diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,12171 @@
+{
+  "best_metric": 0.7390481233596802,
+  "best_model_checkpoint": "./qlora-NeurIPS_repo_rerun3/checkpoint-1984",
+  "epoch": 1.982017982017982,
+  "eval_steps": 64,
+  "global_step": 1984,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.0000000000000002e-07,
+      "loss": 1.2637,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.0000000000000003e-07,
+      "loss": 1.4686,
+      "step": 2
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 6.000000000000001e-07,
+      "loss": 1.5486,
+      "step": 3
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 8.000000000000001e-07,
+      "loss": 1.3901,
+      "step": 4
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 1.4859,
+      "step": 5
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.2000000000000002e-06,
+      "loss": 1.3576,
+      "step": 6
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.4000000000000001e-06,
+      "loss": 1.344,
+      "step": 7
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.6000000000000001e-06,
+      "loss": 1.2745,
+      "step": 8
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 1.8000000000000001e-06,
+      "loss": 1.375,
+      "step": 9
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.4734,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.2e-06,
+      "loss": 1.2147,
+      "step": 11
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.4000000000000003e-06,
+      "loss": 1.3805,
+      "step": 12
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6e-06,
+      "loss": 1.2849,
+      "step": 13
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.8000000000000003e-06,
+      "loss": 1.3019,
+      "step": 14
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 3e-06,
+      "loss": 1.2019,
+      "step": 15
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.2000000000000003e-06,
+      "loss": 1.1411,
+      "step": 16
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.4000000000000005e-06,
+      "loss": 1.3422,
+      "step": 17
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.6000000000000003e-06,
+      "loss": 1.1507,
+      "step": 18
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 3.8000000000000005e-06,
+      "loss": 1.0728,
+      "step": 19
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.0169,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.2000000000000004e-06,
+      "loss": 0.9978,
+      "step": 21
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.4e-06,
+      "loss": 1.177,
+      "step": 22
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.600000000000001e-06,
+      "loss": 1.1552,
+      "step": 23
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.800000000000001e-06,
+      "loss": 1.0804,
+      "step": 24
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 5e-06,
+      "loss": 1.0948,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.2e-06,
+      "loss": 1.0599,
+      "step": 26
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.400000000000001e-06,
+      "loss": 1.0642,
+      "step": 27
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.600000000000001e-06,
+      "loss": 1.0911,
+      "step": 28
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.8e-06,
+      "loss": 1.0905,
+      "step": 29
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6e-06,
+      "loss": 0.982,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.200000000000001e-06,
+      "loss": 1.1842,
+      "step": 31
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.4000000000000006e-06,
+      "loss": 1.0504,
+      "step": 32
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.600000000000001e-06,
+      "loss": 0.9268,
+      "step": 33
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 6.800000000000001e-06,
+      "loss": 1.0374,
+      "step": 34
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 7e-06,
+      "loss": 1.0,
+      "step": 35
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 1.1074,
+      "step": 36
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.4e-06,
+      "loss": 0.925,
+      "step": 37
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 0.991,
+      "step": 38
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 7.800000000000002e-06,
+      "loss": 0.9742,
+      "step": 39
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.0463,
+      "step": 40
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.2e-06,
+      "loss": 0.9318,
+      "step": 41
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 1.0624,
+      "step": 42
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.6e-06,
+      "loss": 0.9967,
+      "step": 43
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 8.8e-06,
+      "loss": 1.0015,
+      "step": 44
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9e-06,
+      "loss": 0.9986,
+      "step": 45
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 0.9808,
+      "step": 46
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.4e-06,
+      "loss": 0.9676,
+      "step": 47
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 0.8693,
+      "step": 48
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.800000000000001e-06,
+      "loss": 1.0204,
+      "step": 49
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1e-05,
+      "loss": 0.872,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.02e-05,
+      "loss": 0.8843,
+      "step": 51
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.04e-05,
+      "loss": 0.9736,
+      "step": 52
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0600000000000002e-05,
+      "loss": 0.9687,
+      "step": 53
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.0800000000000002e-05,
+      "loss": 0.899,
+      "step": 54
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.1000000000000001e-05,
+      "loss": 0.9228,
+      "step": 55
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1200000000000001e-05,
+      "loss": 1.0134,
+      "step": 56
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.14e-05,
+      "loss": 0.7985,
+      "step": 57
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.16e-05,
+      "loss": 0.9846,
+      "step": 58
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.18e-05,
+      "loss": 0.9279,
+      "step": 59
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2e-05,
+      "loss": 0.9569,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.22e-05,
+      "loss": 0.9017,
+      "step": 61
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2400000000000002e-05,
+      "loss": 0.8809,
+      "step": 62
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2600000000000001e-05,
+      "loss": 0.9701,
+      "step": 63
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.2800000000000001e-05,
+      "loss": 0.8555,
+      "step": 64
+    },
+    {
+      "epoch": 0.06,
+      "eval_loss": 0.9695873856544495,
+      "eval_runtime": 88.8625,
+      "eval_samples_per_second": 22.507,
+      "eval_steps_per_second": 11.253,
+      "step": 64
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 0.9658,
+      "step": 65
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.3200000000000002e-05,
+      "loss": 1.069,
+      "step": 66
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.3400000000000002e-05,
+      "loss": 0.8698,
+      "step": 67
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.3600000000000002e-05,
+      "loss": 0.8364,
+      "step": 68
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.38e-05,
+      "loss": 0.9105,
+      "step": 69
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4e-05,
+      "loss": 1.0611,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4200000000000001e-05,
+      "loss": 0.9089,
+      "step": 71
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.4400000000000001e-05,
+      "loss": 0.9374,
+      "step": 72
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.46e-05,
+      "loss": 0.7666,
+      "step": 73
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.48e-05,
+      "loss": 0.943,
+      "step": 74
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.9233,
+      "step": 75
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.5200000000000002e-05,
+      "loss": 0.9147,
+      "step": 76
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.54e-05,
+      "loss": 0.8846,
+      "step": 77
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.5600000000000003e-05,
+      "loss": 0.8737,
+      "step": 78
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.58e-05,
+      "loss": 0.9275,
+      "step": 79
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.9828,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.62e-05,
+      "loss": 0.9551,
+      "step": 81
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.64e-05,
+      "loss": 0.8687,
+      "step": 82
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.66e-05,
+      "loss": 0.8925,
+      "step": 83
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.6800000000000002e-05,
+      "loss": 1.0557,
+      "step": 84
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.7e-05,
+      "loss": 0.8112,
+      "step": 85
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.72e-05,
+      "loss": 0.8072,
+      "step": 86
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.7400000000000003e-05,
+      "loss": 1.07,
+      "step": 87
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.76e-05,
+      "loss": 1.0323,
+      "step": 88
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.7800000000000002e-05,
+      "loss": 0.8928,
+      "step": 89
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8755,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.8200000000000002e-05,
+      "loss": 0.8265,
+      "step": 91
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.8400000000000003e-05,
+      "loss": 0.8791,
+      "step": 92
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.86e-05,
+      "loss": 0.9174,
+      "step": 93
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.88e-05,
+      "loss": 0.9284,
+      "step": 94
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 1.9e-05,
+      "loss": 0.8523,
+      "step": 95
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9200000000000003e-05,
+      "loss": 0.8372,
+      "step": 96
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.94e-05,
+      "loss": 0.964,
+      "step": 97
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9600000000000002e-05,
+      "loss": 0.9476,
+      "step": 98
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.98e-05,
+      "loss": 0.9811,
+      "step": 99
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2e-05,
+      "loss": 0.8888,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999999414434292e-05,
+      "loss": 0.83,
+      "step": 101
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999976577378534e-05,
+      "loss": 0.8956,
+      "step": 102
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999947299127415e-05,
+      "loss": 0.9507,
+      "step": 103
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.9999906309623852e-05,
+      "loss": 1.0065,
+      "step": 104
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.999985360891585e-05,
+      "loss": 0.8595,
+      "step": 105
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999978919706513e-05,
+      "loss": 0.8325,
+      "step": 106
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999713074147126e-05,
+      "loss": 0.8277,
+      "step": 107
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999625240250984e-05,
+      "loss": 0.8909,
+      "step": 108
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999525695479575e-05,
+      "loss": 0.8998,
+      "step": 109
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999941443994947e-05,
+      "loss": 0.7963,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9999291473790975e-05,
+      "loss": 0.763,
+      "step": 111
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999915679714809e-05,
+      "loss": 0.7929,
+      "step": 112
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.999901041017855e-05,
+      "loss": 1.0158,
+      "step": 113
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9998852313053783e-05,
+      "loss": 0.9339,
+      "step": 114
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.9998682505958945e-05,
+      "loss": 0.8396,
+      "step": 115
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9998500989092904e-05,
+      "loss": 1.0167,
+      "step": 116
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9998307762668236e-05,
+      "loss": 0.9133,
+      "step": 117
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9998102826911242e-05,
+      "loss": 0.7984,
+      "step": 118
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999788618206192e-05,
+      "loss": 0.975,
+      "step": 119
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9997657828373996e-05,
+      "loss": 0.9946,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9997417766114896e-05,
+      "loss": 0.9567,
+      "step": 121
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999716599556577e-05,
+      "loss": 0.8376,
+      "step": 122
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9996902517021478e-05,
+      "loss": 1.0126,
+      "step": 123
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.999662733079058e-05,
+      "loss": 0.9379,
+      "step": 124
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 1.9996340437195354e-05,
+      "loss": 0.9359,
+      "step": 125
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9996041836571794e-05,
+      "loss": 0.9905,
+      "step": 126
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9995731529269605e-05,
+      "loss": 0.8234,
+      "step": 127
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999540951565219e-05,
+      "loss": 0.8067,
+      "step": 128
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.9128402471542358,
+      "eval_runtime": 88.8003,
+      "eval_samples_per_second": 22.522,
+      "eval_steps_per_second": 11.261,
+      "step": 128
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9995075796096674e-05,
+      "loss": 1.0005,
+      "step": 129
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9994730370993887e-05,
+      "loss": 0.8068,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9994373240748362e-05,
+      "loss": 0.897,
+      "step": 131
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999400440577835e-05,
+      "loss": 0.8146,
+      "step": 132
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9993623866515804e-05,
+      "loss": 0.9914,
+      "step": 133
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.9993231623406385e-05,
+      "loss": 0.9953,
+      "step": 134
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.999282767690946e-05,
+      "loss": 0.8931,
+      "step": 135
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9992412027498106e-05,
+      "loss": 0.7608,
+      "step": 136
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.99919846756591e-05,
+      "loss": 0.9624,
+      "step": 137
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.999154562189293e-05,
+      "loss": 0.8888,
+      "step": 138
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9991094866713788e-05,
+      "loss": 0.918,
+      "step": 139
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.999063241064956e-05,
+      "loss": 0.913,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.999015825424185e-05,
+      "loss": 0.9005,
+      "step": 141
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9989672398045955e-05,
+      "loss": 0.9251,
+      "step": 142
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9989174842630877e-05,
+      "loss": 0.8851,
+      "step": 143
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9988665588579314e-05,
+      "loss": 0.9148,
+      "step": 144
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 1.9988144636487675e-05,
+      "loss": 0.9347,
+      "step": 145
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9987611986966065e-05,
+      "loss": 0.8989,
+      "step": 146
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.998706764063828e-05,
+      "loss": 0.8544,
+      "step": 147
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9986511598141825e-05,
+      "loss": 0.9415,
+      "step": 148
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9985943860127896e-05,
+      "loss": 1.0248,
+      "step": 149
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9985364427261394e-05,
+      "loss": 0.81,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.998477330022091e-05,
+      "loss": 0.8744,
+      "step": 151
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9984170479698724e-05,
+      "loss": 0.8918,
+      "step": 152
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9983555966400828e-05,
+      "loss": 0.8638,
+      "step": 153
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9982929761046893e-05,
+      "loss": 0.9318,
+      "step": 154
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.9982291864370284e-05,
+      "loss": 0.9215,
+      "step": 155
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.998164227711807e-05,
+      "loss": 0.9694,
+      "step": 156
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9980981000051e-05,
+      "loss": 0.9123,
+      "step": 157
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.998030803394351e-05,
+      "loss": 0.7667,
+      "step": 158
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997962337958374e-05,
+      "loss": 0.7759,
+      "step": 159
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997892703777351e-05,
+      "loss": 0.888,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9978219009328323e-05,
+      "loss": 0.8957,
+      "step": 161
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9977499295077375e-05,
+      "loss": 0.8465,
+      "step": 162
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9976767895863548e-05,
+      "loss": 0.8462,
+      "step": 163
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9976024812543408e-05,
+      "loss": 0.7286,
+      "step": 164
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.997527004598719e-05,
+      "loss": 0.7593,
+      "step": 165
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9974503597078845e-05,
+      "loss": 0.7721,
+      "step": 166
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997372546671597e-05,
+      "loss": 0.8131,
+      "step": 167
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9972935655809868e-05,
+      "loss": 0.8082,
+      "step": 168
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9972134165285504e-05,
+      "loss": 0.837,
+      "step": 169
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997132099608153e-05,
+      "loss": 0.8104,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.997049614915028e-05,
+      "loss": 1.0434,
+      "step": 171
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9969659625457752e-05,
+      "loss": 0.9215,
+      "step": 172
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9968811425983624e-05,
+      "loss": 0.857,
+      "step": 173
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.996795155172125e-05,
+      "loss": 0.8125,
+      "step": 174
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 1.9967080003677664e-05,
+      "loss": 0.8614,
+      "step": 175
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9966196782873553e-05,
+      "loss": 0.9722,
+      "step": 176
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9965301890343293e-05,
+      "loss": 0.86,
+      "step": 177
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9964395327134913e-05,
+      "loss": 0.9142,
+      "step": 178
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996347709431012e-05,
+      "loss": 0.921,
+      "step": 179
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996254719294429e-05,
+      "loss": 0.9794,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996160562412645e-05,
+      "loss": 1.0082,
+      "step": 181
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.996065238895931e-05,
+      "loss": 0.8259,
+      "step": 182
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.995968748855923e-05,
+      "loss": 0.9337,
+      "step": 183
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9958710924056238e-05,
+      "loss": 0.9283,
+      "step": 184
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.9957722696594018e-05,
+      "loss": 0.9883,
+      "step": 185
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9956722807329913e-05,
+      "loss": 0.8241,
+      "step": 186
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9955711257434918e-05,
+      "loss": 0.8128,
+      "step": 187
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9954688048093705e-05,
+      "loss": 0.7865,
+      "step": 188
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9953653180504578e-05,
+      "loss": 0.8958,
+      "step": 189
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9952606655879504e-05,
+      "loss": 0.8916,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.99515484754441e-05,
+      "loss": 0.9097,
+      "step": 191
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9950478640437635e-05,
+      "loss": 0.9674,
+      "step": 192
+    },
+    {
+      "epoch": 0.19,
+      "eval_loss": 0.8853557109832764,
+      "eval_runtime": 88.747,
+      "eval_samples_per_second": 22.536,
+      "eval_steps_per_second": 11.268,
+      "step": 192
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.994939715211303e-05,
+      "loss": 0.8201,
+      "step": 193
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9948304011736845e-05,
+      "loss": 0.9528,
+      "step": 194
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.9947199220589292e-05,
+      "loss": 0.7663,
+      "step": 195
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.994608277996423e-05,
+      "loss": 0.8377,
+      "step": 196
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9944954691169155e-05,
+      "loss": 0.7881,
+      "step": 197
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.994381495552521e-05,
+      "loss": 0.8061,
+      "step": 198
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9942663574367167e-05,
+      "loss": 0.8536,
+      "step": 199
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9941500549043452e-05,
+      "loss": 0.9349,
+      "step": 200
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.994032588091612e-05,
+      "loss": 0.8602,
+      "step": 201
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.993913957136086e-05,
+      "loss": 0.9261,
+      "step": 202
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9937941621766998e-05,
+      "loss": 0.8576,
+      "step": 203
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.993673203353749e-05,
+      "loss": 0.7606,
+      "step": 204
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9935510808088923e-05,
+      "loss": 0.9837,
+      "step": 205
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9934277946851508e-05,
+      "loss": 0.8546,
+      "step": 206
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9933033451269095e-05,
+      "loss": 0.8939,
+      "step": 207
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9931777322799145e-05,
+      "loss": 0.8029,
+      "step": 208
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9930509562912753e-05,
+      "loss": 0.8467,
+      "step": 209
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9929230173094633e-05,
+      "loss": 0.9067,
+      "step": 210
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9927939154843118e-05,
+      "loss": 0.8858,
+      "step": 211
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9926636509670158e-05,
+      "loss": 0.9363,
+      "step": 212
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9925322239101325e-05,
+      "loss": 0.8547,
+      "step": 213
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.9923996344675802e-05,
+      "loss": 0.7235,
+      "step": 214
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.992265882794638e-05,
+      "loss": 0.9026,
+      "step": 215
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.992130969047948e-05,
+      "loss": 0.7974,
+      "step": 216
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9919948933855102e-05,
+      "loss": 1.0075,
+      "step": 217
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9918576559666883e-05,
+      "loss": 0.829,
+      "step": 218
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.991719256952205e-05,
+      "loss": 0.7842,
+      "step": 219
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9915796965041438e-05,
+      "loss": 0.8579,
+      "step": 220
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.991438974785948e-05,
+      "loss": 0.9034,
+      "step": 221
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9912970919624218e-05,
+      "loss": 0.8598,
+      "step": 222
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9911540481997282e-05,
+      "loss": 0.8372,
+      "step": 223
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9910098436653903e-05,
+      "loss": 0.8992,
+      "step": 224
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9908644785282905e-05,
+      "loss": 0.833,
+      "step": 225
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.990717952958671e-05,
+      "loss": 0.7043,
+      "step": 226
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.990570267128132e-05,
+      "loss": 0.8644,
+      "step": 227
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.990421421209633e-05,
+      "loss": 0.7286,
+      "step": 228
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.990271415377492e-05,
+      "loss": 0.8491,
+      "step": 229
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9901202498073863e-05,
+      "loss": 0.9551,
+      "step": 230
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.98996792467635e-05,
+      "loss": 0.7676,
+      "step": 231
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9898144401627755e-05,
+      "loss": 0.8254,
+      "step": 232
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.9896597964464145e-05,
+      "loss": 0.905,
+      "step": 233
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.989503993708374e-05,
+      "loss": 0.9504,
+      "step": 234
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.98934703213112e-05,
+      "loss": 0.9368,
+      "step": 235
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9891889118984753e-05,
+      "loss": 0.9241,
+      "step": 236
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.989029633195619e-05,
+      "loss": 0.8412,
+      "step": 237
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9888691962090877e-05,
+      "loss": 0.7261,
+      "step": 238
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.988707601126774e-05,
+      "loss": 0.9309,
+      "step": 239
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9885448481379272e-05,
+      "loss": 0.9372,
+      "step": 240
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9883809374331522e-05,
+      "loss": 0.9394,
+      "step": 241
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.98821586920441e-05,
+      "loss": 0.8145,
+      "step": 242
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9880496436450175e-05,
+      "loss": 0.6903,
+      "step": 243
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9878822609496466e-05,
+      "loss": 0.8398,
+      "step": 244
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.9877137213143242e-05,
+      "loss": 0.7671,
+      "step": 245
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9875440249364323e-05,
+      "loss": 1.0069,
+      "step": 246
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.987373172014708e-05,
+      "loss": 0.7955,
+      "step": 247
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.987201162749242e-05,
+      "loss": 0.9728,
+      "step": 248
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9870279973414804e-05,
+      "loss": 0.8648,
+      "step": 249
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9868536759942226e-05,
+      "loss": 0.9275,
+      "step": 250
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9866781989116214e-05,
+      "loss": 0.8277,
+      "step": 251
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9865015662991832e-05,
+      "loss": 1.0017,
+      "step": 252
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9863237783637692e-05,
+      "loss": 0.8565,
+      "step": 253
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9861448353135914e-05,
+      "loss": 0.8547,
+      "step": 254
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.9859647373582157e-05,
+      "loss": 0.9581,
+      "step": 255
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.985783484708561e-05,
+      "loss": 0.8914,
+      "step": 256
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.8704828023910522,
+      "eval_runtime": 88.8069,
+      "eval_samples_per_second": 22.521,
+      "eval_steps_per_second": 11.26,
+      "step": 256
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9856010775768974e-05,
+      "loss": 1.011,
+      "step": 257
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9854175161768482e-05,
+      "loss": 0.7789,
+      "step": 258
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9852328007233878e-05,
