diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,7781 @@
+{
+  "best_metric": 0.07883252203464508,
+  "best_model_checkpoint": "segformer-teeth_segment_10ep/checkpoint-1040",
+  "epoch": 24.761904761904763,
+  "eval_steps": 20,
+  "global_step": 1040,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.023809523809523808,
+      "grad_norm": 3.7683873176574707,
+      "learning_rate": 5.995238095238095e-05,
+      "loss": 0.7093,
+      "step": 1
+    },
+    {
+      "epoch": 0.047619047619047616,
+      "grad_norm": 3.244874954223633,
+      "learning_rate": 5.9904761904761905e-05,
+      "loss": 0.6516,
+      "step": 2
+    },
+    {
+      "epoch": 0.07142857142857142,
+      "grad_norm": 3.2348296642303467,
+      "learning_rate": 5.9857142857142856e-05,
+      "loss": 0.6236,
+      "step": 3
+    },
+    {
+      "epoch": 0.09523809523809523,
+      "grad_norm": 3.209193468093872,
+      "learning_rate": 5.9809523809523814e-05,
+      "loss": 0.6066,
+      "step": 4
+    },
+    {
+      "epoch": 0.11904761904761904,
+      "grad_norm": 3.0218300819396973,
+      "learning_rate": 5.9761904761904766e-05,
+      "loss": 0.5764,
+      "step": 5
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 2.9458155632019043,
+      "learning_rate": 5.971428571428572e-05,
+      "loss": 0.5744,
+      "step": 6
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 3.0408496856689453,
+      "learning_rate": 5.966666666666667e-05,
+      "loss": 0.5581,
+      "step": 7
+    },
+    {
+      "epoch": 0.19047619047619047,
+      "grad_norm": 2.8450467586517334,
+      "learning_rate": 5.961904761904762e-05,
+      "loss": 0.5494,
+      "step": 8
+    },
+    {
+      "epoch": 0.21428571428571427,
+      "grad_norm": 2.5434529781341553,
+      "learning_rate": 5.957142857142857e-05,
+      "loss": 0.5207,
+      "step": 9
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 2.4258580207824707,
+      "learning_rate": 5.9523809523809524e-05,
+      "loss": 0.5112,
+      "step": 10
+    },
+    {
+      "epoch": 0.2619047619047619,
+      "grad_norm": 2.4317784309387207,
+      "learning_rate": 5.947619047619048e-05,
+      "loss": 0.5017,
+      "step": 11
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 2.8235714435577393,
+      "learning_rate": 5.9428571428571434e-05,
+      "loss": 0.5308,
+      "step": 12
+    },
+    {
+      "epoch": 0.30952380952380953,
+      "grad_norm": 2.354302406311035,
+      "learning_rate": 5.9380952380952385e-05,
+      "loss": 0.496,
+      "step": 13
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 2.3225038051605225,
+      "learning_rate": 5.933333333333334e-05,
+      "loss": 0.49,
+      "step": 14
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 2.1641042232513428,
+      "learning_rate": 5.928571428571429e-05,
+      "loss": 0.482,
+      "step": 15
+    },
+    {
+      "epoch": 0.38095238095238093,
+      "grad_norm": 2.4745819568634033,
+      "learning_rate": 5.923809523809524e-05,
+      "loss": 0.4914,
+      "step": 16
+    },
+    {
+      "epoch": 0.40476190476190477,
+      "grad_norm": 2.144455909729004,
+      "learning_rate": 5.91904761904762e-05,
+      "loss": 0.4638,
+      "step": 17
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 2.0155744552612305,
+      "learning_rate": 5.914285714285715e-05,
+      "loss": 0.4498,
+      "step": 18
+    },
+    {
+      "epoch": 0.4523809523809524,
+      "grad_norm": 2.136814594268799,
+      "learning_rate": 5.90952380952381e-05,
+      "loss": 0.4435,
+      "step": 19
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 2.1965997219085693,
+      "learning_rate": 5.9047619047619046e-05,
+      "loss": 0.4436,
+      "step": 20
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "eval_dice_score": 0.8391185233064109,
+      "eval_loss": 0.5746188163757324,
+      "eval_runtime": 9.622,
+      "eval_samples_per_second": 9.25,
+      "eval_steps_per_second": 0.935,
+      "step": 20
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.982168197631836,
+      "learning_rate": 5.9e-05,
+      "loss": 0.4275,
+      "step": 21
+    },
+    {
+      "epoch": 0.5238095238095238,
+      "grad_norm": 2.1209304332733154,
+      "learning_rate": 5.895238095238095e-05,
+      "loss": 0.447,
+      "step": 22
+    },
+    {
+      "epoch": 0.5476190476190477,
+      "grad_norm": 2.017097234725952,
+      "learning_rate": 5.89047619047619e-05,
+      "loss": 0.4322,
+      "step": 23
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 1.9348403215408325,
+      "learning_rate": 5.885714285714286e-05,
+      "loss": 0.4141,
+      "step": 24
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 2.1177632808685303,
+      "learning_rate": 5.880952380952381e-05,
+      "loss": 0.4359,
+      "step": 25
+    },
+    {
+      "epoch": 0.6190476190476191,
+      "grad_norm": 2.4067323207855225,
+      "learning_rate": 5.876190476190476e-05,
+      "loss": 0.4383,
+      "step": 26
+    },
+    {
+      "epoch": 0.6428571428571429,
+      "grad_norm": 2.0675342082977295,
+      "learning_rate": 5.8714285714285714e-05,
+      "loss": 0.4188,
+      "step": 27
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 1.8709311485290527,
+      "learning_rate": 5.8666666666666665e-05,
+      "loss": 0.4079,
+      "step": 28
+    },
+    {
+      "epoch": 0.6904761904761905,
+      "grad_norm": 2.5608415603637695,
+      "learning_rate": 5.861904761904762e-05,
+      "loss": 0.4373,
+      "step": 29
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 2.411015510559082,
+      "learning_rate": 5.857142857142857e-05,
+      "loss": 0.3953,
+      "step": 30
+    },
+    {
+      "epoch": 0.7380952380952381,
+      "grad_norm": 1.9586458206176758,
+      "learning_rate": 5.8523809523809527e-05,
+      "loss": 0.3895,
+      "step": 31
+    },
+    {
+      "epoch": 0.7619047619047619,
+      "grad_norm": 2.0025196075439453,
+      "learning_rate": 5.847619047619048e-05,
+      "loss": 0.3839,
+      "step": 32
+    },
+    {
+      "epoch": 0.7857142857142857,
+      "grad_norm": 1.836198329925537,
+      "learning_rate": 5.842857142857143e-05,
+      "loss": 0.3911,
+      "step": 33
+    },
+    {
+      "epoch": 0.8095238095238095,
+      "grad_norm": 1.9035793542861938,
+      "learning_rate": 5.838095238095238e-05,
+      "loss": 0.3941,
+      "step": 34
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 1.7556637525558472,
+      "learning_rate": 5.833333333333333e-05,
+      "loss": 0.3767,
+      "step": 35
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 2.189488649368286,
+      "learning_rate": 5.8285714285714284e-05,
+      "loss": 0.4038,
+      "step": 36
+    },
+    {
+      "epoch": 0.8809523809523809,
+      "grad_norm": 1.6118512153625488,
+      "learning_rate": 5.8238095238095236e-05,
+      "loss": 0.3516,
+      "step": 37
+    },
+    {
+      "epoch": 0.9047619047619048,
+      "grad_norm": 1.7305861711502075,
+      "learning_rate": 5.8190476190476194e-05,
+      "loss": 0.3606,
+      "step": 38
+    },
+    {
+      "epoch": 0.9285714285714286,
+      "grad_norm": 1.6623855829238892,
+      "learning_rate": 5.8142857142857146e-05,
+      "loss": 0.3472,
+      "step": 39
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 1.6707160472869873,
+      "learning_rate": 5.80952380952381e-05,
+      "loss": 0.3473,
+      "step": 40
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "eval_dice_score": 0.8541455296072445,
+      "eval_loss": 0.43145105242729187,
+      "eval_runtime": 8.8181,
+      "eval_samples_per_second": 10.093,
+      "eval_steps_per_second": 1.021,
+      "step": 40
+    },
+    {
+      "epoch": 0.9761904761904762,
+      "grad_norm": 1.6397485733032227,
+      "learning_rate": 5.804761904761905e-05,
+      "loss": 0.342,
+      "step": 41
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.444227457046509,
+      "learning_rate": 5.8e-05,
+      "loss": 0.3436,
+      "step": 42
+    },
+    {
+      "epoch": 1.0238095238095237,
+      "grad_norm": 1.478361964225769,
+      "learning_rate": 5.795238095238095e-05,
+      "loss": 0.3369,
+      "step": 43
+    },
+    {
+      "epoch": 1.0476190476190477,
+      "grad_norm": 1.642811894416809,
+      "learning_rate": 5.790476190476191e-05,
+      "loss": 0.3313,
+      "step": 44
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 1.4398586750030518,
+      "learning_rate": 5.785714285714286e-05,
+      "loss": 0.319,
+      "step": 45
+    },
+    {
+      "epoch": 1.0952380952380953,
+      "grad_norm": 1.407219648361206,
+      "learning_rate": 5.780952380952381e-05,
+      "loss": 0.3163,
+      "step": 46
+    },
+    {
+      "epoch": 1.119047619047619,
+      "grad_norm": 1.5634286403656006,
+      "learning_rate": 5.7761904761904765e-05,
+      "loss": 0.3175,
+      "step": 47
+    },
+    {
+      "epoch": 1.1428571428571428,
+      "grad_norm": 1.5995242595672607,
+      "learning_rate": 5.7714285714285716e-05,
+      "loss": 0.3245,
+      "step": 48
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 1.4526621103286743,
+      "learning_rate": 5.766666666666667e-05,
+      "loss": 0.3046,
+      "step": 49
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 2.028379201889038,
+      "learning_rate": 5.761904761904762e-05,
+      "loss": 0.3497,
+      "step": 50
+    },
+    {
+      "epoch": 1.2142857142857142,
+      "grad_norm": 1.4727197885513306,
+      "learning_rate": 5.757142857142858e-05,
+      "loss": 0.3052,
+      "step": 51
+    },
