diff --git "a/checkpoint-1100/trainer_state.json" "b/checkpoint-1100/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/checkpoint-1100/trainer_state.json"
@@ -0,0 +1,6709 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.4740357681534152,
+  "eval_steps": 100,
+  "global_step": 1100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 5e-05,
+      "loss": 1.5475,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999997705921923e-05,
+      "loss": 1.4738,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9999908236919016e-05,
+      "loss": 1.709,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999979353322567e-05,
+      "loss": 1.3682,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99996329483497e-05,
+      "loss": 1.4525,
+      "step": 5
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999942648258581e-05,
+      "loss": 1.2902,
+      "step": 6
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999917413631294e-05,
+      "loss": 1.2581,
+      "step": 7
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9998875909994206e-05,
+      "loss": 1.1363,
+      "step": 8
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9998531804176926e-05,
+      "loss": 1.0479,
+      "step": 9
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999814181949263e-05,
+      "loss": 1.0362,
+      "step": 10
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999770595665704e-05,
+      "loss": 0.9415,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999722421647006e-05,
+      "loss": 1.0849,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999669659981584e-05,
+      "loss": 0.8538,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999612310766268e-05,
+      "loss": 0.9404,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9995503741063096e-05,
+      "loss": 0.9645,
+      "step": 15
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999483850115379e-05,
+      "loss": 1.0898,
+      "step": 16
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9994127389155623e-05,
+      "loss": 0.9545,
+      "step": 17
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9993370406373705e-05,
+      "loss": 0.8432,
+      "step": 18
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999256755419729e-05,
+      "loss": 0.9946,
+      "step": 19
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999171883409982e-05,
+      "loss": 0.7908,
+      "step": 20
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999082424763891e-05,
+      "loss": 0.8971,
+      "step": 21
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998988379645638e-05,
+      "loss": 0.85,
+      "step": 22
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998889748227819e-05,
+      "loss": 0.8626,
+      "step": 23
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9987865306914495e-05,
+      "loss": 0.8335,
+      "step": 24
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9986787272259596e-05,
+      "loss": 0.9314,
+      "step": 25
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998566338029198e-05,
+      "loss": 0.8645,
+      "step": 26
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998449363307428e-05,
+      "loss": 0.9316,
+      "step": 27
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998327803275328e-05,
+      "loss": 0.805,
+      "step": 28
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998201658155994e-05,
+      "loss": 1.0418,
+      "step": 29
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.998070928180935e-05,
+      "loss": 0.649,
+      "step": 30
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.997935613590075e-05,
+      "loss": 0.8,
+      "step": 31
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.997795714631751e-05,
+      "loss": 0.8014,
+      "step": 32
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.997651231562715e-05,
+      "loss": 0.7854,
+      "step": 33
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.9975021646481315e-05,
+      "loss": 0.8782,
+      "step": 34
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.997348514161576e-05,
+      "loss": 0.7564,
+      "step": 35
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.997190280385039e-05,
+      "loss": 0.9438,
+      "step": 36
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.997027463608921e-05,
+      "loss": 0.9248,
+      "step": 37
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996860064132033e-05,
+      "loss": 0.76,
+      "step": 38
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9966880822615965e-05,
+      "loss": 0.8834,
+      "step": 39
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996511518313244e-05,
+      "loss": 0.8928,
+      "step": 40
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996330372611017e-05,
+      "loss": 0.8857,
+      "step": 41
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996144645487364e-05,
+      "loss": 0.7975,
+      "step": 42
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.995954337283145e-05,
+      "loss": 0.7815,
+      "step": 43
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9957594483476236e-05,
+      "loss": 0.8869,
+      "step": 44
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.995559979038474e-05,
+      "loss": 0.9173,
+      "step": 45
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.995355929721773e-05,
+      "loss": 0.9258,
+      "step": 46
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.995147300772006e-05,
+      "loss": 0.896,
+      "step": 47
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.994934092572061e-05,
+      "loss": 0.893,
+      "step": 48
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.994716305513232e-05,
+      "loss": 0.8257,
+      "step": 49
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.994493939995214e-05,
+      "loss": 0.8759,
+      "step": 50
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.994266996426107e-05,
+      "loss": 0.9216,
+      "step": 51
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.994035475222412e-05,
+      "loss": 0.9097,
+      "step": 52
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9937993768090306e-05,
+      "loss": 1.015,
+      "step": 53
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.993558701619266e-05,
+      "loss": 0.8633,
+      "step": 54
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9933134500948205e-05,
+      "loss": 0.928,
+      "step": 55
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.993063622685794e-05,
+      "loss": 0.8069,
+      "step": 56
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.9928092198506866e-05,
+      "loss": 0.885,
+      "step": 57
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.992550242056393e-05,
+      "loss": 0.7342,
+      "step": 58
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.992286689778207e-05,
+      "loss": 0.8593,
+      "step": 59
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.992018563499814e-05,
+      "loss": 0.9458,
+      "step": 60
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.991745863713299e-05,
+      "loss": 0.9511,
+      "step": 61
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.991468590919135e-05,
+      "loss": 0.8605,
+      "step": 62
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9911867456261916e-05,
+      "loss": 0.76,
+      "step": 63
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9909003283517283e-05,
+      "loss": 0.8037,
+      "step": 64
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.990609339621397e-05,
+      "loss": 0.8945,
+      "step": 65
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.990313779969237e-05,
+      "loss": 0.7774,
+      "step": 66
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.990013649937679e-05,
+      "loss": 0.8737,
+      "step": 67
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.989708950077541e-05,
+      "loss": 0.8914,
+      "step": 68
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9893996809480246e-05,
+      "loss": 0.8682,
+      "step": 69
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9890858431167226e-05,
+      "loss": 0.7745,
+      "step": 70
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.988767437159608e-05,
+      "loss": 1.0306,
+      "step": 71
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.988444463661041e-05,
+      "loss": 0.904,
+      "step": 72
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.98811692321376e-05,
+      "loss": 0.8478,
+      "step": 73
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9877848164188904e-05,
+      "loss": 0.7117,
+      "step": 74
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.987448143885933e-05,
+      "loss": 0.6789,
+      "step": 75
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.987106906232773e-05,
+      "loss": 0.9385,
+      "step": 76
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9867611040856695e-05,
+      "loss": 1.049,
+      "step": 77
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9864107380792594e-05,
+      "loss": 0.8914,
+      "step": 78
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.986055808856558e-05,
+      "loss": 0.8211,
+      "step": 79
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9856963170689525e-05,
+      "loss": 0.9007,
+      "step": 80
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.9853322633762045e-05,
+      "loss": 0.832,
+      "step": 81
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.984963648446449e-05,
+      "loss": 0.807,
+      "step": 82
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.984590472956191e-05,
+      "loss": 0.7031,
+      "step": 83
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.984212737590304e-05,
+      "loss": 0.9261,
+      "step": 84
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.983830443042033e-05,
+      "loss": 0.7759,
+      "step": 85
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.983443590012989e-05,
+      "loss": 0.7147,
+      "step": 86
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9830521792131484e-05,
+      "loss": 0.8554,
+      "step": 87
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.982656211360852e-05,
+      "loss": 0.8052,
+      "step": 88
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.982255687182806e-05,
+      "loss": 0.8408,
+      "step": 89
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.981850607414077e-05,
+      "loss": 0.7086,
+      "step": 90
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.981440972798092e-05,
+      "loss": 0.7808,
+      "step": 91
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.981026784086639e-05,
+      "loss": 0.8632,
+      "step": 92
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.980608042039862e-05,
+      "loss": 0.8798,
+      "step": 93
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.980184747426263e-05,
+      "loss": 0.9253,
+      "step": 94
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9797569010226995e-05,
+      "loss": 0.8305,
+      "step": 95
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.979324503614381e-05,
+      "loss": 0.9474,
+      "step": 96
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.97888755599487e-05,
+      "loss": 0.8102,
+      "step": 97
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.978446058966081e-05,
+      "loss": 0.6445,
+      "step": 98
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.978000013338277e-05,
+      "loss": 0.8123,
+      "step": 99
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.977549419930067e-05,
+      "loss": 0.9368,
+      "step": 100
+    },
+    {
+      "epoch": 0.04,
+      "eval_loss": 0.8984958529472351,
+      "eval_runtime": 10.1685,
+      "eval_samples_per_second": 3.639,
+      "eval_steps_per_second": 0.492,
+      "step": 100
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.97709427956841e-05,
+      "loss": 0.7673,
+      "step": 101
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.976634593088608e-05,
+      "loss": 0.8571,
+      "step": 102
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9761703613343045e-05,
+      "loss": 0.9388,
+      "step": 103
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.975701585157489e-05,
+      "loss": 0.7665,
+      "step": 104
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9752282654184876e-05,
+      "loss": 0.9379,
+      "step": 105
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.974750402985966e-05,
+      "loss": 0.9475,
+      "step": 106
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.974267998736928e-05,
+      "loss": 0.9757,
+      "step": 107
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.973781053556711e-05,
+      "loss": 0.6865,
+      "step": 108
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.973289568338989e-05,
+      "loss": 0.9761,
+      "step": 109
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9727935439857634e-05,
+      "loss": 0.9132,
+      "step": 110
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.972292981407372e-05,
+      "loss": 0.774,
+      "step": 111
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.971787881522477e-05,
+      "loss": 1.0217,
+      "step": 112
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9712782452580695e-05,
+      "loss": 0.9341,
+      "step": 113
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.970764073549467e-05,
+      "loss": 0.7135,
+      "step": 114
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.970245367340307e-05,
+      "loss": 0.7177,
+      "step": 115
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.969722127582553e-05,
+      "loss": 0.9312,
+      "step": 116
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9691943552364885e-05,
+      "loss": 0.8677,
+      "step": 117
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.968662051270711e-05,
+      "loss": 0.7371,
+      "step": 118
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.968125216662142e-05,
+      "loss": 0.7248,
+      "step": 119
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9675838523960114e-05,
+      "loss": 0.7281,
+      "step": 120
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.967037959465865e-05,
+      "loss": 0.8678,
+      "step": 121
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.96648753887356e-05,
+      "loss": 0.8477,
+      "step": 122
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.965932591629262e-05,
+      "loss": 0.7801,
+      "step": 123
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.965373118751446e-05,
+      "loss": 0.8289,
+      "step": 124
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9648091212668904e-05,
+      "loss": 0.7771,
+      "step": 125
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.964240600210679e-05,
+      "loss": 0.7933,
+      "step": 126
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.963667556626198e-05,
+      "loss": 0.8119,
+      "step": 127
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.963089991565131e-05,
+      "loss": 0.9761,
+      "step": 128
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9625079060874634e-05,
+      "loss": 0.7456,
+      "step": 129
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.961921301261474e-05,
+      "loss": 0.7955,
+      "step": 130
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.961330178163736e-05,
+      "loss": 0.8379,
+      "step": 131
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.960734537879117e-05,
+      "loss": 0.793,
+      "step": 132
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.960134381500771e-05,
+      "loss": 0.8465,
+      "step": 133
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.959529710130144e-05,
+      "loss": 0.8813,
+      "step": 134
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.958920524876967e-05,
+      "loss": 0.8148,
+      "step": 135
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9583068268592544e-05,
+      "loss": 0.8524,
+      "step": 136
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.957688617203302e-05,
+      "loss": 0.6524,
+      "step": 137
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.957065897043689e-05,
+      "loss": 0.8021,
+      "step": 138
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9564386675232674e-05,
