diff --git "a/checkpoint-1488/trainer_state.json" "b/checkpoint-1488/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1488/trainer_state.json"
@@ -0,0 +1,9005 @@
+{
+  "best_metric": 1.3443834781646729,
+  "best_model_checkpoint": "./qlora-out/checkpoint-1488",
+  "epoch": 0.7503782148260212,
+  "eval_steps": 248,
+  "global_step": 1488,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.3547,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "eval_loss": 1.5788122415542603,
+      "eval_runtime": 99.5564,
+      "eval_samples_per_second": 1.165,
+      "eval_steps_per_second": 1.165,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.5678,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6e-06,
+      "loss": 1.8406,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 2.2337,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1e-05,
+      "loss": 1.6725,
+      "step": 5
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.2e-05,
+      "loss": 1.4016,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.4e-05,
+      "loss": 1.3171,
+      "step": 7
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 1.4367,
+      "step": 8
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.8e-05,
+      "loss": 1.4082,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2e-05,
+      "loss": 1.5169,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999996846759028e-05,
+      "loss": 1.6735,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.99999873870381e-05,
+      "loss": 1.3341,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999971620843182e-05,
+      "loss": 1.7104,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999949548184215e-05,
+      "loss": 1.2838,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999921169075117e-05,
+      "loss": 1.4659,
+      "step": 15
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999886483533792e-05,
+      "loss": 1.504,
+      "step": 16
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.999984549158211e-05,
+      "loss": 1.7891,
+      "step": 17
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999798193245924e-05,
+      "loss": 1.7854,
+      "step": 18
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999744588555065e-05,
+      "loss": 1.7957,
+      "step": 19
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999684677543332e-05,
+      "loss": 1.5033,
+      "step": 20
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999618460248515e-05,
+      "loss": 1.5491,
+      "step": 21
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999545936712364e-05,
+      "loss": 1.5846,
+      "step": 22
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999467106980627e-05,
+      "loss": 1.6081,
+      "step": 23
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999381971103015e-05,
+      "loss": 1.0283,
+      "step": 24
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999290529133215e-05,
+      "loss": 2.0389,
+      "step": 25
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9999192781128893e-05,
+      "loss": 1.3088,
+      "step": 26
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.99990887271517e-05,
+      "loss": 1.6174,
+      "step": 27
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9998978367267258e-05,
+      "loss": 1.4197,
+      "step": 28
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.9998861701545155e-05,
+      "loss": 1.2337,
+      "step": 29
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9998738730058974e-05,
+      "loss": 1.3482,
+      "step": 30
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999860945288627e-05,
+      "loss": 1.6648,
+      "step": 31
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9998473870108565e-05,
+      "loss": 1.4751,
+      "step": 32
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999833198181137e-05,
+      "loss": 2.0115,
+      "step": 33
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9998183788084155e-05,
+      "loss": 1.1591,
+      "step": 34
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9998029289020388e-05,
+      "loss": 1.4557,
+      "step": 35
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9997868484717504e-05,
+      "loss": 1.8469,
+      "step": 36
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999770137527691e-05,
+      "loss": 1.7395,
+      "step": 37
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9997527960803994e-05,
+      "loss": 1.1644,
+      "step": 38
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999734824140812e-05,
+      "loss": 1.3629,
+      "step": 39
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999716221720263e-05,
+      "loss": 1.1913,
+      "step": 40
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9996969888304835e-05,
+      "loss": 1.9801,
+      "step": 41
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999677125483603e-05,
+      "loss": 1.1722,
+      "step": 42
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9996566316921485e-05,
+      "loss": 1.4999,
+      "step": 43
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9996355074690438e-05,
+      "loss": 1.5593,
+      "step": 44
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.999613752827611e-05,
+      "loss": 1.7799,
+      "step": 45
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9995913677815705e-05,
+      "loss": 1.3714,
+      "step": 46
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9995683523450382e-05,
+      "loss": 1.428,
+      "step": 47
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9995447065325292e-05,
+      "loss": 1.4206,
+      "step": 48
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.9995204303589557e-05,
+      "loss": 1.6583,
+      "step": 49
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9994955238396276e-05,
+      "loss": 1.4349,
+      "step": 50
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9994699869902516e-05,
+      "loss": 1.1203,
+      "step": 51
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999443819826933e-05,
+      "loss": 1.2595,
+      "step": 52
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999417022366174e-05,
+      "loss": 1.7085,
+      "step": 53
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9993895946248744e-05,
+      "loss": 1.4112,
+      "step": 54
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9993615366203313e-05,
+      "loss": 1.1461,
+      "step": 55
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9993328483702393e-05,
+      "loss": 1.3644,
+      "step": 56
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999303529892691e-05,
+      "loss": 1.2273,
+      "step": 57
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9992735812061757e-05,
+      "loss": 1.2104,
+      "step": 58
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999243002329581e-05,
+      "loss": 1.6421,
+      "step": 59
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9992117932821906e-05,
+      "loss": 1.3875,
+      "step": 60
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9991799540836867e-05,
+      "loss": 1.4965,
+      "step": 61
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999147484754149e-05,
+      "loss": 1.304,
+      "step": 62
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9991143853140543e-05,
+      "loss": 1.5476,
+      "step": 63
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9990806557842758e-05,
+      "loss": 1.609,
+      "step": 64
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.999046296186086e-05,
+      "loss": 1.5958,
+      "step": 65
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9990113065411532e-05,
+      "loss": 1.6518,
+      "step": 66
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9989756868715435e-05,
+      "loss": 1.6508,
+      "step": 67
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9989394371997205e-05,
+      "loss": 1.6813,
+      "step": 68
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.9989025575485453e-05,
+      "loss": 1.5673,
+      "step": 69
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9988650479412754e-05,
+      "loss": 1.2511,
+      "step": 70
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9988269084015668e-05,
+      "loss": 1.6433,
+      "step": 71
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9987881389534715e-05,
+      "loss": 1.7642,
+      "step": 72
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.99874873962144e-05,
+      "loss": 1.2926,
+      "step": 73
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9987087104303188e-05,
+      "loss": 1.1941,
+      "step": 74
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9986680514053526e-05,
+      "loss": 1.6356,
+      "step": 75
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998626762572183e-05,
+      "loss": 1.6645,
+      "step": 76
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9985848439568486e-05,
+      "loss": 1.0501,
+      "step": 77
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998542295585785e-05,
+      "loss": 1.1477,
+      "step": 78
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998499117485826e-05,
+      "loss": 1.1104,
+      "step": 79
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998455309684201e-05,
+      "loss": 1.2144,
+      "step": 80
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9984108722085378e-05,
+      "loss": 1.6791,
+      "step": 81
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998365805086861e-05,
+      "loss": 1.2276,
+      "step": 82
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998320108347591e-05,
+      "loss": 1.6222,
+      "step": 83
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998273782019548e-05,
+      "loss": 1.6216,
+      "step": 84
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9982268261319462e-05,
+      "loss": 1.332,
+      "step": 85
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9981792407143988e-05,
+      "loss": 1.3808,
+      "step": 86
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9981310257969158e-05,
+      "loss": 1.4833,
+      "step": 87
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.9980821814099033e-05,
+      "loss": 1.5134,
+      "step": 88
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.998032707584165e-05,
+      "loss": 1.6795,
+      "step": 89
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997982604350902e-05,
+      "loss": 1.1978,
+      "step": 90
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9979318717417112e-05,
+      "loss": 1.6426,
+      "step": 91
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997880509788587e-05,
+      "loss": 1.5072,
+      "step": 92
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9978285185239215e-05,
+      "loss": 1.5113,
+      "step": 93
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997775897980502e-05,
+      "loss": 1.6025,
+      "step": 94
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997722648191514e-05,
+      "loss": 1.5437,
+      "step": 95
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9976687691905394e-05,
+      "loss": 0.8807,
+      "step": 96
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9976142610115567e-05,
+      "loss": 1.2017,
+      "step": 97
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9975591236889414e-05,
+      "loss": 1.571,
+      "step": 98
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997503357257466e-05,
+      "loss": 1.2984,
+      "step": 99
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9974469617522992e-05,
+      "loss": 1.6714,
+      "step": 100
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997389937209007e-05,
+      "loss": 1.8847,
+      "step": 101
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9973322836635517e-05,
+      "loss": 1.2481,
+      "step": 102
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9972740011522927e-05,
+      "loss": 1.3737,
+      "step": 103
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.997215089711985e-05,
+      "loss": 1.1397,
+      "step": 104
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9971555493797817e-05,
+      "loss": 0.7657,
+      "step": 105
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9970953801932313e-05,
+      "loss": 1.2579,
+      "step": 106
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.9970345821902795e-05,
+      "loss": 1.588,
+      "step": 107
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.996973155409269e-05,
+      "loss": 1.0317,
+      "step": 108
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.996911099888938e-05,
+      "loss": 1.6896,
+      "step": 109
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9968484156684215e-05,
+      "loss": 1.7309,
+      "step": 110
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.996785102787252e-05,
+      "loss": 1.0569,
+      "step": 111
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9967211612853566e-05,
+      "loss": 1.8228,
+      "step": 112
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9966565912030607e-05,
+      "loss": 1.4042,
+      "step": 113
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9965913925810847e-05,
+      "loss": 1.2318,
+      "step": 114
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9965255654605466e-05,
+      "loss": 1.6043,
+      "step": 115
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.99645910988296e-05,
+      "loss": 1.7256,
+      "step": 116
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9963920258902344e-05,
+      "loss": 1.477,
+      "step": 117
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.996324313524677e-05,
+      "loss": 1.4139,
+      "step": 118
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.99625597282899e-05,
+      "loss": 1.4781,
+      "step": 119
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9961870038462727e-05,
+      "loss": 1.1951,
+      "step": 120
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.99611740662002e-05,
+      "loss": 1.2997,
+      "step": 121
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.996047181194123e-05,
+      "loss": 1.2891,
+      "step": 122
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.99597632761287e-05,
+      "loss": 2.083,
+      "step": 123
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.995904845920944e-05,
+      "loss": 1.3798,
+      "step": 124
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9958327361634248e-05,
+      "loss": 1.3713,
+      "step": 125
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.995759998385789e-05,
+      "loss": 1.2818,
+      "step": 126
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9956866326339076e-05,
+      "loss": 2.1837,
+      "step": 127
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.9956126389540493e-05,
+      "loss": 1.3562,
+      "step": 128
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9955380173928777e-05,
+      "loss": 1.4653,
+      "step": 129
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.995462767997453e-05,
+      "loss": 1.5638,
+      "step": 130
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.995386890815231e-05,
+      "loss": 1.3003,
+      "step": 131
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9953103858940633e-05,
+      "loss": 1.676,
+      "step": 132
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.995233253282198e-05,
+      "loss": 1.5618,
+      "step": 133
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9951554930282782e-05,
+      "loss": 0.9876,
+      "step": 134
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9950771051813435e-05,
+      "loss": 1.3107,
+      "step": 135
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994998089790829e-05,
+      "loss": 1.3858,
+      "step": 136
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994918446906566e-05,
+      "loss": 1.7285,
+      "step": 137
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9948381765787802e-05,
+      "loss": 1.5864,
+      "step": 138
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994757278858095e-05,
+      "loss": 1.6605,
+      "step": 139
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994675753795528e-05,
+      "loss": 1.3543,
+      "step": 140
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9945936014424924e-05,
+      "loss": 1.481,
+      "step": 141
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9945108218507976e-05,
+      "loss": 1.5271,
+      "step": 142
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994427415072649e-05,
+      "loss": 1.4439,
+      "step": 143
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9943433811606465e-05,
+      "loss": 1.5647,
+      "step": 144
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994258720167786e-05,
+      "loss": 1.2134,
+      "step": 145
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9941734321474586e-05,
+      "loss": 1.3431,
+      "step": 146
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.994087517153451e-05,
+      "loss": 1.6574,
+      "step": 147
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.9940009752399462e-05,
+      "loss": 1.3856,
+      "step": 148
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9939138064615205e-05,
+      "loss": 1.2281,
+      "step": 149
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9938260108731474e-05,
+      "loss": 1.7524,
+      "step": 150
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9937375885301948e-05,
+      "loss": 1.4849,
+      "step": 151
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9936485394884263e-05,
+      "loss": 1.6183,
+      "step": 152
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9935588638040005e-05,
+      "loss": 1.854,
+      "step": 153
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.993468561533471e-05,
+      "loss": 1.6805,
+      "step": 154
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.993377632733787e-05,
+      "loss": 1.3257,
+      "step": 155
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.993286077462292e-05,
+      "loss": 1.5286,
+      "step": 156
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.993193895776726e-05,
+      "loss": 1.7098,
+      "step": 157
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.993101087735223e-05,
+      "loss": 2.0118,
+      "step": 158
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9930076533963117e-05,
+      "loss": 1.0722,
+      "step": 159
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.992913592818917e-05,
+      "loss": 1.1482,
+      "step": 160
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9928189060623574e-05,
+      "loss": 1.5841,
+      "step": 161
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9927235931863477e-05,
+      "loss": 1.3712,
+      "step": 162
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.992627654250996e-05,
+      "loss": 1.563,
+      "step": 163
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.992531089316806e-05,
+      "loss": 1.1601,
+      "step": 164
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9924338984446773e-05,
+      "loss": 1.5141,
+      "step": 165
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9923360816959016e-05,
+      "loss": 1.6872,
+      "step": 166
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.992237639132168e-05,
+      "loss": 1.6658,
+      "step": 167
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.9921385708155588e-05,
+      "loss": 1.5354,
+      "step": 168
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9920388768085513e-05,
+      "loss": 1.457,
+      "step": 169
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9919385571740172e-05,
+      "loss": 1.5823,
+      "step": 170
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.991837611975223e-05,
+      "loss": 1.5392,
+      "step": 171
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9917360412758295e-05,
+      "loss": 1.3985,
+      "step": 172
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9916338451398923e-05,
+      "loss": 1.1896,
+      "step": 173
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9915310236318607e-05,
+      "loss": 1.2652,
+      "step": 174
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9914275768165793e-05,
+      "loss": 1.0745,
+      "step": 175
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.991323504759287e-05,
+      "loss": 1.8471,
+      "step": 176
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.991218807525616e-05,
+      "loss": 1.871,
+      "step": 177
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9911134851815935e-05,
+      "loss": 1.3217,
+      "step": 178
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9910075377936414e-05,
+      "loss": 1.4894,
+      "step": 179
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9909009654285748e-05,
+      "loss": 1.1755,
+      "step": 180
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9907937681536032e-05,
+      "loss": 1.6328,
+      "step": 181
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9906859460363307e-05,
+      "loss": 1.4971,
+      "step": 182
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9905774991447552e-05,
+      "loss": 1.4121,
+      "step": 183
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9904684275472684e-05,
+      "loss": 1.5965,
+      "step": 184
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9903587313126557e-05,
+      "loss": 1.21,
+      "step": 185
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9902484105100974e-05,
+      "loss": 1.8833,
+      "step": 186
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9901374652091666e-05,
+      "loss": 1.0323,
+      "step": 187
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9900258954798315e-05,
+      "loss": 1.5577,
+      "step": 188