+    {
+      "epoch": 1.2380952380952381,
+      "grad_norm": 1.344086766242981,
+      "learning_rate": 5.752380952380953e-05,
+      "loss": 0.2915,
+      "step": 52
+    },
+    {
+      "epoch": 1.2619047619047619,
+      "grad_norm": 2.209981679916382,
+      "learning_rate": 5.747619047619048e-05,
+      "loss": 0.3355,
+      "step": 53
+    },
+    {
+      "epoch": 1.2857142857142856,
+      "grad_norm": 1.6532037258148193,
+      "learning_rate": 5.742857142857143e-05,
+      "loss": 0.3161,
+      "step": 54
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 1.2978978157043457,
+      "learning_rate": 5.7380952380952384e-05,
+      "loss": 0.2843,
+      "step": 55
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 1.4890028238296509,
+      "learning_rate": 5.7333333333333336e-05,
+      "loss": 0.2989,
+      "step": 56
+    },
+    {
+      "epoch": 1.3571428571428572,
+      "grad_norm": 1.4468064308166504,
+      "learning_rate": 5.728571428571429e-05,
+      "loss": 0.2874,
+      "step": 57
+    },
+    {
+      "epoch": 1.380952380952381,
+      "grad_norm": 1.2638118267059326,
+      "learning_rate": 5.7238095238095245e-05,
+      "loss": 0.2877,
+      "step": 58
+    },
+    {
+      "epoch": 1.4047619047619047,
+      "grad_norm": 1.3178492784500122,
+      "learning_rate": 5.71904761904762e-05,
+      "loss": 0.2832,
+      "step": 59
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.2155193090438843,
+      "learning_rate": 5.714285714285714e-05,
+      "loss": 0.2772,
+      "step": 60
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "eval_dice_score": 0.8701758660308139,
+      "eval_loss": 0.2781553268432617,
+      "eval_runtime": 8.8606,
+      "eval_samples_per_second": 10.045,
+      "eval_steps_per_second": 1.016,
+      "step": 60
+    },
+    {
+      "epoch": 1.4523809523809523,
+      "grad_norm": 1.8164597749710083,
+      "learning_rate": 5.709523809523809e-05,
+      "loss": 0.2814,
+      "step": 61
+    },
+    {
+      "epoch": 1.4761904761904763,
+      "grad_norm": 1.2323110103607178,
+      "learning_rate": 5.7047619047619045e-05,
+      "loss": 0.2722,
+      "step": 62
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.6265512704849243,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 0.3031,
+      "step": 63
+    },
+    {
+      "epoch": 1.5238095238095237,
+      "grad_norm": 1.2476142644882202,
+      "learning_rate": 5.6952380952380955e-05,
+      "loss": 0.274,
+      "step": 64
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 1.3125485181808472,
+      "learning_rate": 5.6904761904761906e-05,
+      "loss": 0.2646,
+      "step": 65
+    },
+    {
+      "epoch": 1.5714285714285714,
+      "grad_norm": 1.1948935985565186,
+      "learning_rate": 5.685714285714286e-05,
+      "loss": 0.2549,
+      "step": 66
+    },
+    {
+      "epoch": 1.5952380952380953,
+      "grad_norm": 1.230567216873169,
+      "learning_rate": 5.680952380952381e-05,
+      "loss": 0.2757,
+      "step": 67
+    },
+    {
+      "epoch": 1.619047619047619,
+      "grad_norm": 1.2908153533935547,
+      "learning_rate": 5.676190476190476e-05,
+      "loss": 0.2642,
+      "step": 68
+    },
+    {
+      "epoch": 1.6428571428571428,
+      "grad_norm": 1.659945011138916,
+      "learning_rate": 5.671428571428571e-05,
+      "loss": 0.3091,
+      "step": 69
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.4245527982711792,
+      "learning_rate": 5.6666666666666664e-05,
+      "loss": 0.2727,
+      "step": 70
+    },
+    {
+      "epoch": 1.6904761904761905,
+      "grad_norm": 1.1477965116500854,
+      "learning_rate": 5.661904761904762e-05,
+      "loss": 0.2566,
+      "step": 71
+    },
+    {
+      "epoch": 1.7142857142857144,
+      "grad_norm": 1.1955633163452148,
+      "learning_rate": 5.6571428571428574e-05,
+      "loss": 0.2478,
+      "step": 72
+    },
+    {
+      "epoch": 1.7380952380952381,
+      "grad_norm": 1.092235803604126,
+      "learning_rate": 5.6523809523809525e-05,
+      "loss": 0.2653,
+      "step": 73
+    },
+    {
+      "epoch": 1.7619047619047619,
+      "grad_norm": 1.1495394706726074,
+      "learning_rate": 5.647619047619048e-05,
+      "loss": 0.2476,
+      "step": 74
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 1.333282232284546,
+      "learning_rate": 5.642857142857143e-05,
+      "loss": 0.2454,
+      "step": 75
+    },
+    {
+      "epoch": 1.8095238095238095,
+      "grad_norm": 1.0368735790252686,
+      "learning_rate": 5.638095238095238e-05,
+      "loss": 0.2528,
+      "step": 76
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 1.383416771888733,
+      "learning_rate": 5.633333333333333e-05,
+      "loss": 0.266,
+      "step": 77
+    },
+    {
+      "epoch": 1.8571428571428572,
+      "grad_norm": 1.0640007257461548,
+      "learning_rate": 5.628571428571429e-05,
+      "loss": 0.2319,
+      "step": 78
+    },
+    {
+      "epoch": 1.880952380952381,
+      "grad_norm": 1.1098486185073853,
+      "learning_rate": 5.623809523809524e-05,
+      "loss": 0.2473,
+      "step": 79
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.9411126971244812,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.2367,
+      "step": 80
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "eval_dice_score": 0.8800905080784573,
+      "eval_loss": 0.2220260500907898,
+      "eval_runtime": 8.9078,
+      "eval_samples_per_second": 9.991,
+      "eval_steps_per_second": 1.01,
+      "step": 80
+    },
+    {
+      "epoch": 1.9285714285714286,
+      "grad_norm": 1.6729662418365479,
+      "learning_rate": 5.6142857142857145e-05,
+      "loss": 0.262,
+      "step": 81
+    },
+    {
+      "epoch": 1.9523809523809523,
+      "grad_norm": 1.0404990911483765,
+      "learning_rate": 5.6095238095238096e-05,
+      "loss": 0.2283,
+      "step": 82
+    },
+    {
+      "epoch": 1.9761904761904763,
+      "grad_norm": 1.1858619451522827,
+      "learning_rate": 5.604761904761905e-05,
+      "loss": 0.2301,
+      "step": 83
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9727345108985901,
+      "learning_rate": 5.6e-05,
+      "loss": 0.2281,
+      "step": 84
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 1.208103895187378,
+      "learning_rate": 5.595238095238096e-05,
+      "loss": 0.2194,
+      "step": 85
+    },
+    {
+      "epoch": 2.0476190476190474,
+      "grad_norm": 1.3448512554168701,
+      "learning_rate": 5.590476190476191e-05,
+      "loss": 0.2182,
+      "step": 86
+    },
+    {
+      "epoch": 2.0714285714285716,
+      "grad_norm": 0.9734125137329102,
+      "learning_rate": 5.585714285714286e-05,
+      "loss": 0.2299,
+      "step": 87
+    },
+    {
+      "epoch": 2.0952380952380953,
+      "grad_norm": 0.9757384657859802,
+      "learning_rate": 5.580952380952381e-05,
+      "loss": 0.225,
+      "step": 88
+    },
+    {
+      "epoch": 2.119047619047619,
+      "grad_norm": 1.3605905771255493,
+      "learning_rate": 5.5761904761904764e-05,
+      "loss": 0.2237,
+      "step": 89
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 2.0481603145599365,
+      "learning_rate": 5.5714285714285715e-05,
+      "loss": 0.2233,
+      "step": 90
+    },
+    {
+      "epoch": 2.1666666666666665,
+      "grad_norm": 1.8589457273483276,
+      "learning_rate": 5.5666666666666674e-05,
+      "loss": 0.2201,
+      "step": 91
+    },
+    {
+      "epoch": 2.1904761904761907,
+      "grad_norm": 0.9307376742362976,
+      "learning_rate": 5.5619047619047625e-05,
+      "loss": 0.2077,
+      "step": 92
+    },
+    {
+      "epoch": 2.2142857142857144,
+      "grad_norm": 0.9198131561279297,
+      "learning_rate": 5.557142857142858e-05,
+      "loss": 0.2205,
+      "step": 93
+    },
+    {
+      "epoch": 2.238095238095238,
+      "grad_norm": 0.9321783781051636,
+      "learning_rate": 5.552380952380953e-05,
+      "loss": 0.2054,
+      "step": 94
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 1.1142381429672241,
+      "learning_rate": 5.547619047619048e-05,
+      "loss": 0.2356,
+      "step": 95
+    },
+    {
+      "epoch": 2.2857142857142856,
+      "grad_norm": 0.9719504714012146,
+      "learning_rate": 5.542857142857143e-05,
+      "loss": 0.2095,
+      "step": 96
+    },
+    {
+      "epoch": 2.3095238095238093,
+      "grad_norm": 0.9777172207832336,
+      "learning_rate": 5.538095238095238e-05,
+      "loss": 0.2025,
+      "step": 97
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 1.0100057125091553,
+      "learning_rate": 5.533333333333334e-05,
+      "loss": 0.2022,
+      "step": 98
+    },
+    {
+      "epoch": 2.357142857142857,
+      "grad_norm": 0.8255928754806519,
+      "learning_rate": 5.5285714285714286e-05,
+      "loss": 0.2161,
+      "step": 99
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.8491964936256409,
+      "learning_rate": 5.523809523809524e-05,
+      "loss": 0.1945,
+      "step": 100
+    },
+    {
+      "epoch": 2.380952380952381,
+      "eval_dice_score": 0.8890805782568201,
+      "eval_loss": 0.1874212622642517,
+      "eval_runtime": 9.0005,
+      "eval_samples_per_second": 9.888,
+      "eval_steps_per_second": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 2.4047619047619047,
+      "grad_norm": 1.0802334547042847,
+      "learning_rate": 5.519047619047619e-05,
+      "loss": 0.1996,
+      "step": 101
+    },
+    {
+      "epoch": 2.4285714285714284,
+      "grad_norm": 1.2004603147506714,
+      "learning_rate": 5.514285714285714e-05,