+      "loss": 0.8411,
+      "step": 259
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.985046931432842e-05,
+      "loss": 0.7958,
+      "step": 260
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9848599085228878e-05,
+      "loss": 1.1321,
+      "step": 261
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9846717322125546e-05,
+      "loss": 0.8746,
+      "step": 262
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9844824027222208e-05,
+      "loss": 0.9167,
+      "step": 263
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9842919202736165e-05,
+      "loss": 0.759,
+      "step": 264
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.9841002850898217e-05,
+      "loss": 0.9867,
+      "step": 265
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.983907497395266e-05,
+      "loss": 0.7754,
+      "step": 266
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9837135574157295e-05,
+      "loss": 0.7645,
+      "step": 267
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9835184653783413e-05,
+      "loss": 0.8518,
+      "step": 268
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9833222215115796e-05,
+      "loss": 0.9464,
+      "step": 269
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.983124826045272e-05,
+      "loss": 0.6999,
+      "step": 270
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9829262792105947e-05,
+      "loss": 0.8434,
+      "step": 271
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9827265812400718e-05,
+      "loss": 0.7315,
+      "step": 272
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.982525732367576e-05,
+      "loss": 0.9114,
+      "step": 273
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9823237328283277e-05,
+      "loss": 0.7189,
+      "step": 274
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.9821205828588947e-05,
+      "loss": 0.8595,
+      "step": 275
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.981916282697193e-05,
+      "loss": 0.7794,
+      "step": 276
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.981710832582484e-05,
+      "loss": 0.8662,
+      "step": 277
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9815042327553774e-05,
+      "loss": 0.7789,
+      "step": 278
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9812964834578286e-05,
+      "loss": 0.7742,
+      "step": 279
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.981087584933139e-05,
+      "loss": 0.7995,
+      "step": 280
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.980877537425957e-05,
+      "loss": 0.9092,
+      "step": 281
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9806663411822754e-05,
+      "loss": 0.9108,
+      "step": 282
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9804539964494323e-05,
+      "loss": 0.8854,
+      "step": 283
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.980240503476112e-05,
+      "loss": 1.0661,
+      "step": 284
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.9800258625123426e-05,
+      "loss": 0.8872,
+      "step": 285
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.979810073809497e-05,
+      "loss": 0.8266,
+      "step": 286
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9795931376202917e-05,
+      "loss": 0.9558,
+      "step": 287
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.979375054198788e-05,
+      "loss": 0.9615,
+      "step": 288
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9791558238003895e-05,
+      "loss": 0.9973,
+      "step": 289
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9789354466818446e-05,
+      "loss": 0.8993,
+      "step": 290
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9787139231012437e-05,
+      "loss": 0.783,
+      "step": 291
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9784912533180197e-05,
+      "loss": 0.8554,
+      "step": 292
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9782674375929483e-05,
+      "loss": 0.9284,
+      "step": 293
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.978042476188147e-05,
+      "loss": 0.8688,
+      "step": 294
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.977816369367076e-05,
+      "loss": 0.7681,
+      "step": 295
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9775891173945348e-05,
+      "loss": 0.9067,
+      "step": 296
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9773607205366664e-05,
+      "loss": 0.7519,
+      "step": 297
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9771311790609533e-05,
+      "loss": 0.9549,
+      "step": 298
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.976900493236218e-05,
+      "loss": 0.985,
+      "step": 299
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.976668663332625e-05,
+      "loss": 0.8651,
+      "step": 300
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.976435689621677e-05,
+      "loss": 0.8236,
+      "step": 301
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9762015723762165e-05,
+      "loss": 0.8703,
+      "step": 302
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.9759663118704263e-05,
+      "loss": 0.8845,
+      "step": 303
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.975729908379827e-05,
+      "loss": 1.624,
+      "step": 304
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.975492362181278e-05,
+      "loss": 0.7728,
+      "step": 305
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9752536735529774e-05,
+      "loss": 0.7587,
+      "step": 306
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9750138427744608e-05,
+      "loss": 0.8409,
+      "step": 307
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9747728701266015e-05,
+      "loss": 0.8308,
+      "step": 308
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9745307558916104e-05,
+      "loss": 0.8025,
+      "step": 309
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.974287500353035e-05,
+      "loss": 0.7392,
+      "step": 310
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9740431037957593e-05,
+      "loss": 0.8281,
+      "step": 311
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.973797566506004e-05,
+      "loss": 0.9947,
+      "step": 312
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9735508887713252e-05,
+      "loss": 0.8923,
+      "step": 313
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.9733030708806157e-05,
+      "loss": 0.8989,
+      "step": 314
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.973054113124102e-05,
+      "loss": 0.9855,
+      "step": 315
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9728040157933465e-05,
+      "loss": 0.9106,
+      "step": 316
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9725527791812464e-05,
+      "loss": 0.7688,
+      "step": 317
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9723004035820327e-05,
+      "loss": 0.8705,
+      "step": 318
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.97204688929127e-05,
+      "loss": 0.7196,
+      "step": 319
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9717922366058576e-05,
+      "loss": 0.7742,
+      "step": 320
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 0.8576830625534058,
+      "eval_runtime": 88.7515,
+      "eval_samples_per_second": 22.535,
+      "eval_steps_per_second": 11.267,
+      "step": 320
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9715364458240263e-05,
+      "loss": 0.8914,
+      "step": 321
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9712795172453413e-05,
+      "loss": 0.8865,
+      "step": 322
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9710214511707e-05,
+      "loss": 0.7747,
+      "step": 323
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.970762247902331e-05,
+      "loss": 0.8571,
+      "step": 324
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.9705019077437958e-05,
+      "loss": 0.7771,
+      "step": 325
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.970240430999987e-05,
+      "loss": 0.7629,
+      "step": 326
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.969977817977128e-05,
+      "loss": 0.9229,
+      "step": 327
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9697140689827733e-05,
+      "loss": 0.8372,
+      "step": 328
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9694491843258075e-05,
+      "loss": 0.794,
+      "step": 329
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9691831643164455e-05,
+      "loss": 0.7288,
+      "step": 330
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9689160092662316e-05,
+      "loss": 0.8428,
+      "step": 331
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9686477194880393e-05,
+      "loss": 0.7379,
+      "step": 332
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9683782952960716e-05,
+      "loss": 0.8157,
+      "step": 333
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9681077370058595e-05,
+      "loss": 0.7902,
+      "step": 334
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.9678360449342622e-05,
+      "loss": 0.8268,
+      "step": 335
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9675632193994668e-05,
+      "loss": 0.8701,
+      "step": 336
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.967289260720988e-05,
+      "loss": 0.8959,
+      "step": 337
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9670141692196672e-05,
+      "loss": 0.8847,
+      "step": 338
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9667379452176728e-05,
+      "loss": 0.9147,
+      "step": 339
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9664605890384997e-05,
+      "loss": 0.758,
+      "step": 340
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9661821010069676e-05,
+      "loss": 0.7882,
+      "step": 341
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9659024814492237e-05,
+      "loss": 0.8319,
+      "step": 342
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9656217306927386e-05,
+      "loss": 0.8644,
+      "step": 343
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.965339849066308e-05,
+      "loss": 0.8515,
+      "step": 344
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.9650568369000532e-05,
+      "loss": 0.7463,
+      "step": 345
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9647726945254182e-05,
+      "loss": 0.9596,
+      "step": 346
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9644874222751706e-05,
+      "loss": 0.8504,
+      "step": 347
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9642010204834027e-05,
+      "loss": 0.8578,
+      "step": 348
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9639134894855278e-05,
+      "loss": 0.9057,
+      "step": 349
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.963624829618283e-05,
+      "loss": 0.9923,
+      "step": 350
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.963335041219726e-05,
+      "loss": 0.9635,
+      "step": 351
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9630441246292386e-05,
+      "loss": 0.8831,
+      "step": 352
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.9627520801875215e-05,
+      "loss": 0.8529,
+      "step": 353
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.962458908236597e-05,
+      "loss": 0.7638,
+      "step": 354
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.962164609119808e-05,
+      "loss": 0.9547,
+      "step": 355
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.961869183181818e-05,
+      "loss": 0.8013,
+      "step": 356
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.961572630768609e-05,
+      "loss": 0.7823,
+      "step": 357
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9612749522274834e-05,
+      "loss": 1.0659,
+      "step": 358
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.960976147907061e-05,
+      "loss": 0.8875,
+      "step": 359
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9606762181572814e-05,
+      "loss": 0.8552,
+      "step": 360
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9603751633294025e-05,
+      "loss": 0.8147,
+      "step": 361
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.960072983775998e-05,
+      "loss": 0.8681,
+      "step": 362
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9597696798509604e-05,
+      "loss": 0.8248,
+      "step": 363
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.959465251909498e-05,
+      "loss": 0.8806,
+      "step": 364
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1.9591597003081363e-05,
+      "loss": 0.8216,
+      "step": 365
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9588530254047166e-05,
+      "loss": 0.7652,
+      "step": 366
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.958545227558395e-05,
+      "loss": 0.9276,
+      "step": 367
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9582363071296436e-05,
+      "loss": 0.7528,
+      "step": 368
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9579262644802486e-05,
+      "loss": 0.9386,
+      "step": 369
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9576150999733107e-05,
+      "loss": 0.7804,
+      "step": 370
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9573028139732447e-05,
+      "loss": 0.8627,
+      "step": 371
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9569894068457783e-05,
+      "loss": 0.8881,
+      "step": 372
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9566748789579527e-05,
+      "loss": 0.8651,
+      "step": 373
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.9563592306781207e-05,
+      "loss": 0.8104,
+      "step": 374
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.956042462375949e-05,
+      "loss": 0.7141,
+      "step": 375
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.955724574422414e-05,
+      "loss": 0.847,
+      "step": 376
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.955405567189804e-05,
+      "loss": 0.8132,
+      "step": 377
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9550854410517198e-05,
+      "loss": 0.6904,
+      "step": 378
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9547641963830698e-05,
+      "loss": 0.7857,
+      "step": 379
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9544418335600747e-05,
+      "loss": 0.8009,
+      "step": 380
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9541183529602626e-05,
+      "loss": 0.8208,
+      "step": 381
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9537937549624734e-05,
+      "loss": 0.9208,
+      "step": 382
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.953468039946852e-05,
+      "loss": 0.773,
+      "step": 383
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9531412082948554e-05,
+      "loss": 0.8532,
+      "step": 384
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 0.8470838665962219,
+      "eval_runtime": 88.8518,
+      "eval_samples_per_second": 22.509,
+      "eval_steps_per_second": 11.255,
+      "step": 384
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.9528132603892456e-05,
+      "loss": 0.8237,
+      "step": 385
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9524841966140923e-05,
+      "loss": 0.9781,
+      "step": 386
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.952154017354773e-05,
+      "loss": 0.8461,
+      "step": 387
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.951822722997971e-05,
+      "loss": 0.6528,
+      "step": 388
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9514903139316754e-05,
+      "loss": 0.7626,
+      "step": 389
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9511567905451804e-05,
+      "loss": 0.9037,
+      "step": 390
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9508221532290864e-05,
+      "loss": 0.802,
+      "step": 391
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9504864023752974e-05,
+      "loss": 0.7549,
+      "step": 392
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.950149538377022e-05,
+      "loss": 0.8103,
+      "step": 393
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.949811561628772e-05,
+      "loss": 0.8518,
+      "step": 394
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9494724725263623e-05,
+      "loss": 0.7709,
+      "step": 395
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9491322714669115e-05,
+      "loss": 0.7942,
+      "step": 396
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9487909588488393e-05,
+      "loss": 0.9106,
+      "step": 397
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9484485350718673e-05,
+      "loss": 0.9729,
+      "step": 398
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9481050005370194e-05,
+      "loss": 0.7691,
+      "step": 399
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9477603556466195e-05,
+      "loss": 0.8142,
+      "step": 400
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.947414600804292e-05,
+      "loss": 0.8069,
+      "step": 401
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.947067736414961e-05,
+      "loss": 0.8064,
+      "step": 402
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.946719762884851e-05,
+      "loss": 0.8362,
+      "step": 403
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.946370680621484e-05,
+      "loss": 0.9664,
+      "step": 404
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.9460204900336813e-05,
+      "loss": 0.7727,
+      "step": 405
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9456691915315626e-05,
+      "loss": 0.8621,
+      "step": 406
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9453167855265442e-05,
+      "loss": 0.8006,
+      "step": 407
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9449632724313393e-05,
+      "loss": 0.8032,
+      "step": 408
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9446086526599596e-05,
+      "loss": 0.8343,
+      "step": 409
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9442529266277104e-05,
+      "loss": 0.79,
+      "step": 410
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9438960947511938e-05,
+      "loss": 0.6789,
+      "step": 411
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9435381574483066e-05,
+      "loss": 0.9176,
+      "step": 412
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.9431791151382414e-05,
+      "loss": 1.098,
+      "step": 413
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.942818968241483e-05,
+      "loss": 0.6833,
+      "step": 414
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.942457717179811e-05,
+      "loss": 0.8131,
+      "step": 415
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9420953623762978e-05,
+      "loss": 1.0117,
+      "step": 416
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.941731904255309e-05,
+      "loss": 0.8052,
+      "step": 417
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.941367343242501e-05,
+      "loss": 0.6806,
+      "step": 418
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.941001679764823e-05,
+      "loss": 0.8556,
+      "step": 419
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9406349142505153e-05,
+      "loss": 0.8597,
+      "step": 420
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9402670471291078e-05,
+      "loss": 0.945,
+      "step": 421
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9398980788314218e-05,
+      "loss": 0.9914,
+      "step": 422
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.9395280097895683e-05,
+      "loss": 0.9351,
+      "step": 423
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.939156840436945e-05,
+      "loss": 0.8328,
+      "step": 424
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.938784571208242e-05,
+      "loss": 0.9441,
+      "step": 425
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9384112025394338e-05,
+      "loss": 0.7824,
+      "step": 426
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9380367348677855e-05,
+      "loss": 0.8984,
+      "step": 427
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9376611686318468e-05,
+      "loss": 0.8352,
+      "step": 428
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.937284504271456e-05,
+      "loss": 0.8329,
+      "step": 429
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.936906742227736e-05,
+      "loss": 0.7788,
+      "step": 430
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.936527882943096e-05,
+      "loss": 0.7815,
+      "step": 431
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.93614792686123e-05,
+      "loss": 0.8727,
+      "step": 432
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9357668744271173e-05,
+      "loss": 0.873,
+      "step": 433
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9353847260870193e-05,
+      "loss": 0.7234,
+      "step": 434
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.9350014822884818e-05,
+      "loss": 0.725,
+      "step": 435
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9346171434803342e-05,
+      "loss": 0.8242,
+      "step": 436
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.934231710112688e-05,
+      "loss": 0.7766,
+      "step": 437
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.933845182636936e-05,
+      "loss": 0.7384,
+      "step": 438
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9334575615057523e-05,
+      "loss": 0.9487,
+      "step": 439
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.933068847173093e-05,
+      "loss": 0.886,
+      "step": 440
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9326790400941926e-05,
+      "loss": 0.6873,
+      "step": 441
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9322881407255673e-05,
+      "loss": 0.7381,
+      "step": 442
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9318961495250118e-05,
+      "loss": 0.8391,
+      "step": 443
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.9315030669515983e-05,
+      "loss": 0.7742,
+      "step": 444
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.931108893465679e-05,
+      "loss": 0.7162,
+      "step": 445
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9307136295288823e-05,
+      "loss": 0.6951,
+      "step": 446
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9303172756041146e-05,
+      "loss": 0.756,
+      "step": 447
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9299198321555585e-05,
+      "loss": 0.7503,
+      "step": 448
+    },
+    {
+      "epoch": 0.45,
+      "eval_loss": 0.8380163311958313,
+      "eval_runtime": 88.8662,
+      "eval_samples_per_second": 22.506,
+      "eval_steps_per_second": 11.253,
+      "step": 448
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.929521299648672e-05,
+      "loss": 0.7756,
+      "step": 449
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.92912167855019e-05,
+      "loss": 0.8496,
+      "step": 450
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.92872096932812e-05,
+      "loss": 0.8852,
+      "step": 451
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9283191724517462e-05,
+      "loss": 0.727,
+      "step": 452
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.927916288391625e-05,
+      "loss": 0.7063,
+      "step": 453
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9275123176195868e-05,
+      "loss": 0.8145,
+      "step": 454
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.9271072606087347e-05,
+      "loss": 0.8883,
+      "step": 455
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9267011178334432e-05,
+      "loss": 0.8512,
+      "step": 456
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.926293889769359e-05,
+      "loss": 0.8944,
+      "step": 457
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9258855768934e-05,
+      "loss": 0.7907,
+      "step": 458
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9254761796837535e-05,
+      "loss": 0.8282,
+      "step": 459
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9250656986198785e-05,
+      "loss": 0.8938,
+      "step": 460
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9246541341825016e-05,
+      "loss": 0.8208,
+      "step": 461
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.924241486853619e-05,
+      "loss": 0.8071,
+      "step": 462
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9238277571164947e-05,
+      "loss": 0.8381,
+      "step": 463
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9234129454556608e-05,
+      "loss": 0.8005,
+      "step": 464
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.9229970523569163e-05,
+      "loss": 0.8503,
+      "step": 465
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9225800783073267e-05,
+      "loss": 0.8601,
+      "step": 466
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9221620237952233e-05,
+      "loss": 0.9007,
+      "step": 467
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.921742889310203e-05,
+      "loss": 0.8161,
+      "step": 468
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9213226753431272e-05,
+      "loss": 0.868,
+      "step": 469
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9209013823861217e-05,
+      "loss": 0.7736,
+      "step": 470
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9204790109325762e-05,
+      "loss": 0.6409,
+      "step": 471
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9200555614771428e-05,
+      "loss": 0.8424,
+      "step": 472
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9196310345157366e-05,
+      "loss": 0.8458,
+      "step": 473
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9192054305455347e-05,
+      "loss": 0.8427,
+      "step": 474
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.9187787500649747e-05,
+      "loss": 0.8084,
+      "step": 475
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.918350993573756e-05,
+      "loss": 0.8328,
+      "step": 476
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.917922161572838e-05,
+      "loss": 0.7588,
+      "step": 477
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9174922545644383e-05,
+      "loss": 0.6648,
+      "step": 478
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9170612730520352e-05,
+      "loss": 0.9953,
+      "step": 479
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9166292175403645e-05,
+      "loss": 0.7162,
+      "step": 480
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9161960885354203e-05,
+      "loss": 0.8221,
+      "step": 481
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.915761886544453e-05,
+      "loss": 0.8139,
+      "step": 482
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9153266120759707e-05,
+      "loss": 0.9388,
+      "step": 483
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.914890265639737e-05,
+      "loss": 0.8109,
+      "step": 484
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.9144528477467705e-05,
+      "loss": 0.6884,
+      "step": 485
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9140143589093452e-05,
+      "loss": 0.7648,
+      "step": 486
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9135747996409893e-05,
+      "loss": 0.8567,
+      "step": 487
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9131341704564845e-05,
+      "loss": 0.8423,
+      "step": 488
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9126924718718657e-05,
+      "loss": 0.8746,
+      "step": 489
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.912249704404419e-05,
+      "loss": 0.9769,
+      "step": 490
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9118058685726843e-05,
+      "loss": 0.8649,
+      "step": 491
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9113609648964515e-05,
+      "loss": 0.7338,
+      "step": 492
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.910914993896761e-05,
+      "loss": 0.7923,
+      "step": 493
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.9104679560959038e-05,
+      "loss": 0.6997,
+      "step": 494
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.910019852017419e-05,
+      "loss": 0.735,
+      "step": 495
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9095706821860965e-05,
+      "loss": 0.858,
+      "step": 496
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.909120447127973e-05,
+      "loss": 0.9726,
+      "step": 497
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.908669147370332e-05,
+      "loss": 0.6871,
+      "step": 498
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9082167834417064e-05,
+      "loss": 0.6674,
+      "step": 499
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9077633558718723e-05,
+      "loss": 0.9081,
+      "step": 500
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9073088651918534e-05,
+      "loss": 0.7664,
+      "step": 501
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.906853311933919e-05,
+      "loss": 0.7947,
+      "step": 502
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9063966966315802e-05,
+      "loss": 0.8494,
+      "step": 503
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9059390198195947e-05,
+      "loss": 0.7399,
+      "step": 504
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.9054802820339617e-05,
+      "loss": 0.8816,
+      "step": 505
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9050204838119237e-05,
+      "loss": 0.8472,
+      "step": 506
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9045596256919647e-05,
+      "loss": 0.843,
+      "step": 507
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9040977082138103e-05,
+      "loss": 0.7942,
+      "step": 508