+      "loss": 0.9442,
+      "step": 139
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.955806929793171e-05,
+      "loss": 0.8701,
+      "step": 140
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9551706850128024e-05,
+      "loss": 0.9975,
+      "step": 141
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.954529934349839e-05,
+      "loss": 0.7725,
+      "step": 142
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.953884678980225e-05,
+      "loss": 0.7544,
+      "step": 143
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9532349200881745e-05,
+      "loss": 0.6881,
+      "step": 144
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9525806588661656e-05,
+      "loss": 0.8211,
+      "step": 145
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.951921896514938e-05,
+      "loss": 0.7736,
+      "step": 146
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9512586342434956e-05,
+      "loss": 0.9176,
+      "step": 147
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.950590873269098e-05,
+      "loss": 0.7639,
+      "step": 148
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.94991861481726e-05,
+      "loss": 0.8447,
+      "step": 149
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.949241860121755e-05,
+      "loss": 0.7537,
+      "step": 150
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.948560610424604e-05,
+      "loss": 0.7948,
+      "step": 151
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.94787486697608e-05,
+      "loss": 0.7734,
+      "step": 152
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9471846310347006e-05,
+      "loss": 0.8796,
+      "step": 153
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.946489903867231e-05,
+      "loss": 0.7196,
+      "step": 154
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.945790686748678e-05,
+      "loss": 0.7676,
+      "step": 155
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.945086980962288e-05,
+      "loss": 0.7697,
+      "step": 156
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9443787877995464e-05,
+      "loss": 0.8223,
+      "step": 157
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.943666108560173e-05,
+      "loss": 0.8416,
+      "step": 158
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.942948944552122e-05,
+      "loss": 0.8536,
+      "step": 159
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.942227297091577e-05,
+      "loss": 1.0185,
+      "step": 160
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.94150116750295e-05,
+      "loss": 0.8932,
+      "step": 161
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.94077055711888e-05,
+      "loss": 0.8254,
+      "step": 162
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.940035467280229e-05,
+      "loss": 0.7546,
+      "step": 163
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9392958993360794e-05,
+      "loss": 0.8746,
+      "step": 164
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9385518546437325e-05,
+      "loss": 0.9356,
+      "step": 165
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.937803334568706e-05,
+      "loss": 0.8459,
+      "step": 166
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9370503404847304e-05,
+      "loss": 0.7772,
+      "step": 167
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9362928737737465e-05,
+      "loss": 0.9711,
+      "step": 168
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.935530935825907e-05,
+      "loss": 0.849,
+      "step": 169
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.934764528039566e-05,
+      "loss": 0.759,
+      "step": 170
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.933993651821282e-05,
+      "loss": 0.7088,
+      "step": 171
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.933218308585819e-05,
+      "loss": 0.6487,
+      "step": 172
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9324384997561315e-05,
+      "loss": 0.7925,
+      "step": 173
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.931654226763375e-05,
+      "loss": 0.922,
+      "step": 174
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9308654910468964e-05,
+      "loss": 0.8621,
+      "step": 175
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.930072294054233e-05,
+      "loss": 0.8429,
+      "step": 176
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.929274637241108e-05,
+      "loss": 0.7388,
+      "step": 177
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9284725220714324e-05,
+      "loss": 0.8557,
+      "step": 178
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.927665950017297e-05,
+      "loss": 0.8447,
+      "step": 179
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.926854922558975e-05,
+      "loss": 0.8104,
+      "step": 180
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9260394411849134e-05,
+      "loss": 0.9483,
+      "step": 181
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9252195073917335e-05,
+      "loss": 0.7172,
+      "step": 182
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9243951226842305e-05,
+      "loss": 0.7127,
+      "step": 183
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9235662885753666e-05,
+      "loss": 0.7191,
+      "step": 184
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.92273300658627e-05,
+      "loss": 0.7975,
+      "step": 185
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.921895278246231e-05,
+      "loss": 0.7981,
+      "step": 186
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.921053105092701e-05,
+      "loss": 0.7595,
+      "step": 187
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9202064886712906e-05,
+      "loss": 0.7376,
+      "step": 188
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.919355430535761e-05,
+      "loss": 0.8999,
+      "step": 189
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9184999322480285e-05,
+      "loss": 0.8945,
+      "step": 190
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.917639995378156e-05,
+      "loss": 0.7347,
+      "step": 191
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.916775621504354e-05,
+      "loss": 0.8662,
+      "step": 192
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.915906812212976e-05,
+      "loss": 0.6963,
+      "step": 193
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9150335690985136e-05,
+      "loss": 0.7317,
+      "step": 194
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.9141558937635985e-05,
+      "loss": 0.8358,
+      "step": 195
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.913273787818995e-05,
+      "loss": 0.7879,
+      "step": 196
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.912387252883598e-05,
+      "loss": 0.8523,
+      "step": 197
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.911496290584433e-05,
+      "loss": 0.8389,
+      "step": 198
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.910600902556649e-05,
+      "loss": 0.7412,
+      "step": 199
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.909701090443518e-05,
+      "loss": 0.8253,
+      "step": 200
+    },
+    {
+      "epoch": 0.09,
+      "eval_loss": 0.8050532341003418,
+      "eval_runtime": 10.2691,
+      "eval_samples_per_second": 3.603,
+      "eval_steps_per_second": 0.487,
+      "step": 200
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.908796855896432e-05,
+      "loss": 0.7399,
+      "step": 201
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.907888200574898e-05,
+      "loss": 0.9344,
+      "step": 202
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.906975126146538e-05,
+      "loss": 0.8562,
+      "step": 203
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.906057634287082e-05,
+      "loss": 0.7759,
+      "step": 204
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.905135726680369e-05,
+      "loss": 0.7058,
+      "step": 205
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.9042094050183415e-05,
+      "loss": 0.9686,
+      "step": 206
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.9032786710010424e-05,
+      "loss": 0.9007,
+      "step": 207
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.902343526336614e-05,
+      "loss": 0.8106,
+      "step": 208
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.901403972741291e-05,
+      "loss": 0.8317,
+      "step": 209
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.900460011939402e-05,
+      "loss": 0.9889,
+      "step": 210
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.899511645663361e-05,
+      "loss": 0.7061,
+      "step": 211
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.898558875653671e-05,
+      "loss": 0.9084,
+      "step": 212
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.897601703658914e-05,
+      "loss": 0.8296,
+      "step": 213
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.896640131435751e-05,
+      "loss": 0.7512,
+      "step": 214
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.8956741607489215e-05,
+      "loss": 0.7808,
+      "step": 215
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.8947037933712335e-05,
+      "loss": 0.8168,
+      "step": 216
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.893729031083567e-05,
+      "loss": 0.6904,
+      "step": 217
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.8927498756748664e-05,
+      "loss": 0.8154,
+      "step": 218
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.891766328942138e-05,
+      "loss": 0.732,
+      "step": 219
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.8907783926904495e-05,
+      "loss": 0.833,
+      "step": 220
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.889786068732921e-05,
+      "loss": 0.7308,
+      "step": 221
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8887893588907304e-05,
+      "loss": 0.8428,
+      "step": 222
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8877882649930995e-05,
+      "loss": 0.735,
+      "step": 223
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8867827888773e-05,
+      "loss": 0.6815,
+      "step": 224
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.885772932388642e-05,
+      "loss": 0.8329,
+      "step": 225
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.884758697380481e-05,
+      "loss": 0.7601,
+      "step": 226
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8837400857142015e-05,
+      "loss": 0.7283,
+      "step": 227
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.882717099259224e-05,
+      "loss": 0.7835,
+      "step": 228
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8816897398929975e-05,
+      "loss": 0.7538,
+      "step": 229
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.880658009500995e-05,
+      "loss": 0.6243,
+      "step": 230
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8796219099767136e-05,
+      "loss": 0.7844,
+      "step": 231
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.878581443221668e-05,
+      "loss": 0.8362,
+      "step": 232
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.877536611145388e-05,
+      "loss": 0.7069,
+      "step": 233
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.876487415665413e-05,
+      "loss": 0.7603,
+      "step": 234
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.875433858707293e-05,
+      "loss": 0.8936,
+      "step": 235
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.874375942204583e-05,
+      "loss": 0.7754,
+      "step": 236
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8733136680988355e-05,
+      "loss": 0.7515,
+      "step": 237
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8722470383396024e-05,
+      "loss": 0.7706,
+      "step": 238
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.871176054884431e-05,
+      "loss": 0.7938,
+      "step": 239
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.870100719698854e-05,
+      "loss": 0.7918,
+      "step": 240
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8690210347563975e-05,
+      "loss": 0.9421,
+      "step": 241
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.867937002038564e-05,
+      "loss": 0.7249,
+      "step": 242
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.866848623534839e-05,
+      "loss": 0.7552,
+      "step": 243
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.865755901242682e-05,
+      "loss": 0.7415,
+      "step": 244
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.864658837167526e-05,
+      "loss": 0.7475,
+      "step": 245
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.863557433322771e-05,
+      "loss": 0.8791,
+      "step": 246
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.862451691729783e-05,
+      "loss": 0.8837,
+      "step": 247
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.861341614417887e-05,
+      "loss": 0.7218,
+      "step": 248
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.860227203424367e-05,
+      "loss": 0.8089,
+      "step": 249
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8591084607944595e-05,
+      "loss": 0.9958,
+      "step": 250
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.85798538858135e-05,
+      "loss": 0.861,
+      "step": 251
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.856857988846172e-05,
+      "loss": 0.783,
+      "step": 252
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.855726263657999e-05,
+      "loss": 0.8853,
+      "step": 253
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8545902150938436e-05,
+      "loss": 0.7947,
+      "step": 254
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8534498452386543e-05,
+      "loss": 0.7206,
+      "step": 255
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8523051561853084e-05,
+      "loss": 0.8859,
+      "step": 256
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.851156150034611e-05,
+      "loss": 0.745,
+      "step": 257
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.850002828895289e-05,
+      "loss": 0.9045,
+      "step": 258
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8488451948839905e-05,
+      "loss": 0.6362,
+      "step": 259
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.847683250125277e-05,
+      "loss": 0.8622,
+      "step": 260
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8465169967516235e-05,
+      "loss": 0.7037,
+      "step": 261
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.845346436903408e-05,
+      "loss": 0.9182,
+      "step": 262
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.844171572728919e-05,
+      "loss": 1.0684,
+      "step": 263
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.842992406384338e-05,
+      "loss": 0.7909,
+      "step": 264
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8418089400337444e-05,
+      "loss": 0.8534,
+      "step": 265
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.840621175849112e-05,
+      "loss": 0.8092,
+      "step": 266
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.839429116010297e-05,
+      "loss": 0.7109,
+      "step": 267
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.838232762705044e-05,
+      "loss": 0.5593,
+      "step": 268
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8370321181289754e-05,
+      "loss": 0.6169,
+      "step": 269
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.835827184485587e-05,
+      "loss": 0.7919,
+      "step": 270
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.834617963986251e-05,
+      "loss": 0.8294,
+      "step": 271
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.833404458850203e-05,
+      "loss": 0.9067,
+      "step": 272
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.832186671304543e-05,
+      "loss": 0.8037,
+      "step": 273
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8309646035842316e-05,
+      "loss": 0.7649,
+      "step": 274
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.829738257932084e-05,
+      "loss": 0.736,
+      "step": 275
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8285076365987646e-05,
+      "loss": 0.6717,
+      "step": 276
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.827272741842789e-05,
+      "loss": 0.8744,
+      "step": 277
