529808926,
+      "loss": 1.3201,
+      "step": 790
+    },
+    {
+      "epoch": 12.36,
+      "learning_rate": 0.00027527549791647537,
+      "loss": 1.2734,
+      "step": 791
+    },
+    {
+      "epoch": 12.37,
+      "learning_rate": 0.00027476138256261575,
+      "loss": 1.2887,
+      "step": 792
+    },
+    {
+      "epoch": 12.39,
+      "learning_rate": 0.00027424716143271575,
+      "loss": 1.2598,
+      "step": 793
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 0.0002737328367234331,
+      "loss": 1.3082,
+      "step": 794
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 0.0002732184106318675,
+      "loss": 1.2133,
+      "step": 795
+    },
+    {
+      "epoch": 12.44,
+      "learning_rate": 0.00027270388535555206,
+      "loss": 1.2794,
+      "step": 796
+    },
+    {
+      "epoch": 12.45,
+      "learning_rate": 0.00027218926309244333,
+      "loss": 1.2522,
+      "step": 797
+    },
+    {
+      "epoch": 12.47,
+      "learning_rate": 0.0002716745460409125,
+      "loss": 1.3457,
+      "step": 798
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 0.0002711597363997352,
+      "loss": 1.2477,
+      "step": 799
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 0.00027064483636808314,
+      "loss": 1.2892,
+      "step": 800
+    },
+    {
+      "epoch": 12.5,
+      "eval_gen_len": 805.844,
+      "eval_loss": 1.5368956327438354,
+      "eval_rouge1": 54.9563,
+      "eval_rouge2": 23.5105,
+      "eval_rougeL": 25.8876,
+      "eval_rougeLsum": 51.9568,
+      "eval_runtime": 884.5538,
+      "eval_samples_per_second": 0.283,
+      "eval_steps_per_second": 0.283,
+      "step": 800
+    },
+    {
+      "epoch": 12.52,
+      "learning_rate": 0.00027012984814551365,
+      "loss": 1.3415,
+      "step": 801
+    },
+    {
+      "epoch": 12.53,
+      "learning_rate": 0.00026961477393196127,
+      "loss": 1.2526,
+      "step": 802
+    },
+    {
+      "epoch": 12.55,
+      "learning_rate": 0.00026909961592772746,
+      "loss": 1.2488,
+      "step": 803
+    },
+    {
+      "epoch": 12.56,
+      "learning_rate": 0.0002685843763334719,
+      "loss": 1.2557,
+      "step": 804
+    },
+    {
+      "epoch": 12.58,
+      "learning_rate": 0.00026806905735020267,
+      "loss": 1.2802,
+      "step": 805
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 0.00026755366117926706,
+      "loss": 1.2627,
+      "step": 806
+    },
+    {
+      "epoch": 12.61,
+      "learning_rate": 0.000267038190022342,
+      "loss": 1.1703,
+      "step": 807
+    },
+    {
+      "epoch": 12.62,
+      "learning_rate": 0.00026652264608142484,
+      "loss": 1.2512,
+      "step": 808
+    },
+    {
+      "epoch": 12.64,
+      "learning_rate": 0.0002660070315588238,
+      "loss": 1.3306,
+      "step": 809
+    },
+    {
+      "epoch": 12.66,
+      "learning_rate": 0.0002654913486571487,
+      "loss": 1.2447,
+      "step": 810
+    },
+    {
+      "epoch": 12.67,
+      "learning_rate": 0.00026497559957930117,
+      "loss": 1.2808,
+      "step": 811
+    },
+    {
+      "epoch": 12.69,
+      "learning_rate": 0.00026445978652846604,
+      "loss": 1.248,
+      "step": 812
+    },
+    {
+      "epoch": 12.7,
+      "learning_rate": 0.0002639439117081009,
+      "loss": 1.2949,
+      "step": 813
+    },
+    {
+      "epoch": 12.72,
+      "learning_rate": 0.0002634279773219275,
+      "loss": 1.2704,
+      "step": 814
+    },
+    {
+      "epoch": 12.73,
+      "learning_rate": 0.00026291198557392207,
+      "loss": 1.3479,
+      "step": 815
+    },
+    {
+      "epoch": 12.75,
+      "learning_rate": 0.0002623959386683056,
+      "loss": 1.2575,
+      "step": 816
+    },
+    {
+      "epoch": 12.76,
+      "learning_rate": 0.0002618798388095351,
+      "loss": 1.2874,
+      "step": 817
+    },
+    {
+      "epoch": 12.78,
+      "learning_rate": 0.0002613636882022934,
+      "loss": 1.2523,
+      "step": 818
+    },
+    {
+      "epoch": 12.8,
+      "learning_rate": 0.00026084748905148044,
+      "loss": 1.3145,
+      "step": 819
+    },
+    {
+      "epoch": 12.81,
+      "learning_rate": 0.00026033124356220327,
+      "loss": 1.2299,
+      "step": 820
+    },
+    {
+      "epoch": 12.83,
+      "learning_rate": 0.00025981495393976716,
+      "loss": 1.3173,
+      "step": 821
+    },
+    {
+      "epoch": 12.84,
+      "learning_rate": 0.0002592986223896656,
+      "loss": 1.3262,