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9036347319184262e-05,
+      "loss": 0.9324,
+      "step": 509
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9031706973480183e-05,
+      "loss": 0.973,
+      "step": 510
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.902705605046033e-05,
+      "loss": 0.7232,
+      "step": 511
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.902239455557154e-05,
+      "loss": 0.8105,
+      "step": 512
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.8321256637573242,
+      "eval_runtime": 88.8099,
+      "eval_samples_per_second": 22.52,
+      "eval_steps_per_second": 11.26,
+      "step": 512
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.9017722494273035e-05,
+      "loss": 0.7939,
+      "step": 513
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.901303987203641e-05,
+      "loss": 0.8154,
+      "step": 514
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.900834669434564e-05,
+      "loss": 0.6629,
+      "step": 515
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.9003642966697046e-05,
+      "loss": 0.8167,
+      "step": 516
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8998928694599308e-05,
+      "loss": 0.7762,
+      "step": 517
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8994203883573468e-05,
+      "loss": 0.8324,
+      "step": 518
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8989468539152894e-05,
+      "loss": 0.7046,
+      "step": 519
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8984722666883298e-05,
+      "loss": 0.7878,
+      "step": 520
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8979966272322717e-05,
+      "loss": 0.8183,
+      "step": 521
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.897519936104152e-05,
+      "loss": 0.829,
+      "step": 522
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8970421938622384e-05,
+      "loss": 0.8358,
+      "step": 523
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.89656340106603e-05,
+      "loss": 0.8195,
+      "step": 524
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8960835582762556e-05,
+      "loss": 0.9304,
+      "step": 525
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8956026660548744e-05,
+      "loss": 0.7643,
+      "step": 526
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.895120724965075e-05,
+      "loss": 1.0905,
+      "step": 527
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8946377355712725e-05,
+      "loss": 0.8015,
+      "step": 528
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.894153698439112e-05,
+      "loss": 0.8747,
+      "step": 529
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8936686141354643e-05,
+      "loss": 0.8012,
+      "step": 530
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8931824832284267e-05,
+      "loss": 0.7018,
+      "step": 531
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8926953062873224e-05,
+      "loss": 0.899,
+      "step": 532
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.8922070838827e-05,
+      "loss": 1.0892,
+      "step": 533
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.891717816586332e-05,
+      "loss": 0.675,
+      "step": 534
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.891227504971214e-05,
+      "loss": 0.7579,
+      "step": 535
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8907361496115665e-05,
+      "loss": 0.7148,
+      "step": 536
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8902437510828303e-05,
+      "loss": 0.8114,
+      "step": 537
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8897503099616687e-05,
+      "loss": 0.8054,
+      "step": 538
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.889255826825967e-05,
+      "loss": 0.746,
+      "step": 539
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8887603022548288e-05,
+      "loss": 0.8464,
+      "step": 540
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8882637368285794e-05,
+      "loss": 0.7673,
+      "step": 541
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8877661311287617e-05,
+      "loss": 0.6507,
+      "step": 542
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8872674857381378e-05,
+      "loss": 0.7847,
+      "step": 543
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.8867678012406863e-05,
+      "loss": 0.8659,
+      "step": 544
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 1.886267078221604e-05,
+      "loss": 0.7525,
+      "step": 545
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8857653172673035e-05,
+      "loss": 0.813,
+      "step": 546
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.885262518965412e-05,
+      "loss": 0.7647,
+      "step": 547
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8847586839047728e-05,
+      "loss": 0.9482,
+      "step": 548
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8842538126754434e-05,
+      "loss": 0.9578,
+      "step": 549
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.883747905868694e-05,
+      "loss": 0.7448,
+      "step": 550
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8832409640770075e-05,
+      "loss": 0.856,
+      "step": 551
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8827329878940805e-05,
+      "loss": 0.7813,
+      "step": 552
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8822239779148187e-05,
+      "loss": 0.9091,
+      "step": 553
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8817139347353402e-05,
+      "loss": 0.6702,
+      "step": 554
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8812028589529726e-05,
+      "loss": 0.8261,
+      "step": 555
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8806907511662527e-05,
+      "loss": 0.8176,
+      "step": 556
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8801776119749264e-05,
+      "loss": 0.7524,
+      "step": 557
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.879663441979946e-05,
+      "loss": 1.1004,
+      "step": 558
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8791482417834738e-05,
+      "loss": 0.8084,
+      "step": 559
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8786320119888753e-05,
+      "loss": 0.7584,
+      "step": 560
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.878114753200725e-05,
+      "loss": 0.9296,
+      "step": 561
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8775964660247996e-05,
+      "loss": 0.7239,
+      "step": 562
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.877077151068082e-05,
+      "loss": 0.8256,
+      "step": 563
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8765568089387588e-05,
+      "loss": 0.8633,
+      "step": 564
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.8760354402462177e-05,
+      "loss": 0.8125,
+      "step": 565
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8755130456010512e-05,
+      "loss": 0.7046,
+      "step": 566
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8749896256150516e-05,
+      "loss": 0.8327,
+      "step": 567
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8744651809012128e-05,
+      "loss": 0.7625,
+      "step": 568
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8739397120737276e-05,
+      "loss": 0.64,
+      "step": 569
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.87341321974799e-05,
+      "loss": 0.8003,
+      "step": 570
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.872885704540591e-05,
+      "loss": 0.8428,
+      "step": 571
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.872357167069321e-05,
+      "loss": 0.695,
+      "step": 572
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.871827607953166e-05,
+      "loss": 0.9279,
+      "step": 573
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8712970278123102e-05,
+      "loss": 0.8376,
+      "step": 574
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 1.8707654272681317e-05,
+      "loss": 0.8851,
+      "step": 575
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8702328069432055e-05,
+      "loss": 0.8739,
+      "step": 576
+    },
+    {
+      "epoch": 0.58,
+      "eval_loss": 0.8237754702568054,
+      "eval_runtime": 88.755,
+      "eval_samples_per_second": 22.534,
+      "eval_steps_per_second": 11.267,
+      "step": 576
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8696991674612993e-05,
+      "loss": 0.8256,
+      "step": 577
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8691645094473758e-05,
+      "loss": 0.9051,
+      "step": 578
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.868628833527589e-05,
+      "loss": 0.7162,
+      "step": 579
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.868092140329286e-05,
+      "loss": 0.999,
+      "step": 580
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8675544304810056e-05,
+      "loss": 0.8956,
+      "step": 581
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.867015704612476e-05,
+      "loss": 0.7452,
+      "step": 582
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8664759633546166e-05,
+      "loss": 0.868,
+      "step": 583
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.865935207339535e-05,
+      "loss": 0.9255,
+      "step": 584
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8653934372005277e-05,
+      "loss": 0.7148,
+      "step": 585
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8648506535720783e-05,
+      "loss": 0.9121,
+      "step": 586
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8643068570898584e-05,
+      "loss": 0.7663,
+      "step": 587
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8637620483907247e-05,
+      "loss": 0.7213,
+      "step": 588
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8632162281127196e-05,
+      "loss": 0.7209,
+      "step": 589
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8626693968950712e-05,
+      "loss": 0.8532,
+      "step": 590
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.86212155537819e-05,
+      "loss": 0.7439,
+      "step": 591
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.861572704203671e-05,
+      "loss": 0.6527,
+      "step": 592
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8610228440142904e-05,
+      "loss": 0.9329,
+      "step": 593
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.8604719754540067e-05,
+      "loss": 0.923,
+      "step": 594
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.85992009916796e-05,
+      "loss": 0.7381,
+      "step": 595
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8593672158024697e-05,
+      "loss": 0.7204,
+      "step": 596
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8588133260050346e-05,
+      "loss": 0.6572,
+      "step": 597
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.858258430424333e-05,
+      "loss": 0.6991,
+      "step": 598
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8577025297102204e-05,
+      "loss": 0.8658,
+      "step": 599
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.857145624513729e-05,
+      "loss": 0.8149,
+      "step": 600
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8565877154870686e-05,
+      "loss": 0.8789,
+      "step": 601
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8560288032836235e-05,
+      "loss": 0.7343,
+      "step": 602
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.855468888557954e-05,
+      "loss": 0.7667,
+      "step": 603
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8549079719657933e-05,
+      "loss": 0.7589,
+      "step": 604
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.8543460541640487e-05,
+      "loss": 0.7222,
+      "step": 605
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8537831358107995e-05,
+      "loss": 0.7407,
+      "step": 606
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8532192175652973e-05,
+      "loss": 0.7959,
+      "step": 607
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.852654300087964e-05,
+      "loss": 0.7504,
+      "step": 608
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8520883840403933e-05,
+      "loss": 0.8011,
+      "step": 609
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.851521470085346e-05,
+      "loss": 0.6863,
+      "step": 610
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.850953558886754e-05,
+      "loss": 0.827,
+      "step": 611
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8503846511097153e-05,
+      "loss": 0.882,
+      "step": 612
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8498147474204955e-05,
+      "loss": 0.7463,
+      "step": 613
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.849243848486527e-05,
+      "loss": 0.9459,
+      "step": 614
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8486719549764077e-05,
+      "loss": 0.8591,
+      "step": 615
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8480990675598995e-05,
+      "loss": 0.8529,
+      "step": 616
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8475251869079292e-05,
+      "loss": 0.9178,
+      "step": 617
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8469503136925862e-05,
+      "loss": 0.9267,
+      "step": 618
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8463744485871232e-05,
+      "loss": 0.7316,
+      "step": 619
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8457975922659535e-05,
+      "loss": 0.7726,
+      "step": 620
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8452197454046517e-05,
+      "loss": 0.8172,
+      "step": 621
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.844640908679952e-05,
+      "loss": 0.6132,
+      "step": 622
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.844061082769749e-05,
+      "loss": 0.7698,
+      "step": 623
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.8434802683530947e-05,
+      "loss": 0.7798,
+      "step": 624
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 1.842898466110199e-05,
+      "loss": 0.881,
+      "step": 625
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8423156767224286e-05,
+      "loss": 0.8275,
+      "step": 626
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.841731900872307e-05,
+      "loss": 0.7459,
+      "step": 627
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8411471392435124e-05,
+      "loss": 0.9518,
+      "step": 628
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.840561392520877e-05,
+      "loss": 0.8184,
+      "step": 629
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8399746613903878e-05,
+      "loss": 0.9292,
+      "step": 630
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8393869465391833e-05,
+      "loss": 0.6926,
+      "step": 631
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8387982486555557e-05,
+      "loss": 0.6586,
+      "step": 632
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8382085684289472e-05,
+      "loss": 0.6964,
+      "step": 633
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.83761790654995e-05,
+      "loss": 0.9562,
+      "step": 634
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.8370262637103087e-05,
+      "loss": 0.8209,
+      "step": 635
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.836433640602913e-05,
+      "loss": 0.6311,
+      "step": 636
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8358400379218033e-05,
+      "loss": 0.7725,
+      "step": 637
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8352454563621664e-05,
+      "loss": 0.6367,
+      "step": 638
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.834649896620335e-05,
+      "loss": 0.8048,
+      "step": 639
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.834053359393788e-05,
+      "loss": 0.7057,
+      "step": 640
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.8161675930023193,
+      "eval_runtime": 88.8512,
+      "eval_samples_per_second": 22.51,
+      "eval_steps_per_second": 11.255,
+      "step": 640
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8334558453811492e-05,
+      "loss": 0.8085,
+      "step": 641
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.832857355282186e-05,
+      "loss": 0.8123,
+      "step": 642
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8322578897978082e-05,
+      "loss": 0.8445,
+      "step": 643
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8316574496300692e-05,
+      "loss": 0.8441,
+      "step": 644
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8310560354821635e-05,
+      "loss": 0.7833,
+      "step": 645
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.830453648058426e-05,
+      "loss": 0.7952,
+      "step": 646
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8298502880643316e-05,
+      "loss": 0.7673,
+      "step": 647
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.829245956206494e-05,
+      "loss": 0.8559,
+      "step": 648
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.828640653192665e-05,
+      "loss": 0.7095,
+      "step": 649
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8280343797317344e-05,
+      "loss": 0.7709,
+      "step": 650
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8274271365337277e-05,
+      "loss": 0.7622,
+      "step": 651
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.826818924309807e-05,
+      "loss": 0.6762,
+      "step": 652
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8262097437722684e-05,
+      "loss": 0.7914,
+      "step": 653
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.825599595634542e-05,
+      "loss": 0.817,
+      "step": 654
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.8249884806111927e-05,
+      "loss": 0.797,
+      "step": 655
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.824376399417915e-05,
+      "loss": 0.7553,
+      "step": 656
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8237633527715373e-05,
+      "loss": 0.7762,
+      "step": 657
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8231493413900177e-05,
+      "loss": 0.8008,
+      "step": 658
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.822534365992444e-05,
+      "loss": 0.8322,
+      "step": 659
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8219184272990333e-05,
+      "loss": 0.8479,
+      "step": 660
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8213015260311305e-05,
+      "loss": 0.823,
+      "step": 661
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8206836629112087e-05,
+      "loss": 0.8383,
+      "step": 662
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.8200648386628664e-05,
+      "loss": 0.8086,
+      "step": 663
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.819445054010828e-05,
+      "loss": 0.7702,
+      "step": 664
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.818824309680943e-05,
+      "loss": 0.8369,
+      "step": 665
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8182026064001844e-05,
+      "loss": 0.7653,
+      "step": 666
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8175799448966483e-05,
+      "loss": 0.6766,
+      "step": 667
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8169563258995538e-05,
+      "loss": 0.7201,
+      "step": 668
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.81633175013924e-05,
+      "loss": 0.8977,
+      "step": 669
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8157062183471682e-05,
+      "loss": 0.7354,
+      "step": 670
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8150797312559167e-05,
+      "loss": 0.8104,
+      "step": 671
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8144522895991857e-05,
+      "loss": 0.7339,
+      "step": 672
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.813823894111791e-05,
+      "loss": 0.8117,
+      "step": 673
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8131945455296667e-05,
+      "loss": 0.9314,
+      "step": 674
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 1.8125642445898628e-05,
+      "loss": 0.6697,
+      "step": 675
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8119329920305438e-05,
+      "loss": 0.7683,
+      "step": 676
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8113007885909903e-05,
+      "loss": 0.9921,
+      "step": 677
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.810667635011595e-05,
+      "loss": 0.6722,
+      "step": 678
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8100335320338644e-05,
+      "loss": 0.8275,
+      "step": 679
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8093984804004157e-05,
+      "loss": 0.7587,
+      "step": 680
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.808762480854979e-05,
+      "loss": 0.8,
+      "step": 681
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8081255341423924e-05,
+      "loss": 0.7504,
+      "step": 682
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.807487641008604e-05,
+      "loss": 0.6742,
+      "step": 683
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8068488022006714e-05,
+      "loss": 0.8,
+      "step": 684
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8062090184667586e-05,
+      "loss": 0.759,
+      "step": 685
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8055682905561353e-05,
+      "loss": 0.8355,
+      "step": 686
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8049266192191797e-05,
+      "loss": 0.7003,
+      "step": 687
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.804284005207372e-05,
+      "loss": 0.6923,
+      "step": 688
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8036404492732987e-05,
+      "loss": 0.7639,
+      "step": 689
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8029959521706474e-05,
+      "loss": 0.7846,
+      "step": 690
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.80235051465421e-05,
+      "loss": 0.8217,
+      "step": 691
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8017041374798773e-05,
+      "loss": 0.7391,
+      "step": 692
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.801056821404643e-05,
+      "loss": 0.7639,
+      "step": 693
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.8004085671865987e-05,
+      "loss": 1.0368,
+      "step": 694
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.7997593755849355e-05,
+      "loss": 0.8913,
+      "step": 695
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7991092473599424e-05,
+      "loss": 0.8668,
+      "step": 696
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7984581832730042e-05,
+      "loss": 0.7934,
+      "step": 697
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7978061840866036e-05,
+      "loss": 0.7442,
+      "step": 698
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7971532505643162e-05,
+      "loss": 0.821,
+      "step": 699
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7964993834708132e-05,
+      "loss": 0.7814,
+      "step": 700
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.79584458357186e-05,
+      "loss": 1.0184,
+      "step": 701
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.795188851634312e-05,
+      "loss": 0.8157,
+      "step": 702
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7945321884261177e-05,
+      "loss": 0.9535,
+      "step": 703
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.7938745947163166e-05,
+      "loss": 0.9507,
+      "step": 704
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 0.8084772825241089,
+      "eval_runtime": 88.8087,
+      "eval_samples_per_second": 22.52,
+      "eval_steps_per_second": 11.26,
+      "step": 704
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 1.793216071275037e-05,
+      "loss": 0.6816,
+      "step": 705
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7925566188734963e-05,
+      "loss": 0.7804,
+      "step": 706
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7918962382840005e-05,
+      "loss": 0.6319,
+      "step": 707
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.791234930279941e-05,
+      "loss": 0.7329,
+      "step": 708
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7905726956357973e-05,
+      "loss": 0.7394,
+      "step": 709
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.789909535127133e-05,
+      "loss": 0.8789,
+      "step": 710
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7892454495305958e-05,
+      "loss": 0.6921,
+      "step": 711
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7885804396239175e-05,
+      "loss": 0.8274,
+      "step": 712
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7879145061859124e-05,
+      "loss": 0.8109,
+      "step": 713
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.787247649996475e-05,
+      "loss": 0.6683,
+      "step": 714
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7865798718365833e-05,
+      "loss": 0.8605,
+      "step": 715
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7859111724882915e-05,
+      "loss": 0.7162,
+      "step": 716
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7852415527347354e-05,
+      "loss": 0.8672,
+      "step": 717
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.784571013360128e-05,
+      "loss": 0.8368,
+      "step": 718
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7838995551497578e-05,
+      "loss": 0.73,
+      "step": 719
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7832271788899917e-05,
+      "loss": 0.7813,
+      "step": 720
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.78255388536827e-05,
+      "loss": 0.7463,
+      "step": 721
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7818796753731088e-05,
+      "loss": 0.8529,
+      "step": 722
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7812045496940953e-05,
+      "loss": 0.8398,
+      "step": 723
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7805285091218915e-05,
+      "loss": 0.7883,
+      "step": 724
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.7798515544482293e-05,
+      "loss": 0.6891,
+      "step": 725
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7791736864659116e-05,
+      "loss": 0.6995,
+      "step": 726
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7784949059688112e-05,
+      "loss": 0.8135,
+      "step": 727
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7778152137518686e-05,
+      "loss": 0.7637,
+      "step": 728
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7771346106110934e-05,
+      "loss": 0.7548,
+      "step": 729
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.776453097343561e-05,
+      "loss": 0.7701,
+      "step": 730
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.775770674747413e-05,
+      "loss": 0.7618,
+      "step": 731
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7750873436218555e-05,
+      "loss": 0.7747,
+      "step": 732
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.77440310476716e-05,
+      "loss": 0.7513,
+      "step": 733
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.7737179589846595e-05,
+      "loss": 0.8159,
+      "step": 734
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.77303190707675e-05,
+      "loss": 0.7794,
+      "step": 735
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7723449498468882e-05,
+      "loss": 0.7955,
+      "step": 736
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7716570880995913e-05,
+      "loss": 0.8281,
+      "step": 737
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7709683226404362e-05,
+      "loss": 0.7063,
+      "step": 738
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7702786542760574e-05,
+      "loss": 0.769,
+      "step": 739
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.769588083814147e-05,
+      "loss": 0.797,
+      "step": 740
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7688966120634544e-05,
+      "loss": 0.792,
+      "step": 741
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7682042398337832e-05,
+      "loss": 0.7019,
+      "step": 742
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.767510967935993e-05,
+      "loss": 0.906,
+      "step": 743
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7668167971819956e-05,
+      "loss": 0.8094,
+      "step": 744
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7661217283847567e-05,
+      "loss": 0.868,
+      "step": 745
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7654257623582925e-05,
+      "loss": 0.7853,
+      "step": 746
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7647288999176717e-05,
+      "loss": 0.7488,
+      "step": 747
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.764031141879011e-05,
+      "loss": 0.8747,
+      "step": 748
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7633324890594773e-05,
+      "loss": 0.7876,
+      "step": 749
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7626329422772843e-05,
+      "loss": 0.7974,
+      "step": 750
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7619325023516936e-05,
+      "loss": 0.8486,
+      "step": 751
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7612311701030123e-05,
+      "loss": 0.8405,
+      "step": 752
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7605289463525925e-05,
+      "loss": 0.6657,
+      "step": 753
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7598258319228306e-05,
+      "loss": 0.8031,
+      "step": 754
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7591218276371663e-05,
+      "loss": 0.8462,
+      "step": 755
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7584169343200805e-05,
+      "loss": 0.7834,