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.989913701392453e-05,
+      "loss": 1.8889,
+      "step": 189
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9898008830177856e-05,
+      "loss": 1.6154,
+      "step": 190
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9896874404269786e-05,
+      "loss": 1.5606,
+      "step": 191
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.989573373691574e-05,
+      "loss": 1.2633,
+      "step": 192
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.989458682883508e-05,
+      "loss": 1.2808,
+      "step": 193
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9893433680751105e-05,
+      "loss": 1.5281,
+      "step": 194
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9892274293391035e-05,
+      "loss": 1.1487,
+      "step": 195
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9891108667486047e-05,
+      "loss": 1.1523,
+      "step": 196
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9889936803771237e-05,
+      "loss": 1.3093,
+      "step": 197
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9888758702985637e-05,
+      "loss": 1.249,
+      "step": 198
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9887574365872214e-05,
+      "loss": 1.1357,
+      "step": 199
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.988638379317787e-05,
+      "loss": 1.5416,
+      "step": 200
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.988518698565344e-05,
+      "loss": 1.63,
+      "step": 201
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9883983944053678e-05,
+      "loss": 1.194,
+      "step": 202
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9882774669137293e-05,
+      "loss": 1.281,
+      "step": 203
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9881559161666905e-05,
+      "loss": 1.1447,
+      "step": 204
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.988033742240907e-05,
+      "loss": 1.7913,
+      "step": 205
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9879109452134283e-05,
+      "loss": 1.5167,
+      "step": 206
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9877875251616954e-05,
+      "loss": 1.2428,
+      "step": 207
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9876634821635432e-05,
+      "loss": 1.3816,
+      "step": 208
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9875388162971992e-05,
+      "loss": 1.4307,
+      "step": 209
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9874135276412837e-05,
+      "loss": 1.3444,
+      "step": 210
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.98728761627481e-05,
+      "loss": 1.3973,
+      "step": 211
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9871610822771835e-05,
+      "loss": 1.4477,
+      "step": 212
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9870339257282028e-05,
+      "loss": 1.0952,
+      "step": 213
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9869061467080587e-05,
+      "loss": 1.582,
+      "step": 214
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9867777452973352e-05,
+      "loss": 1.4796,
+      "step": 215
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9866487215770084e-05,
+      "loss": 0.996,
+      "step": 216
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9865190756284467e-05,
+      "loss": 1.2355,
+      "step": 217
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9863888075334113e-05,
+      "loss": 1.3907,
+      "step": 218
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.986257917374055e-05,
+      "loss": 1.5925,
+      "step": 219
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.986126405232924e-05,
+      "loss": 1.3692,
+      "step": 220
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9859942711929557e-05,
+      "loss": 1.2802,
+      "step": 221
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9858615153374808e-05,
+      "loss": 1.3644,
+      "step": 222
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.985728137750221e-05,
+      "loss": 1.1744,
+      "step": 223
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.985594138515291e-05,
+      "loss": 1.4946,
+      "step": 224
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9854595177171968e-05,
+      "loss": 1.6203,
+      "step": 225
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9853242754408376e-05,
+      "loss": 1.5148,
+      "step": 226
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9851884117715027e-05,
+      "loss": 1.7796,
+      "step": 227
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9850519267948747e-05,
+      "loss": 1.0399,
+      "step": 228
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9849148205970275e-05,
+      "loss": 1.1837,
+      "step": 229
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.984777093264427e-05,
+      "loss": 1.3893,
+      "step": 230
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9846387448839308e-05,
+      "loss": 1.2283,
+      "step": 231
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9844997755427875e-05,
+      "loss": 1.0594,
+      "step": 232
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.984360185328639e-05,
+      "loss": 1.5023,
+      "step": 233
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9842199743295164e-05,
+      "loss": 1.619,
+      "step": 234
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.984079142633844e-05,
+      "loss": 0.9526,
+      "step": 235
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983937690330437e-05,
+      "loss": 1.3427,
+      "step": 236
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983795617508502e-05,
+      "loss": 1.1187,
+      "step": 237
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9836529242576373e-05,
+      "loss": 1.2346,
+      "step": 238
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983509610667832e-05,
+      "loss": 1.8011,
+      "step": 239
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983365676829466e-05,
+      "loss": 1.1902,
+      "step": 240
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983221122833312e-05,
+      "loss": 1.3564,
+      "step": 241
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.983075948770532e-05,
+      "loss": 1.2038,
+      "step": 242
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9829301547326794e-05,
+      "loss": 1.3158,
+      "step": 243
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9827837408116996e-05,
+      "loss": 1.6507,
+      "step": 244
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9826367070999284e-05,
+      "loss": 1.8778,
+      "step": 245
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9824890536900917e-05,
+      "loss": 1.3691,
+      "step": 246
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.982340780675307e-05,
+      "loss": 1.3956,
+      "step": 247
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.982191888149083e-05,
+      "loss": 1.275,
+      "step": 248
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 1.4743634462356567,
+      "eval_runtime": 99.6539,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 248
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9820423762053178e-05,
+      "loss": 1.3582,
+      "step": 249
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.981892244938301e-05,
+      "loss": 1.3869,
+      "step": 250
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9817414944427133e-05,
+      "loss": 1.2653,
+      "step": 251
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9815901248136242e-05,
+      "loss": 1.3705,
+      "step": 252
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9814381361464953e-05,
+      "loss": 1.5251,
+      "step": 253
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9812855285371778e-05,
+      "loss": 1.5613,
+      "step": 254
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9811323020819136e-05,
+      "loss": 1.3138,
+      "step": 255
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.980978456877334e-05,
+      "loss": 1.0931,
+      "step": 256
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9808239930204625e-05,
+      "loss": 1.4202,
+      "step": 257
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.98066891060871e-05,
+      "loss": 1.2566,
+      "step": 258
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.98051320973988e-05,
+      "loss": 1.1673,
+      "step": 259
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9803568905121647e-05,
+      "loss": 1.2889,
+      "step": 260
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.980199953024146e-05,
+      "loss": 1.3371,
+      "step": 261
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9800423973747972e-05,
+      "loss": 1.4299,
+      "step": 262
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9798842236634797e-05,
+      "loss": 1.4825,
+      "step": 263
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9797254319899453e-05,
+      "loss": 1.2745,
+      "step": 264
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.979566022454337e-05,
+      "loss": 1.4379,
+      "step": 265
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9794059951571848e-05,
+      "loss": 1.386,
+      "step": 266
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.97924535019941e-05,
+      "loss": 1.2546,
+      "step": 267
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.979084087682323e-05,
+      "loss": 1.2028,
+      "step": 268
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.978922207707624e-05,
+      "loss": 1.1725,
+      "step": 269
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.978759710377402e-05,
+      "loss": 1.6772,
+      "step": 270
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9785965957941362e-05,
+      "loss": 1.555,
+      "step": 271
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.978432864060694e-05,
+      "loss": 1.6195,
+      "step": 272
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9782685152803326e-05,
+      "loss": 1.4591,
+      "step": 273
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.978103549556698e-05,
+      "loss": 1.2727,
+      "step": 274
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9779379669938265e-05,
+      "loss": 1.3214,
+      "step": 275
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9777717676961412e-05,
+      "loss": 1.6267,
+      "step": 276
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.977604951768456e-05,
+      "loss": 1.1487,
+      "step": 277
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.977437519315973e-05,
+      "loss": 1.8019,
+      "step": 278
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9772694704442836e-05,
+      "loss": 1.3772,
+      "step": 279
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.977100805259367e-05,
+      "loss": 1.1348,
+      "step": 280
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9769315238675916e-05,
+      "loss": 1.209,
+      "step": 281
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9767616263757146e-05,
+      "loss": 1.4235,
+      "step": 282
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9765911128908813e-05,
+      "loss": 1.2202,
+      "step": 283
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.976419983520626e-05,
+      "loss": 1.2418,
+      "step": 284
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.976248238372871e-05,
+      "loss": 1.3901,
+      "step": 285
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9760758775559275e-05,
+      "loss": 1.1207,
+      "step": 286
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9759029011784936e-05,
+      "loss": 1.0332,
+      "step": 287
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9757293093496573e-05,
+      "loss": 1.4162,
+      "step": 288
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9755551021788934e-05,
+      "loss": 1.5068,
+      "step": 289
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.975380279776066e-05,
+      "loss": 1.5863,
+      "step": 290
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9752048422514262e-05,
+      "loss": 1.4527,
+      "step": 291
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9750287897156136e-05,
+      "loss": 1.2066,
+      "step": 292
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.974852122279655e-05,
+      "loss": 1.2543,
+      "step": 293
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9746748400549653e-05,
+      "loss": 1.455,
+      "step": 294
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9744969431533474e-05,
+      "loss": 1.4658,
+      "step": 295
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9743184316869924e-05,
+      "loss": 1.2493,
+      "step": 296
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.974139305768477e-05,
+      "loss": 1.4861,
+      "step": 297
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9739595655107675e-05,
+      "loss": 1.6473,
+      "step": 298
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9737792110272167e-05,
+      "loss": 1.4982,
+      "step": 299
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9735982424315642e-05,
+      "loss": 1.4031,
+      "step": 300
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.973416659837938e-05,
+      "loss": 1.1998,
+      "step": 301
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.973234463360853e-05,
+      "loss": 1.0635,
+      "step": 302
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.973051653115211e-05,
+      "loss": 1.29,
+      "step": 303
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9728682292163002e-05,
+      "loss": 1.2678,
+      "step": 304
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9726841917797977e-05,
+      "loss": 1.3882,
+      "step": 305
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9724995409217658e-05,
+      "loss": 1.0476,
+      "step": 306
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.972314276758654e-05,
+      "loss": 1.2119,
+      "step": 307
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9721283994072995e-05,
+      "loss": 1.7283,
+      "step": 308
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.971941908984925e-05,
+      "loss": 1.6008,
+      "step": 309
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.97175480560914e-05,
+      "loss": 1.3317,
+      "step": 310
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9715670893979416e-05,
+      "loss": 1.2612,
+      "step": 311
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9713787604697125e-05,
+      "loss": 1.4356,
+      "step": 312
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9711898189432218e-05,
+      "loss": 1.3431,
+      "step": 313
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9710002649376255e-05,
+      "loss": 0.8521,
+      "step": 314
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9708100985724654e-05,
+      "loss": 1.9041,
+      "step": 315
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.970619319967669e-05,
+      "loss": 1.6982,
+      "step": 316
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.970427929243551e-05,
+      "loss": 1.6813,
+      "step": 317
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9702359265208114e-05,
+      "loss": 1.5686,
+      "step": 318
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9700433119205368e-05,
+      "loss": 1.7669,
+      "step": 319
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9698500855641988e-05,
+      "loss": 0.9086,
+      "step": 320
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9696562475736556e-05,
+      "loss": 1.4198,
+      "step": 321
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9694617980711503e-05,
+      "loss": 1.8693,
+      "step": 322
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9692667371793127e-05,
+      "loss": 0.7734,
+      "step": 323
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9690710650211572e-05,
+      "loss": 1.1205,
+      "step": 324
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.968874781720084e-05,
+      "loss": 0.959,
+      "step": 325
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9686778873998792e-05,
+      "loss": 1.9894,
+      "step": 326
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9684803821847137e-05,
+      "loss": 1.6784,
+      "step": 327
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9682822661991435e-05,
+      "loss": 1.4841,
+      "step": 328
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.968083539568111e-05,
+      "loss": 1.7689,
+      "step": 329
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9678842024169418e-05,
+      "loss": 1.3004,
+      "step": 330
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9676842548713475e-05,
+      "loss": 1.5111,
+      "step": 331
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9674836970574253e-05,
+      "loss": 1.0877,
+      "step": 332
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9672825291016564e-05,
+      "loss": 1.3979,
+      "step": 333
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.967080751130907e-05,
+      "loss": 1.5652,
+      "step": 334
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9668783632724278e-05,
+      "loss": 1.2373,
+      "step": 335
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9666753656538545e-05,
+      "loss": 1.5039,
+      "step": 336
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9664717584032075e-05,
+      "loss": 1.3572,
+      "step": 337
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9662675416488908e-05,
+      "loss": 1.3156,
+      "step": 338
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9660627155196934e-05,
+      "loss": 1.4775,
+      "step": 339
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.965857280144789e-05,
+      "loss": 1.5048,
+      "step": 340
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9656512356537343e-05,
+      "loss": 1.0645,
+      "step": 341
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9654445821764717e-05,
+      "loss": 1.4329,
+      "step": 342
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9652373198433265e-05,
+      "loss": 1.3218,
+      "step": 343
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.965029448785008e-05,
+      "loss": 1.6661,
+      "step": 344
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9648209691326103e-05,
+      "loss": 1.183,
+      "step": 345
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.96461188101761e-05,
+      "loss": 1.2924,
+      "step": 346
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.964402184571869e-05,
+      "loss": 1.0152,
+      "step": 347
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9641918799276313e-05,
+      "loss": 1.3248,
+      "step": 348
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9639809672175253e-05,
+      "loss": 1.2242,
+      "step": 349
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.963769446574563e-05,
+      "loss": 1.2416,
+      "step": 350
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9635573181321394e-05,
+      "loss": 1.4711,
+      "step": 351
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9633445820240323e-05,
+      "loss": 1.454,
+      "step": 352
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.963131238384404e-05,
+      "loss": 1.355,
+      "step": 353
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9629172873477995e-05,
+      "loss": 1.3991,
+      "step": 354
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.962702729049146e-05,
+      "loss": 1.3905,
+      "step": 355
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9624875636237547e-05,
+      "loss": 1.5024,
+      "step": 356
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9622717912073193e-05,
+      "loss": 1.1932,
+      "step": 357
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.962055411935916e-05,
+      "loss": 1.4808,
+      "step": 358
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.961838425946004e-05,
+      "loss": 1.5079,
+      "step": 359
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9616208333744255e-05,
+      "loss": 1.1668,
+      "step": 360
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9614026343584048e-05,
+      "loss": 0.9482,
+      "step": 361
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9611838290355483e-05,
+      "loss": 1.5103,
+      "step": 362
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9609644175438457e-05,
+      "loss": 1.4825,
+      "step": 363
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9607444000216676e-05,
+      "loss": 1.0542,
+      "step": 364
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9605237766077686e-05,
+      "loss": 1.3711,
+      "step": 365
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9603025474412844e-05,
+      "loss": 1.8036,
+      "step": 366
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.960080712661732e-05,
+      "loss": 1.2591,
+      "step": 367
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.959858272409012e-05,
+      "loss": 1.5749,
+      "step": 368
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9596352268234053e-05,
+      "loss": 1.4592,
+      "step": 369
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9594115760455755e-05,
+      "loss": 0.8912,
+      "step": 370
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9591873202165678e-05,
+      "loss": 0.9196,
+      "step": 371
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9589624594778077e-05,
+      "loss": 1.4079,
+      "step": 372
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9587369939711044e-05,
+      "loss": 1.3132,
+      "step": 373
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.958510923838647e-05,
+      "loss": 0.6786,
+      "step": 374
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.958284249223006e-05,