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.826033575930512e-05,
+      "loss": 0.8303,
+      "step": 278
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.824790141136127e-05,
+      "loss": 0.8506,
+      "step": 279
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.823542439741666e-05,
+      "loss": 0.7496,
+      "step": 280
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.822290474036987e-05,
+      "loss": 0.8215,
+      "step": 281
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.821034246319774e-05,
+      "loss": 0.8849,
+      "step": 282
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.819773758895538e-05,
+      "loss": 0.7652,
+      "step": 283
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.818509014077602e-05,
+      "loss": 0.7457,
+      "step": 284
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.817240014187105e-05,
+      "loss": 0.7701,
+      "step": 285
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8159667615529954e-05,
+      "loss": 0.7975,
+      "step": 286
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.814689258512025e-05,
+      "loss": 0.7421,
+      "step": 287
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.813407507408748e-05,
+      "loss": 0.771,
+      "step": 288
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8121215105955145e-05,
+      "loss": 0.7896,
+      "step": 289
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8108312704324654e-05,
+      "loss": 0.6527,
+      "step": 290
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.8095367892875295e-05,
+      "loss": 0.7029,
+      "step": 291
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.80823806953642e-05,
+      "loss": 0.7148,
+      "step": 292
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.80693511356263e-05,
+      "loss": 0.7192,
+      "step": 293
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.8056279237574234e-05,
+      "loss": 0.8897,
+      "step": 294
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.8043165025198375e-05,
+      "loss": 0.9168,
+      "step": 295
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.803000852256675e-05,
+      "loss": 0.6315,
+      "step": 296
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.801680975382499e-05,
+      "loss": 0.8135,
+      "step": 297
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.800356874319629e-05,
+      "loss": 0.8978,
+      "step": 298
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.79902855149814e-05,
+      "loss": 0.8826,
+      "step": 299
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.797696009355851e-05,
+      "loss": 0.8169,
+      "step": 300
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.795865535736084,
+      "eval_runtime": 10.328,
+      "eval_samples_per_second": 3.582,
+      "eval_steps_per_second": 0.484,
+      "step": 300
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.796359250338328e-05,
+      "loss": 0.7584,
+      "step": 301
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.795018276898874e-05,
+      "loss": 0.7721,
+      "step": 302
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.793673091498527e-05,
+      "loss": 0.7665,
+      "step": 303
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7923236966060566e-05,
+      "loss": 0.8482,
+      "step": 304
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7909700946979555e-05,
+      "loss": 0.6749,
+      "step": 305
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7896122882584374e-05,
+      "loss": 0.5408,
+      "step": 306
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7882502797794346e-05,
+      "loss": 0.8174,
+      "step": 307
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7868840717605914e-05,
+      "loss": 0.8283,
+      "step": 308
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.785513666709256e-05,
+      "loss": 0.7986,
+      "step": 309
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.784139067140483e-05,
+      "loss": 0.7339,
+      "step": 310
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7827602755770224e-05,
+      "loss": 0.6881,
+      "step": 311
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.7813772945493185e-05,
+      "loss": 0.7879,
+      "step": 312
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.779990126595505e-05,
+      "loss": 0.9105,
+      "step": 313
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.778598774261399e-05,
+      "loss": 0.6961,
+      "step": 314
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.777203240100497e-05,
+      "loss": 0.7347,
+      "step": 315
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.775803526673971e-05,
+      "loss": 0.8124,
+      "step": 316
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7743996365506614e-05,
+      "loss": 0.8755,
+      "step": 317
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.772991572307076e-05,
+      "loss": 0.6714,
+      "step": 318
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.771579336527383e-05,
+      "loss": 0.7593,
+      "step": 319
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7701629318034034e-05,
+      "loss": 0.7754,
+      "step": 320
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.768742360734614e-05,
+      "loss": 0.8489,
+      "step": 321
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7673176259281336e-05,
+      "loss": 0.7423,
+      "step": 322
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7658887299987265e-05,
+      "loss": 0.9569,
+      "step": 323
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7644556755687905e-05,
+      "loss": 0.6688,
+      "step": 324
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.763018465268356e-05,
+      "loss": 0.688,
+      "step": 325
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7615771017350826e-05,
+      "loss": 0.8766,
+      "step": 326
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.76013158761425e-05,
+      "loss": 0.815,
+      "step": 327
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.758681925558756e-05,
+      "loss": 0.6949,
+      "step": 328
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.75722811822911e-05,
+      "loss": 0.6397,
+      "step": 329
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.755770168293432e-05,
+      "loss": 0.8287,
+      "step": 330
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7543080784274414e-05,
+      "loss": 0.831,
+      "step": 331
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.752841851314457e-05,
+      "loss": 0.8885,
+      "step": 332
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.751371489645391e-05,
+      "loss": 0.7274,
+      "step": 333
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7498969961187424e-05,
+      "loss": 0.6892,
+      "step": 334
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.748418373440594e-05,
+      "loss": 0.7592,
+      "step": 335
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.7469356243246066e-05,
+      "loss": 0.7485,
+      "step": 336
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.745448751492014e-05,
+      "loss": 0.7914,
+      "step": 337
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7439577576716175e-05,
+      "loss": 0.9029,
+      "step": 338
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.742462645599783e-05,
+      "loss": 0.6633,
+      "step": 339
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.740963418020433e-05,
+      "loss": 0.7987,
+      "step": 340
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.739460077685043e-05,
+      "loss": 0.9075,
+      "step": 341
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.737952627352639e-05,
+      "loss": 0.7466,
+      "step": 342
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.736441069789786e-05,
+      "loss": 0.6737,
+      "step": 343
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.734925407770589e-05,
+      "loss": 0.7996,
+      "step": 344
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.733405644076688e-05,
+      "loss": 0.8677,
+      "step": 345
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.731881781497246e-05,
+      "loss": 0.8364,
+      "step": 346
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7303538228289504e-05,
+      "loss": 0.7339,
+      "step": 347
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.728821770876008e-05,
+      "loss": 0.7552,
+      "step": 348
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.727285628450136e-05,
+      "loss": 0.6216,
+      "step": 349
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.725745398370558e-05,
+      "loss": 0.7541,
+      "step": 350
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.724201083464e-05,
+      "loss": 0.7895,
+      "step": 351
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.722652686564687e-05,
+      "loss": 0.8574,
+      "step": 352
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.721100210514333e-05,
+      "loss": 0.7342,
+      "step": 353
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.719543658162139e-05,
+      "loss": 0.676,
+      "step": 354
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.717983032364787e-05,
+      "loss": 0.7245,
+      "step": 355
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.716418335986434e-05,
+      "loss": 0.7016,
+      "step": 356
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7148495718987096e-05,
+      "loss": 0.8397,
+      "step": 357
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.7132767429807076e-05,
+      "loss": 0.9152,
+      "step": 358
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.711699852118981e-05,
+      "loss": 0.7737,
+      "step": 359
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.710118902207541e-05,
+      "loss": 0.7912,
+      "step": 360
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.708533896147842e-05,
+      "loss": 0.711,
+      "step": 361
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.706944836848789e-05,
+      "loss": 0.9113,
+      "step": 362
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.7053517272267214e-05,
+      "loss": 0.736,
+      "step": 363
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.703754570205413e-05,
+      "loss": 0.8455,
+      "step": 364
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.7021533687160676e-05,
+      "loss": 0.9134,
+      "step": 365
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.7005481256973097e-05,
+      "loss": 0.9046,
+      "step": 366
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.698938844095181e-05,
+      "loss": 0.9463,
+      "step": 367
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.6973255268631366e-05,
+      "loss": 0.7204,
+      "step": 368
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.6957081769620356e-05,
+      "loss": 0.7333,
+      "step": 369
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.694086797360141e-05,
+      "loss": 0.8651,
+      "step": 370
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.692461391033109e-05,
+      "loss": 0.8249,
+      "step": 371
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.690831960963988e-05,
+      "loss": 0.6987,
+      "step": 372
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.6891985101432084e-05,
+      "loss": 0.6567,
+      "step": 373
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.687561041568582e-05,
+      "loss": 0.7595,
+      "step": 374
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.685919558245293e-05,
+      "loss": 0.8601,
+      "step": 375
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.684274063185894e-05,
+      "loss": 0.6274,
+      "step": 376
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.682624559410301e-05,
+      "loss": 0.6681,
+      "step": 377
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.680971049945786e-05,
+      "loss": 0.7932,
+      "step": 378
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.679313537826973e-05,
+      "loss": 0.7183,
+      "step": 379
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.677652026095831e-05,
+      "loss": 0.6897,
+      "step": 380
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.675986517801672e-05,
+      "loss": 0.7922,
+      "step": 381
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.674317016001138e-05,
+      "loss": 0.8051,
+      "step": 382
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.672643523758206e-05,
+      "loss": 0.6802,
+      "step": 383
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6709660441441705e-05,
+      "loss": 0.7451,
+      "step": 384
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6692845802376494e-05,
+      "loss": 0.8396,
+      "step": 385
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.66759913512457e-05,
+      "loss": 0.9847,
+      "step": 386
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.665909711898164e-05,
+      "loss": 0.8155,
+      "step": 387
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.664216313658969e-05,
+      "loss": 0.7643,
+      "step": 388
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6625189435148146e-05,
+      "loss": 0.7872,
+      "step": 389
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6608176045808206e-05,
+      "loss": 0.7877,
+      "step": 390
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.659112299979389e-05,
+      "loss": 0.5715,
+      "step": 391
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.657403032840203e-05,
+      "loss": 0.7811,
+      "step": 392
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6556898063002164e-05,
+      "loss": 0.5627,
+      "step": 393
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.653972623503649e-05,
+      "loss": 0.6729,
+      "step": 394
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6522514876019816e-05,
+      "loss": 0.6761,
+      "step": 395
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6505264017539516e-05,
+      "loss": 0.7337,
+      "step": 396
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.648797369125543e-05,
+      "loss": 0.7986,
+      "step": 397
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.647064392889985e-05,
+      "loss": 0.7348,
+      "step": 398
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6453274762277435e-05,
+      "loss": 0.8767,
+      "step": 399
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.643586622326516e-05,
+      "loss": 0.7493,
+      "step": 400
+    },
+    {
+      "epoch": 0.17,
+      "eval_loss": 0.7891781330108643,
+      "eval_runtime": 10.3744,
+      "eval_samples_per_second": 3.566,
+      "eval_steps_per_second": 0.482,
+      "step": 400
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.641841834381228e-05,
+      "loss": 0.9408,
+      "step": 401
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.640093115594022e-05,
+      "loss": 0.7631,
+      "step": 402
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.638340469174256e-05,
+      "loss": 0.8535,
+      "step": 403
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.636583898338497e-05,
+      "loss": 0.7969,
+      "step": 404
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.6348234063105124e-05,
+      "loss": 0.6849,
+      "step": 405
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.633058996321268e-05,
+      "loss": 0.698,
+      "step": 406
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.631290671608919e-05,
+      "loss": 0.7716,
+      "step": 407
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.629518435418805e-05,
+      "loss": 0.7092,
+      "step": 408
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.627742291003445e-05,
+      "loss": 0.6969,
+      "step": 409
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.62596224162253e-05,
+      "loss": 0.6474,
+      "step": 410
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.624178290542917e-05,
+      "loss": 0.6569,
+      "step": 411
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.622390441038627e-05,
+      "loss": 0.8783,
+      "step": 412
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.620598696390831e-05,
+      "loss": 0.7998,
+      "step": 413
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.61880305988785e-05,