+      "step": 756
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7577111527970966e-05,
+      "loss": 0.7107,
+      "step": 757
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.757004483894777e-05,
+      "loss": 0.8267,
+      "step": 758
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.756296928440724e-05,
+      "loss": 0.8583,
+      "step": 759
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.755588487263578e-05,
+      "loss": 0.743,
+      "step": 760
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7548791611930173e-05,
+      "loss": 0.7809,
+      "step": 761
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7541689510597547e-05,
+      "loss": 0.706,
+      "step": 762
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7534578576955404e-05,
+      "loss": 0.688,
+      "step": 763
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7527458819331588e-05,
+      "loss": 0.7293,
+      "step": 764
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.7520330246064258e-05,
+      "loss": 0.9841,
+      "step": 765
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7513192865501916e-05,
+      "loss": 0.7393,
+      "step": 766
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7506046686003375e-05,
+      "loss": 0.7296,
+      "step": 767
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7498891715937744e-05,
+      "loss": 0.8461,
+      "step": 768
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.8009968400001526,
+      "eval_runtime": 88.857,
+      "eval_samples_per_second": 22.508,
+      "eval_steps_per_second": 11.254,
+      "step": 768
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7491727963684442e-05,
+      "loss": 0.7201,
+      "step": 769
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7484555437633157e-05,
+      "loss": 0.6904,
+      "step": 770
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7477374146183862e-05,
+      "loss": 0.7647,
+      "step": 771
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7470184097746793e-05,
+      "loss": 0.8107,
+      "step": 772
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.746298530074244e-05,
+      "loss": 0.8447,
+      "step": 773
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7455777763601544e-05,
+      "loss": 0.8187,
+      "step": 774
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7448561494765078e-05,
+      "loss": 0.735,
+      "step": 775
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7441336502684236e-05,
+      "loss": 0.7639,
+      "step": 776
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.743410279582044e-05,
+      "loss": 0.8614,
+      "step": 777
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7426860382645302e-05,
+      "loss": 0.8216,
+      "step": 778
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.741960927164065e-05,
+      "loss": 0.8295,
+      "step": 779
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.741234947129848e-05,
+      "loss": 0.7761,
+      "step": 780
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.740508099012098e-05,
+      "loss": 0.8462,
+      "step": 781
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.739780383662049e-05,
+      "loss": 0.7543,
+      "step": 782
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7390518019319512e-05,
+      "loss": 0.7176,
+      "step": 783
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7383223546750704e-05,
+      "loss": 0.8237,
+      "step": 784
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.7375920427456846e-05,
+      "loss": 0.6964,
+      "step": 785
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7368608669990847e-05,
+      "loss": 0.6851,
+      "step": 786
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7361288282915743e-05,
+      "loss": 0.7936,
+      "step": 787
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.735395927480467e-05,
+      "loss": 0.8793,
+      "step": 788
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.734662165424085e-05,
+      "loss": 0.8362,
+      "step": 789
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.733927542981761e-05,
+      "loss": 0.8331,
+      "step": 790
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7331920610138344e-05,
+      "loss": 0.7247,
+      "step": 791
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.732455720381651e-05,
+      "loss": 0.6166,
+      "step": 792
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.7317185219475618e-05,
+      "loss": 0.8475,
+      "step": 793
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.730980466574924e-05,
+      "loss": 0.7865,
+      "step": 794
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.730241555128097e-05,
+      "loss": 0.7173,
+      "step": 795
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7295017884724436e-05,
+      "loss": 0.7805,
+      "step": 796
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.728761167474327e-05,
+      "loss": 0.7015,
+      "step": 797
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7280196930011124e-05,
+      "loss": 0.9915,
+      "step": 798
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7272773659211636e-05,
+      "loss": 0.8156,
+      "step": 799
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7265341871038438e-05,
+      "loss": 0.9031,
+      "step": 800
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7257901574195116e-05,
+      "loss": 0.9122,
+      "step": 801
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.725045277739525e-05,
+      "loss": 0.9201,
+      "step": 802
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7242995489362352e-05,
+      "loss": 0.7459,
+      "step": 803
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7235529718829886e-05,
+      "loss": 0.7885,
+      "step": 804
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.7228055474541256e-05,
+      "loss": 0.8055,
+      "step": 805
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.722057276524978e-05,
+      "loss": 0.742,
+      "step": 806
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7213081599718694e-05,
+      "loss": 0.8748,
+      "step": 807
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.720558198672114e-05,
+      "loss": 0.8843,
+      "step": 808
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7198073935040148e-05,
+      "loss": 0.7946,
+      "step": 809
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7190557453468636e-05,
+      "loss": 0.8812,
+      "step": 810
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7183032550809387e-05,
+      "loss": 0.7318,
+      "step": 811
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7175499235875053e-05,
+      "loss": 0.8431,
+      "step": 812
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.716795751748814e-05,
+      "loss": 0.7337,
+      "step": 813
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7160407404480982e-05,
+      "loss": 0.6533,
+      "step": 814
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.715284890569576e-05,
+      "loss": 0.8323,
+      "step": 815
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7145282029984472e-05,
+      "loss": 0.8415,
+      "step": 816
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7137706786208917e-05,
+      "loss": 0.6874,
+      "step": 817
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7130123183240708e-05,
+      "loss": 0.6417,
+      "step": 818
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7122531229961235e-05,
+      "loss": 0.6016,
+      "step": 819
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7114930935261674e-05,
+      "loss": 0.7699,
+      "step": 820
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7107322308042974e-05,
+      "loss": 0.7771,
+      "step": 821
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.709970535721583e-05,
+      "loss": 0.8312,
+      "step": 822
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7092080091700695e-05,
+      "loss": 0.7228,
+      "step": 823
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.708444652042776e-05,
+      "loss": 0.8831,
+      "step": 824
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.7076804652336932e-05,
+      "loss": 0.6569,
+      "step": 825
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7069154496377855e-05,
+      "loss": 0.7842,
+      "step": 826
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.706149606150986e-05,
+      "loss": 0.7687,
+      "step": 827
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.705382935670198e-05,
+      "loss": 0.9303,
+      "step": 828
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7046154390932936e-05,
+      "loss": 0.9519,
+      "step": 829
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.703847117319112e-05,
+      "loss": 0.9367,
+      "step": 830
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.703077971247459e-05,
+      "loss": 0.9492,
+      "step": 831
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7023080017791058e-05,
+      "loss": 0.7635,
+      "step": 832
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.7974832057952881,
+      "eval_runtime": 88.8572,
+      "eval_samples_per_second": 22.508,
+      "eval_steps_per_second": 11.254,
+      "step": 832
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.701537209815788e-05,
+      "loss": 0.8036,
+      "step": 833
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.7007655962602038e-05,
+      "loss": 0.6495,
+      "step": 834
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.6999931620160147e-05,
+      "loss": 0.8569,
+      "step": 835
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.699219907987842e-05,
+      "loss": 0.7404,
+      "step": 836
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6984458350812683e-05,
+      "loss": 0.899,
+      "step": 837
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6976709442028346e-05,
+      "loss": 0.7222,
+      "step": 838
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.69689523626004e-05,
+      "loss": 0.8308,
+      "step": 839
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.69611871216134e-05,
+      "loss": 0.6791,
+      "step": 840
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.695341372816147e-05,
+      "loss": 0.6438,
+      "step": 841
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.694563219134827e-05,
+      "loss": 0.7315,
+      "step": 842
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6937842520287007e-05,
+      "loss": 0.6529,
+      "step": 843
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.693004472410041e-05,
+      "loss": 0.7812,
+      "step": 844
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.6922238811920714e-05,
+      "loss": 0.7516,
+      "step": 845
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6914424792889677e-05,
+      "loss": 0.8681,
+      "step": 846
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6906602676158542e-05,
+      "loss": 0.7108,
+      "step": 847
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.689877247088803e-05,
+      "loss": 0.8584,
+      "step": 848
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6890934186248343e-05,
+      "loss": 0.7973,
+      "step": 849
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6883087831419144e-05,
+      "loss": 0.7312,
+      "step": 850
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6875233415589544e-05,
+      "loss": 0.7586,
+      "step": 851
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.68673709479581e-05,
+      "loss": 0.6344,
+      "step": 852
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6859500437732787e-05,
+      "loss": 0.7058,
+      "step": 853
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6851621894131015e-05,
+      "loss": 0.9206,
+      "step": 854
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6843735326379584e-05,
+      "loss": 0.7999,
+      "step": 855
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6835840743714713e-05,
+      "loss": 0.7462,
+      "step": 856
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6827938155381986e-05,
+      "loss": 0.7467,
+      "step": 857
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6820027570636385e-05,
+      "loss": 0.7297,
+      "step": 858
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6812108998742226e-05,
+      "loss": 0.6459,
+      "step": 859
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6804182448973214e-05,
+      "loss": 0.7415,
+      "step": 860
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.679624793061237e-05,
+      "loss": 0.5994,
+      "step": 861
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6788305452952065e-05,
+      "loss": 0.697,
+      "step": 862
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.678035502529398e-05,
+      "loss": 0.765,
+      "step": 863
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.677239665694911e-05,
+      "loss": 0.8578,
+      "step": 864
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.6764430357237752e-05,
+      "loss": 0.7354,
+      "step": 865
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6756456135489487e-05,
+      "loss": 0.8357,
+      "step": 866
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.674847400104318e-05,
+      "loss": 0.7987,
+      "step": 867
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.674048396324696e-05,
+      "loss": 0.7137,
+      "step": 868
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.673248603145821e-05,
+      "loss": 0.7908,
+      "step": 869
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6724480215043553e-05,
+      "loss": 0.7826,
+      "step": 870
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6716466523378865e-05,
+      "loss": 0.772,
+      "step": 871
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.670844496584922e-05,
+      "loss": 0.8403,
+      "step": 872
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6700415551848926e-05,
+      "loss": 0.8107,
+      "step": 873
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6692378290781474e-05,
+      "loss": 0.7894,
+      "step": 874
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.6684333192059556e-05,
+      "loss": 0.9713,
+      "step": 875
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.667628026510504e-05,
+      "loss": 0.6776,
+      "step": 876
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.666821951934896e-05,
+      "loss": 0.687,
+      "step": 877
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6660150964231512e-05,
+      "loss": 0.6554,
+      "step": 878
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6652074609202035e-05,
+      "loss": 0.6541,
+      "step": 879
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6643990463718996e-05,
+      "loss": 0.7934,
+      "step": 880
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.663589853725e-05,
+      "loss": 0.7852,
+      "step": 881
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6627798839271742e-05,
+      "loss": 0.8408,
+      "step": 882
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6619691379270052e-05,
+      "loss": 0.8435,
+      "step": 883
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6611576166739814e-05,
+      "loss": 0.7853,
+      "step": 884
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.6603453211185022e-05,
+      "loss": 0.8569,
+      "step": 885
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6595322522118715e-05,
+      "loss": 0.7326,
+      "step": 886
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6587184109063002e-05,
+      "loss": 0.7659,
+      "step": 887
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6579037981549036e-05,
+      "loss": 0.7959,
+      "step": 888
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6570884149117e-05,
+      "loss": 0.702,
+      "step": 889
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6562722621316102e-05,
+      "loss": 0.7658,
+      "step": 890
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.655455340770457e-05,
+      "loss": 0.749,
+      "step": 891
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6546376517849617e-05,
+      "loss": 0.6493,
+      "step": 892
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6538191961327466e-05,
+      "loss": 0.7489,
+      "step": 893
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.65299997477233e-05,
+      "loss": 0.7172,
+      "step": 894
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.6521799886631286e-05,
+      "loss": 0.6461,
+      "step": 895
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6513592387654527e-05,
+      "loss": 0.7912,
+      "step": 896
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.7900890111923218,
+      "eval_runtime": 88.8611,
+      "eval_samples_per_second": 22.507,
+      "eval_steps_per_second": 11.254,
+      "step": 896
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6505377260405094e-05,
+      "loss": 0.6839,
+      "step": 897
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6497154514503976e-05,
+      "loss": 0.7584,
+      "step": 898
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.648892415958109e-05,
+      "loss": 0.7187,
+      "step": 899
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.648068620527526e-05,
+      "loss": 0.808,
+      "step": 900
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.647244066123422e-05,
+      "loss": 0.6535,
+      "step": 901
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6464187537114578e-05,
+      "loss": 0.7975,
+      "step": 902
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6455926842581834e-05,
+      "loss": 0.6695,
+      "step": 903
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.644765858731034e-05,
+      "loss": 0.6712,
+      "step": 904
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6439382780983315e-05,
+      "loss": 0.966,
+      "step": 905
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.643109943329281e-05,
+      "loss": 0.7732,
+      "step": 906
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6422808553939723e-05,
+      "loss": 0.7593,
+      "step": 907
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6414510152633752e-05,
+      "loss": 0.7174,
+      "step": 908
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6406204239093426e-05,
+      "loss": 0.851,
+      "step": 909
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6397890823046055e-05,
+      "loss": 0.7415,
+      "step": 910
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.638956991422774e-05,
+      "loss": 0.6796,
+      "step": 911
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6381241522383366e-05,
+      "loss": 0.7013,
+      "step": 912
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6372905657266566e-05,
+      "loss": 0.5842,
+      "step": 913
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.636456232863974e-05,
+      "loss": 0.7475,
+      "step": 914
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 1.6356211546274022e-05,
+      "loss": 0.821,
+      "step": 915
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.634785331994927e-05,
+      "loss": 0.9147,
+      "step": 916
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.633948765945407e-05,
+      "loss": 0.8074,
+      "step": 917
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.633111457458571e-05,
+      "loss": 0.7101,
+      "step": 918
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.632273407515017e-05,
+      "loss": 0.8883,
+      "step": 919
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6314346170962118e-05,
+      "loss": 0.7414,
+      "step": 920
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6305950871844893e-05,
+      "loss": 0.874,
+      "step": 921
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.6297548187630492e-05,
+      "loss": 0.8274,
+      "step": 922
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.628913812815956e-05,
+      "loss": 0.7398,
+      "step": 923
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.628072070328139e-05,
+      "loss": 0.792,
+      "step": 924
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.627229592285388e-05,
+      "loss": 0.6166,
+      "step": 925
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6263863796743572e-05,
+      "loss": 0.8314,
+      "step": 926
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6255424334825577e-05,
+      "loss": 0.6605,
+      "step": 927
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6246977546983624e-05,
+      "loss": 0.693,
+      "step": 928
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.623852344311001e-05,
+      "loss": 0.8222,
+      "step": 929
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6230062033105597e-05,
+      "loss": 0.8411,
+      "step": 930
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.622159332687981e-05,
+      "loss": 0.8027,
+      "step": 931
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6213117334350625e-05,
+      "loss": 0.5805,
+      "step": 932
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6204634065444534e-05,
+      "loss": 0.7377,
+      "step": 933
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.6196143530096563e-05,
+      "loss": 0.8146,
+      "step": 934
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.618764573825024e-05,
+      "loss": 0.7166,
+      "step": 935
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6179140699857607e-05,
+      "loss": 0.749,
+      "step": 936
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6170628424879167e-05,
+      "loss": 0.7324,
+      "step": 937
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6162108923283925e-05,
+      "loss": 0.738,
+      "step": 938
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.615358220504933e-05,
+      "loss": 0.7005,
+      "step": 939
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6145048280161294e-05,
+      "loss": 0.6305,
+      "step": 940
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.613650715861416e-05,
+      "loss": 0.6826,
+      "step": 941
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.612795885041071e-05,
+      "loss": 0.7812,
+      "step": 942
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.611940336556213e-05,
+      "loss": 0.7025,
+      "step": 943
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.611084071408802e-05,
+      "loss": 0.7932,
+      "step": 944
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6102270906016366e-05,
+      "loss": 0.6892,
+      "step": 945
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.609369395138355e-05,
+      "loss": 0.6442,
+      "step": 946
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.60851098602343e-05,
+      "loss": 0.8101,
+      "step": 947
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.607651864262173e-05,
+      "loss": 0.9417,
+      "step": 948
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6067920308607266e-05,
+      "loss": 0.7793,
+      "step": 949
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6059314868260703e-05,
+      "loss": 0.875,
+      "step": 950
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6050702331660132e-05,
+      "loss": 0.8437,
+      "step": 951
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6042082708891974e-05,
+      "loss": 0.9148,
+      "step": 952
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.6033456010050935e-05,
+      "loss": 0.6749,
+      "step": 953
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.602482224524001e-05,
+      "loss": 0.7153,
+      "step": 954
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.601618142457048e-05,
+      "loss": 0.6872,
+      "step": 955
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.6007533558161877e-05,
+      "loss": 0.6991,
+      "step": 956
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.599887865614199e-05,
+      "loss": 0.8357,
+      "step": 957
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5990216728646847e-05,
+      "loss": 0.817,
+      "step": 958
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5981547785820696e-05,
+      "loss": 0.7864,
+      "step": 959
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.597287183781602e-05,
+      "loss": 0.7218,
+      "step": 960
+    },
+    {
+      "epoch": 0.96,
+      "eval_loss": 0.7848817110061646,
+      "eval_runtime": 88.8132,
+      "eval_samples_per_second": 22.519,
+      "eval_steps_per_second": 11.26,
+      "step": 960
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5964188894793492e-05,
+      "loss": 0.8142,
+      "step": 961
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5955498966921975e-05,
+      "loss": 0.721,
+      "step": 962
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5946802064378512e-05,
+      "loss": 0.7197,
+      "step": 963
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5938098197348333e-05,
+      "loss": 0.8155,
+      "step": 964
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.5929387376024798e-05,
+      "loss": 0.7409,
+      "step": 965
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.592066961060943e-05,
+      "loss": 0.6935,
+      "step": 966
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.591194491131187e-05,
+      "loss": 0.6564,
+      "step": 967
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.59032132883499e-05,
+      "loss": 0.8281,
+      "step": 968
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5894474751949386e-05,
+      "loss": 0.6303,
+      "step": 969
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.588572931234431e-05,
+      "loss": 0.8682,
+      "step": 970
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5876976979776727e-05,
+      "loss": 0.7819,
+      "step": 971
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.586821776449677e-05,
+      "loss": 0.8028,
+      "step": 972
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5859451676762634e-05,
+      "loss": 0.6827,
+      "step": 973
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5850678726840555e-05,
+      "loss": 0.6692,
+      "step": 974
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.5841898925004813e-05,
+      "loss": 0.7941,
+      "step": 975
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5833112281537708e-05,
+      "loss": 0.6954,
+      "step": 976
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5824318806729557e-05,
+      "loss": 0.7135,
+      "step": 977
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.581551851087867e-05,
+      "loss": 0.7176,
+      "step": 978
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5806711404291356e-05,
+      "loss": 0.8105,
+      "step": 979
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5797897497281888e-05,
+      "loss": 0.9515,
+      "step": 980
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5789076800172515e-05,
+      "loss": 0.9346,
+      "step": 981
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5780249323293427e-05,
+      "loss": 0.8533,
+      "step": 982
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.577141507698277e-05,
+      "loss": 0.749,
+      "step": 983
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5762574071586592e-05,
+      "loss": 0.7779,
+      "step": 984
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.5753726317458884e-05,
+      "loss": 0.6955,
+      "step": 985
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.574487182496152e-05,
+      "loss": 0.8278,
+      "step": 986
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5736010604464284e-05,
+      "loss": 0.8578,
+      "step": 987
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5727142666344823e-05,
+      "loss": 0.7986,
+      "step": 988
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5718268020988664e-05,
+      "loss": 0.7456,
+      "step": 989
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5709386678789172e-05,
+      "loss": 0.6496,
+      "step": 990
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5700498650147577e-05,
+      "loss": 0.749,
+      "step": 991
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5691603945472923e-05,
+      "loss": 0.8465,
+      "step": 992
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.568270257518208e-05,
+      "loss": 0.6407,
+      "step": 993
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.567379454969972e-05,
+      "loss": 0.7428,
+      "step": 994
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.5664879879458316e-05,
+      "loss": 0.5955,
+      "step": 995
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5655958574898113e-05,
+      "loss": 0.7153,
+      "step": 996
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5647030646467136e-05,
+      "loss": 0.7766,
+      "step": 997
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.563809610462116e-05,
+      "loss": 0.84,
+      "step": 998
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5629154959823706e-05,
+      "loss": 0.7166,
+      "step": 999
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.562020722254603e-05,
+      "loss": 0.8409,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5611252903267112e-05,
+      "loss": 0.7234,
+      "step": 1001
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5602292012473634e-05,
+      "loss": 0.4725,
+      "step": 1002
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.559332456065998e-05,
+      "loss": 0.6734,
+      "step": 1003
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5584350558328204e-05,