+      "loss": 1.4426,
+      "step": 375
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9580569702671332e-05,
+      "loss": 1.1317,
+      "step": 376
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.957829087114362e-05,
+      "loss": 1.4542,
+      "step": 377
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.957600599908406e-05,
+      "loss": 0.807,
+      "step": 378
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.957371508793361e-05,
+      "loss": 1.5726,
+      "step": 379
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9571418139137023e-05,
+      "loss": 1.4531,
+      "step": 380
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9569115154142873e-05,
+      "loss": 1.4213,
+      "step": 381
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9566806134403526e-05,
+      "loss": 1.3222,
+      "step": 382
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9564491081375157e-05,
+      "loss": 0.9039,
+      "step": 383
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.956216999651776e-05,
+      "loss": 1.507,
+      "step": 384
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9559842881295122e-05,
+      "loss": 1.4292,
+      "step": 385
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.955750973717483e-05,
+      "loss": 1.2868,
+      "step": 386
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.955517056562828e-05,
+      "loss": 1.5842,
+      "step": 387
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.955282536813066e-05,
+      "loss": 1.615,
+      "step": 388
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.955047414616097e-05,
+      "loss": 1.3707,
+      "step": 389
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9548116901202006e-05,
+      "loss": 1.4077,
+      "step": 390
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9545753634740358e-05,
+      "loss": 1.4326,
+      "step": 391
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9543384348266415e-05,
+      "loss": 1.3821,
+      "step": 392
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.954100904327436e-05,
+      "loss": 1.4803,
+      "step": 393
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.953862772126218e-05,
+      "loss": 1.5412,
+      "step": 394
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.953624038373165e-05,
+      "loss": 1.7189,
+      "step": 395
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9533847032188337e-05,
+      "loss": 1.1109,
+      "step": 396
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.953144766814161e-05,
+      "loss": 1.237,
+      "step": 397
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.952904229310462e-05,
+      "loss": 1.2492,
+      "step": 398
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.952663090859431e-05,
+      "loss": 1.3605,
+      "step": 399
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.952421351613142e-05,
+      "loss": 1.4423,
+      "step": 400
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9521790117240472e-05,
+      "loss": 1.3342,
+      "step": 401
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9519360713449775e-05,
+      "loss": 1.7274,
+      "step": 402
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9516925306291435e-05,
+      "loss": 1.6019,
+      "step": 403
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.951448389730133e-05,
+      "loss": 1.1267,
+      "step": 404
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9512036488019138e-05,
+      "loss": 1.1338,
+      "step": 405
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9509583079988307e-05,
+      "loss": 1.6472,
+      "step": 406
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9507123674756076e-05,
+      "loss": 1.2389,
+      "step": 407
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9504658273873465e-05,
+      "loss": 1.5887,
+      "step": 408
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9502186878895273e-05,
+      "loss": 1.0987,
+      "step": 409
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9499709491380083e-05,
+      "loss": 1.384,
+      "step": 410
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9497226112890252e-05,
+      "loss": 1.7325,
+      "step": 411
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9494736744991925e-05,
+      "loss": 1.4435,
+      "step": 412
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9492241389255006e-05,
+      "loss": 1.9634,
+      "step": 413
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9489740047253197e-05,
+      "loss": 1.5567,
+      "step": 414
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9487232720563962e-05,
+      "loss": 1.3805,
+      "step": 415
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.948471941076854e-05,
+      "loss": 1.325,
+      "step": 416
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9482200119451945e-05,
+      "loss": 1.3569,
+      "step": 417
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.947967484820297e-05,
+      "loss": 1.1947,
+      "step": 418
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.947714359861416e-05,
+      "loss": 1.231,
+      "step": 419
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9474606372281854e-05,
+      "loss": 1.1062,
+      "step": 420
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9472063170806144e-05,
+      "loss": 1.8761,
+      "step": 421
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.94695139957909e-05,
+      "loss": 1.3697,
+      "step": 422
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9466958848843748e-05,
+      "loss": 1.0142,
+      "step": 423
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9464397731576093e-05,
+      "loss": 1.0189,
+      "step": 424
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.94618306456031e-05,
+      "loss": 1.7268,
+      "step": 425
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9459257592543688e-05,
+      "loss": 1.0269,
+      "step": 426
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9456678574020557e-05,
+      "loss": 1.2833,
+      "step": 427
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9454093591660155e-05,
+      "loss": 1.1967,
+      "step": 428
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.94515026470927e-05,
+      "loss": 1.5793,
+      "step": 429
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9448905741952167e-05,
+      "loss": 1.4184,
+      "step": 430
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.944630287787629e-05,
+      "loss": 1.2217,
+      "step": 431
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9443694056506556e-05,
+      "loss": 1.222,
+      "step": 432
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9441079279488213e-05,
+      "loss": 1.2332,
+      "step": 433
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9438458548470268e-05,
+      "loss": 1.7458,
+      "step": 434
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9435831865105482e-05,
+      "loss": 1.3248,
+      "step": 435
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9433199231050367e-05,
+      "loss": 1.9621,
+      "step": 436
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9430560647965192e-05,
+      "loss": 0.8943,
+      "step": 437
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.942791611751397e-05,
+      "loss": 0.6546,
+      "step": 438
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9425265641364467e-05,
+      "loss": 1.4626,
+      "step": 439
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9422609221188208e-05,
+      "loss": 1.3289,
+      "step": 440
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9419946858660452e-05,
+      "loss": 0.8965,
+      "step": 441
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9417278555460223e-05,
+      "loss": 1.225,
+      "step": 442
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.941460431327027e-05,
+      "loss": 1.713,
+      "step": 443
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.941192413377711e-05,
+      "loss": 1.0535,
+      "step": 444
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9409238018670986e-05,
+      "loss": 1.4671,
+      "step": 445
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9406545969645894e-05,
+      "loss": 1.5399,
+      "step": 446
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.940384798839957e-05,
+      "loss": 1.3484,
+      "step": 447
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.940114407663349e-05,
+      "loss": 1.6319,
+      "step": 448
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9398434236052873e-05,
+      "loss": 1.7901,
+      "step": 449
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9395718468366672e-05,
+      "loss": 1.4023,
+      "step": 450
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9392996775287588e-05,
+      "loss": 1.1306,
+      "step": 451
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9390269158532043e-05,
+      "loss": 1.648,
+      "step": 452
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9387535619820207e-05,
+      "loss": 1.5907,
+      "step": 453
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9384796160875982e-05,
+      "loss": 1.2352,
+      "step": 454
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9382050783427e-05,
+      "loss": 1.6986,
+      "step": 455
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9379299489204634e-05,
+      "loss": 1.4436,
+      "step": 456
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.937654227994398e-05,
+      "loss": 1.8155,
+      "step": 457
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.937377915738386e-05,
+      "loss": 1.4464,
+      "step": 458
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.937101012326685e-05,
+      "loss": 1.5623,
+      "step": 459
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9368235179339217e-05,
+      "loss": 1.6515,
+      "step": 460
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9365454327350984e-05,
+      "loss": 1.3808,
+      "step": 461
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.936266756905589e-05,
+      "loss": 1.708,
+      "step": 462
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.93598749062114e-05,
+      "loss": 1.3696,
+      "step": 463
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9357076340578696e-05,
+      "loss": 1.4497,
+      "step": 464
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9354271873922692e-05,
+      "loss": 1.2003,
+      "step": 465
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.935146150801202e-05,
+      "loss": 1.182,
+      "step": 466
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9348645244619035e-05,
+      "loss": 1.9314,
+      "step": 467
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9345823085519804e-05,
+      "loss": 1.5496,
+      "step": 468
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9342995032494116e-05,
+      "loss": 1.0179,
+      "step": 469
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9340161087325483e-05,
+      "loss": 1.3892,
+      "step": 470
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9337321251801123e-05,
+      "loss": 1.0485,
+      "step": 471
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9334475527711973e-05,
+      "loss": 1.786,
+      "step": 472
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9331623916852683e-05,
+      "loss": 1.3026,
+      "step": 473
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.932876642102162e-05,
+      "loss": 1.2246,
+      "step": 474
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9325903042020856e-05,
+      "loss": 1.7167,
+      "step": 475
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9323033781656178e-05,
+      "loss": 1.5324,
+      "step": 476
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9320158641737077e-05,
+      "loss": 1.51,
+      "step": 477
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9317277624076758e-05,
+      "loss": 1.3634,
+      "step": 478
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.931439073049213e-05,
+      "loss": 1.2313,
+      "step": 479
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.93114979628038e-05,
+      "loss": 1.624,
+      "step": 480
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9308599322836092e-05,
+      "loss": 1.152,
+      "step": 481
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.930569481241703e-05,
+      "loss": 1.2635,
+      "step": 482
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9302784433378333e-05,
+      "loss": 1.921,
+      "step": 483
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.929986818755543e-05,
+      "loss": 1.2884,
+      "step": 484
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9296946076787447e-05,
+      "loss": 1.2914,
+      "step": 485
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9294018102917208e-05,
+      "loss": 1.2168,
+      "step": 486
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.929108426779123e-05,
+      "loss": 1.7837,
+      "step": 487
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9288144573259735e-05,
+      "loss": 1.3335,
+      "step": 488
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9285199021176634e-05,
+      "loss": 1.2004,
+      "step": 489
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9282247613399537e-05,
+      "loss": 1.3216,
+      "step": 490
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9279290351789737e-05,
+      "loss": 1.7597,
+      "step": 491
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9276327238212232e-05,
+      "loss": 1.3228,
+      "step": 492
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9273358274535703e-05,
+      "loss": 1.6335,
+      "step": 493
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9270383462632524e-05,
+      "loss": 1.4952,
+      "step": 494
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.926740280437875e-05,
+      "loss": 1.8452,
+      "step": 495
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.926441630165413e-05,
+      "loss": 1.1917,
+      "step": 496
+    },
+    {
+      "epoch": 0.25,
+      "eval_loss": 1.436846375465393,
+      "eval_runtime": 99.6318,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 496
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9261423956342096e-05,
+      "loss": 1.3163,
+      "step": 497
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9258425770329764e-05,
+      "loss": 1.0938,
+      "step": 498
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9255421745507937e-05,
+      "loss": 1.0377,
+      "step": 499
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.92524118837711e-05,
+      "loss": 1.2684,
+      "step": 500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.924939618701741e-05,
+      "loss": 1.3812,
+      "step": 501
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9246374657148714e-05,
+      "loss": 1.4278,
+      "step": 502
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9243347296070535e-05,
+      "loss": 1.503,
+      "step": 503
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9240314105692074e-05,
+      "loss": 1.4397,
+      "step": 504
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9237275087926206e-05,
+      "loss": 1.3929,
+      "step": 505
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9234230244689484e-05,
+      "loss": 1.4961,
+      "step": 506
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9231179577902125e-05,
+      "loss": 0.7874,
+      "step": 507
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9228123089488034e-05,
+      "loss": 1.4337,
+      "step": 508
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9225060781374782e-05,
+      "loss": 1.2741,
+      "step": 509
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.92219926554936e-05,
+      "loss": 1.049,
+      "step": 510
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9218918713779404e-05,
+      "loss": 1.5724,
+      "step": 511
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9215838958170762e-05,
+      "loss": 1.6128,
+      "step": 512
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.921275339060992e-05,
+      "loss": 1.2712,
+      "step": 513
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9209662013042786e-05,
+      "loss": 1.47,
+      "step": 514
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9206564827418935e-05,
+      "loss": 1.174,
+      "step": 515
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9203461835691596e-05,
+      "loss": 1.5194,
+      "step": 516
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.920035303981767e-05,
+      "loss": 1.3734,
+      "step": 517
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9197238441757705e-05,
+      "loss": 1.559,
+      "step": 518
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9194118043475923e-05,
+      "loss": 1.7758,
+      "step": 519
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9190991846940202e-05,
+      "loss": 1.4306,
+      "step": 520
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9187859854122065e-05,
+      "loss": 0.7942,
+      "step": 521
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9184722066996697e-05,
+      "loss": 1.4129,
+      "step": 522
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.918157848754294e-05,
+      "loss": 1.4661,
+      "step": 523
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9178429117743294e-05,
+      "loss": 1.7917,
+      "step": 524
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9175273959583892e-05,
+      "loss": 1.7892,
+      "step": 525
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.917211301505453e-05,
+      "loss": 1.204,
+      "step": 526
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9168946286148663e-05,
+      "loss": 1.017,
+      "step": 527
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9165773774863367e-05,
+      "loss": 0.9312,
+      "step": 528
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9162595483199396e-05,
+      "loss": 1.2145,
+      "step": 529
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.915941141316112e-05,
+      "loss": 1.6209,
+      "step": 530
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9156221566756578e-05,
+      "loss": 1.6366,
+      "step": 531
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9153025945997437e-05,
+      "loss": 1.6848,
+      "step": 532
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9149824552899008e-05,
+      "loss": 1.2836,
+      "step": 533
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.914661738948024e-05,
+      "loss": 1.2836,
+      "step": 534
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9143404457763738e-05,
+      "loss": 1.598,
+      "step": 535
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.914018575977572e-05,
+      "loss": 1.2498,
+      "step": 536
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9136961297546052e-05,
+      "loss": 1.4344,
+      "step": 537
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9133731073108238e-05,
+      "loss": 1.2924,
+      "step": 538
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9130495088499417e-05,
+      "loss": 1.0017,
+      "step": 539
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9127253345760354e-05,
+      "loss": 1.048,
+      "step": 540
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9124005846935446e-05,
+      "loss": 1.1805,
+      "step": 541
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9120752594072724e-05,
+      "loss": 1.1611,
+      "step": 542
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.911749358922385e-05,
+      "loss": 1.2517,
+      "step": 543
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.91142288344441e-05,
+      "loss": 1.4351,
+      "step": 544
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9110958331792396e-05,
+      "loss": 1.2312,
+      "step": 545
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9107682083331265e-05,
+      "loss": 1.3427,
+      "step": 546
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9104400091126877e-05,
+      "loss": 1.9784,
+      "step": 547
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.910111235724901e-05,
+      "loss": 1.0882,
+      "step": 548
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.909781888377106e-05,
+      "loss": 1.2761,
+      "step": 549
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.909451967277006e-05,
+      "loss": 1.1144,
+      "step": 550
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9091214726326652e-05,
+      "loss": 1.5043,
+      "step": 551
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.908790404652509e-05,
+      "loss": 1.6154,
+      "step": 552
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9084587635453246e-05,
+      "loss": 1.3277,
+      "step": 553
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9081265495202618e-05,
+      "loss": 1.6941,
+      "step": 554
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9077937627868295e-05,
+      "loss": 1.0345,
+      "step": 555
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9074604035549e-05,
+      "loss": 1.4388,
+      "step": 556
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9071264720347057e-05,
+      "loss": 0.8748,
+      "step": 557
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.906791968436839e-05,
+      "loss": 1.4478,
+      "step": 558
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9064568929722553e-05,
+      "loss": 1.1109,
+      "step": 559
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.906121245852268e-05,
+      "loss": 1.4213,
+      "step": 560
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9057850272885538e-05,
+      "loss": 1.3594,
+      "step": 561