+      "step": 822
+    },
+    {
+      "epoch": 12.86,
+      "learning_rate": 0.0002587822511175715,
+      "loss": 1.2522,
+      "step": 823
+    },
+    {
+      "epoch": 12.87,
+      "learning_rate": 0.00025826584232932704,
+      "loss": 1.2762,
+      "step": 824
+    },
+    {
+      "epoch": 12.89,
+      "learning_rate": 0.0002577493982309352,
+      "loss": 1.2757,
+      "step": 825
+    },
+    {
+      "epoch": 12.89,
+      "eval_gen_len": 813.476,
+      "eval_loss": 1.5466704368591309,
+      "eval_rouge1": 56.4728,
+      "eval_rouge2": 24.6875,
+      "eval_rougeL": 26.4415,
+      "eval_rougeLsum": 53.4939,
+      "eval_runtime": 890.3223,
+      "eval_samples_per_second": 0.281,
+      "eval_steps_per_second": 0.281,
+      "step": 825
+    },
+    {
+      "epoch": 12.91,
+      "learning_rate": 0.0002572329210285493,
+      "loss": 1.2617,
+      "step": 826
+    },
+    {
+      "epoch": 12.92,
+      "learning_rate": 0.00025671641292846447,
+      "loss": 1.3063,
+      "step": 827
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 0.00025619987613710756,
+      "loss": 1.275,
+      "step": 828
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 0.00025568331286102803,
+      "loss": 1.3496,
+      "step": 829
+    },
+    {
+      "epoch": 12.97,
+      "learning_rate": 0.00025516672530688864,
+      "loss": 1.2791,
+      "step": 830
+    },
+    {
+      "epoch": 12.98,
+      "learning_rate": 0.00025465011568145555,
+      "loss": 1.268,
+      "step": 831
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 0.0002541334861915897,
+      "loss": 1.3177,
+      "step": 832
+    },
+    {
+      "epoch": 13.02,
+      "learning_rate": 0.00025361683904423624,
+      "loss": 1.3568,
+      "step": 833
+    },
+    {
+      "epoch": 13.03,
+      "learning_rate": 0.00025310017644641637,
+      "loss": 1.2423,
+      "step": 834
+    },
+    {
+      "epoch": 13.05,
+      "learning_rate": 0.00025258350060521686,
+      "loss": 1.2066,
+      "step": 835
+    },
+    {
+      "epoch": 13.06,
+      "learning_rate": 0.00025206681372778124,
+      "loss": 1.2076,
+      "step": 836
+    },
+    {
+      "epoch": 13.08,
+      "learning_rate": 0.00025155011802130016,
+      "loss": 1.258,
+      "step": 837
+    },
+    {
+      "epoch": 13.09,
+      "learning_rate": 0.0002510334156930021,
+      "loss": 1.2864,
+      "step": 838
+    },
+    {
+      "epoch": 13.11,
+      "learning_rate": 0.0002505167089501435,
+      "loss": 1.2762,
+      "step": 839
+    },
+    {
+      "epoch": 13.12,
+      "learning_rate": 0.00025,
+      "loss": 1.2388,
+      "step": 840
+    },
+    {
+      "epoch": 13.14,
+      "learning_rate": 0.0002494832910498565,
+      "loss": 1.2254,
+      "step": 841
+    },
+    {
+      "epoch": 13.16,
+      "learning_rate": 0.000248966584306998,
+      "loss": 1.2315,
+      "step": 842
+    },
+    {
+      "epoch": 13.17,
+      "learning_rate": 0.0002484498819786998,
+      "loss": 1.2186,
+      "step": 843
+    },
+    {
+      "epoch": 13.19,
+      "learning_rate": 0.00024793318627221877,
+      "loss": 1.27,
+      "step": 844
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 0.0002474164993947832,
+      "loss": 1.2413,
+      "step": 845
+    },
+    {
+      "epoch": 13.22,
+      "learning_rate": 0.00024689982355358375,
+      "loss": 1.2759,
+      "step": 846
+    },
+    {
+      "epoch": 13.23,
+      "learning_rate": 0.0002463831609557638,
+      "loss": 1.2606,
+      "step": 847
+    },
+    {
+      "epoch": 13.25,
+      "learning_rate": 0.0002458665138084104,
+      "loss": 1.2803,
+      "step": 848
+    },
+    {
+      "epoch": 13.27,
+      "learning_rate": 0.00024534988431854446,
+      "loss": 1.343,
+      "step": 849
+    },
+    {
+      "epoch": 13.28,
+      "learning_rate": 0.0002448332746931115,
+      "loss": 1.2382,
+      "step": 850
+    },
+    {
+      "epoch": 13.28,
+      "eval_gen_len": 787.34,
+      "eval_loss": 1.5448322296142578,
+      "eval_rouge1": 57.2303,
+      "eval_rouge2": 24.9705,
+      "eval_rougeL": 26.8081,
+      "eval_rougeLsum": 54.2747,
+      "eval_runtime": 870.1853,
+      "eval_samples_per_second": 0.287,
+      "eval_steps_per_second": 0.287,
+      "step": 850
+    }
+  ],
+  "max_steps": 1600,
+  "num_train_epochs": 25,
+  "total_flos": 1.193214367236096e+18,
+  "trial_name": null,
+  "trial_params": null
+}