+      "loss": 0.6656,
+      "step": 1004
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5575370015988056e-05,
+      "loss": 0.5797,
+      "step": 1005
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.5566382944156924e-05,
+      "loss": 0.5901,
+      "step": 1006
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5557389353359846e-05,
+      "loss": 0.8194,
+      "step": 1007
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.554838925412951e-05,
+      "loss": 0.6907,
+      "step": 1008
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5539382657006204e-05,
+      "loss": 0.5917,
+      "step": 1009
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5530369572537844e-05,
+      "loss": 0.615,
+      "step": 1010
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.552135001127993e-05,
+      "loss": 0.6122,
+      "step": 1011
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5512323983795565e-05,
+      "loss": 0.5004,
+      "step": 1012
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.55032915006554e-05,
+      "loss": 0.6083,
+      "step": 1013
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.5494252572437667e-05,
+      "loss": 0.746,
+      "step": 1014
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.548520720972814e-05,
+      "loss": 0.6606,
+      "step": 1015
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 1.547615542312012e-05,
+      "loss": 0.5767,
+      "step": 1016
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5467097223214444e-05,
+      "loss": 0.7176,
+      "step": 1017
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.545803262061946e-05,
+      "loss": 0.7197,
+      "step": 1018
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5448961625950996e-05,
+      "loss": 0.793,
+      "step": 1019
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.543988424983239e-05,
+      "loss": 0.5471,
+      "step": 1020
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5430800502894433e-05,
+      "loss": 0.6157,
+      "step": 1021
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.54217103957754e-05,
+      "loss": 0.8016,
+      "step": 1022
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.541261393912098e-05,
+      "loss": 0.5516,
+      "step": 1023
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5403511143584342e-05,
+      "loss": 0.6016,
+      "step": 1024
+    },
+    {
+      "epoch": 1.02,
+      "eval_loss": 0.7883525490760803,
+      "eval_runtime": 88.8444,
+      "eval_samples_per_second": 22.511,
+      "eval_steps_per_second": 11.256,
+      "step": 1024
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.539440201982604e-05,
+      "loss": 0.6044,
+      "step": 1025
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5385286578514065e-05,
+      "loss": 0.642,
+      "step": 1026
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.537616483032379e-05,
+      "loss": 0.6362,
+      "step": 1027
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5367036785937984e-05,
+      "loss": 0.6033,
+      "step": 1028
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.535790245604678e-05,
+      "loss": 0.7156,
+      "step": 1029
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.534876185134769e-05,
+      "loss": 0.7627,
+      "step": 1030
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5339614982545557e-05,
+      "loss": 0.8591,
+      "step": 1031
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5330461860352566e-05,
+      "loss": 0.6512,
+      "step": 1032
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5321302495488227e-05,
+      "loss": 0.5996,
+      "step": 1033
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.531213689867936e-05,
+      "loss": 0.6221,
+      "step": 1034
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5302965080660085e-05,
+      "loss": 0.6377,
+      "step": 1035
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.5293787052171802e-05,
+      "loss": 0.6357,
+      "step": 1036
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.528460282396319e-05,
+      "loss": 0.6439,
+      "step": 1037
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.527541240679019e-05,
+      "loss": 0.6825,
+      "step": 1038
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5266215811415985e-05,
+      "loss": 0.6813,
+      "step": 1039
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5257013048611003e-05,
+      "loss": 0.6871,
+      "step": 1040
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5247804129152881e-05,
+      "loss": 0.7229,
+      "step": 1041
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5238589063826477e-05,
+      "loss": 0.92,
+      "step": 1042
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5229367863423841e-05,
+      "loss": 0.6344,
+      "step": 1043
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5220140538744215e-05,
+      "loss": 0.7507,
+      "step": 1044
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5210907100594008e-05,
+      "loss": 0.7233,
+      "step": 1045
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.5201667559786784e-05,
+      "loss": 0.7648,
+      "step": 1046
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5192421927143266e-05,
+      "loss": 0.7114,
+      "step": 1047
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5183170213491302e-05,
+      "loss": 0.6684,
+      "step": 1048
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5173912429665865e-05,
+      "loss": 0.9303,
+      "step": 1049
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5164648586509038e-05,
+      "loss": 0.6457,
+      "step": 1050
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5155378694869996e-05,
+      "loss": 0.7582,
+      "step": 1051
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5146102765605001e-05,
+      "loss": 0.6437,
+      "step": 1052
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5136820809577386e-05,
+      "loss": 0.5888,
+      "step": 1053
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.512753283765754e-05,
+      "loss": 0.5791,
+      "step": 1054
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5118238860722903e-05,
+      "loss": 0.6762,
+      "step": 1055
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.5108938889657938e-05,
+      "loss": 0.6464,
+      "step": 1056
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5099632935354136e-05,
+      "loss": 0.661,
+      "step": 1057
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5090321008709992e-05,
+      "loss": 0.6115,
+      "step": 1058
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5081003120630993e-05,
+      "loss": 0.5149,
+      "step": 1059
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5071679282029616e-05,
+      "loss": 0.6131,
+      "step": 1060
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5062349503825298e-05,
+      "loss": 0.6786,
+      "step": 1061
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5053013796944433e-05,
+      "loss": 0.6981,
+      "step": 1062
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5043672172320367e-05,
+      "loss": 0.6555,
+      "step": 1063
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5034324640893365e-05,
+      "loss": 0.675,
+      "step": 1064
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5024971213610618e-05,
+      "loss": 0.5791,
+      "step": 1065
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.5015611901426214e-05,
+      "loss": 0.6364,
+      "step": 1066
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.5006246715301145e-05,
+      "loss": 0.7467,
+      "step": 1067
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4996875666203263e-05,
+      "loss": 0.7618,
+      "step": 1068
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4987498765107309e-05,
+      "loss": 0.5805,
+      "step": 1069
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4978116022994861e-05,
+      "loss": 0.5617,
+      "step": 1070
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4968727450854349e-05,
+      "loss": 0.6871,
+      "step": 1071
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4959333059681016e-05,
+      "loss": 0.4785,
+      "step": 1072
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4949932860476932e-05,
+      "loss": 0.6101,
+      "step": 1073
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4940526864250969e-05,
+      "loss": 0.5874,
+      "step": 1074
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.493111508201878e-05,
+      "loss": 0.6337,
+      "step": 1075
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 1.4921697524802805e-05,
+      "loss": 0.6763,
+      "step": 1076
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4912274203632235e-05,
+      "loss": 0.7148,
+      "step": 1077
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4902845129543017e-05,
+      "loss": 0.5686,
+      "step": 1078
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4893410313577843e-05,
+      "loss": 0.6511,
+      "step": 1079
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4883969766786113e-05,
+      "loss": 0.5792,
+      "step": 1080
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4874523500223958e-05,
+      "loss": 0.6251,
+      "step": 1081
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4865071524954192e-05,
+      "loss": 0.6641,
+      "step": 1082
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4855613852046316e-05,
+      "loss": 0.6197,
+      "step": 1083
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4846150492576517e-05,
+      "loss": 0.5618,
+      "step": 1084
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4836681457627625e-05,
+      "loss": 0.6289,
+      "step": 1085
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.482720675828913e-05,
+      "loss": 0.6836,
+      "step": 1086
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4817726405657145e-05,
+      "loss": 0.6174,
+      "step": 1087
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4808240410834413e-05,
+      "loss": 0.6353,
+      "step": 1088
+    },
+    {
+      "epoch": 1.09,
+      "eval_loss": 0.7834817171096802,
+      "eval_runtime": 88.7572,
+      "eval_samples_per_second": 22.533,
+      "eval_steps_per_second": 11.267,
+      "step": 1088
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4798748784930276e-05,
+      "loss": 0.546,
+      "step": 1089
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.478925153906068e-05,
+      "loss": 0.6468,
+      "step": 1090
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4779748684348147e-05,
+      "loss": 0.5424,
+      "step": 1091
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4770240231921769e-05,
+      "loss": 0.6575,
+      "step": 1092
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4760726192917187e-05,
+      "loss": 0.5227,
+      "step": 1093
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.47512065784766e-05,
+      "loss": 0.6903,
+      "step": 1094
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4741681399748724e-05,
+      "loss": 0.6109,
+      "step": 1095
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 1.4732150667888792e-05,
+      "loss": 0.6478,
+      "step": 1096
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.472261439405855e-05,
+      "loss": 0.6724,
+      "step": 1097
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4713072589426221e-05,
+      "loss": 0.6166,
+      "step": 1098
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4703525265166516e-05,
+      "loss": 0.8078,
+      "step": 1099
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4693972432460607e-05,
+      "loss": 0.6037,
+      "step": 1100
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4684414102496112e-05,
+      "loss": 0.722,
+      "step": 1101
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4674850286467099e-05,
+      "loss": 0.694,
+      "step": 1102
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4665280995574044e-05,
+      "loss": 0.6105,
+      "step": 1103
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4655706241023853e-05,
+      "loss": 0.6524,
+      "step": 1104
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4646126034029814e-05,
+      "loss": 0.7021,
+      "step": 1105
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4636540385811617e-05,
+      "loss": 0.5738,
+      "step": 1106
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.462694930759531e-05,
+      "loss": 0.5898,
+      "step": 1107
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4617352810613307e-05,
+      "loss": 0.6985,
+      "step": 1108
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4607750906104365e-05,
+      "loss": 0.6455,
+      "step": 1109
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4598143605313582e-05,
+      "loss": 0.6854,
+      "step": 1110
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4588530919492364e-05,
+      "loss": 0.5569,
+      "step": 1111
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4578912859898436e-05,
+      "loss": 0.5843,
+      "step": 1112
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4569289437795799e-05,
+      "loss": 0.6106,
+      "step": 1113
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4559660664454753e-05,
+      "loss": 0.6425,
+      "step": 1114
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4550026551151858e-05,
+      "loss": 0.5984,
+      "step": 1115
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.454038710916992e-05,
+      "loss": 0.5803,
+      "step": 1116
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4530742349798e-05,
+      "loss": 0.6532,
+      "step": 1117
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4521092284331371e-05,
+      "loss": 0.5944,
+      "step": 1118
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4511436924071536e-05,
+      "loss": 0.5596,
+      "step": 1119
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4501776280326182e-05,
+      "loss": 0.6126,
+      "step": 1120
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.44921103644092e-05,
+      "loss": 0.5562,
+      "step": 1121
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4482439187640646e-05,
+      "loss": 0.6613,
+      "step": 1122
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4472762761346738e-05,
+      "loss": 0.5612,
+      "step": 1123
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4463081096859837e-05,
+      "loss": 0.5644,
+      "step": 1124
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4453394205518457e-05,
+      "loss": 0.6157,
+      "step": 1125
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.4443702098667208e-05,
+      "loss": 1.298,
+      "step": 1126
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4434004787656832e-05,
+      "loss": 0.6685,
+      "step": 1127
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4424302283844145e-05,
+      "loss": 0.6245,
+      "step": 1128
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.441459459859206e-05,
+      "loss": 0.484,
+      "step": 1129
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4404881743269547e-05,
+      "loss": 0.5202,
+      "step": 1130
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4395163729251645e-05,
+      "loss": 0.6549,
+      "step": 1131
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4385440567919417e-05,
+      "loss": 0.6547,
+      "step": 1132
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4375712270659966e-05,
+      "loss": 0.6661,
+      "step": 1133
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4365978848866403e-05,
+      "loss": 0.549,
+      "step": 1134
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4356240313937853e-05,
+      "loss": 0.5194,
+      "step": 1135
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.434649667727941e-05,
+      "loss": 0.7109,
+      "step": 1136
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4336747950302159e-05,
+      "loss": 0.7517,
+      "step": 1137
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4326994144423138e-05,
+      "loss": 0.6625,
+      "step": 1138
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4317235271065339e-05,
+      "loss": 0.6192,
+      "step": 1139
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4307471341657679e-05,
+      "loss": 0.6246,
+      "step": 1140
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4297702367635009e-05,
+      "loss": 0.5272,
+      "step": 1141
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4287928360438077e-05,
+      "loss": 0.6396,
+      "step": 1142
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4278149331513534e-05,
+      "loss": 0.6444,
+      "step": 1143
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4268365292313899e-05,
+      "loss": 0.563,
+      "step": 1144
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4258576254297579e-05,
+      "loss": 0.6338,
+      "step": 1145
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4248782228928815e-05,
+      "loss": 0.7935,
+      "step": 1146
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4238983227677705e-05,
+      "loss": 0.6255,
+      "step": 1147
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4229179262020162e-05,
+      "loss": 0.521,
+      "step": 1148
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4219370343437924e-05,
+      "loss": 0.6963,
+      "step": 1149
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4209556483418517e-05,
+      "loss": 0.696,
+      "step": 1150
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4199737693455262e-05,
+      "loss": 0.6494,
+      "step": 1151
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4189913985047254e-05,
+      "loss": 0.67,
+      "step": 1152
+    },
+    {
+      "epoch": 1.15,
+      "eval_loss": 0.7798218131065369,
+      "eval_runtime": 88.7999,
+      "eval_samples_per_second": 22.523,
+      "eval_steps_per_second": 11.261,
+      "step": 1152
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4180085369699352e-05,
+      "loss": 0.724,
+      "step": 1153
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4170251858922144e-05,
+      "loss": 0.6832,
+      "step": 1154
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4160413464231975e-05,
+      "loss": 0.5867,
+      "step": 1155
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.4150570197150889e-05,
+      "loss": 0.6907,
+      "step": 1156
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4140722069206651e-05,
+      "loss": 0.6492,
+      "step": 1157
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4130869091932715e-05,
+      "loss": 0.4803,
+      "step": 1158
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4121011276868203e-05,
+      "loss": 0.678,
+      "step": 1159
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4111148635557923e-05,
+      "loss": 0.6601,
+      "step": 1160
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4101281179552316e-05,
+      "loss": 0.6202,
+      "step": 1161
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4091408920407475e-05,
+      "loss": 0.6806,
+      "step": 1162
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4081531869685107e-05,
+      "loss": 0.5971,
+      "step": 1163
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4071650038952542e-05,
+      "loss": 0.6096,
+      "step": 1164
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4061763439782698e-05,
+      "loss": 0.7169,
+      "step": 1165
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4051872083754088e-05,
+      "loss": 0.5959,
+      "step": 1166
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4041975982450782e-05,
+      "loss": 0.5372,
+      "step": 1167
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4032075147462423e-05,
+      "loss": 0.5357,
+      "step": 1168
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4022169590384183e-05,
+      "loss": 0.7213,
+      "step": 1169
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4012259322816775e-05,
+      "loss": 0.7302,
+      "step": 1170
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.4002344356366422e-05,
+      "loss": 0.7545,
+      "step": 1171
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3992424702644857e-05,
+      "loss": 0.5941,
+      "step": 1172
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3982500373269297e-05,
+      "loss": 0.6402,
+      "step": 1173
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3972571379862431e-05,
+      "loss": 0.6084,
+      "step": 1174
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3962637734052418e-05,
+      "loss": 0.5551,
+      "step": 1175
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.3952699447472866e-05,
+      "loss": 0.6604,
+      "step": 1176
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.394275653176281e-05,
+      "loss": 0.5899,
+      "step": 1177
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3932808998566713e-05,
+      "loss": 0.7172,
+      "step": 1178
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3922856859534445e-05,
+      "loss": 0.6294,
+      "step": 1179
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3912900126321268e-05,
+      "loss": 0.6395,
+      "step": 1180
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.390293881058782e-05,
+      "loss": 0.5647,
+      "step": 1181
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3892972924000118e-05,
+      "loss": 0.6165,
+      "step": 1182
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3883002478229521e-05,
+      "loss": 0.5617,
+      "step": 1183
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3873027484952735e-05,
+      "loss": 0.5541,
+      "step": 1184
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3863047955851781e-05,
+      "loss": 0.6338,
+      "step": 1185
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.3853063902614006e-05,
+      "loss": 0.7589,
+      "step": 1186
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3843075336932042e-05,
+      "loss": 0.6101,
+      "step": 1187
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.383308227050382e-05,
+      "loss": 0.6351,
+      "step": 1188
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3823084715032528e-05,
+      "loss": 0.7485,
+      "step": 1189
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.381308268222662e-05,
+      "loss": 0.676,
+      "step": 1190
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3803076183799788e-05,
+      "loss": 0.5848,
+      "step": 1191
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3793065231470961e-05,
+      "loss": 0.6669,
+      "step": 1192
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3783049836964276e-05,
+      "loss": 0.581,
+      "step": 1193
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3773030012009076e-05,
+      "loss": 0.6065,
+      "step": 1194
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.3763005768339897e-05,
+      "loss": 0.5967,
+      "step": 1195
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.375297711769644e-05,
+      "loss": 0.6504,
+      "step": 1196
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3742944071823578e-05,
+      "loss": 0.6016,
+      "step": 1197
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3732906642471324e-05,
+      "loss": 0.6782,
+      "step": 1198
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3722864841394827e-05,
+      "loss": 0.4762,
+      "step": 1199
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3712818680354356e-05,
+      "loss": 0.5421,
+      "step": 1200
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3702768171115283e-05,
+      "loss": 0.7471,
+      "step": 1201
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3692713325448079e-05,
+      "loss": 0.5636,
+      "step": 1202
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.368265415512829e-05,
+      "loss": 0.5299,
+      "step": 1203
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3672590671936524e-05,
+      "loss": 0.7712,
+      "step": 1204
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3662522887658443e-05,
+      "loss": 0.8533,
+      "step": 1205
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.3652450814084741e-05,
+      "loss": 0.6254,
+      "step": 1206
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.364237446301115e-05,
+      "loss": 0.7531,
+      "step": 1207
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3632293846238392e-05,
+      "loss": 0.8079,
+      "step": 1208
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.36222089755722e-05,
+      "loss": 0.6848,
+      "step": 1209
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.361211986282328e-05,
+      "loss": 0.5443,
+      "step": 1210
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3602026519807305e-05,
+      "loss": 0.7703,
+      "step": 1211
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3591928958344914e-05,
+      "loss": 0.6422,
+      "step": 1212
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3581827190261674e-05,
+      "loss": 0.5909,
+      "step": 1213
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3571721227388083e-05,
+      "loss": 0.6457,
+      "step": 1214
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3561611081559556e-05,
+      "loss": 0.5696,
+      "step": 1215
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.3551496764616392e-05,
+      "loss": 0.6306,
+      "step": 1216
+    },
+    {
+      "epoch": 1.21,
+      "eval_loss": 0.7770239114761353,
+      "eval_runtime": 88.8543,
+      "eval_samples_per_second": 22.509,
+      "eval_steps_per_second": 11.254,
+      "step": 1216
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3541378288403797e-05,
+      "loss": 0.6107,
+      "step": 1217
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.353125566477183e-05,
+      "loss": 0.5641,
+      "step": 1218
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3521128905575416e-05,
+      "loss": 0.6169,
+      "step": 1219
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.351099802267432e-05,
+      "loss": 0.6843,
+      "step": 1220
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3500863027933138e-05,
+      "loss": 0.6537,
+      "step": 1221
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3490723933221274e-05,
+      "loss": 0.4883,
+      "step": 1222
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3480580750412952e-05,
+      "loss": 0.6224,
+      "step": 1223
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3470433491387165e-05,
+      "loss": 0.6812,
+      "step": 1224
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3460282168027689e-05,
+      "loss": 0.5927,
+      "step": 1225
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.3450126792223054e-05,
+      "loss": 0.7314,
+      "step": 1226
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3439967375866545e-05,
+      "loss": 0.7307,
+      "step": 1227
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.342980393085617e-05,
+      "loss": 0.702,
+      "step": 1228
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3419636469094659e-05,
+      "loss": 0.5772,
+      "step": 1229
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3409465002489449e-05,
+      "loss": 0.6524,
+      "step": 1230
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.339928954295266e-05,
+      "loss": 0.5867,
+      "step": 1231
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3389110102401092e-05,
+      "loss": 0.4997,
+      "step": 1232
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3378926692756214e-05,
+      "loss": 0.5661,
+      "step": 1233
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.336873932594413e-05,
+      "loss": 0.7669,
+      "step": 1234
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3358548013895591e-05,
+      "loss": 0.5901,
+      "step": 1235
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.3348352768545957e-05,
+      "loss": 0.5802,
+      "step": 1236
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3338153601835203e-05,
+      "loss": 0.7795,
+      "step": 1237
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3327950525707895e-05,
+      "loss": 0.4799,
+      "step": 1238
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3317743552113175e-05,
+      "loss": 0.6811,
+      "step": 1239
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.330753269300475e-05,
+      "loss": 0.6215,
+      "step": 1240
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3297317960340874e-05,
+      "loss": 0.501,
+      "step": 1241
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3287099366084349e-05,
+      "loss": 0.731,
+      "step": 1242
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3276876922202484e-05,
+      "loss": 0.5944,
+      "step": 1243
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3266650640667113e-05,
+      "loss": 0.5098,
+      "step": 1244
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3256420533454549e-05,
+      "loss": 0.6702,
+      "step": 1245
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.3246186612545593e-05,
+      "loss": 0.6353,
+      "step": 1246
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3235948889925511e-05,
+      "loss": 0.5516,
+      "step": 1247
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3225707377584024e-05,
+      "loss": 0.521,
+      "step": 1248
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3215462087515285e-05,
+      "loss": 0.6003,
+      "step": 1249