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.905448237493147e-05,
+      "loss": 1.2485,
+      "step": 562
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.905110876678444e-05,
+      "loss": 1.5978,
+      "step": 563
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9047729450572007e-05,
+      "loss": 1.517,
+      "step": 564
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9044344428425335e-05,
+      "loss": 1.2011,
+      "step": 565
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9040953702479176e-05,
+      "loss": 1.098,
+      "step": 566
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9037557274871893e-05,
+      "loss": 1.1603,
+      "step": 567
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9034155147745424e-05,
+      "loss": 1.5996,
+      "step": 568
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.903074732324533e-05,
+      "loss": 1.0791,
+      "step": 569
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9027333803520736e-05,
+      "loss": 1.4224,
+      "step": 570
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9023914590724383e-05,
+      "loss": 1.3715,
+      "step": 571
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9020489687012582e-05,
+      "loss": 0.907,
+      "step": 572
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9017059094545244e-05,
+      "loss": 1.4973,
+      "step": 573
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9013622815485877e-05,
+      "loss": 1.7539,
+      "step": 574
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9010180852001546e-05,
+      "loss": 1.5102,
+      "step": 575
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9006733206262933e-05,
+      "loss": 1.0709,
+      "step": 576
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9003279880444285e-05,
+      "loss": 1.5776,
+      "step": 577
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.899982087672344e-05,
+      "loss": 1.5828,
+      "step": 578
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8996356197281803e-05,
+      "loss": 0.9347,
+      "step": 579
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8992885844304374e-05,
+      "loss": 1.2101,
+      "step": 580
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8989409819979724e-05,
+      "loss": 1.3425,
+      "step": 581
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.89859281265e-05,
+      "loss": 1.2859,
+      "step": 582
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.898244076606093e-05,
+      "loss": 1.0246,
+      "step": 583
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.897894774086181e-05,
+      "loss": 1.2258,
+      "step": 584
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8975449053105505e-05,
+      "loss": 1.8727,
+      "step": 585
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8971944704998462e-05,
+      "loss": 1.8236,
+      "step": 586
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8968434698750688e-05,
+      "loss": 1.353,
+      "step": 587
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8964919036575764e-05,
+      "loss": 1.6112,
+      "step": 588
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8961397720690837e-05,
+      "loss": 1.0724,
+      "step": 589
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8957870753316617e-05,
+      "loss": 0.7038,
+      "step": 590
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.895433813667738e-05,
+      "loss": 1.3323,
+      "step": 591
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8950799873000964e-05,
+      "loss": 1.5053,
+      "step": 592
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8947255964518766e-05,
+      "loss": 1.8766,
+      "step": 593
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.894370641346575e-05,
+      "loss": 1.5618,
+      "step": 594
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8940151222080432e-05,
+      "loss": 1.4632,
+      "step": 595
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.893659039260489e-05,
+      "loss": 1.3342,
+      "step": 596
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.893302392728475e-05,
+      "loss": 1.6476,
+      "step": 597
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.89294518283692e-05,
+      "loss": 1.6092,
+      "step": 598
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8925874098110976e-05,
+      "loss": 1.4304,
+      "step": 599
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8922290738766368e-05,
+      "loss": 1.1196,
+      "step": 600
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8918701752595213e-05,
+      "loss": 1.3435,
+      "step": 601
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8915107141860902e-05,
+      "loss": 1.0456,
+      "step": 602
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8911506908830367e-05,
+      "loss": 1.4373,
+      "step": 603
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.8907901055774092e-05,
+      "loss": 0.7475,
+      "step": 604
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8904289584966098e-05,
+      "loss": 1.6336,
+      "step": 605
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8900672498683953e-05,
+      "loss": 0.7168,
+      "step": 606
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8897049799208765e-05,
+      "loss": 1.0685,
+      "step": 607
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.889342148882519e-05,
+      "loss": 1.774,
+      "step": 608
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.888978756982141e-05,
+      "loss": 1.2926,
+      "step": 609
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8886148044489145e-05,
+      "loss": 1.387,
+      "step": 610
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8882502915123662e-05,
+      "loss": 1.3983,
+      "step": 611
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8878852184023754e-05,
+      "loss": 1.0143,
+      "step": 612
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8875195853491744e-05,
+      "loss": 1.538,
+      "step": 613
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8871533925833497e-05,
+      "loss": 1.4635,
+      "step": 614
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8867866403358394e-05,
+      "loss": 1.255,
+      "step": 615
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.886419328837936e-05,
+      "loss": 0.8735,
+      "step": 616
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8860514583212825e-05,
+      "loss": 1.2476,
+      "step": 617
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.885683029017877e-05,
+      "loss": 1.2701,
+      "step": 618
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8853140411600685e-05,
+      "loss": 1.107,
+      "step": 619
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8849444949805582e-05,
+      "loss": 0.975,
+      "step": 620
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8845743907123998e-05,
+      "loss": 1.4108,
+      "step": 621
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8842037285889993e-05,
+      "loss": 1.5908,
+      "step": 622
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.8838325088441137e-05,
+      "loss": 1.6427,
+      "step": 623
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.883460731711852e-05,
+      "loss": 1.4362,
+      "step": 624
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8830883974266752e-05,
+      "loss": 1.4124,
+      "step": 625
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8827155062233945e-05,
+      "loss": 0.7645,
+      "step": 626
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8823420583371738e-05,
+      "loss": 1.5692,
+      "step": 627
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8819680540035272e-05,
+      "loss": 1.3647,
+      "step": 628
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8815934934583198e-05,
+      "loss": 1.6264,
+      "step": 629
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8812183769377674e-05,
+      "loss": 1.45,
+      "step": 630
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8808427046784365e-05,
+      "loss": 1.2184,
+      "step": 631
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8804664769172443e-05,
+      "loss": 1.2716,
+      "step": 632
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8800896938914583e-05,
+      "loss": 1.218,
+      "step": 633
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8797123558386954e-05,
+      "loss": 1.3116,
+      "step": 634
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.879334462996924e-05,
+      "loss": 1.3002,
+      "step": 635
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8789560156044612e-05,
+      "loss": 1.1891,
+      "step": 636
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8785770138999737e-05,
+      "loss": 1.2831,
+      "step": 637
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8781974581224792e-05,
+      "loss": 0.9363,
+      "step": 638
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.877817348511343e-05,
+      "loss": 0.6865,
+      "step": 639
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8774366853062807e-05,
+      "loss": 1.1503,
+      "step": 640
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8770554687473574e-05,
+      "loss": 1.2252,
+      "step": 641
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.876673699074986e-05,
+      "loss": 1.3985,
+      "step": 642
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8762913765299294e-05,
+      "loss": 1.3642,
+      "step": 643
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.875908501353298e-05,
+      "loss": 1.0494,
+      "step": 644
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8755250737865518e-05,
+      "loss": 1.2987,
+      "step": 645
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8751410940714986e-05,
+      "loss": 1.0606,
+      "step": 646
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8747565624502946e-05,
+      "loss": 1.7578,
+      "step": 647
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8743714791654438e-05,
+      "loss": 1.3986,
+      "step": 648
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8739858444597984e-05,
+      "loss": 1.5013,
+      "step": 649
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8735996585765578e-05,
+      "loss": 1.5823,
+      "step": 650
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.87321292175927e-05,
+      "loss": 1.3707,
+      "step": 651
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8728256342518302e-05,
+      "loss": 1.6083,
+      "step": 652
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8724377962984795e-05,
+      "loss": 1.6429,
+      "step": 653
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.872049408143808e-05,
+      "loss": 1.289,
+      "step": 654
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8716604700327516e-05,
+      "loss": 1.3554,
+      "step": 655
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8712709822105936e-05,
+      "loss": 1.1375,
+      "step": 656
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8708809449229633e-05,
+      "loss": 1.4046,
+      "step": 657
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.870490358415838e-05,
+      "loss": 0.9515,
+      "step": 658
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8700992229355398e-05,
+      "loss": 1.4147,
+      "step": 659
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8697075387287373e-05,
+      "loss": 2.114,
+      "step": 660
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.869315306042446e-05,
+      "loss": 1.3317,
+      "step": 661
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8689225251240268e-05,
+      "loss": 1.4219,
+      "step": 662
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.8685291962211855e-05,
+      "loss": 1.7639,
+      "step": 663
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.868135319581975e-05,
+      "loss": 1.3751,
+      "step": 664
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8677408954547927e-05,
+      "loss": 1.5809,
+      "step": 665
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.867345924088381e-05,
+      "loss": 1.5636,
+      "step": 666
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8669504057318285e-05,
+      "loss": 1.2585,
+      "step": 667
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.866554340634568e-05,
+      "loss": 1.4658,
+      "step": 668
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.866157729046377e-05,
+      "loss": 1.5769,
+      "step": 669
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8657605712173782e-05,
+      "loss": 1.6577,
+      "step": 670
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8653628673980383e-05,
+      "loss": 1.1018,
+      "step": 671
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.864964617839168e-05,
+      "loss": 1.4336,
+      "step": 672
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8645658227919234e-05,
+      "loss": 1.7202,
+      "step": 673
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.864166482507804e-05,
+      "loss": 2.1243,
+      "step": 674
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8637665972386522e-05,
+      "loss": 1.5908,
+      "step": 675
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.863366167236656e-05,
+      "loss": 1.2659,
+      "step": 676
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8629651927543447e-05,
+      "loss": 1.3677,
+      "step": 677
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8625636740445925e-05,
+      "loss": 1.3144,
+      "step": 678
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.862161611360617e-05,
+      "loss": 1.3658,
+      "step": 679
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8617590049559778e-05,
+      "loss": 1.7923,
+      "step": 680
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.861355855084578e-05,
+      "loss": 1.3498,
+      "step": 681
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8609521620006636e-05,
+      "loss": 1.2324,
+      "step": 682
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.8605479259588225e-05,
+      "loss": 1.4791,
+      "step": 683
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.860143147213986e-05,
+      "loss": 1.743,
+      "step": 684
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.859737826021426e-05,
+      "loss": 2.0427,
+      "step": 685
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.859331962636759e-05,
+      "loss": 1.2335,
+      "step": 686
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8589255573159408e-05,
+      "loss": 1.1923,
+      "step": 687
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8585186103152707e-05,
+      "loss": 1.3751,
+      "step": 688
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.858111121891389e-05,
+      "loss": 1.4027,
+      "step": 689
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.857703092301278e-05,
+      "loss": 1.6319,
+      "step": 690
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.85729452180226e-05,
+      "loss": 1.4707,
+      "step": 691
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.856885410652e-05,
+      "loss": 1.17,
+      "step": 692
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8564757591085027e-05,
+      "loss": 1.6464,
+      "step": 693
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8560655674301144e-05,
+      "loss": 1.4498,
+      "step": 694
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8556548358755217e-05,
+      "loss": 1.668,
+      "step": 695
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8552435647037512e-05,
+      "loss": 1.6958,
+      "step": 696
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.854831754174171e-05,
+      "loss": 1.049,
+      "step": 697
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8544194045464888e-05,
+      "loss": 1.6657,
+      "step": 698
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.854006516080751e-05,
+      "loss": 1.2644,
+      "step": 699
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8535930890373467e-05,
+      "loss": 0.9207,
+      "step": 700
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8531791236770015e-05,
+      "loss": 0.6102,
+      "step": 701
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8527646202607822e-05,
+      "loss": 1.5107,
+      "step": 702
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8523495790500948e-05,
+      "loss": 1.7761,
+      "step": 703
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8519340003066846e-05,
+      "loss": 1.4449,
+      "step": 704
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.851517884292635e-05,
+      "loss": 1.2168,
+      "step": 705
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8511012312703692e-05,
+      "loss": 1.315,
+      "step": 706
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.850684041502649e-05,
+      "loss": 0.9184,
+      "step": 707
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.850266315252573e-05,
+      "loss": 1.4521,
+      "step": 708
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8498480527835812e-05,
+      "loss": 1.3578,
+      "step": 709
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8494292543594488e-05,
+      "loss": 1.4366,
+      "step": 710
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8490099202442912e-05,
+      "loss": 1.5684,
+      "step": 711
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8485900507025603e-05,
+      "loss": 1.5953,
+      "step": 712
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.848169645999046e-05,
+      "loss": 1.4611,
+      "step": 713
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8477487063988757e-05,
+      "loss": 1.0892,
+      "step": 714
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8473272321675143e-05,
+      "loss": 1.2696,
+      "step": 715
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8469052235707637e-05,
+      "loss": 1.297,
+      "step": 716
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.846482680874763e-05,
+      "loss": 1.5286,
+      "step": 717
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.846059604345988e-05,
+      "loss": 1.111,
+      "step": 718
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8456359942512513e-05,
+      "loss": 1.7865,
+      "step": 719
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.845211850857701e-05,
+      "loss": 1.3706,
+      "step": 720
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8447871744328238e-05,
+      "loss": 1.1235,
+      "step": 721
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8443619652444396e-05,
+      "loss": 1.4602,
+      "step": 722
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.8439362235607074e-05,
+      "loss": 1.1644,
+      "step": 723
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.843509949650119e-05,
+      "loss": 1.2309,
+      "step": 724
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8430831437815035e-05,
+      "loss": 1.5331,
+      "step": 725
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8426558062240258e-05,
+      "loss": 1.4842,
+      "step": 726
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8422279372471853e-05,
+      "loss": 1.0548,
+      "step": 727
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.841799537120817e-05,
+      "loss": 1.2995,
+      "step": 728
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8413706061150903e-05,
+      "loss": 1.2846,
+      "step": 729
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.84094114450051e-05,
+      "loss": 1.3393,
+      "step": 730
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8405111525479153e-05,
+      "loss": 0.6297,
+      "step": 731
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8400806305284797e-05,
+      "loss": 1.6738,
+      "step": 732
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8396495787137115e-05,
+      "loss": 1.0798,
+      "step": 733
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8392179973754523e-05,
+      "loss": 1.5205,
+      "step": 734
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.838785886785878e-05,
+      "loss": 1.2577,
+      "step": 735
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8383532472174988e-05,
+      "loss": 1.739,
+      "step": 736
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8379200789431578e-05,
+      "loss": 1.3997,
+      "step": 737
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.837486382236032e-05,
+      "loss": 1.417,
+      "step": 738
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8370521573696317e-05,
+      "loss": 1.3401,
+      "step": 739
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8366174046177993e-05,
+      "loss": 1.2875,
+      "step": 740
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.836182124254711e-05,
+      "loss": 1.1579,
+      "step": 741
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.835746316554876e-05,
+      "loss": 2.2142,
+      "step": 742
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.8353099817931354e-05,
+      "loss": 1.1692,
+      "step": 743
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8348731202446628e-05,
+      "loss": 1.2082,
+      "step": 744
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 1.4074231386184692,
+      "eval_runtime": 99.6576,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 744
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8344357321849642e-05,
+      "loss": 1.6486,
+      "step": 745
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.833997817889878e-05,
+      "loss": 2.2842,
+      "step": 746
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8335593776355737e-05,
+      "loss": 1.2669,
+      "step": 747
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.833120411698553e-05,