+      "loss": 0.218,
+      "step": 102
+    },
+    {
+      "epoch": 2.4523809523809526,
+      "grad_norm": 1.187143325805664,
+      "learning_rate": 5.509523809523809e-05,
+      "loss": 0.2213,
+      "step": 103
+    },
+    {
+      "epoch": 2.4761904761904763,
+      "grad_norm": 0.7938065528869629,
+      "learning_rate": 5.5047619047619044e-05,
+      "loss": 0.1995,
+      "step": 104
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8641159534454346,
+      "learning_rate": 5.5e-05,
+      "loss": 0.2353,
+      "step": 105
+    },
+    {
+      "epoch": 2.5238095238095237,
+      "grad_norm": 0.8983684182167053,
+      "learning_rate": 5.4952380952380954e-05,
+      "loss": 0.192,
+      "step": 106
+    },
+    {
+      "epoch": 2.5476190476190474,
+      "grad_norm": 0.730816125869751,
+      "learning_rate": 5.4904761904761905e-05,
+      "loss": 0.1958,
+      "step": 107
+    },
+    {
+      "epoch": 2.571428571428571,
+      "grad_norm": 0.7313885688781738,
+      "learning_rate": 5.4857142857142857e-05,
+      "loss": 0.2031,
+      "step": 108
+    },
+    {
+      "epoch": 2.5952380952380953,
+      "grad_norm": 0.8542109131813049,
+      "learning_rate": 5.480952380952381e-05,
+      "loss": 0.182,
+      "step": 109
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.8531911373138428,
+      "learning_rate": 5.476190476190476e-05,
+      "loss": 0.2021,
+      "step": 110
+    },
+    {
+      "epoch": 2.642857142857143,
+      "grad_norm": 0.8808167576789856,
+      "learning_rate": 5.471428571428571e-05,
+      "loss": 0.1899,
+      "step": 111
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 1.2466603517532349,
+      "learning_rate": 5.466666666666667e-05,
+      "loss": 0.2161,
+      "step": 112
+    },
+    {
+      "epoch": 2.6904761904761907,
+      "grad_norm": 0.8664019107818604,
+      "learning_rate": 5.461904761904762e-05,
+      "loss": 0.2048,
+      "step": 113
+    },
+    {
+      "epoch": 2.7142857142857144,
+      "grad_norm": 1.0919655561447144,
+      "learning_rate": 5.457142857142857e-05,
+      "loss": 0.2008,
+      "step": 114
+    },
+    {
+      "epoch": 2.738095238095238,
+      "grad_norm": 0.8462991118431091,
+      "learning_rate": 5.4523809523809524e-05,
+      "loss": 0.1982,
+      "step": 115
+    },
+    {
+      "epoch": 2.761904761904762,
+      "grad_norm": 0.7836592197418213,
+      "learning_rate": 5.4476190476190476e-05,
+      "loss": 0.1844,
+      "step": 116
+    },
+    {
+      "epoch": 2.7857142857142856,
+      "grad_norm": 1.0157865285873413,
+      "learning_rate": 5.442857142857143e-05,
+      "loss": 0.1879,
+      "step": 117
+    },
+    {
+      "epoch": 2.8095238095238093,
+      "grad_norm": 0.7947904467582703,
+      "learning_rate": 5.4380952380952386e-05,
+      "loss": 0.1813,
+      "step": 118
+    },
+    {
+      "epoch": 2.8333333333333335,
+      "grad_norm": 0.8740283846855164,
+      "learning_rate": 5.433333333333334e-05,
+      "loss": 0.1743,
+      "step": 119
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.9798213243484497,
+      "learning_rate": 5.428571428571429e-05,
+      "loss": 0.1768,
+      "step": 120
+    },
+    {
+      "epoch": 2.857142857142857,
+      "eval_dice_score": 0.8953037523457541,
+      "eval_loss": 0.1698274165391922,
+      "eval_runtime": 9.057,
+      "eval_samples_per_second": 9.827,
+      "eval_steps_per_second": 0.994,
+      "step": 120
+    },
+    {
+      "epoch": 2.880952380952381,
+      "grad_norm": 0.9734324812889099,
+      "learning_rate": 5.423809523809524e-05,
+      "loss": 0.2079,
+      "step": 121
+    },
+    {
+      "epoch": 2.9047619047619047,
+      "grad_norm": 0.6516209244728088,
+      "learning_rate": 5.419047619047619e-05,
+      "loss": 0.1745,
+      "step": 122
+    },
+    {
+      "epoch": 2.928571428571429,
+      "grad_norm": 0.7925631999969482,
+      "learning_rate": 5.414285714285714e-05,
+      "loss": 0.1802,
+      "step": 123
+    },
+    {
+      "epoch": 2.9523809523809526,
+      "grad_norm": 0.9028489589691162,
+      "learning_rate": 5.4095238095238095e-05,
+      "loss": 0.1625,
+      "step": 124
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 0.70108562707901,
+      "learning_rate": 5.404761904761905e-05,
+      "loss": 0.1718,
+      "step": 125
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7561673521995544,
+      "learning_rate": 5.4000000000000005e-05,
+      "loss": 0.1767,
+      "step": 126
+    },
+    {
+      "epoch": 3.0238095238095237,
+      "grad_norm": 0.8195943832397461,
+      "learning_rate": 5.3952380952380956e-05,
+      "loss": 0.1848,
+      "step": 127
+    },
+    {
+      "epoch": 3.0476190476190474,
+      "grad_norm": 0.7275943756103516,
+      "learning_rate": 5.390476190476191e-05,
+      "loss": 0.1706,
+      "step": 128
+    },
+    {
+      "epoch": 3.0714285714285716,
+      "grad_norm": 0.9391573071479797,
+      "learning_rate": 5.385714285714286e-05,
+      "loss": 0.1782,
+      "step": 129
+    },
+    {
+      "epoch": 3.0952380952380953,
+      "grad_norm": 0.8863614201545715,
+      "learning_rate": 5.380952380952381e-05,
+      "loss": 0.1732,
+      "step": 130
+    },
+    {
+      "epoch": 3.119047619047619,
+      "grad_norm": 0.9662719368934631,
+      "learning_rate": 5.376190476190476e-05,
+      "loss": 0.1547,
+      "step": 131
+    },
+    {
+      "epoch": 3.142857142857143,
+      "grad_norm": 0.6543298363685608,
+      "learning_rate": 5.371428571428572e-05,
+      "loss": 0.1682,
+      "step": 132
+    },
+    {
+      "epoch": 3.1666666666666665,
+      "grad_norm": 0.5799964070320129,
+      "learning_rate": 5.366666666666667e-05,
+      "loss": 0.1807,
+      "step": 133
+    },
+    {
+      "epoch": 3.1904761904761907,
+      "grad_norm": 0.8333367705345154,
+      "learning_rate": 5.3619047619047624e-05,
+      "loss": 0.1669,
+      "step": 134
+    },
+    {
+      "epoch": 3.2142857142857144,
+      "grad_norm": 0.886374831199646,
+      "learning_rate": 5.3571428571428575e-05,
+      "loss": 0.1904,
+      "step": 135
+    },
+    {
+      "epoch": 3.238095238095238,
+      "grad_norm": 0.7203497886657715,
+      "learning_rate": 5.352380952380953e-05,
+      "loss": 0.1707,
+      "step": 136
+    },
+    {
+      "epoch": 3.261904761904762,
+      "grad_norm": 0.7457689046859741,
+      "learning_rate": 5.347619047619048e-05,
+      "loss": 0.1644,
+      "step": 137
+    },
+    {
+      "epoch": 3.2857142857142856,
+      "grad_norm": 1.0430240631103516,
+      "learning_rate": 5.342857142857142e-05,
+      "loss": 0.1887,
+      "step": 138
+    },
+    {
+      "epoch": 3.3095238095238093,
+      "grad_norm": 0.764371931552887,
+      "learning_rate": 5.338095238095238e-05,
+      "loss": 0.1877,
+      "step": 139
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.6030626893043518,
+      "learning_rate": 5.333333333333333e-05,
+      "loss": 0.1616,
+      "step": 140
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "eval_dice_score": 0.8981800042364196,
+      "eval_loss": 0.148634672164917,
+      "eval_runtime": 8.7776,
+      "eval_samples_per_second": 10.139,
+      "eval_steps_per_second": 1.025,
+      "step": 140
+    },
+    {
+      "epoch": 3.357142857142857,
+      "grad_norm": 0.5721175074577332,
+      "learning_rate": 5.3285714285714285e-05,
+      "loss": 0.1484,
+      "step": 141
+    },
+    {
+      "epoch": 3.380952380952381,
+      "grad_norm": 0.5327512621879578,
+      "learning_rate": 5.3238095238095236e-05,
+      "loss": 0.1787,
+      "step": 142
+    },
+    {
+      "epoch": 3.4047619047619047,
+      "grad_norm": 0.6115806698799133,
+      "learning_rate": 5.319047619047619e-05,
+      "loss": 0.1689,
+      "step": 143
+    },
+    {
+      "epoch": 3.4285714285714284,
+      "grad_norm": 0.6276452541351318,
+      "learning_rate": 5.314285714285714e-05,
+      "loss": 0.1657,
+      "step": 144
+    },
+    {
+      "epoch": 3.4523809523809526,
+      "grad_norm": 0.8162292838096619,
+      "learning_rate": 5.30952380952381e-05,
+      "loss": 0.1554,
+      "step": 145
+    },
+    {
+      "epoch": 3.4761904761904763,
+      "grad_norm": 0.6978632211685181,
+      "learning_rate": 5.304761904761905e-05,
+      "loss": 0.1632,
+      "step": 146
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.866811990737915,
+      "learning_rate": 5.3e-05,
+      "loss": 0.1835,
+      "step": 147
+    },
+    {
+      "epoch": 3.5238095238095237,
+      "grad_norm": 0.625978410243988,
+      "learning_rate": 5.295238095238095e-05,
+      "loss": 0.1505,
+      "step": 148
+    },
+    {
+      "epoch": 3.5476190476190474,
+      "grad_norm": 0.5491347908973694,
+      "learning_rate": 5.2904761904761904e-05,
+      "loss": 0.1485,
+      "step": 149
+    },
+    {
+      "epoch": 3.571428571428571,
+      "grad_norm": 0.62246173620224,
+      "learning_rate": 5.2857142857142855e-05,
+      "loss": 0.1534,
+      "step": 150
+    },
+    {
+      "epoch": 3.5952380952380953,
+      "grad_norm": 0.5550210475921631,
+      "learning_rate": 5.280952380952381e-05,
+      "loss": 0.1489,
+      "step": 151
+    },
+    {
+      "epoch": 3.619047619047619,
+      "grad_norm": 0.6564983129501343,
+      "learning_rate": 5.2761904761904765e-05,
+      "loss": 0.1616,
+      "step": 152
+    },
+    {
+      "epoch": 3.642857142857143,
+      "grad_norm": 0.7625710368156433,
+      "learning_rate": 5.271428571428572e-05,
+      "loss": 0.1382,
+      "step": 153