+      "loss": 0.7375,
+      "step": 414
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.61700353482515e-05,
+      "loss": 0.7961,
+      "step": 415
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.615200124505331e-05,
+      "loss": 0.7244,
+      "step": 416
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.613392832238124e-05,
+      "loss": 0.8139,
+      "step": 417
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.611581661340386e-05,
+      "loss": 0.7845,
+      "step": 418
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.609766615136089e-05,
+      "loss": 0.7492,
+      "step": 419
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.607947696956321e-05,
+      "loss": 0.7373,
+      "step": 420
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.606124910139272e-05,
+      "loss": 0.7485,
+      "step": 421
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.604298258030237e-05,
+      "loss": 0.568,
+      "step": 422
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.6024677439816e-05,
+      "loss": 0.8351,
+      "step": 423
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.600633371352836e-05,
+      "loss": 0.6451,
+      "step": 424
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.598795143510499e-05,
+      "loss": 0.7567,
+      "step": 425
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.59695306382822e-05,
+      "loss": 0.728,
+      "step": 426
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.595107135686699e-05,
+      "loss": 0.8118,
+      "step": 427
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.5932573624737e-05,
+      "loss": 0.7179,
+      "step": 428
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.5914037475840385e-05,
+      "loss": 1.0522,
+      "step": 429
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.589546294419589e-05,
+      "loss": 0.8786,
+      "step": 430
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.587685006389262e-05,
+      "loss": 0.6876,
+      "step": 431
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.58581988690901e-05,
+      "loss": 0.6063,
+      "step": 432
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.583950939401819e-05,
+      "loss": 0.7466,
+      "step": 433
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5820781672976955e-05,
+      "loss": 0.7848,
+      "step": 434
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.58020157403367e-05,
+      "loss": 0.837,
+      "step": 435
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.578321163053783e-05,
+      "loss": 0.7999,
+      "step": 436
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.576436937809081e-05,
+      "loss": 0.7283,
+      "step": 437
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5745489017576124e-05,
+      "loss": 0.8953,
+      "step": 438
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5726570583644205e-05,
+      "loss": 0.7776,
+      "step": 439
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.570761411101533e-05,
+      "loss": 0.7233,
+      "step": 440
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5688619634479604e-05,
+      "loss": 0.8226,
+      "step": 441
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.566958718889688e-05,
+      "loss": 0.6558,
+      "step": 442
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.565051680919669e-05,
+      "loss": 0.7776,
+      "step": 443
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.563140853037817e-05,
+      "loss": 0.6725,
+      "step": 444
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.561226238751005e-05,
+      "loss": 0.7942,
+      "step": 445
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5593078415730515e-05,
+      "loss": 0.7764,
+      "step": 446
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.55738566502472e-05,
+      "loss": 0.9007,
+      "step": 447
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.5554597126337076e-05,
+      "loss": 0.7221,
+      "step": 448
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.553529987934643e-05,
+      "loss": 0.809,
+      "step": 449
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.551596494469077e-05,
+      "loss": 0.7276,
+      "step": 450
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.549659235785478e-05,
+      "loss": 0.7519,
+      "step": 451
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.547718215439224e-05,
+      "loss": 0.7447,
+      "step": 452
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.545773436992598e-05,
+      "loss": 0.6342,
+      "step": 453
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.543824904014777e-05,
+      "loss": 0.6875,
+      "step": 454
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5418726200818316e-05,
+      "loss": 0.7269,
+      "step": 455
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.539916588776715e-05,
+      "loss": 0.7012,
+      "step": 456
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5379568136892567e-05,
+      "loss": 0.8766,
+      "step": 457
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5359932984161605e-05,
+      "loss": 0.7997,
+      "step": 458
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.534026046560992e-05,
+      "loss": 0.709,
+      "step": 459
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5320550617341726e-05,
+      "loss": 0.7729,
+      "step": 460
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.530080347552979e-05,
+      "loss": 0.7965,
+      "step": 461
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.528101907641529e-05,
+      "loss": 0.8593,
+      "step": 462
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5261197456307794e-05,
+      "loss": 0.7322,
+      "step": 463
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.524133865158518e-05,
+      "loss": 0.7928,
+      "step": 464
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.522144269869355e-05,
+      "loss": 0.7176,
+      "step": 465
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5201509634147224e-05,
+      "loss": 0.7317,
+      "step": 466
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.518153949452858e-05,
+      "loss": 0.8163,
+      "step": 467
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.51615323164881e-05,
+      "loss": 0.7572,
+      "step": 468
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5141488136744174e-05,
+      "loss": 0.7409,
+      "step": 469
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.512140699208315e-05,
+      "loss": 0.8279,
+      "step": 470
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.51012889193592e-05,
+      "loss": 0.6798,
+      "step": 471
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.508113395549426e-05,
+      "loss": 0.6618,
+      "step": 472
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.506094213747798e-05,
+      "loss": 0.8773,
+      "step": 473
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.504071350236766e-05,
+      "loss": 0.8261,
+      "step": 474
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.5020448087288126e-05,
+      "loss": 0.6215,
+      "step": 475
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.5000145929431756e-05,
+      "loss": 0.8396,
+      "step": 476
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.497980706605833e-05,
+      "loss": 0.7632,
+      "step": 477
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.495943153449501e-05,
+      "loss": 0.7404,
+      "step": 478
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.4939019372136224e-05,
+      "loss": 0.8391,
+      "step": 479
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.491857061644367e-05,
+      "loss": 0.7786,
+      "step": 480
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.489808530494617e-05,
+      "loss": 0.7974,
+      "step": 481
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.487756347523965e-05,
+      "loss": 0.8695,
+      "step": 482
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.4857005164987044e-05,
+      "loss": 0.6575,
+      "step": 483
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.483641041191826e-05,
+      "loss": 0.8302,
+      "step": 484
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.481577925383007e-05,
+      "loss": 0.8316,
+      "step": 485
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.479511172858607e-05,
+      "loss": 0.8911,
+      "step": 486
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.4774407874116584e-05,
+      "loss": 0.7963,
+      "step": 487
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.475366772841862e-05,
+      "loss": 0.7823,
+      "step": 488
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.47328913295558e-05,
+      "loss": 0.8895,
+      "step": 489
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.4712078715658254e-05,
+      "loss": 0.7182,
+      "step": 490
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.46912299249226e-05,
+      "loss": 0.8522,
+      "step": 491
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.467034499561185e-05,
+      "loss": 0.7836,
+      "step": 492
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.46494239660553e-05,
+      "loss": 0.6369,
+      "step": 493
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.462846687464857e-05,
+      "loss": 0.7994,
+      "step": 494
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.4607473759853384e-05,
+      "loss": 0.7834,
+      "step": 495
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.458644466019764e-05,
+      "loss": 0.741,
+      "step": 496
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.456537961427526e-05,
+      "loss": 0.7608,
+      "step": 497
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.454427866074612e-05,
+      "loss": 0.8057,
+      "step": 498
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.452314183833601e-05,
+      "loss": 0.8335,
+      "step": 499
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.450196918583655e-05,
+      "loss": 0.7366,
+      "step": 500
+    },
+    {
+      "epoch": 0.22,
+      "eval_loss": 0.7860936522483826,
+      "eval_runtime": 10.1703,
+      "eval_samples_per_second": 3.638,
+      "eval_steps_per_second": 0.492,
+      "step": 500
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.448076074210511e-05,
+      "loss": 0.703,
+      "step": 501
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.445951654606476e-05,
+      "loss": 0.8081,
+      "step": 502
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.443823663670416e-05,
+      "loss": 0.5891,
+      "step": 503
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.441692105307754e-05,
+      "loss": 0.7695,
+      "step": 504
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.43955698343046e-05,
+      "loss": 0.7281,
+      "step": 505
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4374183019570404e-05,
+      "loss": 0.8165,
+      "step": 506
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.435276064812539e-05,
+      "loss": 0.9928,
+      "step": 507
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4331302759285225e-05,
+      "loss": 0.698,
+      "step": 508
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4309809392430776e-05,
+      "loss": 0.6933,
+      "step": 509
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4288280587007994e-05,
+      "loss": 0.7442,
+      "step": 510
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4266716382527904e-05,
+      "loss": 0.9552,
+      "step": 511
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.424511681856648e-05,
+      "loss": 0.7013,
+      "step": 512
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.422348193476458e-05,
+      "loss": 0.648,
+      "step": 513
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4201811770827905e-05,
+      "loss": 0.6661,
+      "step": 514
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.418010636652689e-05,
+      "loss": 0.8251,
+      "step": 515
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.415836576169664e-05,
+      "loss": 0.6331,
+      "step": 516
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.413658999623689e-05,
+      "loss": 0.79,
+      "step": 517
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.4114779110111866e-05,
+      "loss": 0.7234,
+      "step": 518
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.409293314335028e-05,
+      "loss": 0.8827,
+      "step": 519
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.40710521360452e-05,
+      "loss": 0.7483,
+      "step": 520
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.404913612835404e-05,
+      "loss": 0.8601,
+      "step": 521
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.402718516049842e-05,
+      "loss": 0.7445,
+      "step": 522
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.4005199272764105e-05,
+      "loss": 0.8318,
+      "step": 523
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.3983178505501e-05,
+      "loss": 0.8368,
+      "step": 524
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.3961122899122975e-05,
+      "loss": 0.7101,
+      "step": 525
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.3939032494107855e-05,
+      "loss": 0.7236,
+      "step": 526
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.391690733099735e-05,
+      "loss": 0.7242,
+      "step": 527
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.389474745039691e-05,
+      "loss": 0.7646,
+      "step": 528
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.387255289297576e-05,
+      "loss": 0.7416,
+      "step": 529
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.385032369946673e-05,
+      "loss": 0.8072,
+      "step": 530
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.382805991066622e-05,
+      "loss": 0.7637,
+      "step": 531
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.380576156743412e-05,
+      "loss": 0.8242,
+      "step": 532
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.378342871069376e-05,
+      "loss": 0.7322,
+      "step": 533
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.376106138143178e-05,
+      "loss": 0.7279,
+      "step": 534
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.373865962069811e-05,
+      "loss": 0.8358,
+      "step": 535
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.371622346960584e-05,
+      "loss": 0.8739,
+      "step": 536
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.3693752969331224e-05,
+      "loss": 0.6559,
+      "step": 537
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.367124816111351e-05,
+      "loss": 0.7295,
+      "step": 538
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.364870908625493e-05,
+      "loss": 0.8219,
+      "step": 539
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.36261357861206e-05,
+      "loss": 0.7993,
+      "step": 540
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.360352830213846e-05,
+      "loss": 0.7536,
+      "step": 541
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.358088667579917e-05,
+      "loss": 0.8467,
+      "step": 542
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.355821094865606e-05,
+      "loss": 0.8671,
+      "step": 543
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.3535501162325034e-05,
+      "loss": 0.7898,
+      "step": 544
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.351275735848451e-05,
+      "loss": 0.8136,
+      "step": 545
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.348997957887535e-05,
+      "loss": 0.8536,
+      "step": 546
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.346716786530075e-05,
+      "loss": 0.9403,
+      "step": 547
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.344432225962619e-05,
+      "loss": 0.7021,
+      "step": 548
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.342144280377936e-05,
+      "loss": 0.7304,
+      "step": 549
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.339852953975005e-05,
+      "loss": 0.7546,
+      "step": 550
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.337558250959013e-05,
+      "loss": 0.7948,
+      "step": 551
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.335260175541342e-05,
+      "loss": 0.8696,
+      "step": 552