+      "loss": 1.1171,
+      "step": 748
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8326809203556485e-05,
+      "loss": 0.8204,
+      "step": 749
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8322409038840253e-05,
+      "loss": 1.5321,
+      "step": 750
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8318003625611784e-05,
+      "loss": 1.271,
+      "step": 751
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8313592966649346e-05,
+      "loss": 1.5388,
+      "step": 752
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8309177064734515e-05,
+      "loss": 0.7276,
+      "step": 753
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.830475592265217e-05,
+      "loss": 1.5962,
+      "step": 754
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8300329543190497e-05,
+      "loss": 1.4854,
+      "step": 755
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8295897929140988e-05,
+      "loss": 0.9288,
+      "step": 756
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8291461083298422e-05,
+      "loss": 1.1828,
+      "step": 757
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.82870190084609e-05,
+      "loss": 1.5382,
+      "step": 758
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8282571707429796e-05,
+      "loss": 1.355,
+      "step": 759
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8278119183009802e-05,
+      "loss": 1.3523,
+      "step": 760
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8273661438008893e-05,
+      "loss": 1.2927,
+      "step": 761
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8269198475238334e-05,
+      "loss": 1.6289,
+      "step": 762
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.8264730297512686e-05,
+      "loss": 1.2027,
+      "step": 763
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.82602569076498e-05,
+      "loss": 1.5092,
+      "step": 764
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.825577830847081e-05,
+      "loss": 1.3273,
+      "step": 765
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8251294502800136e-05,
+      "loss": 1.5111,
+      "step": 766
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8246805493465477e-05,
+      "loss": 1.2229,
+      "step": 767
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8242311283297826e-05,
+      "loss": 1.2548,
+      "step": 768
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8237811875131446e-05,
+      "loss": 1.5801,
+      "step": 769
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.823330727180388e-05,
+      "loss": 0.7902,
+      "step": 770
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.822879747615595e-05,
+      "loss": 0.941,
+      "step": 771
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8224282491031747e-05,
+      "loss": 0.8148,
+      "step": 772
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8219762319278638e-05,
+      "loss": 1.2423,
+      "step": 773
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8215236963747264e-05,
+      "loss": 1.2113,
+      "step": 774
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8210706427291535e-05,
+      "loss": 1.3838,
+      "step": 775
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.820617071276862e-05,
+      "loss": 1.1883,
+      "step": 776
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8201629823038958e-05,
+      "loss": 1.4372,
+      "step": 777
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8197083760966257e-05,
+      "loss": 1.1854,
+      "step": 778
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8192532529417486e-05,
+      "loss": 1.5133,
+      "step": 779
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8187976131262863e-05,
+      "loss": 1.5026,
+      "step": 780
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.818341456937588e-05,
+      "loss": 1.6061,
+      "step": 781
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.817884784663327e-05,
+      "loss": 1.2502,
+      "step": 782
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.817427596591504e-05,
+      "loss": 1.3701,
+      "step": 783
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.816969893010442e-05,
+      "loss": 1.3572,
+      "step": 784
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8165116742087927e-05,
+      "loss": 0.8913,
+      "step": 785
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.81605294047553e-05,
+      "loss": 1.351,
+      "step": 786
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8155936920999542e-05,
+      "loss": 1.6656,
+      "step": 787
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8151339293716884e-05,
+      "loss": 1.1011,
+      "step": 788
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8146736525806824e-05,
+      "loss": 1.2324,
+      "step": 789
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.814212862017208e-05,
+      "loss": 1.8689,
+      "step": 790
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.813751557971862e-05,
+      "loss": 1.0252,
+      "step": 791
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8132897407355657e-05,
+      "loss": 1.349,
+      "step": 792
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8128274105995626e-05,
+      "loss": 0.9465,
+      "step": 793
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.81236456785542e-05,
+      "loss": 1.0772,
+      "step": 794
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8119012127950303e-05,
+      "loss": 1.7306,
+      "step": 795
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.811437345710606e-05,
+      "loss": 1.2599,
+      "step": 796
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8109729668946848e-05,
+      "loss": 1.2942,
+      "step": 797
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8105080766401268e-05,
+      "loss": 1.565,
+      "step": 798
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.810042675240113e-05,
+      "loss": 1.5088,
+      "step": 799
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8095767629881486e-05,
+      "loss": 1.4096,
+      "step": 800
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8091103401780604e-05,
+      "loss": 1.3494,
+      "step": 801
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8086434071039972e-05,
+      "loss": 1.4677,
+      "step": 802
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.8081759640604294e-05,
+      "loss": 1.3054,
+      "step": 803
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.807708011342149e-05,
+      "loss": 0.9348,
+      "step": 804
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8072395492442695e-05,
+      "loss": 1.8479,
+      "step": 805
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8067705780622254e-05,
+      "loss": 1.7246,
+      "step": 806
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8063010980917733e-05,
+      "loss": 1.3406,
+      "step": 807
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8058311096289892e-05,
+      "loss": 1.1065,
+      "step": 808
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.805360612970271e-05,
+      "loss": 1.2406,
+      "step": 809
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.804889608412336e-05,
+      "loss": 1.1292,
+      "step": 810
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8044180962522228e-05,
+      "loss": 1.3689,
+      "step": 811
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.80394607678729e-05,
+      "loss": 1.2584,
+      "step": 812
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8034735503152147e-05,
+      "loss": 1.1992,
+      "step": 813
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.803000517133996e-05,
+      "loss": 1.4389,
+      "step": 814
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.802526977541951e-05,
+      "loss": 1.4263,
+      "step": 815
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8020529318377165e-05,
+      "loss": 1.4402,
+      "step": 816
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8015783803202483e-05,
+      "loss": 1.3552,
+      "step": 817
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8011033232888224e-05,
+      "loss": 1.0657,
+      "step": 818
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.8006277610430317e-05,
+      "loss": 1.5182,
+      "step": 819
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.800151693882789e-05,
+      "loss": 1.2039,
+      "step": 820
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.7996751221083256e-05,
+      "loss": 1.3726,
+      "step": 821
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.79919804602019e-05,
+      "loss": 1.1037,
+      "step": 822
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7987204659192495e-05,
+      "loss": 1.6955,
+      "step": 823
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7982423821066893e-05,
+      "loss": 1.6935,
+      "step": 824
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7977637948840124e-05,
+      "loss": 1.4399,
+      "step": 825
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7972847045530383e-05,
+      "loss": 1.1159,
+      "step": 826
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7968051114159046e-05,
+      "loss": 1.5111,
+      "step": 827
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7963250157750665e-05,
+      "loss": 1.0608,
+      "step": 828
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7958444179332946e-05,
+      "loss": 1.3674,
+      "step": 829
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7953633181936776e-05,
+      "loss": 1.1912,
+      "step": 830
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.79488171685962e-05,
+      "loss": 1.3468,
+      "step": 831
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.794399614234843e-05,
+      "loss": 1.0985,
+      "step": 832
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7939170106233833e-05,
+      "loss": 1.3871,
+      "step": 833
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7934339063295942e-05,
+      "loss": 1.3854,
+      "step": 834
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7929503016581446e-05,
+      "loss": 1.2603,
+      "step": 835
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.792466196914019e-05,
+      "loss": 1.3454,
+      "step": 836
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7919815924025174e-05,
+      "loss": 1.3331,
+      "step": 837
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7914964884292543e-05,
+      "loss": 1.1845,
+      "step": 838
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7910108853001597e-05,
+      "loss": 1.2957,
+      "step": 839
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7905247833214785e-05,
+      "loss": 1.2642,
+      "step": 840
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.79003818279977e-05,
+      "loss": 1.3585,
+      "step": 841
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7895510840419083e-05,
+      "loss": 1.3701,
+      "step": 842
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7890634873550805e-05,
+      "loss": 1.5067,
+      "step": 843
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7885753930467894e-05,
+      "loss": 1.3115,
+      "step": 844
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.78808680142485e-05,
+      "loss": 1.3628,
+      "step": 845
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7875977127973925e-05,
+      "loss": 1.0624,
+      "step": 846
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7871081274728594e-05,
+      "loss": 1.3664,
+      "step": 847
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.786618045760007e-05,
+      "loss": 1.3321,
+      "step": 848
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.786127467967904e-05,
+      "loss": 1.418,
+      "step": 849
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7856363944059327e-05,
+      "loss": 1.2677,
+      "step": 850
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.785144825383788e-05,
+      "loss": 1.2668,
+      "step": 851
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7846527612114766e-05,
+      "loss": 1.1138,
+      "step": 852
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7841602021993177e-05,
+      "loss": 1.4431,
+      "step": 853
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7836671486579434e-05,
+      "loss": 1.113,
+      "step": 854
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7831736008982967e-05,
+      "loss": 1.2348,
+      "step": 855
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7826795592316326e-05,
+      "loss": 1.1089,
+      "step": 856
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.782185023969518e-05,
+      "loss": 1.9396,
+      "step": 857
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.78168999542383e-05,
+      "loss": 1.4311,
+      "step": 858
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.781194473906757e-05,
+      "loss": 1.3562,
+      "step": 859
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7806984597308004e-05,
+      "loss": 1.748,
+      "step": 860
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.7802019532087692e-05,
+      "loss": 1.4277,
+      "step": 861
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.779704954653785e-05,
+      "loss": 1.6625,
+      "step": 862
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7792074643792793e-05,
+      "loss": 1.6347,
+      "step": 863
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7787094826989927e-05,
+      "loss": 1.7164,
+      "step": 864
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.778211009926977e-05,
+      "loss": 1.1276,
+      "step": 865
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7777120463775927e-05,
+      "loss": 1.1386,
+      "step": 866
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7772125923655104e-05,
+      "loss": 0.7429,
+      "step": 867
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7767126482057102e-05,
+      "loss": 1.7416,
+      "step": 868
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.776212214213481e-05,
+      "loss": 1.0909,
+      "step": 869
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.77571129070442e-05,
+      "loss": 1.2001,
+      "step": 870
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.775209877994434e-05,
+      "loss": 1.1985,
+      "step": 871
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7747079763997383e-05,
+      "loss": 1.3927,
+      "step": 872
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.774205586236856e-05,
+      "loss": 1.2791,
+      "step": 873
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7737027078226187e-05,
+      "loss": 1.329,
+      "step": 874
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7731993414741656e-05,
+      "loss": 1.845,
+      "step": 875
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.772695487508944e-05,
+      "loss": 1.4102,
+      "step": 876
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7721911462447074e-05,
+      "loss": 1.004,
+      "step": 877
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7716863179995194e-05,
+      "loss": 1.7119,
+      "step": 878
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.771181003091748e-05,
+      "loss": 1.5352,
+      "step": 879
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7706752018400693e-05,
+      "loss": 1.1019,
+      "step": 880
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7701689145634664e-05,
+      "loss": 1.6929,
+      "step": 881
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.7696621415812275e-05,
+      "loss": 1.3236,
+      "step": 882
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.769154883212949e-05,
+      "loss": 1.0412,
+      "step": 883
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7686471397785322e-05,
+      "loss": 1.3958,
+      "step": 884
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7681389115981843e-05,
+      "loss": 1.8158,
+      "step": 885
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.767630198992419e-05,
+      "loss": 0.9901,
+      "step": 886
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7671210022820543e-05,
+      "loss": 1.5822,
+      "step": 887
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7666113217882145e-05,
+      "loss": 1.0834,
+      "step": 888
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.766101157832329e-05,
+      "loss": 1.2856,
+      "step": 889
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7655905107361313e-05,
+      "loss": 1.3248,
+      "step": 890
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7650793808216603e-05,
+      "loss": 1.2248,
+      "step": 891
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.764567768411259e-05,
+      "loss": 1.1503,
+      "step": 892
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7640556738275755e-05,
+      "loss": 1.1089,
+      "step": 893
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7635430973935606e-05,
+      "loss": 1.0886,
+      "step": 894
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.76303003943247e-05,
+      "loss": 1.2826,
+      "step": 895
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7625165002678623e-05,
+      "loss": 1.5564,
+      "step": 896
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7620024802236004e-05,
+      "loss": 1.4393,
+      "step": 897
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7614879796238505e-05,
+      "loss": 1.3972,
+      "step": 898
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.760972998793081e-05,
+      "loss": 1.121,
+      "step": 899
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7604575380560637e-05,
+      "loss": 1.1306,
+      "step": 900
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.759941597737873e-05,
+      "loss": 1.4002,
+      "step": 901
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7594251781638857e-05,
+      "loss": 1.1716,
+      "step": 902
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7589082796597812e-05,
+      "loss": 1.4862,
+      "step": 903
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.75839090255154e-05,
+      "loss": 1.1721,
+      "step": 904
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.757873047165446e-05,
+      "loss": 1.4819,
+      "step": 905
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7573547138280828e-05,
+      "loss": 0.9485,
+      "step": 906
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7568359028663365e-05,
+      "loss": 1.1956,
+      "step": 907
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7563166146073947e-05,
+      "loss": 0.8966,
+      "step": 908
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7557968493787455e-05,
+      "loss": 1.3303,
+      "step": 909
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7552766075081776e-05,
+      "loss": 1.485,
+      "step": 910
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7547558893237808e-05,
+      "loss": 1.3002,
+      "step": 911
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7542346951539454e-05,
+      "loss": 1.4948,
+      "step": 912
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7537130253273613e-05,
+      "loss": 1.7421,
+      "step": 913
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7531908801730185e-05,
+      "loss": 1.2823,
+      "step": 914
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7526682600202068e-05,
+      "loss": 1.3982,
+      "step": 915
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.752145165198516e-05,
+      "loss": 1.8014,
+      "step": 916
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7516215960378347e-05,
+      "loss": 1.5365,
+      "step": 917
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7510975528683512e-05,
+      "loss": 1.3348,
+      "step": 918
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.750573036020552e-05,
+      "loss": 1.2711,
+      "step": 919
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7500480458252226e-05,
+      "loss": 1.1657,
+      "step": 920
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7495225826134476e-05,
+      "loss": 1.3426,
+      "step": 921
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.7489966467166088e-05,
+      "loss": 1.9029,
+      "step": 922
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7484702384663873e-05,
+      "loss": 1.0706,
+      "step": 923
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7479433581947614e-05,
+      "loss": 1.5385,
+      "step": 924
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7474160062340066e-05,
+      "loss": 1.3825,
+      "step": 925
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7468881829166972e-05,
+      "loss": 1.3206,
+      "step": 926
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7463598885757038e-05,
+      "loss": 0.9489,
+      "step": 927
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7458311235441944e-05,
+      "loss": 1.2469,
+      "step": 928
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7453018881556333e-05,
+      "loss": 1.5614,
+      "step": 929
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.744772182743782e-05,
+      "loss": 1.6497,
+      "step": 930
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7442420076426986e-05,
+      "loss": 1.1453,
+      "step": 931
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7437113631867366e-05,
+      "loss": 1.1558,
+      "step": 932
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.743180249710546e-05,
+      "loss": 1.0424,
+      "step": 933
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7426486675490732e-05,
+      "loss": 1.4262,
+      "step": 934
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7421166170375587e-05,
+      "loss": 1.3042,
+      "step": 935