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.6809992790222168,
+      "learning_rate": 5.266666666666667e-05,
+      "loss": 0.1593,
+      "step": 154
+    },
+    {
+      "epoch": 3.6904761904761907,
+      "grad_norm": 0.5852124094963074,
+      "learning_rate": 5.261904761904762e-05,
+      "loss": 0.1675,
+      "step": 155
+    },
+    {
+      "epoch": 3.7142857142857144,
+      "grad_norm": 0.5885810256004333,
+      "learning_rate": 5.257142857142857e-05,
+      "loss": 0.1443,
+      "step": 156
+    },
+    {
+      "epoch": 3.738095238095238,
+      "grad_norm": 0.8580813407897949,
+      "learning_rate": 5.252380952380952e-05,
+      "loss": 0.1471,
+      "step": 157
+    },
+    {
+      "epoch": 3.761904761904762,
+      "grad_norm": 0.7032930254936218,
+      "learning_rate": 5.2476190476190475e-05,
+      "loss": 0.1403,
+      "step": 158
+    },
+    {
+      "epoch": 3.7857142857142856,
+      "grad_norm": 0.6581481099128723,
+      "learning_rate": 5.242857142857143e-05,
+      "loss": 0.1555,
+      "step": 159
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.9175595641136169,
+      "learning_rate": 5.2380952380952384e-05,
+      "loss": 0.1749,
+      "step": 160
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "eval_dice_score": 0.9037976009755364,
+      "eval_loss": 0.1529674082994461,
+      "eval_runtime": 8.7408,
+      "eval_samples_per_second": 10.182,
+      "eval_steps_per_second": 1.03,
+      "step": 160
+    },
+    {
+      "epoch": 3.8333333333333335,
+      "grad_norm": 0.5896248817443848,
+      "learning_rate": 5.2333333333333336e-05,
+      "loss": 0.1709,
+      "step": 161
+    },
+    {
+      "epoch": 3.857142857142857,
+      "grad_norm": 0.5807002782821655,
+      "learning_rate": 5.228571428571429e-05,
+      "loss": 0.1436,
+      "step": 162
+    },
+    {
+      "epoch": 3.880952380952381,
+      "grad_norm": 0.6526125073432922,
+      "learning_rate": 5.223809523809524e-05,
+      "loss": 0.1327,
+      "step": 163
+    },
+    {
+      "epoch": 3.9047619047619047,
+      "grad_norm": 0.5510716438293457,
+      "learning_rate": 5.219047619047619e-05,
+      "loss": 0.1627,
+      "step": 164
+    },
+    {
+      "epoch": 3.928571428571429,
+      "grad_norm": 0.5205186009407043,
+      "learning_rate": 5.214285714285714e-05,
+      "loss": 0.1322,
+      "step": 165
+    },
+    {
+      "epoch": 3.9523809523809526,
+      "grad_norm": 0.5087710618972778,
+      "learning_rate": 5.20952380952381e-05,
+      "loss": 0.1546,
+      "step": 166
+    },
+    {
+      "epoch": 3.9761904761904763,
+      "grad_norm": 0.5551671981811523,
+      "learning_rate": 5.204761904761905e-05,
+      "loss": 0.1584,
+      "step": 167
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6679753065109253,
+      "learning_rate": 5.2000000000000004e-05,
+      "loss": 0.1515,
+      "step": 168
+    },
+    {
+      "epoch": 4.023809523809524,
+      "grad_norm": 0.547113299369812,
+      "learning_rate": 5.1952380952380955e-05,
+      "loss": 0.1424,
+      "step": 169
+    },
+    {
+      "epoch": 4.0476190476190474,
+      "grad_norm": 0.43713295459747314,
+      "learning_rate": 5.190476190476191e-05,
+      "loss": 0.1534,
+      "step": 170
+    },
+    {
+      "epoch": 4.071428571428571,
+      "grad_norm": 0.5712536573410034,
+      "learning_rate": 5.185714285714286e-05,
+      "loss": 0.1674,
+      "step": 171
+    },
+    {
+      "epoch": 4.095238095238095,
+      "grad_norm": 0.5892536044120789,
+      "learning_rate": 5.1809523809523817e-05,
+      "loss": 0.1374,
+      "step": 172
+    },
+    {
+      "epoch": 4.119047619047619,
+      "grad_norm": 0.5541051626205444,
+      "learning_rate": 5.176190476190477e-05,
+      "loss": 0.154,
+      "step": 173
+    },
+    {
+      "epoch": 4.142857142857143,
+      "grad_norm": 0.6557343602180481,
+      "learning_rate": 5.171428571428572e-05,
+      "loss": 0.1421,
+      "step": 174
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 0.48764219880104065,
+      "learning_rate": 5.166666666666667e-05,
+      "loss": 0.1351,
+      "step": 175
+    },
+    {
+      "epoch": 4.190476190476191,
+      "grad_norm": 0.4162357747554779,
+      "learning_rate": 5.161904761904762e-05,
+      "loss": 0.1515,
+      "step": 176
+    },
+    {
+      "epoch": 4.214285714285714,
+      "grad_norm": 0.5261713862419128,
+      "learning_rate": 5.1571428571428574e-05,
+      "loss": 0.1518,
+      "step": 177
+    },
+    {
+      "epoch": 4.238095238095238,
+      "grad_norm": 0.5337617993354797,
+      "learning_rate": 5.152380952380952e-05,
+      "loss": 0.1217,
+      "step": 178
+    },
+    {
+      "epoch": 4.261904761904762,
+      "grad_norm": 0.6286489963531494,
+      "learning_rate": 5.147619047619048e-05,
+      "loss": 0.1409,
+      "step": 179
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.6179653406143188,
+      "learning_rate": 5.142857142857143e-05,
+      "loss": 0.1358,
+      "step": 180
+    },
+    {
+      "epoch": 4.285714285714286,
+      "eval_dice_score": 0.9069158827058441,
+      "eval_loss": 0.13152673840522766,
+      "eval_runtime": 8.7958,
+      "eval_samples_per_second": 10.118,
+      "eval_steps_per_second": 1.023,
+      "step": 180
+    },
+    {
+      "epoch": 4.309523809523809,
+      "grad_norm": 0.4932554066181183,
+      "learning_rate": 5.138095238095238e-05,
+      "loss": 0.1561,
+      "step": 181
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 0.5466799139976501,
+      "learning_rate": 5.133333333333333e-05,
+      "loss": 0.1548,
+      "step": 182
+    },
+    {
+      "epoch": 4.357142857142857,
+      "grad_norm": 0.44180309772491455,
+      "learning_rate": 5.1285714285714284e-05,
+      "loss": 0.1482,
+      "step": 183
+    },
+    {
+      "epoch": 4.380952380952381,
+      "grad_norm": 0.7849823832511902,
+      "learning_rate": 5.1238095238095235e-05,
+      "loss": 0.1794,
+      "step": 184
+    },
+    {
+      "epoch": 4.404761904761905,
+      "grad_norm": 0.406508207321167,
+      "learning_rate": 5.119047619047619e-05,
+      "loss": 0.1294,
+      "step": 185
+    },
+    {
+      "epoch": 4.428571428571429,
+      "grad_norm": 0.8998963236808777,
+      "learning_rate": 5.1142857142857145e-05,
+      "loss": 0.1358,
+      "step": 186
+    },
+    {
+      "epoch": 4.4523809523809526,
+      "grad_norm": 0.41310542821884155,
+      "learning_rate": 5.1095238095238097e-05,
+      "loss": 0.134,
+      "step": 187
+    },
+    {
+      "epoch": 4.476190476190476,
+      "grad_norm": 0.41451480984687805,
+      "learning_rate": 5.104761904761905e-05,
+      "loss": 0.1419,
+      "step": 188
+    },
+    {
+      "epoch": 4.5,
+      "grad_norm": 0.5282092094421387,
+      "learning_rate": 5.1e-05,
+      "loss": 0.136,
+      "step": 189
+    },
+    {
+      "epoch": 4.523809523809524,
+      "grad_norm": 0.5178981423377991,
+      "learning_rate": 5.095238095238095e-05,
+      "loss": 0.1274,
+      "step": 190
+    },
+    {
+      "epoch": 4.5476190476190474,
+      "grad_norm": 0.5173990726470947,
+      "learning_rate": 5.09047619047619e-05,
+      "loss": 0.1391,
+      "step": 191
+    },
+    {
+      "epoch": 4.571428571428571,
+      "grad_norm": 0.479457825422287,
+      "learning_rate": 5.0857142857142854e-05,
+      "loss": 0.1332,
+      "step": 192
+    },
+    {
+      "epoch": 4.595238095238095,
+      "grad_norm": 0.4143787622451782,
+      "learning_rate": 5.080952380952381e-05,
+      "loss": 0.1571,
+      "step": 193
+    },
+    {
+      "epoch": 4.619047619047619,
+      "grad_norm": 0.4726862609386444,
+      "learning_rate": 5.0761904761904764e-05,
+      "loss": 0.1373,
+      "step": 194
+    },
+    {
+      "epoch": 4.642857142857143,
+      "grad_norm": 0.42671865224838257,
+      "learning_rate": 5.0714285714285716e-05,
+      "loss": 0.1316,
+      "step": 195
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.5958961844444275,
+      "learning_rate": 5.066666666666667e-05,
+      "loss": 0.1563,
+      "step": 196
+    },
+    {
+      "epoch": 4.690476190476191,
+      "grad_norm": 0.4654116928577423,
+      "learning_rate": 5.061904761904762e-05,
+      "loss": 0.1426,
+      "step": 197
+    },
+    {
+      "epoch": 4.714285714285714,
+      "grad_norm": 0.6422296762466431,
+      "learning_rate": 5.057142857142857e-05,
+      "loss": 0.1278,
+      "step": 198
+    },
+    {
+      "epoch": 4.738095238095238,
+      "grad_norm": 0.4659450352191925,
+      "learning_rate": 5.052380952380953e-05,
+      "loss": 0.12,
+      "step": 199
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.4322237968444824,
+      "learning_rate": 5.047619047619048e-05,
+      "loss": 0.1166,
+      "step": 200
+    },
+    {
+      "epoch": 4.761904761904762,
+      "eval_dice_score": 0.9074214584101227,
+      "eval_loss": 0.12160411477088928,
+      "eval_runtime": 8.9569,
+      "eval_samples_per_second": 9.937,
+      "eval_steps_per_second": 1.005,
+      "step": 200
+    },
+    {
+      "epoch": 4.785714285714286,
+      "grad_norm": 0.6910778284072876,
+      "learning_rate": 5.042857142857143e-05,
+      "loss": 0.1562,
+      "step": 201
+    },
+    {
+      "epoch": 4.809523809523809,
+      "grad_norm": 0.4675683081150055,
+      "learning_rate": 5.038095238095238e-05,
+      "loss": 0.1547,
+      "step": 202
+    },
+    {
+      "epoch": 4.833333333333333,
+      "grad_norm": 0.4359719753265381,
+      "learning_rate": 5.0333333333333335e-05,