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.332958731939562e-05,
+      "loss": 0.8357,
+      "step": 553
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.330653924377428e-05,
+      "loss": 0.7318,
+      "step": 554
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.328345757084866e-05,
+      "loss": 0.7246,
+      "step": 555
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.32603423429797e-05,
+      "loss": 0.7,
+      "step": 556
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3237193602589885e-05,
+      "loss": 0.7644,
+      "step": 557
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3214011392163244e-05,
+      "loss": 0.7991,
+      "step": 558
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.319079575424523e-05,
+      "loss": 0.8297,
+      "step": 559
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3167546731442605e-05,
+      "loss": 0.6975,
+      "step": 560
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3144264366423445e-05,
+      "loss": 0.7131,
+      "step": 561
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3120948701917e-05,
+      "loss": 0.7741,
+      "step": 562
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3097599780713625e-05,
+      "loss": 0.8072,
+      "step": 563
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.307421764566473e-05,
+      "loss": 0.7055,
+      "step": 564
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3050802339682665e-05,
+      "loss": 0.8643,
+      "step": 565
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.302735390574066e-05,
+      "loss": 0.6969,
+      "step": 566
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.3003872386872744e-05,
+      "loss": 0.7327,
+      "step": 567
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.2980357826173665e-05,
+      "loss": 0.8235,
+      "step": 568
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.295681026679883e-05,
+      "loss": 0.7095,
+      "step": 569
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2933229751964175e-05,
+      "loss": 0.7745,
+      "step": 570
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.290961632494614e-05,
+      "loss": 0.7294,
+      "step": 571
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2885970029081555e-05,
+      "loss": 0.7303,
+      "step": 572
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2862290907767585e-05,
+      "loss": 0.7418,
+      "step": 573
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.283857900446163e-05,
+      "loss": 0.8628,
+      "step": 574
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2814834362681255e-05,
+      "loss": 0.7715,
+      "step": 575
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.279105702600412e-05,
+      "loss": 0.675,
+      "step": 576
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.276724703806786e-05,
+      "loss": 0.7764,
+      "step": 577
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2743404442570065e-05,
+      "loss": 0.6354,
+      "step": 578
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2719529283268156e-05,
+      "loss": 0.8398,
+      "step": 579
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.269562160397931e-05,
+      "loss": 0.836,
+      "step": 580
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2671681448580404e-05,
+      "loss": 0.6738,
+      "step": 581
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.264770886100789e-05,
+      "loss": 0.7566,
+      "step": 582
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2623703885257774e-05,
+      "loss": 0.8301,
+      "step": 583
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.259966656538548e-05,
+      "loss": 0.7379,
+      "step": 584
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2575596945505794e-05,
+      "loss": 0.6814,
+      "step": 585
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.255149506979279e-05,
+      "loss": 0.7451,
+      "step": 586
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.2527360982479735e-05,
+      "loss": 0.6818,
+      "step": 587
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.250319472785902e-05,
+      "loss": 0.8787,
+      "step": 588
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.247899635028206e-05,
+      "loss": 0.7882,
+      "step": 589
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.245476589415923e-05,
+      "loss": 0.7852,
+      "step": 590
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.243050340395977e-05,
+      "loss": 0.8433,
+      "step": 591
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.240620892421172e-05,
+      "loss": 0.8116,
+      "step": 592
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.238188249950185e-05,
+      "loss": 0.7848,
+      "step": 593
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.235752417447549e-05,
+      "loss": 0.8715,
+      "step": 594
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.233313399383659e-05,
+      "loss": 0.6482,
+      "step": 595
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.230871200234754e-05,
+      "loss": 0.6538,
+      "step": 596
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2284258244829076e-05,
+      "loss": 0.7536,
+      "step": 597
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2259772766160276e-05,
+      "loss": 0.7742,
+      "step": 598
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2235255611278425e-05,
+      "loss": 0.7807,
+      "step": 599
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.221070682517894e-05,
+      "loss": 0.7651,
+      "step": 600
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.7837547063827515,
+      "eval_runtime": 10.1825,
+      "eval_samples_per_second": 3.634,
+      "eval_steps_per_second": 0.491,
+      "step": 600
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2186126452915256e-05,
+      "loss": 0.8788,
+      "step": 601
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.216151453959885e-05,
+      "loss": 0.7653,
+      "step": 602
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2136871130399015e-05,
+      "loss": 0.8086,
+      "step": 603
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.211219627054288e-05,
+      "loss": 0.6593,
+      "step": 604
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.2087490005315293e-05,
+      "loss": 0.864,
+      "step": 605
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.206275238005873e-05,
+      "loss": 0.7781,
+      "step": 606
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.203798344017324e-05,
+      "loss": 0.8283,
+      "step": 607
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.201318323111631e-05,
+      "loss": 0.8298,
+      "step": 608
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.1988351798402846e-05,
+      "loss": 0.7659,
+      "step": 609
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.1963489187605034e-05,
+      "loss": 0.8664,
+      "step": 610
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.193859544435229e-05,
+      "loss": 0.7306,
+      "step": 611
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.191367061433118e-05,
+      "loss": 0.7797,
+      "step": 612
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.18887147432853e-05,
+      "loss": 0.6415,
+      "step": 613
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.186372787701522e-05,
+      "loss": 0.8717,
+      "step": 614
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.183871006137841e-05,
+      "loss": 0.7534,
+      "step": 615
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.181366134228912e-05,
+      "loss": 0.7723,
+      "step": 616
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.1788581765718316e-05,
+      "loss": 0.7216,
+      "step": 617
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.176347137769362e-05,
+      "loss": 0.8546,
+      "step": 618
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.1738330224299165e-05,
+      "loss": 0.7798,
+      "step": 619
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.171315835167558e-05,
+      "loss": 0.7227,
+      "step": 620
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.1687955806019864e-05,
+      "loss": 0.6509,
+      "step": 621
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.166272263358528e-05,
+      "loss": 0.8394,
+      "step": 622
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.163745888068134e-05,
+      "loss": 0.7652,
+      "step": 623
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.161216459367366e-05,
+      "loss": 0.7998,
+      "step": 624
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.158683981898389e-05,
+      "loss": 0.801,
+      "step": 625
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.156148460308964e-05,
+      "loss": 0.6806,
+      "step": 626
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.153609899252439e-05,
+      "loss": 0.7841,
+      "step": 627
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.15106830338774e-05,
+      "loss": 0.7259,
+      "step": 628
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.148523677379361e-05,
+      "loss": 0.6297,
+      "step": 629
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.14597602589736e-05,
+      "loss": 0.8397,
+      "step": 630
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.143425353617346e-05,
+      "loss": 0.7199,
+      "step": 631
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.1408716652204716e-05,
+      "loss": 0.759,
+      "step": 632
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.138314965393425e-05,
+      "loss": 0.6946,
+      "step": 633
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.135755258828422e-05,
+      "loss": 0.7367,
+      "step": 634
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.133192550223196e-05,
+      "loss": 0.7487,
+      "step": 635
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.130626844280989e-05,
+      "loss": 0.7692,
+      "step": 636
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.128058145710547e-05,
+      "loss": 0.6799,
+      "step": 637
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.125486459226104e-05,
+      "loss": 0.7387,
+      "step": 638
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.12291178954738e-05,
+      "loss": 0.8323,
+      "step": 639
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.12033414139957e-05,
+      "loss": 0.7543,
+      "step": 640
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.117753519513336e-05,
+      "loss": 0.6675,
+      "step": 641
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.115169928624794e-05,
+      "loss": 0.7366,
+      "step": 642
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.1125833734755114e-05,
+      "loss": 0.7067,
+      "step": 643
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.109993858812499e-05,
+      "loss": 0.8237,
+      "step": 644
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.107401389388193e-05,
+      "loss": 0.7359,
+      "step": 645
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.104805969960456e-05,
+      "loss": 0.7633,
+      "step": 646
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.1022076052925637e-05,
+      "loss": 0.7281,
+      "step": 647
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.099606300153198e-05,
+      "loss": 0.7322,
+      "step": 648
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.097002059316436e-05,
+      "loss": 0.905,
+      "step": 649
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0943948875617433e-05,
+      "loss": 0.6952,
+      "step": 650
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0917847896739644e-05,
+      "loss": 0.7417,
+      "step": 651
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.089171770443313e-05,
+      "loss": 0.7173,
+      "step": 652
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0865558346653664e-05,
+      "loss": 0.8249,
+      "step": 653
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0839369871410524e-05,
+      "loss": 0.7726,
+      "step": 654
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.081315232676646e-05,
+      "loss": 0.7356,
+      "step": 655
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0786905760837514e-05,
+      "loss": 0.7732,
+      "step": 656
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0760630221793036e-05,
+      "loss": 0.9094,
+      "step": 657
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.073432575785554e-05,
+      "loss": 0.8432,
+      "step": 658
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.0707992417300616e-05,
+      "loss": 0.7868,
+      "step": 659
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.068163024845686e-05,
+      "loss": 0.8738,
+      "step": 660
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.065523929970578e-05,
+      "loss": 0.7762,
+      "step": 661
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.062881961948167e-05,
+      "loss": 0.9031,
+      "step": 662
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.06023712562716e-05,
+      "loss": 0.8335,
+      "step": 663
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.057589425861525e-05,
+      "loss": 0.7018,
+      "step": 664
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.054938867510485e-05,
+      "loss": 0.785,
+      "step": 665
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.0522854554385116e-05,
+      "loss": 0.7981,
+      "step": 666
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.049629194515312e-05,
+      "loss": 0.8843,
+      "step": 667
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.046970089615822e-05,
+      "loss": 0.737,
+      "step": 668
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.044308145620197e-05,
+      "loss": 0.8855,
+      "step": 669
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.0416433674138024e-05,
+      "loss": 0.7986,
+      "step": 670
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.038975759887207e-05,
+      "loss": 0.8071,
+      "step": 671
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.0363053279361696e-05,
+      "loss": 0.8288,
+      "step": 672
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.033632076461634e-05,
+      "loss": 0.8119,
+      "step": 673
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.0309560103697184e-05,
+      "loss": 0.7528,
+      "step": 674
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.028277134571706e-05,
+      "loss": 0.7865,
+      "step": 675
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.025595453984038e-05,
+      "loss": 0.7289,
+      "step": 676
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.022910973528301e-05,
+      "loss": 0.8222,
+      "step": 677
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.020223698131221e-05,
+      "loss": 0.8173,
+      "step": 678
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.017533632724656e-05,
+      "loss": 0.7644,
+      "step": 679
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.0148407822455794e-05,
+      "loss": 0.8409,
+      "step": 680
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.012145151636079e-05,
+      "loss": 0.665,
+      "step": 681
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.009446745843346e-05,
+      "loss": 0.8076,
+      "step": 682
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.006745569819662e-05,
+      "loss": 0.7214,
+      "step": 683
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.004041628522394e-05,
+      "loss": 0.7404,
+      "step": 684
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.001334926913985e-05,
+      "loss": 0.6325,
+      "step": 685
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.998625469961942e-05,
+      "loss": 0.7213,
+      "step": 686
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.9959132626388294e-05,
+      "loss": 0.6753,
+      "step": 687
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.9931983099222595e-05,
+      "loss": 0.8213,
+      "step": 688
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.990480616794884e-05,
+      "loss": 0.6437,
+      "step": 689
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.987760188244382e-05,
+      "loss": 0.8144,
+      "step": 690
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.985037029263454e-05,