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7415840985115394e-05,
+      "loss": 1.3541,
+      "step": 936
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.741051112306848e-05,
+      "loss": 1.2677,
+      "step": 937
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.74051765875961e-05,
+      "loss": 1.1085,
+      "step": 938
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.7399837382062478e-05,
+      "loss": 1.4783,
+      "step": 939
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.739449350983477e-05,
+      "loss": 1.7204,
+      "step": 940
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.738914497428308e-05,
+      "loss": 1.9833,
+      "step": 941
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7383791778780455e-05,
+      "loss": 1.4038,
+      "step": 942
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7378433926702877e-05,
+      "loss": 1.4205,
+      "step": 943
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.737307142142926e-05,
+      "loss": 1.2557,
+      "step": 944
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7367704266341468e-05,
+      "loss": 1.4119,
+      "step": 945
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7362332464824284e-05,
+      "loss": 1.1406,
+      "step": 946
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.735695602026542e-05,
+      "loss": 1.6188,
+      "step": 947
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7351574936055528e-05,
+      "loss": 1.2576,
+      "step": 948
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.734618921558817e-05,
+      "loss": 0.953,
+      "step": 949
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.734079886225985e-05,
+      "loss": 1.1006,
+      "step": 950
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7335403879469984e-05,
+      "loss": 0.7349,
+      "step": 951
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7330004270620904e-05,
+      "loss": 1.2366,
+      "step": 952
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7324600039117862e-05,
+      "loss": 1.3368,
+      "step": 953
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7319191188369032e-05,
+      "loss": 1.1208,
+      "step": 954
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7313777721785496e-05,
+      "loss": 1.1123,
+      "step": 955
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.730835964278124e-05,
+      "loss": 1.2072,
+      "step": 956
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.730293695477318e-05,
+      "loss": 1.5776,
+      "step": 957
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7297509661181104e-05,
+      "loss": 1.2845,
+      "step": 958
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.729207776542774e-05,
+      "loss": 1.4605,
+      "step": 959
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.72866412709387e-05,
+      "loss": 0.9502,
+      "step": 960
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.72812001811425e-05,
+      "loss": 0.9166,
+      "step": 961
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7275754499470547e-05,
+      "loss": 1.2824,
+      "step": 962
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.727030422935716e-05,
+      "loss": 1.1884,
+      "step": 963
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7264849374239538e-05,
+      "loss": 1.3889,
+      "step": 964
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.725938993755777e-05,
+      "loss": 1.316,
+      "step": 965
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7253925922754846e-05,
+      "loss": 1.1767,
+      "step": 966
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7248457333276638e-05,
+      "loss": 1.2903,
+      "step": 967
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.72429841725719e-05,
+      "loss": 1.4666,
+      "step": 968
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7237506444092268e-05,
+      "loss": 1.2527,
+      "step": 969
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7232024151292267e-05,
+      "loss": 1.4831,
+      "step": 970
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.722653729762929e-05,
+      "loss": 1.385,
+      "step": 971
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7221045886563615e-05,
+      "loss": 1.3572,
+      "step": 972
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.721554992155839e-05,
+      "loss": 1.6375,
+      "step": 973
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7210049406079633e-05,
+      "loss": 1.4267,
+      "step": 974
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.720454434359624e-05,
+      "loss": 1.7291,
+      "step": 975
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7199034737579962e-05,
+      "loss": 1.4631,
+      "step": 976
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.719352059150542e-05,
+      "loss": 1.1362,
+      "step": 977
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.718800190885011e-05,
+      "loss": 1.3162,
+      "step": 978
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7182478693094372e-05,
+      "loss": 1.7466,
+      "step": 979
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.717695094772142e-05,
+      "loss": 1.314,
+      "step": 980
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.7171418676217304e-05,
+      "loss": 1.6715,
+      "step": 981
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7165881882070952e-05,
+      "loss": 1.0099,
+      "step": 982
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.716034056877412e-05,
+      "loss": 1.0538,
+      "step": 983
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7154794739821447e-05,
+      "loss": 1.3831,
+      "step": 984
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7149244398710387e-05,
+      "loss": 0.8075,
+      "step": 985
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7143689548941253e-05,
+      "loss": 1.3607,
+      "step": 986
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7138130194017203e-05,
+      "loss": 1.6853,
+      "step": 987
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7132566337444238e-05,
+      "loss": 1.8848,
+      "step": 988
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7126997982731185e-05,
+      "loss": 1.3898,
+      "step": 989
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7121425133389725e-05,
+      "loss": 1.462,
+      "step": 990
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.711584779293436e-05,
+      "loss": 1.3396,
+      "step": 991
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7110265964882432e-05,
+      "loss": 0.9142,
+      "step": 992
+    },
+    {
+      "epoch": 0.5,
+      "eval_loss": 1.3848216533660889,
+      "eval_runtime": 99.6301,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 992
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.710467965275411e-05,
+      "loss": 1.3565,
+      "step": 993
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7099088860072397e-05,
+      "loss": 1.2052,
+      "step": 994
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7093493590363112e-05,
+      "loss": 1.2579,
+      "step": 995
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7087893847154897e-05,
+      "loss": 1.5135,
+      "step": 996
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7082289633979225e-05,
+      "loss": 1.6288,
+      "step": 997
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7076680954370383e-05,
+      "loss": 1.2431,
+      "step": 998
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7071067811865477e-05,
+      "loss": 2.0285,
+      "step": 999
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7065450210004422e-05,
+      "loss": 1.4954,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.7059828152329943e-05,
+      "loss": 1.3016,
+      "step": 1001
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7054201642387595e-05,
+      "loss": 1.2285,
+      "step": 1002
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.704857068372571e-05,
+      "loss": 1.3029,
+      "step": 1003
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7042935279895452e-05,
+      "loss": 1.5233,
+      "step": 1004
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7037295434450774e-05,
+      "loss": 1.4802,
+      "step": 1005
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.703165115094844e-05,
+      "loss": 1.1672,
+      "step": 1006
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7026002432948004e-05,
+      "loss": 1.2408,
+      "step": 1007
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7020349284011815e-05,
+      "loss": 0.5283,
+      "step": 1008
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7014691707705027e-05,
+      "loss": 1.3596,
+      "step": 1009
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7009029707595578e-05,
+      "loss": 0.9959,
+      "step": 1010
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.70033632872542e-05,
+      "loss": 1.5955,
+      "step": 1011
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6997692450254407e-05,
+      "loss": 1.0983,
+      "step": 1012
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.699201720017251e-05,
+      "loss": 1.4687,
+      "step": 1013
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6986337540587586e-05,
+      "loss": 1.2817,
+      "step": 1014
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6980653475081505e-05,
+      "loss": 1.2472,
+      "step": 1015
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6974965007238917e-05,
+      "loss": 1.5873,
+      "step": 1016
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6969272140647236e-05,
+      "loss": 1.0519,
+      "step": 1017
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6963574878896663e-05,
+      "loss": 0.9721,
+      "step": 1018
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.695787322558017e-05,
+      "loss": 1.6194,
+      "step": 1019
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.6952167184293485e-05,
+      "loss": 1.1158,
+      "step": 1020
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.694645675863512e-05,
+      "loss": 0.9189,
+      "step": 1021
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6940741952206342e-05,
+      "loss": 1.3312,
+      "step": 1022
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.693502276861118e-05,
+      "loss": 1.0799,
+      "step": 1023
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6929299211456434e-05,
+      "loss": 1.6904,
+      "step": 1024
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.692357128435165e-05,
+      "loss": 1.5506,
+      "step": 1025
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6917838990909133e-05,
+      "loss": 1.2288,
+      "step": 1026
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.691210233474395e-05,
+      "loss": 1.3026,
+      "step": 1027
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.690636131947391e-05,
+      "loss": 1.4344,
+      "step": 1028
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.690061594871957e-05,
+      "loss": 1.4413,
+      "step": 1029
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6894866226104248e-05,
+      "loss": 1.989,
+      "step": 1030
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.688911215525398e-05,
+      "loss": 1.6863,
+      "step": 1031
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.688335373979758e-05,
+      "loss": 1.6185,
+      "step": 1032
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.687759098336656e-05,
+      "loss": 1.1258,
+      "step": 1033
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.687182388959521e-05,
+      "loss": 1.1172,
+      "step": 1034
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6866052462120532e-05,
+      "loss": 1.4315,
+      "step": 1035
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6860276704582266e-05,
+      "loss": 1.0994,
+      "step": 1036
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.685449662062288e-05,
+      "loss": 0.7539,
+      "step": 1037
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.684871221388758e-05,
+      "loss": 1.4692,
+      "step": 1038
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.684292348802428e-05,
+      "loss": 1.2738,
+      "step": 1039
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6837130446683642e-05,
+      "loss": 1.0591,
+      "step": 1040
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.6831333093519037e-05,
+      "loss": 1.1646,
+      "step": 1041
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6825531432186545e-05,
+      "loss": 1.0403,
+      "step": 1042
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.681972546634498e-05,
+      "loss": 1.1038,
+      "step": 1043
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.681391519965586e-05,
+      "loss": 0.8883,
+      "step": 1044
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.680810063578342e-05,
+      "loss": 0.8832,
+      "step": 1045
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.680228177839461e-05,
+      "loss": 1.5098,
+      "step": 1046
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.679645863115908e-05,
+      "loss": 1.2526,
+      "step": 1047
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6790631197749182e-05,
+      "loss": 1.4754,
+      "step": 1048
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6784799481839984e-05,
+      "loss": 1.2071,
+      "step": 1049
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.677896348710924e-05,
+      "loss": 1.3081,
+      "step": 1050
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6773123217237412e-05,
+      "loss": 1.1485,
+      "step": 1051
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6767278675907654e-05,
+      "loss": 1.7123,
+      "step": 1052
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6761429866805818e-05,
+      "loss": 2.1722,
+      "step": 1053
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6755576793620446e-05,
+      "loss": 0.5983,
+      "step": 1054
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.674971946004277e-05,
+      "loss": 1.4454,
+      "step": 1055
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6743857869766698e-05,
+      "loss": 1.4261,
+      "step": 1056
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.6737992026488834e-05,
+      "loss": 1.8654,
+      "step": 1057
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.673212193390847e-05,
+      "loss": 1.4037,
+      "step": 1058
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.672624759572756e-05,
+      "loss": 1.3131,
+      "step": 1059
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.672036901565075e-05,
+      "loss": 1.1671,
+      "step": 1060
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.671448619738535e-05,
+      "loss": 1.1584,
+      "step": 1061
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6708599144641357e-05,
+      "loss": 1.2702,
+      "step": 1062
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6702707861131422e-05,
+      "loss": 1.3125,
+      "step": 1063
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.669681235057088e-05,
+      "loss": 1.3498,
+      "step": 1064
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.669091261667772e-05,
+      "loss": 1.3509,
+      "step": 1065
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6685008663172597e-05,
+      "loss": 1.1434,
+      "step": 1066
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.667910049377883e-05,
+      "loss": 1.3289,
+      "step": 1067
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6673188112222394e-05,
+      "loss": 1.2959,
+      "step": 1068
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6667271522231926e-05,
+      "loss": 1.2411,
+      "step": 1069
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6661350727538712e-05,
+      "loss": 1.0544,
+      "step": 1070
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6655425731876685e-05,
+      "loss": 1.1301,
+      "step": 1071
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6649496538982443e-05,
+      "loss": 1.1511,
+      "step": 1072
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6643563152595204e-05,
+      "loss": 1.232,
+      "step": 1073
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6637625576456868e-05,
+      "loss": 1.193,
+      "step": 1074
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.663168381431194e-05,
+      "loss": 1.135,
+      "step": 1075
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.662573786990759e-05,
+      "loss": 1.1136,
+      "step": 1076
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6619787746993617e-05,
+      "loss": 0.9565,
+      "step": 1077
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.661383344932245e-05,
+      "loss": 1.489,
+      "step": 1078
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.6607874980649162e-05,
+      "loss": 1.3864,
+      "step": 1079
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.660191234473145e-05,
+      "loss": 1.4215,
+      "step": 1080
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.659594554532963e-05,
+      "loss": 1.6097,
+      "step": 1081
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.658997458620667e-05,
+      "loss": 1.2021,
+      "step": 1082
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6583999471128128e-05,
+      "loss": 1.696,
+      "step": 1083
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.657802020386221e-05,
+      "loss": 0.3453,
+      "step": 1084
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6572036788179728e-05,
+      "loss": 1.247,
+      "step": 1085
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.656604922785411e-05,
+      "loss": 1.5292,
+      "step": 1086
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.65600575266614e-05,
+      "loss": 1.2324,
+      "step": 1087
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6554061688380256e-05,
+      "loss": 0.6247,
+      "step": 1088
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6548061716791943e-05,
+      "loss": 1.1605,
+      "step": 1089
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.654205761568033e-05,
+      "loss": 1.2244,
+      "step": 1090
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6536049388831897e-05,
+      "loss": 1.2709,
+      "step": 1091
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.653003704003571e-05,
+      "loss": 0.67,
+      "step": 1092
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.652402057308346e-05,
+      "loss": 1.3283,
+      "step": 1093
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.651799999176941e-05,
+      "loss": 1.1593,
+      "step": 1094
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6511975299890438e-05,
+      "loss": 1.5307,
+      "step": 1095
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6505946501246e-05,
+      "loss": 1.0621,
+      "step": 1096
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6499913599638147e-05,
+      "loss": 1.8225,
+      "step": 1097
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6493876598871517e-05,
+      "loss": 1.1734,
+      "step": 1098
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6487835502753333e-05,
+      "loss": 0.8684,
+      "step": 1099
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.6481790315093405e-05,
+      "loss": 1.0447,
+      "step": 1100
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6475741039704118e-05,
+      "loss": 1.2504,
+      "step": 1101
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6469687680400435e-05,
+      "loss": 1.5288,
+      "step": 1102
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.64636302409999e-05,
+      "loss": 1.5258,
+      "step": 1103
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6457568725322623e-05,
+      "loss": 0.825,
+      "step": 1104
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6451503137191285e-05,
+      "loss": 0.9596,
+      "step": 1105
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6445433480431145e-05,
+      "loss": 1.2489,
+      "step": 1106
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6439359758870016e-05,
+      "loss": 1.4619,
+      "step": 1107
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6433281976338285e-05,
+      "loss": 1.7392,
+      "step": 1108
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6427200136668886e-05,
+      "loss": 1.2478,
+      "step": 1109
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6421114243697328e-05,
+      "loss": 1.1571,
+      "step": 1110
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6415024301261664e-05,
+      "loss": 1.4271,
+      "step": 1111
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6408930313202508e-05,
+      "loss": 1.6418,
+      "step": 1112
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.640283228336302e-05,
+      "loss": 0.8528,
+      "step": 1113
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6396730215588913e-05,
+      "loss": 0.4624,
+      "step": 1114
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6390624113728448e-05,
+      "loss": 1.2621,
+      "step": 1115
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.638451398163242e-05,
+      "loss": 1.369,
+      "step": 1116
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.637839982315418e-05,
+      "loss": 1.0964,
+      "step": 1117
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6372281642149603e-05,
+      "loss": 1.2814,
+      "step": 1118
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.636615944247712e-05,
+      "loss": 1.1635,
+      "step": 1119
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.6360033227997674e-05,
+      "loss": 1.1724,
+      "step": 1120
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6353903002574758e-05,