+      "loss": 0.1196,
+      "step": 203
+    },
+    {
+      "epoch": 4.857142857142857,
+      "grad_norm": 0.4786078929901123,
+      "learning_rate": 5.0285714285714286e-05,
+      "loss": 0.1441,
+      "step": 204
+    },
+    {
+      "epoch": 4.880952380952381,
+      "grad_norm": 0.5416626334190369,
+      "learning_rate": 5.023809523809524e-05,
+      "loss": 0.1219,
+      "step": 205
+    },
+    {
+      "epoch": 4.904761904761905,
+      "grad_norm": 0.5497027039527893,
+      "learning_rate": 5.0190476190476196e-05,
+      "loss": 0.1221,
+      "step": 206
+    },
+    {
+      "epoch": 4.928571428571429,
+      "grad_norm": 0.5406143069267273,
+      "learning_rate": 5.014285714285715e-05,
+      "loss": 0.1211,
+      "step": 207
+    },
+    {
+      "epoch": 4.9523809523809526,
+      "grad_norm": 0.472383052110672,
+      "learning_rate": 5.00952380952381e-05,
+      "loss": 0.1459,
+      "step": 208
+    },
+    {
+      "epoch": 4.976190476190476,
+      "grad_norm": 0.4744516611099243,
+      "learning_rate": 5.004761904761905e-05,
+      "loss": 0.1488,
+      "step": 209
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.40707042813301086,
+      "learning_rate": 5e-05,
+      "loss": 0.1423,
+      "step": 210
+    },
+    {
+      "epoch": 5.023809523809524,
+      "grad_norm": 0.3842790424823761,
+      "learning_rate": 4.9952380952380954e-05,
+      "loss": 0.1329,
+      "step": 211
+    },
+    {
+      "epoch": 5.0476190476190474,
+      "grad_norm": 0.45681342482566833,
+      "learning_rate": 4.9904761904761905e-05,
+      "loss": 0.1366,
+      "step": 212
+    },
+    {
+      "epoch": 5.071428571428571,
+      "grad_norm": 0.3685499429702759,
+      "learning_rate": 4.9857142857142864e-05,
+      "loss": 0.1247,
+      "step": 213
+    },
+    {
+      "epoch": 5.095238095238095,
+      "grad_norm": 0.44698742032051086,
+      "learning_rate": 4.9809523809523815e-05,
+      "loss": 0.1252,
+      "step": 214
+    },
+    {
+      "epoch": 5.119047619047619,
+      "grad_norm": 0.49723654985427856,
+      "learning_rate": 4.976190476190477e-05,
+      "loss": 0.1311,
+      "step": 215
+    },
+    {
+      "epoch": 5.142857142857143,
+      "grad_norm": 0.6091992855072021,
+      "learning_rate": 4.971428571428572e-05,
+      "loss": 0.1711,
+      "step": 216
+    },
+    {
+      "epoch": 5.166666666666667,
+      "grad_norm": 0.5066311955451965,
+      "learning_rate": 4.966666666666666e-05,
+      "loss": 0.1323,
+      "step": 217
+    },
+    {
+      "epoch": 5.190476190476191,
+      "grad_norm": 0.3507075011730194,
+      "learning_rate": 4.9619047619047615e-05,
+      "loss": 0.1135,
+      "step": 218
+    },
+    {
+      "epoch": 5.214285714285714,
+      "grad_norm": 0.4396760165691376,
+      "learning_rate": 4.957142857142857e-05,
+      "loss": 0.1474,
+      "step": 219
+    },
+    {
+      "epoch": 5.238095238095238,
+      "grad_norm": 0.39729997515678406,
+      "learning_rate": 4.9523809523809525e-05,
+      "loss": 0.1426,
+      "step": 220
+    },
+    {
+      "epoch": 5.238095238095238,
+      "eval_dice_score": 0.9099840335920737,
+      "eval_loss": 0.11943376809358597,
+      "eval_runtime": 8.7914,
+      "eval_samples_per_second": 10.124,
+      "eval_steps_per_second": 1.024,
+      "step": 220
+    },
+    {
+      "epoch": 5.261904761904762,
+      "grad_norm": 0.49762511253356934,
+      "learning_rate": 4.9476190476190476e-05,
+      "loss": 0.1177,
+      "step": 221
+    },
+    {
+      "epoch": 5.285714285714286,
+      "grad_norm": 0.33840247988700867,
+      "learning_rate": 4.942857142857143e-05,
+      "loss": 0.1343,
+      "step": 222
+    },
+    {
+      "epoch": 5.309523809523809,
+      "grad_norm": 0.3600093722343445,
+      "learning_rate": 4.938095238095238e-05,
+      "loss": 0.113,
+      "step": 223
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.45041218400001526,
+      "learning_rate": 4.933333333333333e-05,
+      "loss": 0.1359,
+      "step": 224
+    },
+    {
+      "epoch": 5.357142857142857,
+      "grad_norm": 0.3925110697746277,
+      "learning_rate": 4.928571428571428e-05,
+      "loss": 0.1125,
+      "step": 225
+    },
+    {
+      "epoch": 5.380952380952381,
+      "grad_norm": 0.3413161039352417,
+      "learning_rate": 4.923809523809524e-05,
+      "loss": 0.1173,
+      "step": 226
+    },
+    {
+      "epoch": 5.404761904761905,
+      "grad_norm": 0.4640910029411316,
+      "learning_rate": 4.919047619047619e-05,
+      "loss": 0.1395,
+      "step": 227
+    },
+    {
+      "epoch": 5.428571428571429,
+      "grad_norm": 0.44285720586776733,
+      "learning_rate": 4.9142857142857144e-05,
+      "loss": 0.1347,
+      "step": 228
+    },
+    {
+      "epoch": 5.4523809523809526,
+      "grad_norm": 0.4267982244491577,
+      "learning_rate": 4.9095238095238095e-05,
+      "loss": 0.1234,
+      "step": 229
+    },
+    {
+      "epoch": 5.476190476190476,
+      "grad_norm": 0.4833822548389435,
+      "learning_rate": 4.904761904761905e-05,
+      "loss": 0.1146,
+      "step": 230
+    },
+    {
+      "epoch": 5.5,
+      "grad_norm": 0.4235381782054901,
+      "learning_rate": 4.9e-05,
+      "loss": 0.127,
+      "step": 231
+    },
+    {
+      "epoch": 5.523809523809524,
+      "grad_norm": 0.36395061016082764,
+      "learning_rate": 4.895238095238095e-05,
+      "loss": 0.1132,
+      "step": 232
+    },
+    {
+      "epoch": 5.5476190476190474,
+      "grad_norm": 0.3754006326198578,
+      "learning_rate": 4.890476190476191e-05,
+      "loss": 0.1476,
+      "step": 233
+    },
+    {
+      "epoch": 5.571428571428571,
+      "grad_norm": 0.5128763318061829,
+      "learning_rate": 4.885714285714286e-05,
+      "loss": 0.1052,
+      "step": 234
+    },
+    {
+      "epoch": 5.595238095238095,
+      "grad_norm": 0.381115198135376,
+      "learning_rate": 4.880952380952381e-05,
+      "loss": 0.1229,
+      "step": 235
+    },
+    {
+      "epoch": 5.619047619047619,
+      "grad_norm": 0.42561641335487366,
+      "learning_rate": 4.876190476190476e-05,
+      "loss": 0.1178,
+      "step": 236
+    },
+    {
+      "epoch": 5.642857142857143,
+      "grad_norm": 0.4684426486492157,
+      "learning_rate": 4.8714285714285714e-05,
+      "loss": 0.1256,
+      "step": 237
+    },
+    {
+      "epoch": 5.666666666666667,
+      "grad_norm": 0.3221814036369324,
+      "learning_rate": 4.8666666666666666e-05,
+      "loss": 0.1299,
+      "step": 238
+    },
+    {
+      "epoch": 5.690476190476191,
+      "grad_norm": 0.3549591898918152,
+      "learning_rate": 4.861904761904762e-05,
+      "loss": 0.1103,
+      "step": 239
+    },
+    {
+      "epoch": 5.714285714285714,
+      "grad_norm": 0.4589278995990753,
+      "learning_rate": 4.8571428571428576e-05,
+      "loss": 0.1199,
+      "step": 240
+    },
+    {
+      "epoch": 5.714285714285714,
+      "eval_dice_score": 0.9114496222860314,
+      "eval_loss": 0.12006842344999313,
+      "eval_runtime": 8.774,
+      "eval_samples_per_second": 10.144,
+      "eval_steps_per_second": 1.026,
+      "step": 240
+    },
+    {
+      "epoch": 5.738095238095238,
+      "grad_norm": 0.46148115396499634,
+      "learning_rate": 4.852380952380953e-05,
+      "loss": 0.1248,
+      "step": 241
+    },
+    {
+      "epoch": 5.761904761904762,
+      "grad_norm": 0.535234808921814,
+      "learning_rate": 4.847619047619048e-05,
+      "loss": 0.1367,
+      "step": 242
+    },
+    {
+      "epoch": 5.785714285714286,
+      "grad_norm": 0.4368140399456024,
+      "learning_rate": 4.842857142857143e-05,
+      "loss": 0.1282,
+      "step": 243
+    },
+    {
+      "epoch": 5.809523809523809,
+      "grad_norm": 0.41349226236343384,
+      "learning_rate": 4.838095238095238e-05,
+      "loss": 0.1089,
+      "step": 244
+    },
+    {
+      "epoch": 5.833333333333333,
+      "grad_norm": 0.3823813199996948,
+      "learning_rate": 4.8333333333333334e-05,
+      "loss": 0.1409,
+      "step": 245
+    },
+    {
+      "epoch": 5.857142857142857,
+      "grad_norm": 0.3816284239292145,
+      "learning_rate": 4.828571428571429e-05,
+      "loss": 0.1056,
+      "step": 246
+    },
+    {
+      "epoch": 5.880952380952381,
+      "grad_norm": 0.5315767526626587,
+      "learning_rate": 4.8238095238095243e-05,
+      "loss": 0.1201,
+      "step": 247
+    },
+    {
+      "epoch": 5.904761904761905,
+      "grad_norm": 0.46191665530204773,
+      "learning_rate": 4.8190476190476195e-05,
+      "loss": 0.1306,
+      "step": 248
+    },
+    {
+      "epoch": 5.928571428571429,
+      "grad_norm": 0.41716304421424866,
+      "learning_rate": 4.8142857142857147e-05,
+      "loss": 0.1361,
+      "step": 249
+    },
+    {
+      "epoch": 5.9523809523809526,
+      "grad_norm": 0.5382347106933594,
+      "learning_rate": 4.80952380952381e-05,
+      "loss": 0.1499,
+      "step": 250
+    },
+    {
+      "epoch": 5.976190476190476,
+      "grad_norm": 0.4075874090194702,
+      "learning_rate": 4.804761904761905e-05,
+      "loss": 0.1221,
+      "step": 251
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.30484116077423096,
+      "learning_rate": 4.8e-05,
+      "loss": 0.1365,
+      "step": 252
+    },
+    {
+      "epoch": 6.023809523809524,
+      "grad_norm": 0.46580827236175537,
+      "learning_rate": 4.795238095238096e-05,
+      "loss": 0.1254,
+      "step": 253
+    },
+    {
+      "epoch": 6.0476190476190474,
+      "grad_norm": 0.41510361433029175,
+      "learning_rate": 4.790476190476191e-05,