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3205213031717881e-05,
+      "loss": 0.4752,
+      "step": 1250
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3194960222194796e-05,
+      "loss": 0.534,
+      "step": 1251
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3184703670953425e-05,
+      "loss": 0.5437,
+      "step": 1252
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3174443390005529e-05,
+      "loss": 0.5653,
+      "step": 1253
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3164179391367253e-05,
+      "loss": 0.5956,
+      "step": 1254
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3153911687059084e-05,
+      "loss": 0.7131,
+      "step": 1255
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.3143640289105855e-05,
+      "loss": 0.7432,
+      "step": 1256
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.313336520953672e-05,
+      "loss": 0.7356,
+      "step": 1257
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3123086460385148e-05,
+      "loss": 0.57,
+      "step": 1258
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3112804053688907e-05,
+      "loss": 0.5759,
+      "step": 1259
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3102518001490049e-05,
+      "loss": 0.6925,
+      "step": 1260
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3092228315834889e-05,
+      "loss": 0.6121,
+      "step": 1261
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3081935008774003e-05,
+      "loss": 0.6483,
+      "step": 1262
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.30716380923622e-05,
+      "loss": 0.6398,
+      "step": 1263
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3061337578658533e-05,
+      "loss": 0.627,
+      "step": 1264
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3051033479726249e-05,
+      "loss": 0.577,
+      "step": 1265
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.3040725807632805e-05,
+      "loss": 0.7611,
+      "step": 1266
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.3030414574449842e-05,
+      "loss": 0.5717,
+      "step": 1267
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.3020099792253163e-05,
+      "loss": 0.6522,
+      "step": 1268
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.3009781473122738e-05,
+      "loss": 0.5659,
+      "step": 1269
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2999459629142672e-05,
+      "loss": 0.7753,
+      "step": 1270
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2989134272401205e-05,
+      "loss": 0.7436,
+      "step": 1271
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2978805414990683e-05,
+      "loss": 0.5635,
+      "step": 1272
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2968473069007556e-05,
+      "loss": 0.7438,
+      "step": 1273
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.295813724655236e-05,
+      "loss": 0.6258,
+      "step": 1274
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2947797959729702e-05,
+      "loss": 0.6669,
+      "step": 1275
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.2937455220648244e-05,
+      "loss": 0.5492,
+      "step": 1276
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2927109041420694e-05,
+      "loss": 0.7355,
+      "step": 1277
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2916759434163784e-05,
+      "loss": 0.6018,
+      "step": 1278
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2906406410998269e-05,
+      "loss": 0.6431,
+      "step": 1279
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2896049984048897e-05,
+      "loss": 0.6572,
+      "step": 1280
+    },
+    {
+      "epoch": 1.28,
+      "eval_loss": 0.7731580138206482,
+      "eval_runtime": 88.8257,
+      "eval_samples_per_second": 22.516,
+      "eval_steps_per_second": 11.258,
+      "step": 1280
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2885690165444407e-05,
+      "loss": 0.6016,
+      "step": 1281
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2875326967317502e-05,
+      "loss": 0.5719,
+      "step": 1282
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2864960401804858e-05,
+      "loss": 0.5386,
+      "step": 1283
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2854590481047078e-05,
+      "loss": 0.6155,
+      "step": 1284
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2844217217188705e-05,
+      "loss": 0.5936,
+      "step": 1285
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.2833840622378197e-05,
+      "loss": 0.5908,
+      "step": 1286
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2823460708767908e-05,
+      "loss": 0.8534,
+      "step": 1287
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2813077488514077e-05,
+      "loss": 0.4497,
+      "step": 1288
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2802690973776823e-05,
+      "loss": 0.5683,
+      "step": 1289
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2792301176720118e-05,
+      "loss": 0.6186,
+      "step": 1290
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2781908109511788e-05,
+      "loss": 0.5901,
+      "step": 1291
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2771511784323467e-05,
+      "loss": 0.7563,
+      "step": 1292
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2761112213330628e-05,
+      "loss": 0.5137,
+      "step": 1293
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.275070940871253e-05,
+      "loss": 0.6203,
+      "step": 1294
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2740303382652229e-05,
+      "loss": 0.5984,
+      "step": 1295
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.2729894147336543e-05,
+      "loss": 0.5278,
+      "step": 1296
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.271948171495606e-05,
+      "loss": 0.6971,
+      "step": 1297
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2709066097705103e-05,
+      "loss": 0.5036,
+      "step": 1298
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2698647307781733e-05,
+      "loss": 0.5206,
+      "step": 1299
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2688225357387715e-05,
+      "loss": 0.767,
+      "step": 1300
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.267780025872853e-05,
+      "loss": 0.7009,
+      "step": 1301
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2667372024013335e-05,
+      "loss": 0.6248,
+      "step": 1302
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2656940665454967e-05,
+      "loss": 0.5873,
+      "step": 1303
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.264650619526991e-05,
+      "loss": 0.5126,
+      "step": 1304
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.2636068625678307e-05,
+      "loss": 0.7721,
+      "step": 1305
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.262562796890392e-05,
+      "loss": 0.6494,
+      "step": 1306
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2615184237174127e-05,
+      "loss": 0.5494,
+      "step": 1307
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.260473744271992e-05,
+      "loss": 0.6009,
+      "step": 1308
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2594287597775859e-05,
+      "loss": 0.5642,
+      "step": 1309
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.258383471458009e-05,
+      "loss": 0.4378,
+      "step": 1310
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2573378805374314e-05,
+      "loss": 0.8139,
+      "step": 1311
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2562919882403774e-05,
+      "loss": 0.7645,
+      "step": 1312
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2552457957917239e-05,
+      "loss": 0.59,
+      "step": 1313
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2541993044167002e-05,
+      "loss": 0.4468,
+      "step": 1314
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.253152515340885e-05,
+      "loss": 0.5286,
+      "step": 1315
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.2521054297902062e-05,
+      "loss": 0.6379,
+      "step": 1316
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2510580489909383e-05,
+      "loss": 0.7154,
+      "step": 1317
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.250010374169702e-05,
+      "loss": 0.5415,
+      "step": 1318
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2489624065534618e-05,
+      "loss": 0.6146,
+      "step": 1319
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.247914147369526e-05,
+      "loss": 0.6026,
+      "step": 1320
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2468655978455437e-05,
+      "loss": 0.6495,
+      "step": 1321
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2458167592095043e-05,
+      "loss": 0.5561,
+      "step": 1322
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.244767632689735e-05,
+      "loss": 0.5976,
+      "step": 1323
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2437182195149018e-05,
+      "loss": 0.5668,
+      "step": 1324
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2426685209140048e-05,
+      "loss": 0.5744,
+      "step": 1325
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.2416185381163795e-05,
+      "loss": 0.57,
+      "step": 1326
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2405682723516932e-05,
+      "loss": 0.7311,
+      "step": 1327
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2395177248499456e-05,
+      "loss": 0.4972,
+      "step": 1328
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2384668968414655e-05,
+      "loss": 0.5834,
+      "step": 1329
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2374157895569107e-05,
+      "loss": 0.4149,
+      "step": 1330
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.236364404227266e-05,
+      "loss": 0.5383,
+      "step": 1331
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2353127420838422e-05,
+      "loss": 0.5928,
+      "step": 1332
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2342608043582731e-05,
+      "loss": 0.6482,
+      "step": 1333
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2332085922825165e-05,
+      "loss": 0.5729,
+      "step": 1334
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.232156107088851e-05,
+      "loss": 0.8532,
+      "step": 1335
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.231103350009875e-05,
+      "loss": 0.6038,
+      "step": 1336
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2300503222785054e-05,
+      "loss": 0.5516,
+      "step": 1337
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2289970251279761e-05,
+      "loss": 0.6649,
+      "step": 1338
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.227943459791836e-05,
+      "loss": 0.6155,
+      "step": 1339
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2268896275039495e-05,
+      "loss": 0.6069,
+      "step": 1340
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2258355294984919e-05,
+      "loss": 0.5592,
+      "step": 1341
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.224781167009951e-05,
+      "loss": 0.6174,
+      "step": 1342
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2237265412731237e-05,
+      "loss": 0.6242,
+      "step": 1343
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.222671653523115e-05,
+      "loss": 0.4869,
+      "step": 1344
+    },
+    {
+      "epoch": 1.34,
+      "eval_loss": 0.7695678472518921,
+      "eval_runtime": 88.7912,
+      "eval_samples_per_second": 22.525,
+      "eval_steps_per_second": 11.262,
+      "step": 1344
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2216165049953374e-05,
+      "loss": 0.7009,
+      "step": 1345
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2205610969255085e-05,
+      "loss": 0.5575,
+      "step": 1346
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2195054305496499e-05,
+      "loss": 0.5356,
+      "step": 1347
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2184495071040851e-05,
+      "loss": 0.6166,
+      "step": 1348
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2173933278254402e-05,
+      "loss": 0.7019,
+      "step": 1349
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.216336893950639e-05,
+      "loss": 0.7516,
+      "step": 1350
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2152802067169049e-05,
+      "loss": 0.6056,
+      "step": 1351
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2142232673617576e-05,
+      "loss": 0.6766,
+      "step": 1352
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2131660771230115e-05,
+      "loss": 0.5917,
+      "step": 1353
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2121086372387757e-05,
+      "loss": 0.6151,
+      "step": 1354
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2110509489474514e-05,
+      "loss": 0.7001,
+      "step": 1355
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.2099930134877298e-05,
+      "loss": 0.7416,
+      "step": 1356
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2089348320985935e-05,
+      "loss": 0.6118,
+      "step": 1357
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.207876406019311e-05,
+      "loss": 0.5516,
+      "step": 1358
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2068177364894388e-05,
+      "loss": 0.6856,
+      "step": 1359
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2057588247488179e-05,
+      "loss": 0.8338,
+      "step": 1360
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2046996720375732e-05,
+      "loss": 0.6367,
+      "step": 1361
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2036402795961117e-05,
+      "loss": 0.675,
+      "step": 1362
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2025806486651212e-05,
+      "loss": 0.6259,
+      "step": 1363
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.2015207804855686e-05,
+      "loss": 0.7106,
+      "step": 1364
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.200460676298699e-05,
+      "loss": 0.6907,
+      "step": 1365
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.1994003373460337e-05,
+      "loss": 0.63,
+      "step": 1366
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1983397648693687e-05,
+      "loss": 0.7309,
+      "step": 1367
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1972789601107742e-05,
+      "loss": 0.5566,
+      "step": 1368
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1962179243125918e-05,
+      "loss": 0.7016,
+      "step": 1369
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1951566587174335e-05,
+      "loss": 0.6313,
+      "step": 1370
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1940951645681814e-05,
+      "loss": 0.5732,
+      "step": 1371
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.193033443107984e-05,
+      "loss": 0.626,
+      "step": 1372
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1919714955802573e-05,
+      "loss": 0.5893,
+      "step": 1373
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1909093232286809e-05,
+      "loss": 0.5866,
+      "step": 1374
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1898469272971985e-05,
+      "loss": 0.6023,
+      "step": 1375
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1887843090300149e-05,
+      "loss": 0.5154,
+      "step": 1376
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1877214696715963e-05,
+      "loss": 0.6451,
+      "step": 1377
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1866584104666672e-05,
+      "loss": 0.5227,
+      "step": 1378
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1855951326602095e-05,
+      "loss": 0.6644,
+      "step": 1379
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.184531637497461e-05,
+      "loss": 0.6916,
+      "step": 1380
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1834679262239149e-05,
+      "loss": 0.6205,
+      "step": 1381
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.182404000085316e-05,
+      "loss": 0.6818,
+      "step": 1382
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1813398603276625e-05,
+      "loss": 0.6786,
+      "step": 1383
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1802755081972015e-05,
+      "loss": 0.6402,
+      "step": 1384
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1792109449404292e-05,
+      "loss": 0.5516,
+      "step": 1385
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.1781461718040892e-05,
+      "loss": 0.6143,
+      "step": 1386
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1770811900351704e-05,
+      "loss": 0.5959,
+      "step": 1387
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1760160008809073e-05,
+      "loss": 0.5255,
+      "step": 1388
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1749506055887754e-05,
+      "loss": 0.5965,
+      "step": 1389
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1738850054064933e-05,
+      "loss": 0.565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1728192015820179e-05,
+      "loss": 0.5639,
+      "step": 1391
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1717531953635468e-05,
+      "loss": 0.5898,
+      "step": 1392
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1706869879995126e-05,
+      "loss": 0.7147,
+      "step": 1393
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1696205807385846e-05,
+      "loss": 0.7228,
+      "step": 1394
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1685539748296653e-05,
+      "loss": 0.7181,
+      "step": 1395
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1674871715218908e-05,
+      "loss": 0.5916,
+      "step": 1396
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1664201720646277e-05,
+      "loss": 0.5727,
+      "step": 1397
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1653529777074733e-05,
+      "loss": 0.5375,
+      "step": 1398
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1642855897002516e-05,
+      "loss": 0.5874,
+      "step": 1399
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1632180092930147e-05,
+      "loss": 0.6945,
+      "step": 1400
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1621502377360392e-05,
+      "loss": 0.6145,
+      "step": 1401
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1610822762798264e-05,
+      "loss": 0.7252,
+      "step": 1402
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1600141261750992e-05,
+      "loss": 0.6266,
+      "step": 1403
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1589457886728017e-05,
+      "loss": 0.5836,
+      "step": 1404
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1578772650240975e-05,
+      "loss": 0.5595,
+      "step": 1405
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.1568085564803684e-05,
+      "loss": 0.569,
+      "step": 1406
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1557396642932123e-05,
+      "loss": 0.5619,
+      "step": 1407
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.154670589714443e-05,
+      "loss": 0.6385,
+      "step": 1408
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.7667235136032104,
+      "eval_runtime": 88.7544,
+      "eval_samples_per_second": 22.534,
+      "eval_steps_per_second": 11.267,
+      "step": 1408
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1536013339960868e-05,
+      "loss": 0.5646,
+      "step": 1409
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1525318983903826e-05,
+      "loss": 0.4848,
+      "step": 1410
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.15146228414978e-05,
+      "loss": 0.6505,
+      "step": 1411
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1503924925269385e-05,
+      "loss": 0.7031,
+      "step": 1412
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1493225247747237e-05,
+      "loss": 0.7346,
+      "step": 1413
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1482523821462097e-05,
+      "loss": 0.7447,
+      "step": 1414
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1471820658946729e-05,
+      "loss": 0.5145,
+      "step": 1415
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 1.1461115772735949e-05,
+      "loss": 0.58,
+      "step": 1416
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1450409175366584e-05,
+      "loss": 0.5342,
+      "step": 1417
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.143970087937747e-05,
+      "loss": 0.7084,
+      "step": 1418
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1428990897309424e-05,
+      "loss": 0.6313,
+      "step": 1419
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1418279241705244e-05,
+      "loss": 0.5863,
+      "step": 1420
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.140756592510969e-05,
+      "loss": 0.6794,
+      "step": 1421
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1396850960069456e-05,
+      "loss": 0.6021,
+      "step": 1422
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1386134359133183e-05,
+      "loss": 0.5103,
+      "step": 1423
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1375416134851413e-05,
+      "loss": 0.5243,
+      "step": 1424
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1364696299776599e-05,
+      "loss": 0.6185,
+      "step": 1425
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1353974866463069e-05,
+      "loss": 0.6217,
+      "step": 1426
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1343251847467041e-05,
+      "loss": 0.6739,
+      "step": 1427
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1332527255346572e-05,
+      "loss": 0.6818,
+      "step": 1428
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1321801102661576e-05,
+      "loss": 0.6843,
+      "step": 1429
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1311073401973775e-05,
+      "loss": 0.5697,
+      "step": 1430
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.130034416584673e-05,
+      "loss": 0.733,
+      "step": 1431
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1289613406845776e-05,
+      "loss": 0.5466,
+      "step": 1432
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1278881137538052e-05,
+      "loss": 0.633,
+      "step": 1433
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1268147370492447e-05,
+      "loss": 0.6818,
+      "step": 1434
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1257412118279616e-05,
+      "loss": 0.5666,
+      "step": 1435
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.1246675393471946e-05,
+      "loss": 0.6373,
+      "step": 1436
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1235937208643564e-05,
+      "loss": 0.7219,
+      "step": 1437
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1225197576370282e-05,
+      "loss": 0.7533,
+      "step": 1438
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.121445650922963e-05,
+      "loss": 0.6122,
+      "step": 1439
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1203714019800807e-05,
+      "loss": 0.5603,
+      "step": 1440
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1192970120664679e-05,
+      "loss": 0.5279,
+      "step": 1441
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1182224824403763e-05,
+      "loss": 0.7464,
+      "step": 1442
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1171478143602215e-05,
+      "loss": 0.6312,
+      "step": 1443
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.116073009084581e-05,
+      "loss": 0.659,
+      "step": 1444
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.114998067872193e-05,
+      "loss": 0.6779,
+      "step": 1445
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 1.1139229919819545e-05,
+      "loss": 0.5554,
+      "step": 1446
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1128477826729214e-05,
+      "loss": 0.5,
+      "step": 1447
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1117724412043046e-05,
+      "loss": 0.5753,
+      "step": 1448
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1106969688354705e-05,
+      "loss": 0.6264,
+      "step": 1449
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1096213668259386e-05,
+      "loss": 0.6996,
+      "step": 1450
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1085456364353801e-05,
+      "loss": 0.5337,
+      "step": 1451
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1074697789236167e-05,
+      "loss": 0.6704,
+      "step": 1452
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1063937955506188e-05,
+      "loss": 0.6089,
+      "step": 1453
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1053176875765042e-05,
+      "loss": 0.6685,
+      "step": 1454
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1042414562615375e-05,
+      "loss": 0.562,
+      "step": 1455
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1031651028661261e-05,
+      "loss": 0.5573,
+      "step": 1456
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.1020886286508218e-05,
+      "loss": 0.6688,
+      "step": 1457
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.1010120348763171e-05,
+      "loss": 0.5279,
+      "step": 1458
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.099935322803445e-05,
+      "loss": 0.6145,
+      "step": 1459
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.098858493693177e-05,
+      "loss": 0.5528,
+      "step": 1460
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.097781548806621e-05,
+      "loss": 0.5533,
+      "step": 1461
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0967044894050214e-05,
+      "loss": 0.7398,
+      "step": 1462
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.095627316749756e-05,
+      "loss": 0.5449,
+      "step": 1463
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.094550032102336e-05,
+      "loss": 0.5651,
+      "step": 1464
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0934726367244026e-05,
+      "loss": 0.7718,
+      "step": 1465
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.0923951318777282e-05,
+      "loss": 0.5764,
+      "step": 1466
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0913175188242118e-05,
+      "loss": 0.568,
+      "step": 1467
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0902397988258804e-05,
+      "loss": 0.5253,
+      "step": 1468
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0891619731448855e-05,
+      "loss": 0.5567,
+      "step": 1469
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0880840430435029e-05,
+      "loss": 0.5784,
+      "step": 1470
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.08700600978413e-05,
+      "loss": 0.6781,
+      "step": 1471
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0859278746292858e-05,
+      "loss": 0.5973,
+      "step": 1472
+    },
+    {
+      "epoch": 1.47,
+      "eval_loss": 0.7631298303604126,
+      "eval_runtime": 88.8206,
+      "eval_samples_per_second": 22.517,
+      "eval_steps_per_second": 11.259,
+      "step": 1472
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0848496388416079e-05,
+      "loss": 0.6337,
+      "step": 1473
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0837713036838524e-05,
+      "loss": 0.6297,
+      "step": 1474
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.0826928704188914e-05,
+      "loss": 0.6019,
+      "step": 1475
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.081614340309712e-05,
+      "loss": 0.5909,
+      "step": 1476
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0805357146194143e-05,
+      "loss": 0.7228,
+      "step": 1477
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0794569946112114e-05,
+      "loss": 0.6465,
+      "step": 1478
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0783781815484258e-05,
+      "loss": 0.7455,
+      "step": 1479
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0772992766944894e-05,
+      "loss": 0.6007,
+      "step": 1480
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0762202813129419e-05,
+      "loss": 0.5761,
+      "step": 1481
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0751411966674281e-05,
+      "loss": 0.6086,
+      "step": 1482
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0740620240216982e-05,
+      "loss": 0.5995,
+      "step": 1483
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0729827646396054e-05,
+      "loss": 0.6325,
+      "step": 1484
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.071903419785104e-05,
+      "loss": 0.7608,
+      "step": 1485
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.0708239907222487e-05,
+      "loss": 0.5423,
+      "step": 1486
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0697444787151927e-05,
+      "loss": 0.7455,
+      "step": 1487
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0686648850281869e-05,
+      "loss": 0.6594,
+      "step": 1488
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0675852109255767e-05,
+      "loss": 0.7232,
+      "step": 1489
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0665054576718028e-05,
+      "loss": 0.5721,
+      "step": 1490
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0654256265313982e-05,
+      "loss": 0.6063,
+      "step": 1491
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.064345718768987e-05,
+      "loss": 0.5577,
+      "step": 1492
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0632657356492828e-05,
+      "loss": 0.5984,
+      "step": 1493
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0621856784370883e-05,
+      "loss": 0.597,
+      "step": 1494
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.0611055483972916e-05,