+      "loss": 0.6676,
+      "step": 691
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.982311144849813e-05,
+      "loss": 0.7082,
+      "step": 692
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.97958254000617e-05,
+      "loss": 0.7795,
+      "step": 693
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.976851219740234e-05,
+      "loss": 0.88,
+      "step": 694
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.974117189064691e-05,
+      "loss": 0.7574,
+      "step": 695
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.971380452997208e-05,
+      "loss": 0.7663,
+      "step": 696
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.9686410165604124e-05,
+      "loss": 0.6625,
+      "step": 697
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.965898884781889e-05,
+      "loss": 0.8158,
+      "step": 698
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.963154062694169e-05,
+      "loss": 0.7573,
+      "step": 699
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.9604065553347224e-05,
+      "loss": 0.6229,
+      "step": 700
+    },
+    {
+      "epoch": 0.3,
+      "eval_loss": 0.7801283001899719,
+      "eval_runtime": 10.3197,
+      "eval_samples_per_second": 3.585,
+      "eval_steps_per_second": 0.485,
+      "step": 700
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.957656367745947e-05,
+      "loss": 0.6979,
+      "step": 701
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.954903504975156e-05,
+      "loss": 0.7313,
+      "step": 702
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.952147972074578e-05,
+      "loss": 0.723,
+      "step": 703
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.949389774101337e-05,
+      "loss": 0.7319,
+      "step": 704
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.946628916117452e-05,
+      "loss": 0.651,
+      "step": 705
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.9438654031898204e-05,
+      "loss": 1.0151,
+      "step": 706
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 3.941099240390215e-05,
+      "loss": 0.6635,
+      "step": 707
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9383304327952696e-05,
+      "loss": 0.7306,
+      "step": 708
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.935558985486474e-05,
+      "loss": 0.7589,
+      "step": 709
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9327849035501604e-05,
+      "loss": 0.7903,
+      "step": 710
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9300081920774966e-05,
+      "loss": 0.7878,
+      "step": 711
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9272288561644786e-05,
+      "loss": 0.8944,
+      "step": 712
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9244469009119165e-05,
+      "loss": 1.1365,
+      "step": 713
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.921662331425429e-05,
+      "loss": 0.747,
+      "step": 714
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.918875152815431e-05,
+      "loss": 0.8827,
+      "step": 715
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.916085370197127e-05,
+      "loss": 0.7942,
+      "step": 716
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9132929886905e-05,
+      "loss": 0.8337,
+      "step": 717
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.910498013420304e-05,
+      "loss": 0.7707,
+      "step": 718
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.907700449516051e-05,
+      "loss": 0.8093,
+      "step": 719
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9049003021120055e-05,
+      "loss": 0.7472,
+      "step": 720
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.9020975763471734e-05,
+      "loss": 0.9015,
+      "step": 721
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.8992922773652915e-05,
+      "loss": 0.6248,
+      "step": 722
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.89648441031482e-05,
+      "loss": 0.815,
+      "step": 723
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.893673980348932e-05,
+      "loss": 0.6683,
+      "step": 724
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.890860992625504e-05,
+      "loss": 0.7272,
+      "step": 725
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.8880454523071064e-05,
+      "loss": 0.7075,
+      "step": 726
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.885227364560995e-05,
+      "loss": 0.6867,
+      "step": 727
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.882406734559101e-05,
+      "loss": 0.7838,
+      "step": 728
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.879583567478019e-05,
+      "loss": 0.8641,
+      "step": 729
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 3.876757868499003e-05,
+      "loss": 0.6906,
+      "step": 730
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.873929642807953e-05,
+      "loss": 0.8569,
+      "step": 731
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.871098895595404e-05,
+      "loss": 0.7502,
+      "step": 732
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8682656320565204e-05,
+      "loss": 0.8316,
+      "step": 733
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.865429857391084e-05,
+      "loss": 0.7405,
+      "step": 734
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8625915768034873e-05,
+      "loss": 0.756,
+      "step": 735
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.859750795502718e-05,
+      "loss": 0.7492,
+      "step": 736
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8569075187023575e-05,
+      "loss": 0.8525,
+      "step": 737
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8540617516205624e-05,
+      "loss": 0.9735,
+      "step": 738
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.851213499480064e-05,
+      "loss": 0.7838,
+      "step": 739
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8483627675081524e-05,
+      "loss": 0.9609,
+      "step": 740
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8455095609366693e-05,
+      "loss": 0.8497,
+      "step": 741
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8426538850019964e-05,
+      "loss": 0.7637,
+      "step": 742
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.83979574494505e-05,
+      "loss": 0.6037,
+      "step": 743
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.836935146011266e-05,
+      "loss": 0.8975,
+      "step": 744
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.834072093450594e-05,
+      "loss": 0.6858,
+      "step": 745
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.831206592517488e-05,
+      "loss": 0.8235,
+      "step": 746
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.828338648470895e-05,
+      "loss": 0.7814,
+      "step": 747
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.825468266574243e-05,
+      "loss": 0.7573,
+      "step": 748
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.8225954520954364e-05,
+      "loss": 0.755,
+      "step": 749
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.819720210306845e-05,
+      "loss": 0.7047,
+      "step": 750
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.816842546485292e-05,
+      "loss": 0.7636,
+      "step": 751
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.813962465912045e-05,
+      "loss": 0.8064,
+      "step": 752
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.811079973872809e-05,
+      "loss": 0.8318,
+      "step": 753
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 3.808195075657711e-05,
+      "loss": 0.9713,
+      "step": 754
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.805307776561299e-05,
+      "loss": 0.7643,
+      "step": 755
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.802418081882524e-05,
+      "loss": 0.7808,
+      "step": 756
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.799525996924733e-05,
+      "loss": 0.7491,
+      "step": 757
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.796631526995663e-05,
+      "loss": 0.8423,
+      "step": 758
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.793734677407423e-05,
+      "loss": 0.7888,
+      "step": 759
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.790835453476495e-05,
+      "loss": 0.77,
+      "step": 760
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7879338605237146e-05,
+      "loss": 0.7197,
+      "step": 761
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7850299038742673e-05,
+      "loss": 0.7985,
+      "step": 762
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.782123588857675e-05,
+      "loss": 0.8538,
+      "step": 763
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.779214920807788e-05,
+      "loss": 0.7602,
+      "step": 764
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7763039050627767e-05,
+      "loss": 0.8949,
+      "step": 765
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.773390546965119e-05,
+      "loss": 0.7133,
+      "step": 766
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7704748518615915e-05,
+      "loss": 0.854,
+      "step": 767
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.767556825103259e-05,
+      "loss": 0.6997,
+      "step": 768
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7646364720454675e-05,
+      "loss": 0.9395,
+      "step": 769
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.761713798047831e-05,
+      "loss": 0.6775,
+      "step": 770
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.758788808474224e-05,
+      "loss": 0.8521,
+      "step": 771
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.755861508692769e-05,
+      "loss": 0.8262,
+      "step": 772
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.75293190407583e-05,
+      "loss": 0.7014,
+      "step": 773
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.6883,
+      "step": 774
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.747065801846093e-05,
+      "loss": 0.8857,
+      "step": 775
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.744129314999133e-05,
+      "loss": 0.7464,
+      "step": 776
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 3.741190544848343e-05,
+      "loss": 0.7916,
+      "step": 777
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.7382494967871386e-05,
+      "loss": 0.7719,
+      "step": 778
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.735306176213114e-05,
+      "loss": 0.6991,
+      "step": 779
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.732360588528036e-05,
+      "loss": 0.6965,
+      "step": 780
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.72941273913783e-05,
+      "loss": 0.8097,
+      "step": 781
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.726462633452574e-05,
+      "loss": 0.7607,
+      "step": 782
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.723510276886487e-05,
+      "loss": 0.7492,
+      "step": 783
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.720555674857916e-05,
+      "loss": 0.7734,
+      "step": 784
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.7175988327893317e-05,
+      "loss": 0.8125,
+      "step": 785
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.714639756107317e-05,
+      "loss": 0.7121,
+      "step": 786
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.711678450242554e-05,
+      "loss": 0.6998,
+      "step": 787
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.7087149206298154e-05,
+      "loss": 0.7277,
+      "step": 788
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.7057491727079554e-05,
+      "loss": 0.6993,
+      "step": 789
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.702781211919901e-05,
+      "loss": 0.7536,
+      "step": 790
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6998110437126384e-05,
+      "loss": 0.7398,
+      "step": 791
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6968386735372055e-05,
+      "loss": 0.7772,
+      "step": 792
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6938641068486826e-05,
+      "loss": 0.6984,
+      "step": 793
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.69088734910618e-05,
+      "loss": 0.6349,
+      "step": 794
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6879084057728304e-05,
+      "loss": 0.7508,
+      "step": 795
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.684927282315774e-05,
+      "loss": 0.7468,
+      "step": 796
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.681943984206158e-05,
+      "loss": 0.655,
+      "step": 797
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.678958516919114e-05,
+      "loss": 0.7449,
+      "step": 798
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6759708859337614e-05,
+      "loss": 0.9635,
+      "step": 799
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 3.6729810967331856e-05,
+      "loss": 0.8153,
+      "step": 800
+    },
+    {
+      "epoch": 0.34,
+      "eval_loss": 0.7773626446723938,
+      "eval_runtime": 10.1729,
+      "eval_samples_per_second": 3.637,
+      "eval_steps_per_second": 0.492,
+      "step": 800
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6699891548044336e-05,
+      "loss": 0.8183,
+      "step": 801
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6669950656385065e-05,
+      "loss": 0.7709,
+      "step": 802
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.663998834730341e-05,
+      "loss": 0.6341,
+      "step": 803
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.66100046757881e-05,
+      "loss": 0.826,
+      "step": 804
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.657999969686702e-05,
+      "loss": 0.7063,
+      "step": 805
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.654997346560719e-05,
+      "loss": 0.6664,
+      "step": 806
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.651992603711462e-05,
+      "loss": 0.8019,
+      "step": 807
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.648985746653424e-05,
+      "loss": 0.7915,
+      "step": 808
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.645976780904975e-05,
+      "loss": 0.776,
+      "step": 809
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6429657119883585e-05,
+      "loss": 0.8638,
+      "step": 810
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.639952545429676e-05,
+      "loss": 0.7179,
+      "step": 811
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.636937286758878e-05,
+      "loss": 0.705,
+      "step": 812
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.633919941509757e-05,
+      "loss": 0.8691,
+      "step": 813
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.630900515219932e-05,
+      "loss": 0.792,
+      "step": 814
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6278790134308446e-05,
+      "loss": 0.8007,
+      "step": 815
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6248554416877414e-05,
+      "loss": 0.8518,
+      "step": 816
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.621829805539671e-05,
+      "loss": 0.7176,
+      "step": 817
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.618802110539471e-05,
+      "loss": 0.6979,
+      "step": 818
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6157723622437544e-05,
+      "loss": 0.7744,
+      "step": 819
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.612740566212906e-05,
+      "loss": 0.7161,
+      "step": 820
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.6097067280110665e-05,
+      "loss": 0.8242,
+      "step": 821
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.606670853206125e-05,
+      "loss": 0.7249,
+      "step": 822
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 3.60363294736971e-05,
+      "loss": 0.7814,
+      "step": 823
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.6005930160771755e-05,
+      "loss": 0.74,
+      "step": 824
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5975510649075915e-05,
+      "loss": 0.8421,
+      "step": 825
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.594507099443739e-05,
+      "loss": 0.7091,
+      "step": 826
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.591461125272093e-05,
+      "loss": 0.8217,
+      "step": 827
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5884131479828155e-05,