+      "loss": 0.1411,
+      "step": 254
+    },
+    {
+      "epoch": 6.071428571428571,
+      "grad_norm": 0.37615931034088135,
+      "learning_rate": 4.785714285714286e-05,
+      "loss": 0.1101,
+      "step": 255
+    },
+    {
+      "epoch": 6.095238095238095,
+      "grad_norm": 0.43453407287597656,
+      "learning_rate": 4.780952380952381e-05,
+      "loss": 0.1227,
+      "step": 256
+    },
+    {
+      "epoch": 6.119047619047619,
+      "grad_norm": 0.3365417718887329,
+      "learning_rate": 4.776190476190476e-05,
+      "loss": 0.1497,
+      "step": 257
+    },
+    {
+      "epoch": 6.142857142857143,
+      "grad_norm": 0.3814278244972229,
+      "learning_rate": 4.771428571428571e-05,
+      "loss": 0.1164,
+      "step": 258
+    },
+    {
+      "epoch": 6.166666666666667,
+      "grad_norm": 0.4330274760723114,
+      "learning_rate": 4.766666666666666e-05,
+      "loss": 0.1056,
+      "step": 259
+    },
+    {
+      "epoch": 6.190476190476191,
+      "grad_norm": 0.3585191071033478,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 0.142,
+      "step": 260
+    },
+    {
+      "epoch": 6.190476190476191,
+      "eval_dice_score": 0.9102984988362752,
+      "eval_loss": 0.11125586926937103,
+      "eval_runtime": 8.8105,
+      "eval_samples_per_second": 10.102,
+      "eval_steps_per_second": 1.022,
+      "step": 260
+    },
+    {
+      "epoch": 6.214285714285714,
+      "grad_norm": 0.38072383403778076,
+      "learning_rate": 4.757142857142857e-05,
+      "loss": 0.1254,
+      "step": 261
+    },
+    {
+      "epoch": 6.238095238095238,
+      "grad_norm": 0.439165323972702,
+      "learning_rate": 4.7523809523809523e-05,
+      "loss": 0.1129,
+      "step": 262
+    },
+    {
+      "epoch": 6.261904761904762,
+      "grad_norm": 0.4433733820915222,
+      "learning_rate": 4.7476190476190475e-05,
+      "loss": 0.1194,
+      "step": 263
+    },
+    {
+      "epoch": 6.285714285714286,
+      "grad_norm": 0.36688342690467834,
+      "learning_rate": 4.7428571428571427e-05,
+      "loss": 0.1261,
+      "step": 264
+    },
+    {
+      "epoch": 6.309523809523809,
+      "grad_norm": 0.41297096014022827,
+      "learning_rate": 4.738095238095238e-05,
+      "loss": 0.1227,
+      "step": 265
+    },
+    {
+      "epoch": 6.333333333333333,
+      "grad_norm": 0.4545275866985321,
+      "learning_rate": 4.733333333333333e-05,
+      "loss": 0.1067,
+      "step": 266
+    },
+    {
+      "epoch": 6.357142857142857,
+      "grad_norm": 0.40353894233703613,
+      "learning_rate": 4.728571428571429e-05,
+      "loss": 0.122,
+      "step": 267
+    },
+    {
+      "epoch": 6.380952380952381,
+      "grad_norm": 0.36850976943969727,
+      "learning_rate": 4.723809523809524e-05,
+      "loss": 0.1036,
+      "step": 268
+    },
+    {
+      "epoch": 6.404761904761905,
+      "grad_norm": 0.4579184651374817,
+      "learning_rate": 4.719047619047619e-05,
+      "loss": 0.1225,
+      "step": 269
+    },
+    {
+      "epoch": 6.428571428571429,
+      "grad_norm": 0.3441743850708008,
+      "learning_rate": 4.714285714285714e-05,
+      "loss": 0.1114,
+      "step": 270
+    },
+    {
+      "epoch": 6.4523809523809526,
+      "grad_norm": 0.3042502999305725,
+      "learning_rate": 4.7095238095238094e-05,
+      "loss": 0.1069,
+      "step": 271
+    },
+    {
+      "epoch": 6.476190476190476,
+      "grad_norm": 0.34080275893211365,
+      "learning_rate": 4.7047619047619046e-05,
+      "loss": 0.113,
+      "step": 272
+    },
+    {
+      "epoch": 6.5,
+      "grad_norm": 0.31439319252967834,
+      "learning_rate": 4.7000000000000004e-05,
+      "loss": 0.1119,
+      "step": 273
+    },
+    {
+      "epoch": 6.523809523809524,
+      "grad_norm": 0.36084115505218506,
+      "learning_rate": 4.6952380952380956e-05,
+      "loss": 0.1064,
+      "step": 274
+    },
+    {
+      "epoch": 6.5476190476190474,
+      "grad_norm": 0.38591280579566956,
+      "learning_rate": 4.690476190476191e-05,
+      "loss": 0.1303,
+      "step": 275
+    },
+    {
+      "epoch": 6.571428571428571,
+      "grad_norm": 0.32427167892456055,
+      "learning_rate": 4.685714285714286e-05,
+      "loss": 0.1243,
+      "step": 276
+    },
+    {
+      "epoch": 6.595238095238095,
+      "grad_norm": 0.36310875415802,
+      "learning_rate": 4.680952380952381e-05,
+      "loss": 0.1252,
+      "step": 277
+    },
+    {
+      "epoch": 6.619047619047619,
+      "grad_norm": 0.3615739047527313,
+      "learning_rate": 4.676190476190476e-05,
+      "loss": 0.1272,
+      "step": 278
+    },
+    {
+      "epoch": 6.642857142857143,
+      "grad_norm": 0.4673764109611511,
+      "learning_rate": 4.671428571428571e-05,
+      "loss": 0.1072,
+      "step": 279
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 0.43281465768814087,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.1104,
+      "step": 280
+    },
+    {
+      "epoch": 6.666666666666667,
+      "eval_dice_score": 0.91285228294862,
+      "eval_loss": 0.11082390695810318,
+      "eval_runtime": 8.8851,
+      "eval_samples_per_second": 10.017,
+      "eval_steps_per_second": 1.013,
+      "step": 280
+    },
+    {
+      "epoch": 6.690476190476191,
+      "grad_norm": 0.2869553565979004,
+      "learning_rate": 4.661904761904762e-05,
+      "loss": 0.1214,
+      "step": 281
+    },
+    {
+      "epoch": 6.714285714285714,
+      "grad_norm": 0.30401700735092163,
+      "learning_rate": 4.6571428571428575e-05,
+      "loss": 0.1001,
+      "step": 282
+    },
+    {
+      "epoch": 6.738095238095238,
+      "grad_norm": 0.2841688096523285,
+      "learning_rate": 4.6523809523809526e-05,
+      "loss": 0.1104,
+      "step": 283
+    },
+    {
+      "epoch": 6.761904761904762,
+      "grad_norm": 0.3578402101993561,
+      "learning_rate": 4.647619047619048e-05,
+      "loss": 0.1234,
+      "step": 284
+    },
+    {
+      "epoch": 6.785714285714286,
+      "grad_norm": 0.3647173047065735,
+      "learning_rate": 4.642857142857143e-05,
+      "loss": 0.1239,
+      "step": 285
+    },
+    {
+      "epoch": 6.809523809523809,
+      "grad_norm": 0.34969446063041687,
+      "learning_rate": 4.638095238095238e-05,
+      "loss": 0.1133,
+      "step": 286
+    },
+    {
+      "epoch": 6.833333333333333,
+      "grad_norm": 0.48136159777641296,
+      "learning_rate": 4.633333333333334e-05,
+      "loss": 0.1228,
+      "step": 287
+    },
+    {
+      "epoch": 6.857142857142857,
+      "grad_norm": 0.4168005585670471,
+      "learning_rate": 4.628571428571429e-05,
+      "loss": 0.1239,
+      "step": 288
+    },
+    {
+      "epoch": 6.880952380952381,
+      "grad_norm": 0.37365084886550903,
+      "learning_rate": 4.623809523809524e-05,
+      "loss": 0.1417,
+      "step": 289
+    },
+    {
+      "epoch": 6.904761904761905,
+      "grad_norm": 0.3516845703125,
+      "learning_rate": 4.6190476190476194e-05,
+      "loss": 0.1439,
+      "step": 290
+    },
+    {
+      "epoch": 6.928571428571429,
+      "grad_norm": 0.3463188111782074,
+      "learning_rate": 4.6142857142857145e-05,
+      "loss": 0.1258,
+      "step": 291
+    },
+    {
+      "epoch": 6.9523809523809526,
+      "grad_norm": 0.3746688961982727,
+      "learning_rate": 4.60952380952381e-05,
+      "loss": 0.1208,
+      "step": 292
+    },
+    {
+      "epoch": 6.976190476190476,
+      "grad_norm": 0.4245692491531372,
+      "learning_rate": 4.604761904761905e-05,
+      "loss": 0.1017,
+      "step": 293
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.46867018938064575,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.0981,
+      "step": 294
+    },
+    {
+      "epoch": 7.023809523809524,
+      "grad_norm": 0.308337539434433,
+      "learning_rate": 4.595238095238096e-05,
+      "loss": 0.1375,
+      "step": 295
+    },
+    {
+      "epoch": 7.0476190476190474,
+      "grad_norm": 0.4807004928588867,
+      "learning_rate": 4.59047619047619e-05,
+      "loss": 0.1427,
+      "step": 296
+    },
+    {
+      "epoch": 7.071428571428571,
+      "grad_norm": 0.5126243233680725,
+      "learning_rate": 4.5857142857142855e-05,
+      "loss": 0.0971,
+      "step": 297
+    },
+    {
+      "epoch": 7.095238095238095,
+      "grad_norm": 0.37334510684013367,
+      "learning_rate": 4.5809523809523806e-05,
+      "loss": 0.1248,
+      "step": 298
+    },
+    {
+      "epoch": 7.119047619047619,
+      "grad_norm": 0.41911178827285767,
+      "learning_rate": 4.576190476190476e-05,
+      "loss": 0.1111,
+      "step": 299
+    },
+    {
+      "epoch": 7.142857142857143,
+      "grad_norm": 0.3633350133895874,
+      "learning_rate": 4.5714285714285716e-05,
+      "loss": 0.0931,
+      "step": 300
+    },
+    {
+      "epoch": 7.142857142857143,
+      "eval_dice_score": 0.9114141838731112,
+      "eval_loss": 0.10547726601362228,
+      "eval_runtime": 8.8954,
+      "eval_samples_per_second": 10.005,
+      "eval_steps_per_second": 1.012,
+      "step": 300
+    },
+    {
+      "epoch": 7.166666666666667,
+      "grad_norm": 0.3622587323188782,
+      "learning_rate": 4.566666666666667e-05,
+      "loss": 0.0989,
+      "step": 301
+    },
+    {
+      "epoch": 7.190476190476191,
+      "grad_norm": 0.53045254945755,
+      "learning_rate": 4.561904761904762e-05,
+      "loss": 0.1146,
+      "step": 302
+    },
+    {
+      "epoch": 7.214285714285714,
+      "grad_norm": 0.2951499819755554,
+      "learning_rate": 4.557142857142857e-05,
+      "loss": 0.12,