+      "loss": 1.484,
+      "step": 1121
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6347768770074385e-05,
+      "loss": 1.2226,
+      "step": 1122
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6341630534365095e-05,
+      "loss": 1.3481,
+      "step": 1123
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.633548829931796e-05,
+      "loss": 1.006,
+      "step": 1124
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.632934206880657e-05,
+      "loss": 1.5531,
+      "step": 1125
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.632319184670703e-05,
+      "loss": 1.0621,
+      "step": 1126
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6317037636897972e-05,
+      "loss": 1.3661,
+      "step": 1127
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.631087944326053e-05,
+      "loss": 1.0466,
+      "step": 1128
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6304717269678363e-05,
+      "loss": 1.3364,
+      "step": 1129
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6298551120037633e-05,
+      "loss": 1.412,
+      "step": 1130
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.629238099822701e-05,
+      "loss": 1.3741,
+      "step": 1131
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.628620690813767e-05,
+      "loss": 1.3023,
+      "step": 1132
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6280028853663297e-05,
+      "loss": 1.4314,
+      "step": 1133
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6273846838700063e-05,
+      "loss": 0.7864,
+      "step": 1134
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.626766086714665e-05,
+      "loss": 1.6701,
+      "step": 1135
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6261470942904224e-05,
+      "loss": 1.5097,
+      "step": 1136
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6255277069876454e-05,
+      "loss": 0.8726,
+      "step": 1137
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.62490792519695e-05,
+      "loss": 1.3155,
+      "step": 1138
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.624287749309199e-05,
+      "loss": 1.0042,
+      "step": 1139
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.6236671797155064e-05,
+      "loss": 1.6354,
+      "step": 1140
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.623046216807233e-05,
+      "loss": 1.744,
+      "step": 1141
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6224248609759877e-05,
+      "loss": 0.9126,
+      "step": 1142
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.621803112613628e-05,
+      "loss": 0.7313,
+      "step": 1143
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6211809721122575e-05,
+      "loss": 1.2979,
+      "step": 1144
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6205584398642286e-05,
+      "loss": 1.2585,
+      "step": 1145
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.61993551626214e-05,
+      "loss": 1.6595,
+      "step": 1146
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6193122016988377e-05,
+      "loss": 1.5178,
+      "step": 1147
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6186884965674134e-05,
+      "loss": 1.4424,
+      "step": 1148
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6180644012612056e-05,
+      "loss": 1.3139,
+      "step": 1149
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6174399161737993e-05,
+      "loss": 1.4371,
+      "step": 1150
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6168150416990246e-05,
+      "loss": 1.1936,
+      "step": 1151
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6161897782309574e-05,
+      "loss": 1.4718,
+      "step": 1152
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.615564126163919e-05,
+      "loss": 1.4299,
+      "step": 1153
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.614938085892476e-05,
+      "loss": 1.3404,
+      "step": 1154
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6143116578114395e-05,
+      "loss": 1.8379,
+      "step": 1155
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6136848423158652e-05,
+      "loss": 1.356,
+      "step": 1156
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6130576398010528e-05,
+      "loss": 1.3782,
+      "step": 1157
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.612430050662547e-05,
+      "loss": 1.3236,
+      "step": 1158
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.611802075296135e-05,
+      "loss": 1.5099,
+      "step": 1159
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.6111737140978495e-05,
+      "loss": 0.8791,
+      "step": 1160
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6105449674639642e-05,
+      "loss": 1.6677,
+      "step": 1161
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.609915835790998e-05,
+      "loss": 1.4567,
+      "step": 1162
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.609286319475711e-05,
+      "loss": 1.0328,
+      "step": 1163
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.608656418915106e-05,
+      "loss": 1.1495,
+      "step": 1164
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6080261345064298e-05,
+      "loss": 1.438,
+      "step": 1165
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.60739546664717e-05,
+      "loss": 0.9594,
+      "step": 1166
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6067644157350555e-05,
+      "loss": 1.3485,
+      "step": 1167
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6061329821680578e-05,
+      "loss": 1.5222,
+      "step": 1168
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6055011663443894e-05,
+      "loss": 0.5173,
+      "step": 1169
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6048689686625033e-05,
+      "loss": 1.4507,
+      "step": 1170
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6042363895210948e-05,
+      "loss": 1.4224,
+      "step": 1171
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6036034293190978e-05,
+      "loss": 1.6695,
+      "step": 1172
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.602970088455688e-05,
+      "loss": 1.1182,
+      "step": 1173
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6023363673302802e-05,
+      "loss": 1.052,
+      "step": 1174
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6017022663425304e-05,
+      "loss": 0.8028,
+      "step": 1175
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6010677858923328e-05,
+      "loss": 1.2943,
+      "step": 1176
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.6004329263798208e-05,
+      "loss": 1.4253,
+      "step": 1177
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.599797688205368e-05,
+      "loss": 1.0901,
+      "step": 1178
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.599162071769586e-05,
+      "loss": 1.5041,
+      "step": 1179
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5985260774733252e-05,
+      "loss": 0.9272,
+      "step": 1180
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5978897057176746e-05,
+      "loss": 0.8361,
+      "step": 1181
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5972529569039603e-05,
+      "loss": 4.7241,
+      "step": 1182
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5966158314337472e-05,
+      "loss": 1.0232,
+      "step": 1183
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.595978329708837e-05,
+      "loss": 0.9797,
+      "step": 1184
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5953404521312695e-05,
+      "loss": 1.0088,
+      "step": 1185
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5947021991033206e-05,
+      "loss": 1.5512,
+      "step": 1186
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5940635710275033e-05,
+      "loss": 1.1678,
+      "step": 1187
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5934245683065682e-05,
+      "loss": 1.3641,
+      "step": 1188
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5927851913435e-05,
+      "loss": 1.7091,
+      "step": 1189
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.592145440541521e-05,
+      "loss": 1.1663,
+      "step": 1190
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5915053163040894e-05,
+      "loss": 1.668,
+      "step": 1191
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.590864819034898e-05,
+      "loss": 1.3709,
+      "step": 1192
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5902239491378756e-05,
+      "loss": 1.4036,
+      "step": 1193
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.589582707017185e-05,
+      "loss": 0.7514,
+      "step": 1194
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5889410930772245e-05,
+      "loss": 0.9069,
+      "step": 1195
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.588299107722627e-05,
+      "loss": 1.0832,
+      "step": 1196
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5876567513582598e-05,
+      "loss": 1.4737,
+      "step": 1197
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.5870140243892233e-05,
+      "loss": 1.0943,
+      "step": 1198
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.586370927220852e-05,
+      "loss": 1.3604,
+      "step": 1199
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5857274602587137e-05,
+      "loss": 1.5542,
+      "step": 1200
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5850836239086107e-05,
+      "loss": 1.0561,
+      "step": 1201
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5844394185765764e-05,
+      "loss": 1.4651,
+      "step": 1202
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5837948446688782e-05,
+      "loss": 1.3101,
+      "step": 1203
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5831499025920153e-05,
+      "loss": 1.1301,
+      "step": 1204
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.582504592752719e-05,
+      "loss": 1.8708,
+      "step": 1205
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.581858915557953e-05,
+      "loss": 1.287,
+      "step": 1206
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5812128714149125e-05,
+      "loss": 1.2097,
+      "step": 1207
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5805664607310238e-05,
+      "loss": 0.9074,
+      "step": 1208
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5799196839139456e-05,
+      "loss": 1.4384,
+      "step": 1209
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.579272541371565e-05,
+      "loss": 1.3117,
+      "step": 1210
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5786250335120022e-05,
+      "loss": 0.4972,
+      "step": 1211
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5779771607436073e-05,
+      "loss": 1.0646,
+      "step": 1212
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5773289234749588e-05,
+      "loss": 1.5521,
+      "step": 1213
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5766803221148676e-05,
+      "loss": 1.3221,
+      "step": 1214
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5760313570723716e-05,
+      "loss": 1.2105,
+      "step": 1215
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.575382028756741e-05,
+      "loss": 0.2071,
+      "step": 1216
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.574732337577473e-05,
+      "loss": 1.3853,
+      "step": 1217
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5740822839442936e-05,
+      "loss": 1.0781,
+      "step": 1218
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.5734318682671586e-05,
+      "loss": 1.1282,
+      "step": 1219
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5727810909562508e-05,
+      "loss": 1.0346,
+      "step": 1220
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5721299524219823e-05,
+      "loss": 1.2719,
+      "step": 1221
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5714784530749927e-05,
+      "loss": 0.631,
+      "step": 1222
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.570826593326148e-05,
+      "loss": 1.5083,
+      "step": 1223
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.570174373586543e-05,
+      "loss": 1.2729,
+      "step": 1224
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.569521794267499e-05,
+      "loss": 1.5531,
+      "step": 1225
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.568868855780563e-05,
+      "loss": 1.0676,
+      "step": 1226
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.568215558537511e-05,
+      "loss": 1.0867,
+      "step": 1227
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5675619029503426e-05,
+      "loss": 1.2246,
+      "step": 1228
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5669078894312848e-05,
+      "loss": 1.6245,
+      "step": 1229
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.56625351839279e-05,
+      "loss": 1.1104,
+      "step": 1230
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5655987902475365e-05,
+      "loss": 1.5263,
+      "step": 1231
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5649437054084267e-05,
+      "loss": 0.6199,
+      "step": 1232
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5642882642885894e-05,
+      "loss": 1.3261,
+      "step": 1233
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.563632467301377e-05,
+      "loss": 1.3702,
+      "step": 1234
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5629763148603666e-05,
+      "loss": 1.5479,
+      "step": 1235
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.56231980737936e-05,
+      "loss": 1.3279,
+      "step": 1236
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5616629452723818e-05,
+      "loss": 1.7209,
+      "step": 1237
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.5610057289536814e-05,
+      "loss": 1.3794,
+      "step": 1238
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.560348158837731e-05,
+      "loss": 1.1775,
+      "step": 1239
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5596902353392258e-05,
+      "loss": 1.3188,
+      "step": 1240
+    },
+    {
+      "epoch": 0.63,
+      "eval_loss": 1.3631528615951538,
+      "eval_runtime": 99.6336,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 1240
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5590319588730844e-05,
+      "loss": 1.1441,
+      "step": 1241
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5583733298544472e-05,
+      "loss": 1.1252,
+      "step": 1242
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5577143486986778e-05,
+      "loss": 1.2104,
+      "step": 1243
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5570550158213614e-05,
+      "loss": 1.6675,
+      "step": 1244
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5563953316383052e-05,
+      "loss": 0.8392,
+      "step": 1245
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.555735296565537e-05,
+      "loss": 1.0056,
+      "step": 1246
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.555074911019308e-05,
+      "loss": 1.1701,
+      "step": 1247
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5544141754160885e-05,
+      "loss": 1.0574,
+      "step": 1248
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5537530901725702e-05,
+      "loss": 1.1025,
+      "step": 1249
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5530916557056653e-05,
+      "loss": 1.3033,
+      "step": 1250
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5524298724325064e-05,
+      "loss": 1.4389,
+      "step": 1251
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.551767740770446e-05,
+      "loss": 1.1951,
+      "step": 1252
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.551105261137056e-05,
+      "loss": 1.2612,
+      "step": 1253
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5504424339501284e-05,
+      "loss": 1.195,
+      "step": 1254
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.549779259627673e-05,
+      "loss": 1.3947,
+      "step": 1255
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5491157385879206e-05,
+      "loss": 1.0962,
+      "step": 1256
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5484518712493188e-05,
+      "loss": 1.1659,
+      "step": 1257
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.547787658030535e-05,
+      "loss": 1.6983,
+      "step": 1258
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.5471230993504535e-05,
+      "loss": 1.4831,
+      "step": 1259
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.546458195628177e-05,
+      "loss": 1.4953,
+      "step": 1260
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5457929472830263e-05,
+      "loss": 1.0964,
+      "step": 1261
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5451273547345387e-05,
+      "loss": 1.2131,
+      "step": 1262
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5444614184024694e-05,
+      "loss": 0.727,
+      "step": 1263
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.543795138706789e-05,
+      "loss": 1.263,
+      "step": 1264
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5431285160676866e-05,
+      "loss": 0.8979,
+      "step": 1265
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5424615509055664e-05,
+      "loss": 1.1657,
+      "step": 1266
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5417942436410482e-05,
+      "loss": 1.6339,
+      "step": 1267
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5411265946949686e-05,
+      "loss": 0.7919,
+      "step": 1268
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5404586044883788e-05,
+      "loss": 1.6113,
+      "step": 1269
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.539790273442546e-05,
+      "loss": 1.0389,
+      "step": 1270
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5391216019789522e-05,
+      "loss": 1.1089,
+      "step": 1271
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.538452590519293e-05,
+      "loss": 1.1045,
+      "step": 1272
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.53778323948548e-05,
+      "loss": 1.4433,
+      "step": 1273
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5371135492996382e-05,
+      "loss": 1.0384,
+      "step": 1274
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5364435203841058e-05,
+      "loss": 1.2829,
+      "step": 1275
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5357731531614363e-05,
+      "loss": 1.6972,
+      "step": 1276
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5351024480543948e-05,
+      "loss": 1.0734,
+      "step": 1277
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5344314054859606e-05,
+      "loss": 1.5467,
+      "step": 1278
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.5337600258793255e-05,
+      "loss": 1.0984,
+      "step": 1279
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5330883096578937e-05,
+      "loss": 1.3236,
+      "step": 1280
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.532416257245282e-05,
+      "loss": 1.1772,
+      "step": 1281
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.531743869065319e-05,
+      "loss": 1.708,
+      "step": 1282
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.531071145542045e-05,
+      "loss": 1.5264,
+      "step": 1283
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.530398087099712e-05,
+      "loss": 1.1115,
+      "step": 1284
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5297246941627827e-05,
+      "loss": 1.1771,
+      "step": 1285
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5290509671559315e-05,
+      "loss": 1.3713,
+      "step": 1286
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5283769065040434e-05,
+      "loss": 1.205,
+      "step": 1287
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5277025126322128e-05,
+      "loss": 1.148,
+      "step": 1288
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5270277859657457e-05,
+      "loss": 1.64,
+      "step": 1289
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5263527269301567e-05,
+      "loss": 1.3575,
+      "step": 1290
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.525677335951171e-05,
+      "loss": 1.4293,
+      "step": 1291
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5250016134547223e-05,
+      "loss": 1.019,
+      "step": 1292
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.524325559866954e-05,
+      "loss": 1.6696,
+      "step": 1293
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5236491756142183e-05,
+      "loss": 1.6222,
+      "step": 1294
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5229724611230752e-05,
+      "loss": 0.9186,
+      "step": 1295
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5222954168202938e-05,
+      "loss": 0.3488,
+      "step": 1296
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5216180431328506e-05,
+      "loss": 1.9509,
+      "step": 1297
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.5209403404879305e-05,
+      "loss": 1.1477,
+      "step": 1298
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.520262309312925e-05,
+      "loss": 1.1513,
+      "step": 1299
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5195839500354337e-05,
+      "loss": 1.3353,
+      "step": 1300
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5189052630832626e-05,
+      "loss": 1.2326,
+      "step": 1301
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5182262488844237e-05,
+      "loss": 1.4649,
+      "step": 1302
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.517546907867137e-05,
+      "loss": 1.4153,
+      "step": 1303
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5168672404598273e-05,
+      "loss": 1.2863,
+      "step": 1304
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5161872470911252e-05,
+      "loss": 1.4202,
+      "step": 1305
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.515506928189868e-05,