+      "loss": 0.7085,
+      "step": 1495
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.060025346794868e-05,
+      "loss": 0.5722,
+      "step": 1496
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0589450748948749e-05,
+      "loss": 0.6626,
+      "step": 1497
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0578647339624526e-05,
+      "loss": 0.5803,
+      "step": 1498
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0567843252628225e-05,
+      "loss": 0.5419,
+      "step": 1499
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0557038500612853e-05,
+      "loss": 0.5597,
+      "step": 1500
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0546233096232192e-05,
+      "loss": 0.5009,
+      "step": 1501
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.053542705214079e-05,
+      "loss": 0.6316,
+      "step": 1502
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0524620380993946e-05,
+      "loss": 0.6555,
+      "step": 1503
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0513813095447694e-05,
+      "loss": 0.64,
+      "step": 1504
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.0503005208158783e-05,
+      "loss": 0.6681,
+      "step": 1505
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.049219673178467e-05,
+      "loss": 0.5214,
+      "step": 1506
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0481387678983503e-05,
+      "loss": 0.6576,
+      "step": 1507
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0470578062414098e-05,
+      "loss": 0.6077,
+      "step": 1508
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0459767894735945e-05,
+      "loss": 0.6132,
+      "step": 1509
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0448957188609162e-05,
+      "loss": 0.5449,
+      "step": 1510
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0438145956694514e-05,
+      "loss": 0.6,
+      "step": 1511
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.042733421165337e-05,
+      "loss": 0.5882,
+      "step": 1512
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0416521966147706e-05,
+      "loss": 0.6499,
+      "step": 1513
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.040570923284008e-05,
+      "loss": 0.4968,
+      "step": 1514
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0394896024393626e-05,
+      "loss": 0.5561,
+      "step": 1515
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.0384082353472033e-05,
+      "loss": 0.6406,
+      "step": 1516
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0373268232739531e-05,
+      "loss": 0.6541,
+      "step": 1517
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.036245367486087e-05,
+      "loss": 0.6876,
+      "step": 1518
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0351638692501327e-05,
+      "loss": 0.5352,
+      "step": 1519
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0340823298326663e-05,
+      "loss": 0.5774,
+      "step": 1520
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0330007505003127e-05,
+      "loss": 0.5215,
+      "step": 1521
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0319191325197436e-05,
+      "loss": 0.5971,
+      "step": 1522
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0308374771576755e-05,
+      "loss": 0.6049,
+      "step": 1523
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0297557856808689e-05,
+      "loss": 0.6297,
+      "step": 1524
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.028674059356127e-05,
+      "loss": 0.6259,
+      "step": 1525
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.0275922994502936e-05,
+      "loss": 0.6841,
+      "step": 1526
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0265105072302512e-05,
+      "loss": 0.6333,
+      "step": 1527
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0254286839629211e-05,
+      "loss": 0.5563,
+      "step": 1528
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0243468309152603e-05,
+      "loss": 0.5459,
+      "step": 1529
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0232649493542605e-05,
+      "loss": 0.5272,
+      "step": 1530
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0221830405469481e-05,
+      "loss": 0.5555,
+      "step": 1531
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.02110110576038e-05,
+      "loss": 0.6171,
+      "step": 1532
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0200191462616437e-05,
+      "loss": 0.5179,
+      "step": 1533
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0189371633178567e-05,
+      "loss": 0.6832,
+      "step": 1534
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0178551581961622e-05,
+      "loss": 0.565,
+      "step": 1535
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.0167731321637315e-05,
+      "loss": 0.6059,
+      "step": 1536
+    },
+    {
+      "epoch": 1.53,
+      "eval_loss": 0.7608674168586731,
+      "eval_runtime": 88.7412,
+      "eval_samples_per_second": 22.537,
+      "eval_steps_per_second": 11.269,
+      "step": 1536
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0156910864877587e-05,
+      "loss": 0.573,
+      "step": 1537
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0146090224354617e-05,
+      "loss": 0.4602,
+      "step": 1538
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0135269412740791e-05,
+      "loss": 0.581,
+      "step": 1539
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.012444844270871e-05,
+      "loss": 0.6084,
+      "step": 1540
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0113627326931148e-05,
+      "loss": 0.4998,
+      "step": 1541
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0102806078081053e-05,
+      "loss": 0.6679,
+      "step": 1542
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0091984708831528e-05,
+      "loss": 0.6294,
+      "step": 1543
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0081163231855825e-05,
+      "loss": 0.5248,
+      "step": 1544
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0070341659827305e-05,
+      "loss": 0.6199,
+      "step": 1545
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 1.0059520005419463e-05,
+      "loss": 0.5435,
+      "step": 1546
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0048698281305871e-05,
+      "loss": 0.7399,
+      "step": 1547
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.003787650016019e-05,
+      "loss": 0.7372,
+      "step": 1548
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0027054674656152e-05,
+      "loss": 0.7732,
+      "step": 1549
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0016232817467535e-05,
+      "loss": 0.5891,
+      "step": 1550
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.0005410941268153e-05,
+      "loss": 0.5504,
+      "step": 1551
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.994589058731849e-06,
+      "loss": 0.5565,
+      "step": 1552
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.983767182532468e-06,
+      "loss": 0.7935,
+      "step": 1553
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.972945325343851e-06,
+      "loss": 0.5711,
+      "step": 1554
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.962123499839813e-06,
+      "loss": 0.5962,
+      "step": 1555
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 9.951301718694134e-06,
+      "loss": 0.634,
+      "step": 1556
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.94047999458054e-06,
+      "loss": 0.5461,
+      "step": 1557
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.929658340172698e-06,
+      "loss": 0.686,
+      "step": 1558
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.918836768144182e-06,
+      "loss": 0.6784,
+      "step": 1559
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.908015291168472e-06,
+      "loss": 0.8519,
+      "step": 1560
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.89719392191895e-06,
+      "loss": 0.8432,
+      "step": 1561
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.886372673068856e-06,
+      "loss": 0.6555,
+      "step": 1562
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.875551557291296e-06,
+      "loss": 0.6079,
+      "step": 1563
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.864730587259209e-06,
+      "loss": 0.5165,
+      "step": 1564
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.853909775645387e-06,
+      "loss": 0.7481,
+      "step": 1565
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.843089135122415e-06,
+      "loss": 0.5803,
+      "step": 1566
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.832268678362687e-06,
+      "loss": 0.7437,
+      "step": 1567
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.821448418038378e-06,
+      "loss": 0.8093,
+      "step": 1568
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.810628366821437e-06,
+      "loss": 0.5449,
+      "step": 1569
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.799808537383566e-06,
+      "loss": 0.5401,
+      "step": 1570
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.788988942396204e-06,
+      "loss": 0.5308,
+      "step": 1571
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.778169594530522e-06,
+      "loss": 0.775,
+      "step": 1572
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.767350506457397e-06,
+      "loss": 0.5956,
+      "step": 1573
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.756531690847402e-06,
+      "loss": 0.6567,
+      "step": 1574
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.74571316037079e-06,
+      "loss": 0.5888,
+      "step": 1575
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 9.73489492769749e-06,
+      "loss": 0.6409,
+      "step": 1576
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.724077005497067e-06,
+      "loss": 0.6608,
+      "step": 1577
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.713259406438734e-06,
+      "loss": 0.7137,
+      "step": 1578
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.702442143191311e-06,
+      "loss": 0.8226,
+      "step": 1579
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.691625228423248e-06,
+      "loss": 0.7155,
+      "step": 1580
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.680808674802568e-06,
+      "loss": 0.5869,
+      "step": 1581
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.669992494996876e-06,
+      "loss": 0.5947,
+      "step": 1582
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.659176701673339e-06,
+      "loss": 0.7556,
+      "step": 1583
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.648361307498675e-06,
+      "loss": 0.5863,
+      "step": 1584
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.637546325139133e-06,
+      "loss": 0.533,
+      "step": 1585
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 9.626731767260474e-06,
+      "loss": 0.5881,
+      "step": 1586
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.615917646527969e-06,
+      "loss": 0.5451,
+      "step": 1587
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.605103975606376e-06,
+      "loss": 0.5595,
+      "step": 1588
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.594290767159924e-06,
+      "loss": 0.5347,
+      "step": 1589
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.583478033852299e-06,
+      "loss": 0.5816,
+      "step": 1590
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.572665788346632e-06,
+      "loss": 0.6841,
+      "step": 1591
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.561854043305488e-06,
+      "loss": 0.5027,
+      "step": 1592
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.551042811390841e-06,
+      "loss": 0.5617,
+      "step": 1593
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.54023210526406e-06,
+      "loss": 0.6192,
+      "step": 1594
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.529421937585902e-06,
+      "loss": 0.747,
+      "step": 1595
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 9.5186123210165e-06,
+      "loss": 0.6287,
+      "step": 1596
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.507803268215331e-06,
+      "loss": 0.5114,
+      "step": 1597
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.49699479184122e-06,
+      "loss": 0.566,
+      "step": 1598
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.486186904552308e-06,
+      "loss": 0.493,
+      "step": 1599
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.475379619006055e-06,
+      "loss": 0.6974,
+      "step": 1600
+    },
+    {
+      "epoch": 1.6,
+      "eval_loss": 0.7587617039680481,
+      "eval_runtime": 88.8538,
+      "eval_samples_per_second": 22.509,
+      "eval_steps_per_second": 11.254,
+      "step": 1600
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.464572947859213e-06,
+      "loss": 0.6029,
+      "step": 1601
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.453766903767813e-06,
+      "loss": 0.5481,
+      "step": 1602
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.44296149938715e-06,
+      "loss": 0.6451,
+      "step": 1603
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.432156747371778e-06,
+      "loss": 0.6104,
+      "step": 1604
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.421352660375478e-06,
+      "loss": 0.4375,
+      "step": 1605
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 9.410549251051256e-06,
+      "loss": 0.6665,
+      "step": 1606
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.39974653205132e-06,
+      "loss": 0.5493,
+      "step": 1607
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.388944516027086e-06,
+      "loss": 0.5839,
+      "step": 1608
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.378143215629123e-06,
+      "loss": 0.6281,
+      "step": 1609
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.367342643507172e-06,
+      "loss": 0.5644,
+      "step": 1610
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.356542812310132e-06,
+      "loss": 0.5767,
+      "step": 1611
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.345743734686021e-06,
+      "loss": 0.5628,
+      "step": 1612
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.334945423281973e-06,
+      "loss": 0.6156,
+      "step": 1613
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.324147890744235e-06,
+      "loss": 0.5009,
+      "step": 1614
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.313351149718134e-06,
+      "loss": 0.47,
+      "step": 1615
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 9.302555212848075e-06,
+      "loss": 0.6736,
+      "step": 1616
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.291760092777518e-06,
+      "loss": 0.6041,
+      "step": 1617
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.280965802148963e-06,
+      "loss": 0.6009,
+      "step": 1618
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.27017235360395e-06,
+      "loss": 0.5804,
+      "step": 1619
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.259379759783022e-06,
+      "loss": 0.6353,
+      "step": 1620
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.248588033325724e-06,
+      "loss": 0.6637,
+      "step": 1621
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.237797186870583e-06,
+      "loss": 0.7938,
+      "step": 1622
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.227007233055107e-06,
+      "loss": 0.6423,
+      "step": 1623
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.216218184515746e-06,
+      "loss": 0.5514,
+      "step": 1624
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.20543005388789e-06,
+      "loss": 0.5953,
+      "step": 1625
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.194642853805856e-06,
+      "loss": 0.6358,
+      "step": 1626
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.183856596902883e-06,
+      "loss": 0.7492,
+      "step": 1627
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.17307129581109e-06,
+      "loss": 0.618,
+      "step": 1628
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.162286963161478e-06,
+      "loss": 0.6069,
+      "step": 1629
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.151503611583923e-06,
+      "loss": 0.648,
+      "step": 1630
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.140721253707145e-06,
+      "loss": 0.5628,
+      "step": 1631
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.129939902158702e-06,
+      "loss": 0.4935,
+      "step": 1632
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.119159569564976e-06,
+      "loss": 0.5897,
+      "step": 1633
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.108380268551146e-06,
+      "loss": 0.5328,
+      "step": 1634
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.0976020117412e-06,
+      "loss": 0.4907,
+      "step": 1635
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 9.086824811757886e-06,
+      "loss": 0.6008,
+      "step": 1636
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.076048681222723e-06,
+      "loss": 0.5307,
+      "step": 1637
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.065273632755976e-06,
+      "loss": 0.5802,
+      "step": 1638
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.054499678976645e-06,
+      "loss": 0.5346,
+      "step": 1639
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.043726832502443e-06,
+      "loss": 0.6459,
+      "step": 1640
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.032955105949791e-06,
+      "loss": 0.6614,
+      "step": 1641
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.022184511933792e-06,
+      "loss": 0.6337,
+      "step": 1642
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.011415063068233e-06,
+      "loss": 0.6669,
+      "step": 1643
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.000646771965551e-06,
+      "loss": 0.566,
+      "step": 1644
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.989879651236834e-06,
+      "loss": 0.5571,
+      "step": 1645
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.979113713491785e-06,
+      "loss": 0.5685,
+      "step": 1646
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.968348971338742e-06,
+      "loss": 0.6185,
+      "step": 1647
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.95758543738463e-06,
+      "loss": 0.5125,
+      "step": 1648
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.94682312423496e-06,
+      "loss": 0.6122,
+      "step": 1649
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.936062044493816e-06,
+      "loss": 0.7062,
+      "step": 1650
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.925302210763838e-06,
+      "loss": 0.6403,
+      "step": 1651
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.914543635646204e-06,
+      "loss": 0.5356,
+      "step": 1652
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.903786331740616e-06,
+      "loss": 0.542,
+      "step": 1653
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.893030311645297e-06,
+      "loss": 0.7083,
+      "step": 1654
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.882275587956956e-06,
+      "loss": 0.5536,
+      "step": 1655
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.871522173270791e-06,
+      "loss": 0.6085,
+      "step": 1656
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.860770080180455e-06,
+      "loss": 0.8508,
+      "step": 1657
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.850019321278074e-06,
+      "loss": 0.627,
+      "step": 1658
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.839269909154193e-06,
+      "loss": 0.648,
+      "step": 1659
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.82852185639779e-06,
+      "loss": 0.6204,
+      "step": 1660
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.817775175596237e-06,
+      "loss": 0.6613,
+      "step": 1661
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.807029879335323e-06,
+      "loss": 0.5378,
+      "step": 1662
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.796285980199195e-06,
+      "loss": 0.5844,
+      "step": 1663
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.785543490770371e-06,
+      "loss": 0.6088,
+      "step": 1664
+    },
+    {
+      "epoch": 1.66,
+      "eval_loss": 0.7548692226409912,
+      "eval_runtime": 88.8179,
+      "eval_samples_per_second": 22.518,
+      "eval_steps_per_second": 11.259,
+      "step": 1664
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.774802423629718e-06,
+      "loss": 0.6722,
+      "step": 1665
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.76406279135644e-06,
+      "loss": 0.5064,
+      "step": 1666
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.753324606528056e-06,
+      "loss": 0.669,
+      "step": 1667
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.742587881720391e-06,
+      "loss": 0.4997,
+      "step": 1668
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.731852629507557e-06,
+      "loss": 0.5546,
+      "step": 1669
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.721118862461952e-06,
+      "loss": 0.6507,
+      "step": 1670
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.710386593154226e-06,
+      "loss": 0.6357,
+      "step": 1671
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.699655834153275e-06,
+      "loss": 0.5246,
+      "step": 1672
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.688926598026223e-06,
+      "loss": 0.5624,
+      "step": 1673
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.678198897338428e-06,
+      "loss": 0.7684,
+      "step": 1674
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.667472744653432e-06,
+      "loss": 0.7572,
+      "step": 1675
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.656748152532964e-06,
+      "loss": 0.626,
+      "step": 1676
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.646025133536931e-06,
+      "loss": 0.5223,
+      "step": 1677
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.635303700223405e-06,
+      "loss": 0.6707,
+      "step": 1678
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.62458386514859e-06,
+      "loss": 0.5932,
+      "step": 1679
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.61386564086682e-06,
+      "loss": 0.6159,
+      "step": 1680
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.603149039930546e-06,
+      "loss": 0.5853,
+      "step": 1681
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.592434074890315e-06,
+      "loss": 0.6028,
+      "step": 1682
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.581720758294758e-06,
+      "loss": 0.7374,
+      "step": 1683
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.571009102690579e-06,
+      "loss": 0.5373,
+      "step": 1684
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.560299120622532e-06,
+      "loss": 0.692,
+      "step": 1685
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.549590824633418e-06,
+      "loss": 0.5999,
+      "step": 1686
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.538884227264054e-06,
+      "loss": 0.5304,
+      "step": 1687
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.528179341053273e-06,
+      "loss": 0.6552,
+      "step": 1688
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.517476178537905e-06,
+      "loss": 0.6678,
+      "step": 1689
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.506774752252764e-06,
+      "loss": 0.4632,
+      "step": 1690
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.496075074730622e-06,
+      "loss": 0.5528,
+      "step": 1691
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.4853771585022e-06,
+      "loss": 0.7617,
+      "step": 1692
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.474681016096177e-06,
+      "loss": 0.731,
+      "step": 1693
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.463986660039136e-06,
+      "loss": 0.6924,
+      "step": 1694
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.453294102855574e-06,
+      "loss": 0.5638,
+      "step": 1695
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.442603357067877e-06,
+      "loss": 0.5566,
+      "step": 1696
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.431914435196317e-06,
+      "loss": 0.6574,
+      "step": 1697
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.421227349759029e-06,
+      "loss": 0.6087,
+      "step": 1698
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.410542113271986e-06,
+      "loss": 0.5996,
+      "step": 1699
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.39985873824901e-06,
+      "loss": 0.6194,
+      "step": 1700
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.389177237201739e-06,
+      "loss": 0.5909,
+      "step": 1701
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.37849762263961e-06,
+      "loss": 0.635,
+      "step": 1702
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.367819907069857e-06,
+      "loss": 0.5957,
+      "step": 1703
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.357144102997484e-06,
+      "loss": 0.5106,
+      "step": 1704
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.34647022292527e-06,
+      "loss": 0.5881,
+      "step": 1705
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 8.335798279353724e-06,
+      "loss": 0.6912,
+      "step": 1706
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.325128284781097e-06,
+      "loss": 0.5616,
+      "step": 1707
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.314460251703349e-06,
+      "loss": 0.7468,
+      "step": 1708
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.303794192614159e-06,
+      "loss": 0.4598,
+      "step": 1709
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.293130120004876e-06,
+      "loss": 0.6906,
+      "step": 1710
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.282468046364535e-06,
+      "loss": 0.6242,
+      "step": 1711
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.271807984179821e-06,
+      "loss": 0.6417,
+      "step": 1712
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.261149945935072e-06,
+      "loss": 0.5574,
+      "step": 1713
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.25049394411225e-06,
+      "loss": 0.6588,
+      "step": 1714
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.239839991190932e-06,
+      "loss": 0.6316,
+      "step": 1715
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 8.229188099648295e-06,
+      "loss": 0.543,
+      "step": 1716
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.218538281959113e-06,
+      "loss": 0.5491,
+      "step": 1717
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.207890550595711e-06,
+      "loss": 0.5504,
+      "step": 1718
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.19724491802799e-06,
+      "loss": 0.617,
+      "step": 1719
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.186601396723378e-06,
+      "loss": 0.5903,
+      "step": 1720
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.175959999146843e-06,
+      "loss": 0.4883,
+      "step": 1721
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.165320737760858e-06,
+      "loss": 0.746,
+      "step": 1722
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.15468362502539e-06,
+      "loss": 0.6079,
+      "step": 1723
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.144048673397907e-06,
+      "loss": 0.5689,
+      "step": 1724
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.133415895333329e-06,
+      "loss": 0.642,
+      "step": 1725
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 8.122785303284038e-06,
+      "loss": 0.6023,
+      "step": 1726
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.112156909699853e-06,
+      "loss": 0.6083,
+      "step": 1727
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.10153072702802e-06,
+      "loss": 0.536,
+      "step": 1728
+    },
+    {
+      "epoch": 1.73,
+      "eval_loss": 0.7491377592086792,
+      "eval_runtime": 88.8028,
+      "eval_samples_per_second": 22.522,
+      "eval_steps_per_second": 11.261,
+      "step": 1728
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.090906767713196e-06,
+      "loss": 0.5937,
+      "step": 1729
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.080285044197432e-06,
+      "loss": 0.6362,
+      "step": 1730
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.069665568920161e-06,
+      "loss": 0.6098,
+      "step": 1731
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.059048354318191e-06,
+      "loss": 0.5175,
+      "step": 1732
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.048433412825669e-06,
+      "loss": 0.4532,
+      "step": 1733
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.037820756874087e-06,
+      "loss": 0.5399,
+      "step": 1734
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.02721039889226e-06,
+      "loss": 0.6447,
+      "step": 1735
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.016602351306314e-06,
+      "loss": 0.6332,
+      "step": 1736
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 8.005996626539668e-06,
+      "loss": 0.7203,
+      "step": 1737
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.995393237013014e-06,
+      "loss": 0.5252,
+      "step": 1738
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.984792195144314e-06,
+      "loss": 0.5284,
+      "step": 1739
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.97419351334879e-06,
+      "loss": 0.6464,
+      "step": 1740