+      "loss": 0.797,
+      "step": 828
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.585363173169744e-05,
+      "loss": 0.8585,
+      "step": 829
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.582311206430383e-05,
+      "loss": 0.7903,
+      "step": 830
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.579257253365894e-05,
+      "loss": 0.9192,
+      "step": 831
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.57620131958108e-05,
+      "loss": 0.8264,
+      "step": 832
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5731434106843836e-05,
+      "loss": 1.0079,
+      "step": 833
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.57008353228787e-05,
+      "loss": 0.7485,
+      "step": 834
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.567021690007218e-05,
+      "loss": 0.7599,
+      "step": 835
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.563957889461713e-05,
+      "loss": 0.6383,
+      "step": 836
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.560892136274232e-05,
+      "loss": 0.6461,
+      "step": 837
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5578244360712373e-05,
+      "loss": 0.7783,
+      "step": 838
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.5547547944827644e-05,
+      "loss": 0.7189,
+      "step": 839
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.55168321714241e-05,
+      "loss": 0.7878,
+      "step": 840
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.548609709687326e-05,
+      "loss": 0.8456,
+      "step": 841
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.545534277758205e-05,
+      "loss": 0.8697,
+      "step": 842
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.542456926999271e-05,
+      "loss": 0.9097,
+      "step": 843
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.539377663058271e-05,
+      "loss": 0.7874,
+      "step": 844
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.536296491586463e-05,
+      "loss": 0.8056,
+      "step": 845
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 3.533213418238605e-05,
+      "loss": 0.7522,
+      "step": 846
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.530128448672944e-05,
+      "loss": 0.7501,
+      "step": 847
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.527041588551212e-05,
+      "loss": 0.7787,
+      "step": 848
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.523952843538604e-05,
+      "loss": 0.7066,
+      "step": 849
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.520862219303781e-05,
+      "loss": 0.7545,
+      "step": 850
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.5177697215188475e-05,
+      "loss": 0.8294,
+      "step": 851
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.51467535585935e-05,
+      "loss": 0.7757,
+      "step": 852
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.5115791280042606e-05,
+      "loss": 0.7384,
+      "step": 853
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.508481043635971e-05,
+      "loss": 0.8955,
+      "step": 854
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.505381108440279e-05,
+      "loss": 0.6803,
+      "step": 855
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.502279328106377e-05,
+      "loss": 0.8034,
+      "step": 856
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.49917570832685e-05,
+      "loss": 0.8621,
+      "step": 857
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.496070254797652e-05,
+      "loss": 0.782,
+      "step": 858
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.4929629732181064e-05,
+      "loss": 0.7774,
+      "step": 859
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.489853869290889e-05,
+      "loss": 0.7672,
+      "step": 860
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.486742948722024e-05,
+      "loss": 0.7518,
+      "step": 861
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.4836302172208645e-05,
+      "loss": 0.9012,
+      "step": 862
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.480515680500092e-05,
+      "loss": 0.8715,
+      "step": 863
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.477399344275698e-05,
+      "loss": 0.7544,
+      "step": 864
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.474281214266977e-05,
+      "loss": 0.6645,
+      "step": 865
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.471161296196516e-05,
+      "loss": 0.9114,
+      "step": 866
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.468039595790184e-05,
+      "loss": 0.7907,
+      "step": 867
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.4649161187771206e-05,
+      "loss": 0.9009,
+      "step": 868
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.461790870889725e-05,
+      "loss": 0.6086,
+      "step": 869
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 3.458663857863649e-05,
+      "loss": 0.8777,
+      "step": 870
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.455535085437781e-05,
+      "loss": 0.6795,
+      "step": 871
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4524045593542386e-05,
+      "loss": 0.6594,
+      "step": 872
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.449272285358361e-05,
+      "loss": 0.6547,
+      "step": 873
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.446138269198692e-05,
+      "loss": 0.8464,
+      "step": 874
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4430025166269744e-05,
+      "loss": 0.7462,
+      "step": 875
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4398650333981354e-05,
+      "loss": 0.8541,
+      "step": 876
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.436725825270283e-05,
+      "loss": 0.8847,
+      "step": 877
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.433584898004685e-05,
+      "loss": 0.6683,
+      "step": 878
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.43044225736577e-05,
+      "loss": 0.7771,
+      "step": 879
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.427297909121107e-05,
+      "loss": 0.7601,
+      "step": 880
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4241518590414e-05,
+      "loss": 0.8268,
+      "step": 881
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.421004112900477e-05,
+      "loss": 0.7158,
+      "step": 882
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.417854676475279e-05,
+      "loss": 0.7148,
+      "step": 883
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4147035555458477e-05,
+      "loss": 0.7187,
+      "step": 884
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4115507558953174e-05,
+      "loss": 0.8365,
+      "step": 885
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.408396283309903e-05,
+      "loss": 0.7431,
+      "step": 886
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4052401435788885e-05,
+      "loss": 0.7332,
+      "step": 887
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.4020823424946205e-05,
+      "loss": 0.8885,
+      "step": 888
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.3989228858524914e-05,
+      "loss": 0.6457,
+      "step": 889
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.3957617794509344e-05,
+      "loss": 0.7648,
+      "step": 890
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.392599029091409e-05,
+      "loss": 1.0406,
+      "step": 891
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.389434640578392e-05,
+      "loss": 0.7927,
+      "step": 892
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 3.386268619719366e-05,
+      "loss": 0.8087,
+      "step": 893
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.383100972324812e-05,
+      "loss": 0.7671,
+      "step": 894
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.379931704208193e-05,
+      "loss": 0.8655,
+      "step": 895
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3767608211859495e-05,
+      "loss": 0.7414,
+      "step": 896
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.373588329077482e-05,
+      "loss": 0.7659,
+      "step": 897
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.370414233705147e-05,
+      "loss": 0.6748,
+      "step": 898
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.367238540894243e-05,
+      "loss": 0.6292,
+      "step": 899
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3640612564729996e-05,
+      "loss": 0.6958,
+      "step": 900
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.775384783744812,
+      "eval_runtime": 10.201,
+      "eval_samples_per_second": 3.627,
+      "eval_steps_per_second": 0.49,
+      "step": 900
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3608823862725675e-05,
+      "loss": 0.6352,
+      "step": 901
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.357701936127008e-05,
+      "loss": 0.7488,
+      "step": 902
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.354519911873282e-05,
+      "loss": 0.6704,
+      "step": 903
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.351336319351238e-05,
+      "loss": 0.8388,
+      "step": 904
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.348151164403605e-05,
+      "loss": 0.7291,
+      "step": 905
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.344964452875978e-05,
+      "loss": 0.7755,
+      "step": 906
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.34177619061681e-05,
+      "loss": 0.8571,
+      "step": 907
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3385863834773975e-05,
+      "loss": 0.767,
+      "step": 908
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.335395037311874e-05,
+      "loss": 0.8152,
+      "step": 909
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.332202157977198e-05,
+      "loss": 0.6353,
+      "step": 910
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.329007751333142e-05,
+      "loss": 0.9138,
+      "step": 911
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3258118232422787e-05,
+      "loss": 0.8154,
+      "step": 912
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.322614379569975e-05,
+      "loss": 0.7702,
+      "step": 913
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.319415426184381e-05,
+      "loss": 0.7814,
+      "step": 914
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3162149689564135e-05,
+      "loss": 0.788,
+      "step": 915
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 3.3130130137597546e-05,
+      "loss": 0.932,
+      "step": 916
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.309809566470829e-05,
+      "loss": 0.8445,
+      "step": 917
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.3066046329688056e-05,
+      "loss": 0.8815,
+      "step": 918
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.303398219135578e-05,
+      "loss": 0.861,
+      "step": 919
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.300190330855756e-05,
+      "loss": 0.7331,
+      "step": 920
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2969809740166583e-05,
+      "loss": 0.7393,
+      "step": 921
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.293770154508296e-05,
+      "loss": 0.8168,
+      "step": 922
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2905578782233665e-05,
+      "loss": 0.8113,
+      "step": 923
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.287344151057239e-05,
+      "loss": 0.7994,
+      "step": 924
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2841289789079465e-05,
+      "loss": 0.7081,
+      "step": 925
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.280912367676173e-05,
+      "loss": 0.6252,
+      "step": 926
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2776943232652465e-05,
+      "loss": 0.7764,
+      "step": 927
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.27447485158112e-05,
+      "loss": 0.7369,
+      "step": 928
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.271253958532372e-05,
+      "loss": 0.7788,
+      "step": 929
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.268031650030186e-05,
+      "loss": 0.7606,
+      "step": 930
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.264807931988342e-05,
+      "loss": 0.9283,
+      "step": 931
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.261582810323211e-05,
+      "loss": 0.7162,
+      "step": 932
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.258356290953736e-05,
+      "loss": 0.8287,
+      "step": 933
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2551283798014285e-05,
+      "loss": 0.7492,
+      "step": 934
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.25189908279035e-05,
+      "loss": 0.7372,
+      "step": 935
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.248668405847111e-05,
+      "loss": 0.6945,
+      "step": 936
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.24543635490085e-05,
+      "loss": 0.6628,
+      "step": 937
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.24220293588323e-05,
+      "loss": 0.7582,
+      "step": 938
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 3.2389681547284224e-05,
+      "loss": 0.8779,
+      "step": 939
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.2357320173731e-05,
+      "loss": 0.6828,
+      "step": 940
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.2324945297564236e-05,
+      "loss": 0.7707,
+      "step": 941
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.229255697820034e-05,
+      "loss": 0.8004,
+      "step": 942
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.2260155275080363e-05,
+      "loss": 0.6871,
+      "step": 943
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.222774024766995e-05,
+      "loss": 0.9631,
+      "step": 944
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.219531195545918e-05,
+      "loss": 0.7378,
+      "step": 945
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.216287045796247e-05,
+      "loss": 0.8937,
+      "step": 946
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.213041581471849e-05,
+      "loss": 0.7671,
+      "step": 947
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.209794808529004e-05,
+      "loss": 0.6912,
+      "step": 948
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.20654673292639e-05,
+      "loss": 0.8395,
+      "step": 949
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.2032973606250804e-05,
+      "loss": 0.8311,
+      "step": 950
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.200046697588526e-05,
+      "loss": 0.7332,
+      "step": 951
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.196794749782544e-05,
+      "loss": 0.6453,
+      "step": 952
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.1935415231753155e-05,
+      "loss": 0.7365,
+      "step": 953
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.190287023737364e-05,
+      "loss": 0.8021,
+      "step": 954
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.187031257441549e-05,
+      "loss": 0.7917,
+      "step": 955
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.183774230263058e-05,
+      "loss": 0.813,
+      "step": 956
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.18051594817939e-05,
+      "loss": 0.768,
+      "step": 957
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.1772564171703475e-05,
+      "loss": 0.7123,
+      "step": 958
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.173995643218025e-05,
+      "loss": 0.836,
+      "step": 959
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.1707336323068e-05,
+      "loss": 0.7068,
+      "step": 960
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.167470390423317e-05,
+      "loss": 0.6678,
+      "step": 961
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.164205923556483e-05,
+      "loss": 0.699,
+      "step": 962
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 3.1609402376974495e-05,
+      "loss": 0.8364,
+      "step": 963
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.157673338839609e-05,
+      "loss": 0.6276,
+      "step": 964
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.154405232978577e-05,
+      "loss": 0.627,
+      "step": 965
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.151135926112187e-05,
+      "loss": 0.6027,