+      "loss": 1.231,
+      "step": 1306
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5148262841850972e-05,
+      "loss": 0.758,
+      "step": 1307
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5141453155060601e-05,
+      "loss": 1.247,
+      "step": 1308
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5134640225822077e-05,
+      "loss": 0.9016,
+      "step": 1309
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5127824058431967e-05,
+      "loss": 1.6539,
+      "step": 1310
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5121004657188871e-05,
+      "loss": 1.4608,
+      "step": 1311
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5114182026393434e-05,
+      "loss": 1.7002,
+      "step": 1312
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.510735617034834e-05,
+      "loss": 1.2566,
+      "step": 1313
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5100527093358295e-05,
+      "loss": 1.1942,
+      "step": 1314
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5093694799730046e-05,
+      "loss": 0.9594,
+      "step": 1315
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5086859293772369e-05,
+      "loss": 1.2609,
+      "step": 1316
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5080020579796066e-05,
+      "loss": 1.5419,
+      "step": 1317
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.5073178662113952e-05,
+      "loss": 1.4504,
+      "step": 1318
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5066333545040879e-05,
+      "loss": 1.3445,
+      "step": 1319
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.50594852328937e-05,
+      "loss": 1.1537,
+      "step": 1320
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5052633729991296e-05,
+      "loss": 1.051,
+      "step": 1321
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.504577904065455e-05,
+      "loss": 0.9607,
+      "step": 1322
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5038921169206361e-05,
+      "loss": 1.2126,
+      "step": 1323
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.503206011997164e-05,
+      "loss": 1.1248,
+      "step": 1324
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5025195897277284e-05,
+      "loss": 1.2748,
+      "step": 1325
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5018328505452213e-05,
+      "loss": 0.9557,
+      "step": 1326
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5011457948827328e-05,
+      "loss": 1.2038,
+      "step": 1327
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.5004584231735536e-05,
+      "loss": 1.3273,
+      "step": 1328
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4997707358511734e-05,
+      "loss": 1.3941,
+      "step": 1329
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4990827333492812e-05,
+      "loss": 0.4418,
+      "step": 1330
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.498394416101764e-05,
+      "loss": 0.946,
+      "step": 1331
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4977057845427083e-05,
+      "loss": 1.737,
+      "step": 1332
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4970168391063982e-05,
+      "loss": 0.7765,
+      "step": 1333
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4963275802273156e-05,
+      "loss": 1.2591,
+      "step": 1334
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4956380083401412e-05,
+      "loss": 1.3889,
+      "step": 1335
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4949481238797512e-05,
+      "loss": 1.1744,
+      "step": 1336
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4942579272812208e-05,
+      "loss": 1.1486,
+      "step": 1337
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.4935674189798209e-05,
+      "loss": 0.9002,
+      "step": 1338
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4928765994110192e-05,
+      "loss": 0.9724,
+      "step": 1339
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.49218546901048e-05,
+      "loss": 0.9945,
+      "step": 1340
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4914940282140635e-05,
+      "loss": 1.1417,
+      "step": 1341
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4908022774578254e-05,
+      "loss": 1.4544,
+      "step": 1342
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4901102171780175e-05,
+      "loss": 1.2665,
+      "step": 1343
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4894178478110856e-05,
+      "loss": 1.2536,
+      "step": 1344
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.488725169793672e-05,
+      "loss": 1.5239,
+      "step": 1345
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4880321835626118e-05,
+      "loss": 1.0851,
+      "step": 1346
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.487338889554937e-05,
+      "loss": 1.1498,
+      "step": 1347
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.486645288207871e-05,
+      "loss": 1.5201,
+      "step": 1348
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4859513799588326e-05,
+      "loss": 1.4594,
+      "step": 1349
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4852571652454343e-05,
+      "loss": 1.6581,
+      "step": 1350
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4845626445054802e-05,
+      "loss": 0.9147,
+      "step": 1351
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4838678181769695e-05,
+      "loss": 1.1244,
+      "step": 1352
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4831726866980929e-05,
+      "loss": 1.3587,
+      "step": 1353
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4824772505072335e-05,
+      "loss": 1.1802,
+      "step": 1354
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4817815100429678e-05,
+      "loss": 1.4132,
+      "step": 1355
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4810854657440622e-05,
+      "loss": 1.3665,
+      "step": 1356
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4803891180494763e-05,
+      "loss": 1.3553,
+      "step": 1357
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4796924673983604e-05,
+      "loss": 1.2812,
+      "step": 1358
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.478995514230056e-05,
+      "loss": 1.6723,
+      "step": 1359
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4782982589840957e-05,
+      "loss": 1.3414,
+      "step": 1360
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4776007021002019e-05,
+      "loss": 0.8794,
+      "step": 1361
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4769028440182874e-05,
+      "loss": 1.7376,
+      "step": 1362
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4762046851784554e-05,
+      "loss": 1.1508,
+      "step": 1363
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4755062260209985e-05,
+      "loss": 0.9689,
+      "step": 1364
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4748074669863986e-05,
+      "loss": 0.928,
+      "step": 1365
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.474108408515327e-05,
+      "loss": 1.2703,
+      "step": 1366
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4734090510486435e-05,
+      "loss": 1.7092,
+      "step": 1367
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4727093950273967e-05,
+      "loss": 1.4311,
+      "step": 1368
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4720094408928239e-05,
+      "loss": 1.1123,
+      "step": 1369
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4713091890863487e-05,
+      "loss": 1.3895,
+      "step": 1370
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4706086400495844e-05,
+      "loss": 1.2201,
+      "step": 1371
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4699077942243315e-05,
+      "loss": 0.9592,
+      "step": 1372
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4692066520525763e-05,
+      "loss": 0.9824,
+      "step": 1373
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.468505213976493e-05,
+      "loss": 1.277,
+      "step": 1374
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4678034804384424e-05,
+      "loss": 1.4331,
+      "step": 1375
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4671014518809717e-05,
+      "loss": 0.6717,
+      "step": 1376
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4663991287468136e-05,
+      "loss": 1.1993,
+      "step": 1377
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.465696511478887e-05,
+      "loss": 1.2738,
+      "step": 1378
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4649936005202965e-05,
+      "loss": 1.45,
+      "step": 1379
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4642903963143309e-05,
+      "loss": 1.2078,
+      "step": 1380
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4635868993044655e-05,
+      "loss": 0.8517,
+      "step": 1381
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.462883109934359e-05,
+      "loss": 1.329,
+      "step": 1382
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4621790286478547e-05,
+      "loss": 1.3455,
+      "step": 1383
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.461474655888981e-05,
+      "loss": 1.4645,
+      "step": 1384
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4607699921019483e-05,
+      "loss": 1.2868,
+      "step": 1385
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4600650377311523e-05,
+      "loss": 1.3007,
+      "step": 1386
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4593597932211709e-05,
+      "loss": 0.9934,
+      "step": 1387
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4586542590167648e-05,
+      "loss": 1.0202,
+      "step": 1388
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4579484355628788e-05,
+      "loss": 1.344,
+      "step": 1389
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4572423233046386e-05,
+      "loss": 0.8531,
+      "step": 1390
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4565359226873529e-05,
+      "loss": 1.2904,
+      "step": 1391
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4558292341565117e-05,
+      "loss": 1.2244,
+      "step": 1392
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.455122258157787e-05,
+      "loss": 1.0088,
+      "step": 1393
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.454414995137032e-05,
+      "loss": 1.2391,
+      "step": 1394
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4537074455402806e-05,
+      "loss": 0.6694,
+      "step": 1395
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4529996098137478e-05,
+      "loss": 1.2455,
+      "step": 1396
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4522914884038293e-05,
+      "loss": 0.9336,
+      "step": 1397
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.4515830817570998e-05,
+      "loss": 1.1965,
+      "step": 1398
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4508743903203156e-05,
+      "loss": 1.4426,
+      "step": 1399
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.450165414540411e-05,
+      "loss": 1.6646,
+      "step": 1400
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4494561548645e-05,
+      "loss": 0.3554,
+      "step": 1401
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4487466117398772e-05,
+      "loss": 1.0771,
+      "step": 1402
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4480367856140136e-05,
+      "loss": 0.8783,
+      "step": 1403
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.44732667693456e-05,
+      "loss": 1.7154,
+      "step": 1404
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4466162861493453e-05,
+      "loss": 1.2587,
+      "step": 1405
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4459056137063763e-05,
+      "loss": 1.2324,
+      "step": 1406
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.445194660053837e-05,
+      "loss": 0.9895,
+      "step": 1407
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.444483425640089e-05,
+      "loss": 1.2503,
+      "step": 1408
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4437719109136712e-05,
+      "loss": 1.3562,
+      "step": 1409
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4430601163232991e-05,
+      "loss": 2.0011,
+      "step": 1410
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4423480423178643e-05,
+      "loss": 1.3415,
+      "step": 1411
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4416356893464355e-05,
+      "loss": 1.2739,
+      "step": 1412
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4409230578582566e-05,
+      "loss": 1.2871,
+      "step": 1413
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.440210148302747e-05,
+      "loss": 1.2031,
+      "step": 1414
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4394969611295029e-05,
+      "loss": 1.3758,
+      "step": 1415
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4387834967882934e-05,
+      "loss": 1.2098,
+      "step": 1416
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4380697557290638e-05,
+      "loss": 1.0952,
+      "step": 1417
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4373557384019337e-05,
+      "loss": 1.3265,
+      "step": 1418
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4366414452571969e-05,
+      "loss": 1.5613,
+      "step": 1419
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4359268767453206e-05,
+      "loss": 0.7024,
+      "step": 1420
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4352120333169472e-05,
+      "loss": 1.3177,
+      "step": 1421
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4344969154228903e-05,
+      "loss": 1.6341,
+      "step": 1422
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4337815235141386e-05,
+      "loss": 1.2655,
+      "step": 1423
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4330658580418519e-05,
+      "loss": 1.843,
+      "step": 1424
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4323499194573637e-05,
+      "loss": 1.1735,
+      "step": 1425
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.43163370821218e-05,
+      "loss": 0.3937,
+      "step": 1426
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4309172247579769e-05,
+      "loss": 1.2832,
+      "step": 1427
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4302004695466041e-05,
+      "loss": 1.5502,
+      "step": 1428
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4294834430300822e-05,
+      "loss": 1.1838,
+      "step": 1429
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4287661456606021e-05,
+      "loss": 1.2188,
+      "step": 1430
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4280485778905265e-05,
+      "loss": 1.1253,
+      "step": 1431
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4273307401723881e-05,
+      "loss": 1.308,
+      "step": 1432
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4266126329588899e-05,
+      "loss": 1.8793,
+      "step": 1433
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.425894256702905e-05,
+      "loss": 1.1015,
+      "step": 1434
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4251756118574763e-05,
+      "loss": 1.2162,
+      "step": 1435
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4244566988758152e-05,
+      "loss": 1.8843,
+      "step": 1436
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.4237375182113036e-05,
+      "loss": 1.0567,
+      "step": 1437
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.423018070317491e-05,
+      "loss": 1.7976,
+      "step": 1438
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4222983556480964e-05,
+      "loss": 1.2846,
+      "step": 1439
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.421578374657006e-05,
+      "loss": 1.3551,
+      "step": 1440
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4208581277982752e-05,
+      "loss": 0.8113,
+      "step": 1441
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4201376155261256e-05,
+      "loss": 1.3371,
+      "step": 1442
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4194168382949472e-05,
+      "loss": 1.6013,
+      "step": 1443
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4186957965592973e-05,
+      "loss": 1.7967,
+      "step": 1444
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.417974490773899e-05,
+      "loss": 1.525,
+      "step": 1445
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.417252921393643e-05,
+      "loss": 1.355,
+      "step": 1446
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4165310888735854e-05,
+      "loss": 1.0396,
+      "step": 1447
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4158089936689485e-05,
+      "loss": 1.3804,
+      "step": 1448
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4150866362351205e-05,
+      "loss": 1.6258,
+      "step": 1449
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.414364017027655e-05,
+      "loss": 0.8329,
+      "step": 1450
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4136411365022702e-05,
+      "loss": 1.8108,
+      "step": 1451
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4129179951148495e-05,
+      "loss": 1.3163,
+      "step": 1452
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4121945933214407e-05,
+      "loss": 1.172,
+      "step": 1453
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.411470931578256e-05,
+      "loss": 1.2028,
+      "step": 1454
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.410747010341671e-05,
+      "loss": 1.1196,
+      "step": 1455
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4100228300682252e-05,
+      "loss": 1.1249,
+      "step": 1456
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4092983912146226e-05,
+      "loss": 0.9167,
+      "step": 1457
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4085736942377285e-05,
+      "loss": 1.2677,
+      "step": 1458
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4078487395945712e-05,
+      "loss": 1.2615,
+      "step": 1459
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4071235277423432e-05,
+      "loss": 1.6524,
+      "step": 1460
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4063980591383972e-05,
+      "loss": 1.132,
+      "step": 1461
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.405672334240249e-05,
+      "loss": 0.9321,
+      "step": 1462
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4049463535055758e-05,
+      "loss": 0.7317,
+      "step": 1463
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4042201173922158e-05,
+      "loss": 1.2999,
+      "step": 1464
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4034936263581686e-05,
+      "loss": 1.2965,
+      "step": 1465
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4027668808615944e-05,
+      "loss": 0.7941,
+      "step": 1466
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4020398813608141e-05,
+      "loss": 1.2406,
+      "step": 1467
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4013126283143084e-05,
+      "loss": 1.3433,
+      "step": 1468
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.400585122180718e-05,
+      "loss": 1.0819,
+      "step": 1469
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3998573634188441e-05,
+      "loss": 1.3964,
+      "step": 1470
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3991293524876456e-05,
+      "loss": 1.2857,
+      "step": 1471
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3984010898462417e-05,
+      "loss": 1.2419,
+      "step": 1472
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3976725759539095e-05,
+      "loss": 1.6901,
+      "step": 1473
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3969438112700857e-05,
+      "loss": 1.2265,
+      "step": 1474
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3962147962543635e-05,
+      "loss": 1.5302,
+      "step": 1475
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3954855313664958e-05,
+      "loss": 1.5453,
+      "step": 1476
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.3947560170663916e-05,
+      "loss": 0.8828,
+      "step": 1477
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3940262538141182e-05,
+      "loss": 1.4679,
+      "step": 1478
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3932962420698993e-05,
+      "loss": 1.3953,
+      "step": 1479
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3925659822941149e-05,
+      "loss": 0.9066,
+      "step": 1480
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.391835474947303e-05,
+      "loss": 1.576,
+      "step": 1481
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.391104720490156e-05,
+      "loss": 1.3168,
+      "step": 1482
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3903737193835236e-05,
+      "loss": 1.0702,
+      "step": 1483
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3896424720884097e-05,
+      "loss": 1.5452,
+      "step": 1484
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3889109790659739e-05,
+      "loss": 1.9422,
+      "step": 1485
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3881792407775318e-05,
+      "loss": 1.1427,
+      "step": 1486
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.3874472576845523e-05,
+      "loss": 1.3092,
+      "step": 1487
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.386715030248659e-05,
+      "loss": 1.1538,
+      "step": 1488
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.3443834781646729,
+      "eval_runtime": 99.6391,
+      "eval_samples_per_second": 1.164,
+      "eval_steps_per_second": 1.164,
+      "step": 1488
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3966,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 496,
+  "total_flos": 2.630979922302075e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}