+      "step": 303
+    },
+    {
+      "epoch": 7.238095238095238,
+      "grad_norm": 0.29450082778930664,
+      "learning_rate": 4.552380952380952e-05,
+      "loss": 0.0994,
+      "step": 304
+    },
+    {
+      "epoch": 7.261904761904762,
+      "grad_norm": 0.35549187660217285,
+      "learning_rate": 4.5476190476190474e-05,
+      "loss": 0.1136,
+      "step": 305
+    },
+    {
+      "epoch": 7.285714285714286,
+      "grad_norm": 0.4132053256034851,
+      "learning_rate": 4.5428571428571425e-05,
+      "loss": 0.1176,
+      "step": 306
+    },
+    {
+      "epoch": 7.309523809523809,
+      "grad_norm": 0.540023922920227,
+      "learning_rate": 4.5380952380952384e-05,
+      "loss": 0.1395,
+      "step": 307
+    },
+    {
+      "epoch": 7.333333333333333,
+      "grad_norm": 0.3276040852069855,
+      "learning_rate": 4.5333333333333335e-05,
+      "loss": 0.1218,
+      "step": 308
+    },
+    {
+      "epoch": 7.357142857142857,
+      "grad_norm": 0.2875545024871826,
+      "learning_rate": 4.528571428571429e-05,
+      "loss": 0.1031,
+      "step": 309
+    },
+    {
+      "epoch": 7.380952380952381,
+      "grad_norm": 0.3251423239707947,
+      "learning_rate": 4.523809523809524e-05,
+      "loss": 0.1086,
+      "step": 310
+    },
+    {
+      "epoch": 7.404761904761905,
+      "grad_norm": 0.3128550946712494,
+      "learning_rate": 4.519047619047619e-05,
+      "loss": 0.1142,
+      "step": 311
+    },
+    {
+      "epoch": 7.428571428571429,
+      "grad_norm": 0.29231375455856323,
+      "learning_rate": 4.514285714285714e-05,
+      "loss": 0.1177,
+      "step": 312
+    },
+    {
+      "epoch": 7.4523809523809526,
+      "grad_norm": 0.27210715413093567,
+      "learning_rate": 4.509523809523809e-05,
+      "loss": 0.0969,
+      "step": 313
+    },
+    {
+      "epoch": 7.476190476190476,
+      "grad_norm": 0.35545647144317627,
+      "learning_rate": 4.504761904761905e-05,
+      "loss": 0.1007,
+      "step": 314
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 0.37293487787246704,
+      "learning_rate": 4.5e-05,
+      "loss": 0.113,
+      "step": 315
+    },
+    {
+      "epoch": 7.523809523809524,
+      "grad_norm": 0.45221376419067383,
+      "learning_rate": 4.4952380952380954e-05,
+      "loss": 0.1136,
+      "step": 316
+    },
+    {
+      "epoch": 7.5476190476190474,
+      "grad_norm": 0.4034024775028229,
+      "learning_rate": 4.4904761904761906e-05,
+      "loss": 0.1063,
+      "step": 317
+    },
+    {
+      "epoch": 7.571428571428571,
+      "grad_norm": 0.35284796357154846,
+      "learning_rate": 4.485714285714286e-05,
+      "loss": 0.1058,
+      "step": 318
+    },
+    {
+      "epoch": 7.595238095238095,
+      "grad_norm": 0.31462305784225464,
+      "learning_rate": 4.480952380952381e-05,
+      "loss": 0.097,
+      "step": 319
+    },
+    {
+      "epoch": 7.619047619047619,
+      "grad_norm": 0.35902947187423706,
+      "learning_rate": 4.476190476190476e-05,
+      "loss": 0.0987,
+      "step": 320
+    },
+    {
+      "epoch": 7.619047619047619,
+      "eval_dice_score": 0.9146509876386677,
+      "eval_loss": 0.10323517769575119,
+      "eval_runtime": 8.7797,
+      "eval_samples_per_second": 10.137,
+      "eval_steps_per_second": 1.025,
+      "step": 320
+    },
+    {
+      "epoch": 7.642857142857143,
+      "grad_norm": 0.4236834943294525,
+      "learning_rate": 4.471428571428572e-05,
+      "loss": 0.1157,
+      "step": 321
+    },
+    {
+      "epoch": 7.666666666666667,
+      "grad_norm": 0.3109116554260254,
+      "learning_rate": 4.466666666666667e-05,
+      "loss": 0.11,
+      "step": 322
+    },
+    {
+      "epoch": 7.690476190476191,
+      "grad_norm": 0.3016926944255829,
+      "learning_rate": 4.461904761904762e-05,
+      "loss": 0.123,
+      "step": 323
+    },
+    {
+      "epoch": 7.714285714285714,
+      "grad_norm": 0.27012699842453003,
+      "learning_rate": 4.4571428571428574e-05,
+      "loss": 0.1271,
+      "step": 324
+    },
+    {
+      "epoch": 7.738095238095238,
+      "grad_norm": 0.38100895285606384,
+      "learning_rate": 4.4523809523809525e-05,
+      "loss": 0.1305,
+      "step": 325
+    },
+    {
+      "epoch": 7.761904761904762,
+      "grad_norm": 0.3512851595878601,
+      "learning_rate": 4.447619047619048e-05,
+      "loss": 0.1106,
+      "step": 326
+    },
+    {
+      "epoch": 7.785714285714286,
+      "grad_norm": 0.5073042511940002,
+      "learning_rate": 4.4428571428571435e-05,
+      "loss": 0.117,
+      "step": 327
+    },
+    {
+      "epoch": 7.809523809523809,
+      "grad_norm": 0.2925731837749481,
+      "learning_rate": 4.4380952380952386e-05,
+      "loss": 0.0982,
+      "step": 328
+    },
+    {
+      "epoch": 7.833333333333333,
+      "grad_norm": 0.2976164221763611,
+      "learning_rate": 4.433333333333334e-05,
+      "loss": 0.1039,
+      "step": 329
+    },
+    {
+      "epoch": 7.857142857142857,
+      "grad_norm": 0.35800570249557495,
+      "learning_rate": 4.428571428571429e-05,
+      "loss": 0.1358,
+      "step": 330
+    },
+    {
+      "epoch": 7.880952380952381,
+      "grad_norm": 0.35599473118782043,
+      "learning_rate": 4.423809523809524e-05,
+      "loss": 0.1201,
+      "step": 331
+    },
+    {
+      "epoch": 7.904761904761905,
+      "grad_norm": 0.35861942172050476,
+      "learning_rate": 4.419047619047619e-05,
+      "loss": 0.1029,
+      "step": 332
+    },
+    {
+      "epoch": 7.928571428571429,
+      "grad_norm": 0.29498741030693054,
+      "learning_rate": 4.4142857142857144e-05,
+      "loss": 0.0926,
+      "step": 333
+    },
+    {
+      "epoch": 7.9523809523809526,
+      "grad_norm": 0.37311461567878723,
+      "learning_rate": 4.40952380952381e-05,
+      "loss": 0.1318,
+      "step": 334
+    },
+    {
+      "epoch": 7.976190476190476,
+      "grad_norm": 0.3335290551185608,
+      "learning_rate": 4.404761904761905e-05,
+      "loss": 0.1129,
+      "step": 335
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.3926750123500824,
+      "learning_rate": 4.4e-05,
+      "loss": 0.0913,
+      "step": 336
+    },
+    {
+      "epoch": 8.023809523809524,
+      "grad_norm": 0.2891796827316284,
+      "learning_rate": 4.395238095238095e-05,
+      "loss": 0.1046,
+      "step": 337
+    },
+    {
+      "epoch": 8.047619047619047,
+      "grad_norm": 0.3221498429775238,
+      "learning_rate": 4.39047619047619e-05,
+      "loss": 0.111,
+      "step": 338
+    },
+    {
+      "epoch": 8.071428571428571,
+      "grad_norm": 0.3588779866695404,
+      "learning_rate": 4.3857142857142853e-05,
+      "loss": 0.1027,
+      "step": 339
+    },
+    {
+      "epoch": 8.095238095238095,
+      "grad_norm": 0.3896290361881256,
+      "learning_rate": 4.3809523809523805e-05,
+      "loss": 0.1011,
+      "step": 340
+    },
+    {
+      "epoch": 8.095238095238095,
+      "eval_dice_score": 0.9160615055344299,
+      "eval_loss": 0.10283860564231873,
+      "eval_runtime": 9.022,
+      "eval_samples_per_second": 9.865,
+      "eval_steps_per_second": 0.998,
+      "step": 340
+    },
+    {
+      "epoch": 8.119047619047619,
+      "grad_norm": 0.2446659505367279,
+      "learning_rate": 4.376190476190476e-05,
+      "loss": 0.0999,
+      "step": 341
+    },
+    {
+      "epoch": 8.142857142857142,
+      "grad_norm": 0.27880406379699707,
+      "learning_rate": 4.3714285714285715e-05,
+      "loss": 0.0976,
+      "step": 342
+    },
+    {
+      "epoch": 8.166666666666666,
+      "grad_norm": 0.3263701796531677,
+      "learning_rate": 4.3666666666666666e-05,
+      "loss": 0.1059,
+      "step": 343
+    },
+    {
+      "epoch": 8.19047619047619,
+      "grad_norm": 0.31584370136260986,
+      "learning_rate": 4.361904761904762e-05,
+      "loss": 0.1334,
+      "step": 344
+    },
+    {
+      "epoch": 8.214285714285714,
+      "grad_norm": 0.37839260697364807,
+      "learning_rate": 4.357142857142857e-05,
+      "loss": 0.1344,
+      "step": 345
+    },
+    {
+      "epoch": 8.238095238095237,
+      "grad_norm": 0.33257240056991577,
+      "learning_rate": 4.352380952380952e-05,
+      "loss": 0.094,
+      "step": 346
+    },
+    {
+      "epoch": 8.261904761904763,
+      "grad_norm": 0.28752750158309937,
+      "learning_rate": 4.347619047619048e-05,
+      "loss": 0.1036,
+      "step": 347
+    },
+    {
+      "epoch": 8.285714285714286,
+      "grad_norm": 0.34251728653907776,
+      "learning_rate": 4.342857142857143e-05,
+      "loss": 0.1096,
+      "step": 348
+    },
+    {
+      "epoch": 8.30952380952381,
+      "grad_norm": 0.3254129886627197,
+      "learning_rate": 4.338095238095238e-05,
+      "loss": 0.0908,
+      "step": 349
+    },
+    {
+      "epoch": 8.333333333333334,
+      "grad_norm": 0.3975524306297302,
+      "learning_rate": 4.3333333333333334e-05,
+      "loss": 0.1199,
+      "step": 350
+    },
+    {
+      "epoch": 8.357142857142858,
+      "grad_norm": 0.3707966208457947,
+      "learning_rate": 4.3285714285714286e-05,
+      "loss": 0.101,
+      "step": 351
+    },
+    {
+      "epoch": 8.380952380952381,
+      "grad_norm": 0.29389944672584534,
+      "learning_rate": 4.323809523809524e-05,
+      "loss": 0.1054,
+      "step": 352
+    },
+    {
+      "epoch": 8.404761904761905,
+      "grad_norm": 0.4392881691455841,
+      "learning_rate": 4.319047619047619e-05,
+      "loss": 0.1212,
+      "step": 353
+    },
+    {
+      "epoch": 8.428571428571429,
+      "grad_norm": 0.34391042590141296,