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.963597204038885e-06,
+      "loss": 0.8598,
+      "step": 1741
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.953003279624271e-06,
+      "loss": 0.5534,
+      "step": 1742
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.942411752511821e-06,
+      "loss": 0.5558,
+      "step": 1743
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.931822635105613e-06,
+      "loss": 0.6868,
+      "step": 1744
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.921235939806893e-06,
+      "loss": 0.5553,
+      "step": 1745
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.91065167901407e-06,
+      "loss": 0.4812,
+      "step": 1746
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.900069865122702e-06,
+      "loss": 0.5457,
+      "step": 1747
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.889490510525491e-06,
+      "loss": 0.6085,
+      "step": 1748
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.878913627612246e-06,
+      "loss": 0.4942,
+      "step": 1749
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.868339228769888e-06,
+      "loss": 0.6985,
+      "step": 1750
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.857767326382428e-06,
+      "loss": 0.7335,
+      "step": 1751
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.847197932830953e-06,
+      "loss": 0.7159,
+      "step": 1752
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.836631060493613e-06,
+      "loss": 0.5312,
+      "step": 1753
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.826066721745604e-06,
+      "loss": 0.7012,
+      "step": 1754
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.815504928959147e-06,
+      "loss": 0.5625,
+      "step": 1755
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.804945694503505e-06,
+      "loss": 0.6072,
+      "step": 1756
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.794389030744919e-06,
+      "loss": 0.5545,
+      "step": 1757
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.78383495004663e-06,
+      "loss": 0.4992,
+      "step": 1758
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.773283464768851e-06,
+      "loss": 0.4736,
+      "step": 1759
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.762734587268767e-06,
+      "loss": 0.5492,
+      "step": 1760
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.752188329900492e-06,
+      "loss": 0.5832,
+      "step": 1761
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.741644705015081e-06,
+      "loss": 0.4772,
+      "step": 1762
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.731103724960506e-06,
+      "loss": 0.4948,
+      "step": 1763
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.720565402081642e-06,
+      "loss": 0.6937,
+      "step": 1764
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.710029748720245e-06,
+      "loss": 0.5376,
+      "step": 1765
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.69949677721495e-06,
+      "loss": 0.5318,
+      "step": 1766
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.688966499901253e-06,
+      "loss": 0.718,
+      "step": 1767
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.678438929111493e-06,
+      "loss": 0.4683,
+      "step": 1768
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.667914077174839e-06,
+      "loss": 0.728,
+      "step": 1769
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.65739195641727e-06,
+      "loss": 0.7365,
+      "step": 1770
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.646872579161581e-06,
+      "loss": 0.5135,
+      "step": 1771
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.636355957727343e-06,
+      "loss": 0.7423,
+      "step": 1772
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.625842104430897e-06,
+      "loss": 0.4843,
+      "step": 1773
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.6153310315853465e-06,
+      "loss": 0.6236,
+      "step": 1774
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.604822751500547e-06,
+      "loss": 0.6085,
+      "step": 1775
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.5943172764830695e-06,
+      "loss": 0.6936,
+      "step": 1776
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.583814618836208e-06,
+      "loss": 0.4644,
+      "step": 1777
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.573314790859952e-06,
+      "loss": 0.6331,
+      "step": 1778
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.562817804850984e-06,
+      "loss": 0.551,
+      "step": 1779
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.552323673102651e-06,
+      "loss": 0.575,
+      "step": 1780
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.541832407904963e-06,
+      "loss": 0.4448,
+      "step": 1781
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.531344021544566e-06,
+      "loss": 0.6688,
+      "step": 1782
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.520858526304743e-06,
+      "loss": 0.5302,
+      "step": 1783
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.510375934465385e-06,
+      "loss": 0.5422,
+      "step": 1784
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.499896258302985e-06,
+      "loss": 0.5832,
+      "step": 1785
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.489419510090618e-06,
+      "loss": 0.5146,
+      "step": 1786
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.47894570209794e-06,
+      "loss": 0.7329,
+      "step": 1787
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.468474846591151e-06,
+      "loss": 0.4832,
+      "step": 1788
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.458006955833001e-06,
+      "loss": 0.5192,
+      "step": 1789
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.447542042082761e-06,
+      "loss": 0.5345,
+      "step": 1790
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.437080117596229e-06,
+      "loss": 0.5503,
+      "step": 1791
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.426621194625687e-06,
+      "loss": 0.5856,
+      "step": 1792
+    },
+    {
+      "epoch": 1.79,
+      "eval_loss": 0.7484854459762573,
+      "eval_runtime": 88.7999,
+      "eval_samples_per_second": 22.523,
+      "eval_steps_per_second": 11.261,
+      "step": 1792
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.416165285419912e-06,
+      "loss": 0.6397,
+      "step": 1793
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.405712402224142e-06,
+      "loss": 0.6284,
+      "step": 1794
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.395262557280083e-06,
+      "loss": 0.5389,
+      "step": 1795
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.384815762825875e-06,
+      "loss": 0.4726,
+      "step": 1796
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.374372031096084e-06,
+      "loss": 0.6793,
+      "step": 1797
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.363931374321698e-06,
+      "loss": 0.709,
+      "step": 1798
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.353493804730094e-06,
+      "loss": 0.5181,
+      "step": 1799
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.343059334545039e-06,
+      "loss": 0.5778,
+      "step": 1800
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.332627975986664e-06,
+      "loss": 0.6572,
+      "step": 1801
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.322199741271472e-06,
+      "loss": 0.6334,
+      "step": 1802
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.311774642612287e-06,
+      "loss": 0.6137,
+      "step": 1803
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.301352692218272e-06,
+      "loss": 0.6953,
+      "step": 1804
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.2909339022948965e-06,
+      "loss": 0.6722,
+      "step": 1805
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.2805182850439425e-06,
+      "loss": 0.7714,
+      "step": 1806
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.2701058526634594e-06,
+      "loss": 0.6731,
+      "step": 1807
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.259696617347776e-06,
+      "loss": 0.6471,
+      "step": 1808
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.2492905912874725e-06,
+      "loss": 0.5229,
+      "step": 1809
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.238887786669375e-06,
+      "loss": 0.7308,
+      "step": 1810
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.228488215676536e-06,
+      "loss": 0.5403,
+      "step": 1811
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.218091890488217e-06,
+      "loss": 0.661,
+      "step": 1812
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.207698823279882e-06,
+      "loss": 0.5324,
+      "step": 1813
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.1973090262231805e-06,
+      "loss": 0.6866,
+      "step": 1814
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.186922511485927e-06,
+      "loss": 0.6685,
+      "step": 1815
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 7.176539291232098e-06,
+      "loss": 0.5176,
+      "step": 1816
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.166159377621806e-06,
+      "loss": 0.5331,
+      "step": 1817
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.155782782811297e-06,
+      "loss": 0.5978,
+      "step": 1818
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.145409518952926e-06,
+      "loss": 0.6114,
+      "step": 1819
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.135039598195148e-06,
+      "loss": 0.5659,
+      "step": 1820
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.124673032682499e-06,
+      "loss": 0.6158,
+      "step": 1821
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.114309834555597e-06,
+      "loss": 0.4863,
+      "step": 1822
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.103950015951107e-06,
+      "loss": 0.572,
+      "step": 1823
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.093593589001737e-06,
+      "loss": 0.835,
+      "step": 1824
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.083240565836217e-06,
+      "loss": 0.5681,
+      "step": 1825
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 7.07289095857931e-06,
+      "loss": 0.6087,
+      "step": 1826
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.062544779351759e-06,
+      "loss": 0.601,
+      "step": 1827
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.052202040270301e-06,
+      "loss": 0.6498,
+      "step": 1828
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.04186275344764e-06,
+      "loss": 0.6743,
+      "step": 1829
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.031526930992445e-06,
+      "loss": 0.6173,
+      "step": 1830
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.02119458500932e-06,
+      "loss": 0.6137,
+      "step": 1831
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.010865727598799e-06,
+      "loss": 0.499,
+      "step": 1832
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 7.000540370857329e-06,
+      "loss": 0.5361,
+      "step": 1833
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.990218526877266e-06,
+      "loss": 0.5872,
+      "step": 1834
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.979900207746841e-06,
+      "loss": 0.6394,
+      "step": 1835
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 6.969585425550161e-06,
+      "loss": 0.5442,
+      "step": 1836
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.9592741923671956e-06,
+      "loss": 0.6822,
+      "step": 1837
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.948966520273753e-06,
+      "loss": 0.5973,
+      "step": 1838
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.938662421341473e-06,
+      "loss": 0.6411,
+      "step": 1839
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.928361907637799e-06,
+      "loss": 0.6912,
+      "step": 1840
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.9180649912260016e-06,
+      "loss": 0.6327,
+      "step": 1841
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.907771684165114e-06,
+      "loss": 0.629,
+      "step": 1842
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.897481998509954e-06,
+      "loss": 0.6142,
+      "step": 1843
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.887195946311093e-06,
+      "loss": 0.5129,
+      "step": 1844
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.876913539614853e-06,
+      "loss": 0.5561,
+      "step": 1845
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.8666347904632845e-06,
+      "loss": 0.4456,
+      "step": 1846
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.8563597108941516e-06,
+      "loss": 0.5976,
+      "step": 1847
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.846088312940919e-06,
+      "loss": 0.6724,
+      "step": 1848
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.835820608632751e-06,
+      "loss": 0.6095,
+      "step": 1849
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.825556609994475e-06,
+      "loss": 0.5615,
+      "step": 1850
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.815296329046581e-06,
+      "loss": 0.6272,
+      "step": 1851
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.805039777805205e-06,
+      "loss": 0.8037,
+      "step": 1852
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.794786968282121e-06,
+      "loss": 0.5641,
+      "step": 1853
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.784537912484719e-06,
+      "loss": 0.5288,
+      "step": 1854
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.7742926224159815e-06,
+      "loss": 0.5845,
+      "step": 1855
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.76405111007449e-06,
+      "loss": 0.585,
+      "step": 1856
+    },
+    {
+      "epoch": 1.85,
+      "eval_loss": 0.7450958490371704,
+      "eval_runtime": 88.8258,
+      "eval_samples_per_second": 22.516,
+      "eval_steps_per_second": 11.258,
+      "step": 1856
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.753813387454409e-06,
+      "loss": 0.6582,
+      "step": 1857
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.743579466545454e-06,
+      "loss": 0.5221,
+      "step": 1858
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.73334935933289e-06,
+      "loss": 0.5411,
+      "step": 1859
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.723123077797515e-06,
+      "loss": 0.5382,
+      "step": 1860
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.712900633915653e-06,
+      "loss": 0.4572,
+      "step": 1861
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.702682039659127e-06,
+      "loss": 0.5233,
+      "step": 1862
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.692467306995255e-06,
+      "loss": 0.6738,
+      "step": 1863
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.682256447886828e-06,
+      "loss": 0.7235,
+      "step": 1864
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.672049474292107e-06,
+      "loss": 0.6002,
+      "step": 1865
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.6618463981648e-06,
+      "loss": 0.6311,
+      "step": 1866
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.651647231454046e-06,
+      "loss": 0.6925,
+      "step": 1867
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.64145198610441e-06,
+      "loss": 0.4247,
+      "step": 1868
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.631260674055873e-06,
+      "loss": 0.7743,
+      "step": 1869
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.62107330724379e-06,
+      "loss": 0.7996,
+      "step": 1870
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.610889897598911e-06,
+      "loss": 0.6315,
+      "step": 1871
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.600710457047341e-06,
+      "loss": 0.435,
+      "step": 1872
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.590534997510554e-06,
+      "loss": 0.6108,
+      "step": 1873
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.580363530905343e-06,
+      "loss": 0.62,
+      "step": 1874
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.5701960691438315e-06,
+      "loss": 0.6667,
+      "step": 1875
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 6.560032624133458e-06,
+      "loss": 0.6659,
+      "step": 1876
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.549873207776949e-06,
+      "loss": 0.5416,
+      "step": 1877
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.539717831972315e-06,
+      "loss": 0.5268,
+      "step": 1878
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.529566508612837e-06,
+      "loss": 0.7636,
+      "step": 1879
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.51941924958705e-06,
+      "loss": 0.6423,
+      "step": 1880
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.509276066778727e-06,
+      "loss": 0.628,
+      "step": 1881
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.499136972066868e-06,
+      "loss": 0.7284,
+      "step": 1882
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.489001977325681e-06,
+      "loss": 0.4847,
+      "step": 1883
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.478871094424587e-06,
+      "loss": 0.4753,
+      "step": 1884
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.468744335228173e-06,
+      "loss": 0.5437,
+      "step": 1885
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.458621711596207e-06,
+      "loss": 0.5686,
+      "step": 1886
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.448503235383608e-06,
+      "loss": 0.5128,
+      "step": 1887
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.438388918440448e-06,
+      "loss": 0.5876,
+      "step": 1888
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.428278772611919e-06,
+      "loss": 0.5346,
+      "step": 1889
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.41817280973833e-06,
+      "loss": 0.5722,
+      "step": 1890
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.4080710416550885e-06,
+      "loss": 0.5282,
+      "step": 1891
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.3979734801926965e-06,
+      "loss": 0.5896,
+      "step": 1892
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.3878801371767255e-06,
+      "loss": 0.5756,
+      "step": 1893
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.377791024427804e-06,
+      "loss": 0.5612,
+      "step": 1894
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.36770615376161e-06,
+      "loss": 0.6084,
+      "step": 1895
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 6.357625536988854e-06,
+      "loss": 0.8479,
+      "step": 1896
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.34754918591526e-06,
+      "loss": 0.5956,
+      "step": 1897
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.3374771123415625e-06,
+      "loss": 0.6559,
+      "step": 1898
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.327409328063479e-06,
+      "loss": 0.5256,
+      "step": 1899
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.317345844871713e-06,
+      "loss": 0.5459,
+      "step": 1900
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.307286674551923e-06,
+      "loss": 0.5671,
+      "step": 1901
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.29723182888472e-06,
+      "loss": 0.4931,
+      "step": 1902
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.287181319645645e-06,
+      "loss": 0.7066,
+      "step": 1903
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.277135158605174e-06,
+      "loss": 0.6499,
+      "step": 1904
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.267093357528677e-06,
+      "loss": 0.5588,
+      "step": 1905
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 6.257055928176425e-06,
+      "loss": 0.5938,
+      "step": 1906
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.247022882303559e-06,
+      "loss": 0.6139,
+      "step": 1907
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.236994231660105e-06,
+      "loss": 0.5468,
+      "step": 1908
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.226969987990926e-06,
+      "loss": 0.6266,
+      "step": 1909
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.216950163035727e-06,
+      "loss": 0.5614,
+      "step": 1910
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.206934768529043e-06,
+      "loss": 0.6777,
+      "step": 1911
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.196923816200216e-06,
+      "loss": 0.5049,
+      "step": 1912
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.186917317773384e-06,
+      "loss": 0.5514,
+      "step": 1913
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.176915284967474e-06,
+      "loss": 0.5614,
+      "step": 1914
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.166917729496182e-06,
+      "loss": 0.5849,
+      "step": 1915
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 6.1569246630679605e-06,
+      "loss": 0.6687,
+      "step": 1916
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.146936097385999e-06,
+      "loss": 0.6321,
+      "step": 1917
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.13695204414822e-06,
+      "loss": 0.6551,
+      "step": 1918
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.126972515047269e-06,
+      "loss": 0.5871,
+      "step": 1919
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.1169975217704805e-06,
+      "loss": 0.5143,
+      "step": 1920
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.7439701557159424,
+      "eval_runtime": 88.8154,
+      "eval_samples_per_second": 22.519,
+      "eval_steps_per_second": 11.259,
+      "step": 1920
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.107027075999887e-06,
+      "loss": 0.5085,
+      "step": 1921
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.097061189412181e-06,
+      "loss": 0.5354,
+      "step": 1922
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.0870998736787355e-06,
+      "loss": 0.7136,
+      "step": 1923
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.077143140465558e-06,
+      "loss": 0.6739,
+      "step": 1924
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.0671910014332895e-06,
+      "loss": 0.4988,
+      "step": 1925
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 6.057243468237192e-06,
+      "loss": 0.6326,
+      "step": 1926
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.0473005525271365e-06,
+      "loss": 0.4922,
+      "step": 1927
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.037362265947584e-06,
+      "loss": 0.5836,
+      "step": 1928
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.027428620137574e-06,
+      "loss": 0.7251,
+      "step": 1929
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.017499626730707e-06,
+      "loss": 0.6268,
+      "step": 1930
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 6.007575297355146e-06,
+      "loss": 0.5628,
+      "step": 1931
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.99765564363358e-06,
+      "loss": 0.541,
+      "step": 1932
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.987740677183229e-06,
+      "loss": 0.5807,
+      "step": 1933
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.977830409615817e-06,
+      "loss": 0.6026,
+      "step": 1934
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.967924852537579e-06,
+      "loss": 0.6233,
+      "step": 1935
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.95802401754922e-06,
+      "loss": 0.5217,
+      "step": 1936
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.9481279162459164e-06,
+      "loss": 0.6669,
+      "step": 1937
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.938236560217301e-06,
+      "loss": 0.5845,
+      "step": 1938
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.928349961047459e-06,
+      "loss": 0.61,
+      "step": 1939
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.918468130314895e-06,
+      "loss": 0.6591,
+      "step": 1940
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.908591079592529e-06,
+      "loss": 0.4925,
+      "step": 1941
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.898718820447685e-06,
+      "loss": 0.5516,
+      "step": 1942
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.88885136444208e-06,
+      "loss": 0.4537,
+      "step": 1943
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.878988723131797e-06,
+      "loss": 0.5765,
+      "step": 1944
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.869130908067291e-06,
+      "loss": 0.5357,
+      "step": 1945
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.85927793079335e-06,
+      "loss": 0.5888,
+      "step": 1946
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.849429802849114e-06,
+      "loss": 0.627,
+      "step": 1947
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.839586535768032e-06,
+      "loss": 0.4781,
+      "step": 1948
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.829748141077855e-06,
+      "loss": 0.4761,
+      "step": 1949
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.819914630300653e-06,
+      "loss": 0.4859,
+      "step": 1950
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.810086014952747e-06,
+      "loss": 0.5839,
+      "step": 1951
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.800262306544743e-06,
+      "loss": 0.568,
+      "step": 1952
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.790443516581488e-06,
+      "loss": 0.6449,
+      "step": 1953
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.780629656562079e-06,
+      "loss": 0.7283,
+      "step": 1954
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.770820737979841e-06,
+      "loss": 0.6201,
+      "step": 1955
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 5.7610167723222976e-06,
+      "loss": 0.7511,
+      "step": 1956
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.751217771071185e-06,
+      "loss": 0.5401,
+      "step": 1957
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.741423745702422e-06,
+      "loss": 0.5501,
+      "step": 1958
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.731634707686104e-06,
+      "loss": 0.6013,
+      "step": 1959
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.721850668486475e-06,
+      "loss": 0.5687,
+      "step": 1960
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.712071639561923e-06,
+      "loss": 0.5671,
+      "step": 1961
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.7022976323649944e-06,
+      "loss": 0.6546,
+      "step": 1962
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.692528658342322e-06,
+      "loss": 0.5525,
+      "step": 1963
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.682764728934667e-06,
+      "loss": 0.5449,
+      "step": 1964
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.673005855576863e-06,
+      "loss": 0.4965,
+      "step": 1965
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.663252049697843e-06,
+      "loss": 0.572,
+      "step": 1966
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.653503322720594e-06,
+      "loss": 0.7223,
+      "step": 1967
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.643759686062152e-06,
+      "loss": 0.5907,
+      "step": 1968
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.634021151133597e-06,
+      "loss": 0.5713,
+      "step": 1969
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.624287729340039e-06,
+      "loss": 0.5979,
+      "step": 1970
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.6145594320805865e-06,
+      "loss": 0.7585,
+      "step": 1971
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.604836270748362e-06,
+      "loss": 0.461,
+      "step": 1972
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.595118256730452e-06,
+      "loss": 0.6431,
+      "step": 1973
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.585405401407944e-06,
+      "loss": 0.4972,
+      "step": 1974
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.575697716155861e-06,
+      "loss": 0.5496,
+      "step": 1975
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.5659952123431734e-06,
+      "loss": 0.5657,
+      "step": 1976
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.556297901332794e-06,
+      "loss": 0.6389,
+      "step": 1977
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.546605794481546e-06,
+      "loss": 0.5551,
+      "step": 1978
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.536918903140165e-06,
+      "loss": 0.4569,
+      "step": 1979
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.5272372386532665e-06,
+      "loss": 0.5818,
+      "step": 1980
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.517560812359355e-06,
+      "loss": 0.6564,
+      "step": 1981
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.507889635590802e-06,
+      "loss": 0.6242,
+      "step": 1982
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.498223719673818e-06,
+      "loss": 0.7349,
+      "step": 1983
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 5.48856307592847e-06,
+      "loss": 0.5408,
+      "step": 1984
+    },
+    {
+      "epoch": 1.98,
+      "eval_loss": 0.7390481233596802,
+      "eval_runtime": 88.8545,
+      "eval_samples_per_second": 22.509,
+      "eval_steps_per_second": 11.254,
+      "step": 1984
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3003,
+  "num_train_epochs": 3,
+  "save_steps": 64,
+  "total_flos": 5.809085963833442e+18,
+  "trial_name": null,
+  "trial_params": null
+}