+      "step": 966
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.147865424240474e-05,
+      "loss": 0.6496,
+      "step": 967
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.144593733365667e-05,
+      "loss": 0.7317,
+      "step": 968
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.14132085949218e-05,
+      "loss": 0.7783,
+      "step": 969
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.138046808626591e-05,
+      "loss": 0.7837,
+      "step": 970
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.134771586777647e-05,
+      "loss": 0.7783,
+      "step": 971
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.131495199956237e-05,
+      "loss": 0.7335,
+      "step": 972
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.128217654175393e-05,
+      "loss": 0.7112,
+      "step": 973
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.124938955450269e-05,
+      "loss": 0.8552,
+      "step": 974
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.1216591097981394e-05,
+      "loss": 0.7491,
+      "step": 975
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.118378123238383e-05,
+      "loss": 0.8029,
+      "step": 976
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.1150960017924685e-05,
+      "loss": 0.6775,
+      "step": 977
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.111812751483952e-05,
+      "loss": 0.8973,
+      "step": 978
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.1085283783384586e-05,
+      "loss": 0.7634,
+      "step": 979
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.1052428883836764e-05,
+      "loss": 0.8199,
+      "step": 980
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.1019562876493397e-05,
+      "loss": 0.6868,
+      "step": 981
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.0986685821672264e-05,
+      "loss": 0.6868,
+      "step": 982
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.095379777971135e-05,
+      "loss": 0.8666,
+      "step": 983
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.092089881096888e-05,
+      "loss": 0.6998,
+      "step": 984
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.088798897582308e-05,
+      "loss": 0.6144,
+      "step": 985
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 3.085506833467214e-05,
+      "loss": 0.9385,
+      "step": 986
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.082213694793407e-05,
+      "loss": 0.7249,
+      "step": 987
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0789194876046614e-05,
+      "loss": 0.8582,
+      "step": 988
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.075624217946712e-05,
+      "loss": 0.7414,
+      "step": 989
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.072327891867243e-05,
+      "loss": 0.8248,
+      "step": 990
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.069030515415878e-05,
+      "loss": 0.7254,
+      "step": 991
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0657320946441686e-05,
+      "loss": 0.6965,
+      "step": 992
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0624326356055824e-05,
+      "loss": 0.7979,
+      "step": 993
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.059132144355494e-05,
+      "loss": 0.6902,
+      "step": 994
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0558306269511686e-05,
+      "loss": 0.7549,
+      "step": 995
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.052528089451759e-05,
+      "loss": 0.757,
+      "step": 996
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0492245379182877e-05,
+      "loss": 0.7158,
+      "step": 997
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.045919978413639e-05,
+      "loss": 0.7439,
+      "step": 998
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0426144170025456e-05,
+      "loss": 0.7263,
+      "step": 999
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0393078597515827e-05,
+      "loss": 0.8714,
+      "step": 1000
+    },
+    {
+      "epoch": 0.43,
+      "eval_loss": 0.7731763124465942,
+      "eval_runtime": 10.1599,
+      "eval_samples_per_second": 3.642,
+      "eval_steps_per_second": 0.492,
+      "step": 1000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0360003127291494e-05,
+      "loss": 0.7357,
+      "step": 1001
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0326917820054622e-05,
+      "loss": 0.6703,
+      "step": 1002
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.029382273652544e-05,
+      "loss": 0.7843,
+      "step": 1003
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0260717937442113e-05,
+      "loss": 0.8579,
+      "step": 1004
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.022760348356063e-05,
+      "loss": 0.8031,
+      "step": 1005
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.019447943565471e-05,
+      "loss": 0.763,
+      "step": 1006
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0161345854515672e-05,
+      "loss": 0.9019,
+      "step": 1007
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.012820280095234e-05,
+      "loss": 0.7761,
+      "step": 1008
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 3.0095050335790903e-05,
+      "loss": 0.6707,
+      "step": 1009
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.006188851987485e-05,
+      "loss": 0.7175,
+      "step": 1010
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 3.0028717414064805e-05,
+      "loss": 0.8008,
+      "step": 1011
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9995537079238467e-05,
+      "loss": 0.7714,
+      "step": 1012
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9962347576290444e-05,
+      "loss": 0.6841,
+      "step": 1013
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.99291489661322e-05,
+      "loss": 0.8408,
+      "step": 1014
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9895941309691883e-05,
+      "loss": 0.8279,
+      "step": 1015
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9862724667914265e-05,
+      "loss": 0.7386,
+      "step": 1016
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.982949910176061e-05,
+      "loss": 0.7823,
+      "step": 1017
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9796264672208535e-05,
+      "loss": 0.7185,
+      "step": 1018
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9763021440251954e-05,
+      "loss": 0.7604,
+      "step": 1019
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9729769466900918e-05,
+      "loss": 0.8062,
+      "step": 1020
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9696508813181534e-05,
+      "loss": 0.6379,
+      "step": 1021
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9663239540135818e-05,
+      "loss": 0.6967,
+      "step": 1022
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9629961708821618e-05,
+      "loss": 0.695,
+      "step": 1023
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9596675380312507e-05,
+      "loss": 0.8031,
+      "step": 1024
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9563380615697605e-05,
+      "loss": 0.7674,
+      "step": 1025
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.953007747608158e-05,
+      "loss": 0.8916,
+      "step": 1026
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9496766022584398e-05,
+      "loss": 0.8159,
+      "step": 1027
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9463446316341346e-05,
+      "loss": 0.7979,
+      "step": 1028
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.943011841850281e-05,
+      "loss": 0.729,
+      "step": 1029
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9396782390234245e-05,
+      "loss": 0.9,
+      "step": 1030
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.9363438292716004e-05,
+      "loss": 0.8511,
+      "step": 1031
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.933008618714327e-05,
+      "loss": 0.923,
+      "step": 1032
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.9296726134725887e-05,
+      "loss": 0.8954,
+      "step": 1033
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.926335819668832e-05,
+      "loss": 0.8655,
+      "step": 1034
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.922998243426951e-05,
+      "loss": 0.8657,
+      "step": 1035
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.919659890872271e-05,
+      "loss": 0.8267,
+      "step": 1036
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.9163207681315476e-05,
+      "loss": 0.6383,
+      "step": 1037
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.912980881332945e-05,
+      "loss": 0.7495,
+      "step": 1038
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.909640236606035e-05,
+      "loss": 0.748,
+      "step": 1039
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.906298840081775e-05,
+      "loss": 0.8319,
+      "step": 1040
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.9029566978925055e-05,
+      "loss": 0.8384,
+      "step": 1041
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.899613816171935e-05,
+      "loss": 0.8418,
+      "step": 1042
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8962702010551285e-05,
+      "loss": 0.6754,
+      "step": 1043
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8929258586784974e-05,
+      "loss": 0.8605,
+      "step": 1044
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8895807951797866e-05,
+      "loss": 0.7574,
+      "step": 1045
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8862350166980673e-05,
+      "loss": 0.6856,
+      "step": 1046
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8828885293737206e-05,
+      "loss": 0.676,
+      "step": 1047
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8795413393484288e-05,
+      "loss": 0.7304,
+      "step": 1048
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8761934527651642e-05,
+      "loss": 0.8223,
+      "step": 1049
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8728448757681775e-05,
+      "loss": 0.9364,
+      "step": 1050
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.869495614502986e-05,
+      "loss": 0.8179,
+      "step": 1051
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8661456751163633e-05,
+      "loss": 0.8226,
+      "step": 1052
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8627950637563283e-05,
+      "loss": 0.8599,
+      "step": 1053
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8594437865721303e-05,
+      "loss": 0.8034,
+      "step": 1054
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.8560918497142443e-05,
+      "loss": 0.7842,
+      "step": 1055
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.852739259334354e-05,
+      "loss": 0.8364,
+      "step": 1056
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8493860215853413e-05,
+      "loss": 0.6798,
+      "step": 1057
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8460321426212793e-05,
+      "loss": 0.6865,
+      "step": 1058
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.842677628597415e-05,
+      "loss": 0.7019,
+      "step": 1059
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.839322485670163e-05,
+      "loss": 0.769,
+      "step": 1060
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.83596671999709e-05,
+      "loss": 0.7563,
+      "step": 1061
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8326103377369073e-05,
+      "loss": 0.7969,
+      "step": 1062
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.829253345049458e-05,
+      "loss": 0.7958,
+      "step": 1063
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.825895748095703e-05,
+      "loss": 0.7038,
+      "step": 1064
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8225375530377162e-05,
+      "loss": 0.7533,
+      "step": 1065
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8191787660386653e-05,
+      "loss": 0.7245,
+      "step": 1066
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8158193932628068e-05,
+      "loss": 0.6946,
+      "step": 1067
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.812459440875471e-05,
+      "loss": 0.7737,
+      "step": 1068
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8090989150430523e-05,
+      "loss": 0.6086,
+      "step": 1069
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.8057378219329983e-05,
+      "loss": 0.7691,
+      "step": 1070
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.802376167713796e-05,
+      "loss": 0.6517,
+      "step": 1071
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.7990139585549646e-05,
+      "loss": 0.7769,
+      "step": 1072
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.7956512006270387e-05,
+      "loss": 0.7431,
+      "step": 1073
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.7922879001015632e-05,
+      "loss": 0.7643,
+      "step": 1074
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.788924063151076e-05,
+      "loss": 0.8355,
+      "step": 1075
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.7855596959491015e-05,
+      "loss": 0.6971,
+      "step": 1076
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.782194804670136e-05,
+      "loss": 0.8161,
+      "step": 1077
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.77882939548964e-05,
+      "loss": 0.6974,
+      "step": 1078
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 2.77546347458402e-05,
+      "loss": 0.7395,
+      "step": 1079
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7720970481306258e-05,
+      "loss": 0.671,
+      "step": 1080
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7687301223077332e-05,
+      "loss": 0.64,
+      "step": 1081
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7653627032945346e-05,
+      "loss": 0.7831,
+      "step": 1082
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.761994797271129e-05,
+      "loss": 0.755,
+      "step": 1083
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7586264104185055e-05,
+      "loss": 0.656,
+      "step": 1084
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.75525754891854e-05,
+      "loss": 0.7147,
+      "step": 1085
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7518882189539762e-05,
+      "loss": 0.7379,
+      "step": 1086
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.74851842670842e-05,
+      "loss": 0.6838,
+      "step": 1087
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7451481783663246e-05,
+      "loss": 0.8398,
+      "step": 1088
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.741777480112979e-05,
+      "loss": 0.7378,
+      "step": 1089
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7384063381345017e-05,
+      "loss": 0.9643,
+      "step": 1090
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7350347586178194e-05,
+      "loss": 0.8147,
+      "step": 1091
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7316627477506688e-05,
+      "loss": 0.6773,
+      "step": 1092
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7282903117215724e-05,
+      "loss": 0.7694,
+      "step": 1093
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.724917456719837e-05,
+      "loss": 0.82,
+      "step": 1094
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7215441889355364e-05,
+      "loss": 0.642,
+      "step": 1095
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.718170514559502e-05,
+      "loss": 0.7938,
+      "step": 1096
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7147964397833127e-05,
+      "loss": 0.8053,
+      "step": 1097
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7114219707992798e-05,
+      "loss": 0.7755,
+      "step": 1098
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7080471138004405e-05,
+      "loss": 0.7722,
+      "step": 1099
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7046718749805422e-05,
+      "loss": 0.8225,
+      "step": 1100
+    },
+    {
+      "epoch": 0.47,
+      "eval_loss": 0.7722584009170532,
+      "eval_runtime": 10.1954,
+      "eval_samples_per_second": 3.629,
+      "eval_steps_per_second": 0.49,
+      "step": 1100
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2320,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 1.5468848970989568e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}