+      "learning_rate": 4.314285714285715e-05,
+      "loss": 0.0878,
+      "step": 354
+    },
+    {
+      "epoch": 8.452380952380953,
+      "grad_norm": 0.2981056272983551,
+      "learning_rate": 4.30952380952381e-05,
+      "loss": 0.1015,
+      "step": 355
+    },
+    {
+      "epoch": 8.476190476190476,
+      "grad_norm": 0.589784562587738,
+      "learning_rate": 4.304761904761905e-05,
+      "loss": 0.1483,
+      "step": 356
+    },
+    {
+      "epoch": 8.5,
+      "grad_norm": 0.34352627396583557,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0927,
+      "step": 357
+    },
+    {
+      "epoch": 8.523809523809524,
+      "grad_norm": 0.3901398181915283,
+      "learning_rate": 4.295238095238095e-05,
+      "loss": 0.1039,
+      "step": 358
+    },
+    {
+      "epoch": 8.547619047619047,
+      "grad_norm": 0.48700276017189026,
+      "learning_rate": 4.2904761904761905e-05,
+      "loss": 0.1376,
+      "step": 359
+    },
+    {
+      "epoch": 8.571428571428571,
+      "grad_norm": 0.40282323956489563,
+      "learning_rate": 4.2857142857142856e-05,
+      "loss": 0.118,
+      "step": 360
+    },
+    {
+      "epoch": 8.571428571428571,
+      "eval_dice_score": 0.9140197492850877,
+      "eval_loss": 0.09799595177173615,
+      "eval_runtime": 8.8085,
+      "eval_samples_per_second": 10.104,
+      "eval_steps_per_second": 1.022,
+      "step": 360
+    },
+    {
+      "epoch": 8.595238095238095,
+      "grad_norm": 0.28458479046821594,
+      "learning_rate": 4.2809523809523815e-05,
+      "loss": 0.0886,
+      "step": 361
+    },
+    {
+      "epoch": 8.619047619047619,
+      "grad_norm": 0.36778727173805237,
+      "learning_rate": 4.2761904761904766e-05,
+      "loss": 0.0932,
+      "step": 362
+    },
+    {
+      "epoch": 8.642857142857142,
+      "grad_norm": 0.3417188227176666,
+      "learning_rate": 4.271428571428572e-05,
+      "loss": 0.1171,
+      "step": 363
+    },
+    {
+      "epoch": 8.666666666666666,
+      "grad_norm": 0.32228532433509827,
+      "learning_rate": 4.266666666666667e-05,
+      "loss": 0.108,
+      "step": 364
+    },
+    {
+      "epoch": 8.69047619047619,
+      "grad_norm": 0.2850722372531891,
+      "learning_rate": 4.261904761904762e-05,
+      "loss": 0.1081,
+      "step": 365
+    },
+    {
+      "epoch": 8.714285714285714,
+      "grad_norm": 0.42572009563446045,
+      "learning_rate": 4.257142857142857e-05,
+      "loss": 0.1087,
+      "step": 366
+    },
+    {
+      "epoch": 8.738095238095237,
+      "grad_norm": 0.42448675632476807,
+      "learning_rate": 4.2523809523809524e-05,
+      "loss": 0.0925,
+      "step": 367
+    },
+    {
+      "epoch": 8.761904761904763,
+      "grad_norm": 0.3276635706424713,
+      "learning_rate": 4.247619047619048e-05,
+      "loss": 0.1016,
+      "step": 368
+    },
+    {
+      "epoch": 8.785714285714286,
+      "grad_norm": 0.3028334081172943,
+      "learning_rate": 4.2428571428571434e-05,
+      "loss": 0.1055,
+      "step": 369
+    },
+    {
+      "epoch": 8.80952380952381,
+      "grad_norm": 0.2703326642513275,
+      "learning_rate": 4.2380952380952385e-05,
+      "loss": 0.0981,
+      "step": 370
+    },
+    {
+      "epoch": 8.833333333333334,
+      "grad_norm": 0.41063007712364197,
+      "learning_rate": 4.233333333333334e-05,
+      "loss": 0.119,
+      "step": 371
+    },
+    {
+      "epoch": 8.857142857142858,
+      "grad_norm": 0.4103991985321045,
+      "learning_rate": 4.228571428571429e-05,
+      "loss": 0.1243,
+      "step": 372
+    },
+    {
+      "epoch": 8.880952380952381,
+      "grad_norm": 0.4108157753944397,
+      "learning_rate": 4.223809523809524e-05,
+      "loss": 0.1145,
+      "step": 373
+    },
+    {
+      "epoch": 8.904761904761905,
+      "grad_norm": 0.29548731446266174,
+      "learning_rate": 4.219047619047619e-05,
+      "loss": 0.095,
+      "step": 374
+    },
+    {
+      "epoch": 8.928571428571429,
+      "grad_norm": 0.35359784960746765,
+      "learning_rate": 4.214285714285714e-05,
+      "loss": 0.104,
+      "step": 375
+    },
+    {
+      "epoch": 8.952380952380953,
+      "grad_norm": 0.8022054433822632,
+      "learning_rate": 4.2095238095238095e-05,
+      "loss": 0.1198,
+      "step": 376
+    },
+    {
+      "epoch": 8.976190476190476,
+      "grad_norm": 0.3940359354019165,
+      "learning_rate": 4.2047619047619046e-05,
+      "loss": 0.1185,
+      "step": 377
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.3221697509288788,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0962,
+      "step": 378
+    },
+    {
+      "epoch": 9.023809523809524,
+      "grad_norm": 0.4048317074775696,
+      "learning_rate": 4.195238095238095e-05,
+      "loss": 0.1212,
+      "step": 379
+    },
+    {
+      "epoch": 9.047619047619047,
+      "grad_norm": 0.2531253397464752,
+      "learning_rate": 4.19047619047619e-05,
+      "loss": 0.0929,
+      "step": 380
+    },
+    {
+      "epoch": 9.047619047619047,
+      "eval_dice_score": 0.9176058599475401,
+      "eval_loss": 0.09571261703968048,
+      "eval_runtime": 8.8625,
+      "eval_samples_per_second": 10.042,
+      "eval_steps_per_second": 1.016,
+      "step": 380
+    },
+    {
+      "epoch": 9.071428571428571,
+      "grad_norm": 0.27786773443222046,
+      "learning_rate": 4.185714285714286e-05,
+      "loss": 0.1012,
+      "step": 381
+    },
+    {
+      "epoch": 9.095238095238095,
+      "grad_norm": 0.3440767526626587,
+      "learning_rate": 4.180952380952381e-05,
+      "loss": 0.0946,
+      "step": 382
+    },
+    {
+      "epoch": 9.119047619047619,
+      "grad_norm": 0.2682666778564453,
+      "learning_rate": 4.176190476190476e-05,
+      "loss": 0.0949,
+      "step": 383
+    },
+    {
+      "epoch": 9.142857142857142,
+      "grad_norm": 0.4733286499977112,
+      "learning_rate": 4.1714285714285714e-05,
+      "loss": 0.1239,
+      "step": 384
+    },
+    {
+      "epoch": 9.166666666666666,
+      "grad_norm": 0.24958311021327972,
+      "learning_rate": 4.1666666666666665e-05,
+      "loss": 0.0937,
+      "step": 385
+    },
+    {
+      "epoch": 9.19047619047619,
+      "grad_norm": 0.23216277360916138,
+      "learning_rate": 4.161904761904762e-05,
+      "loss": 0.0918,
+      "step": 386
+    },
+    {
+      "epoch": 9.214285714285714,
+      "grad_norm": 0.3856890797615051,
+      "learning_rate": 4.157142857142857e-05,
+      "loss": 0.0795,
+      "step": 387
+    },
+    {
+      "epoch": 9.238095238095237,
+      "grad_norm": 0.29770028591156006,
+      "learning_rate": 4.152380952380953e-05,
+      "loss": 0.1268,
+      "step": 388
+    },
+    {
+      "epoch": 9.261904761904763,
+      "grad_norm": 0.26541242003440857,
+      "learning_rate": 4.147619047619048e-05,
+      "loss": 0.1055,
+      "step": 389
+    },
+    {
+      "epoch": 9.285714285714286,
+      "grad_norm": 0.3441533148288727,
+      "learning_rate": 4.142857142857143e-05,
+      "loss": 0.1188,
+      "step": 390
+    },
+    {
+      "epoch": 9.30952380952381,
+      "grad_norm": 0.4041042625904083,
+      "learning_rate": 4.138095238095238e-05,
+      "loss": 0.1069,
+      "step": 391
+    },
+    {
+      "epoch": 9.333333333333334,
+      "grad_norm": 0.25647029280662537,
+      "learning_rate": 4.133333333333333e-05,
+      "loss": 0.0989,
+      "step": 392
+    },
+    {
+      "epoch": 9.357142857142858,
+      "grad_norm": 0.2511276602745056,
+      "learning_rate": 4.1285714285714284e-05,
+      "loss": 0.0903,
+      "step": 393
+    },
+    {
+      "epoch": 9.380952380952381,
+      "grad_norm": 0.26701053977012634,
+      "learning_rate": 4.1238095238095236e-05,
+      "loss": 0.0932,
+      "step": 394
+    },
+    {
+      "epoch": 9.404761904761905,
+      "grad_norm": 0.3096204102039337,
+      "learning_rate": 4.1190476190476194e-05,
+      "loss": 0.0974,
+      "step": 395
+    },
+    {
+      "epoch": 9.428571428571429,
+      "grad_norm": 0.5277116894721985,
+      "learning_rate": 4.1142857142857146e-05,
+      "loss": 0.1466,
+      "step": 396
+    },
+    {
+      "epoch": 9.452380952380953,
+      "grad_norm": 0.4356262981891632,
+      "learning_rate": 4.10952380952381e-05,
+      "loss": 0.1091,
+      "step": 397
+    },
+    {
+      "epoch": 9.476190476190476,
+      "grad_norm": 0.28454017639160156,
+      "learning_rate": 4.104761904761905e-05,
+      "loss": 0.0975,
+      "step": 398
+    },
+    {
+      "epoch": 9.5,
+      "grad_norm": 0.2634621262550354,
+      "learning_rate": 4.1e-05,
+      "loss": 0.1062,
+      "step": 399
+    },
+    {
+      "epoch": 9.523809523809524,
+      "grad_norm": 0.26061224937438965,
+      "learning_rate": 4.095238095238095e-05,
+      "loss": 0.1092,
+      "step": 400
+    },
+    {
+      "epoch": 9.523809523809524,
+      "eval_dice_score": 0.9180726876748045,
+      "eval_loss": 0.09548372775316238,
+      "eval_runtime": 8.7658,
+      "eval_samples_per_second": 10.153,
+      "eval_steps_per_second": 1.027,
+      "step": 400
+    },
+    {
+      "epoch": 9.547619047619047,
+      "grad_norm": 0.24562084674835205,
+      "learning_rate": 4.090476190476191e-05,
+      "loss": 0.1044,
+      "step": 401
+    },
+    {
+      "epoch": 9.571428571428571,
+      "grad_norm": 0.3777954876422882,
+      "learning_rate": 4.085714285714286e-05,
+      "loss": 0.1181,
+      "step": 402
+    },
+    {
+      "epoch": 9.595238095238095,
+      "grad_norm": 0.3255544602870941,
+      "learning_rate": 4.0809523809523813e-05,
+      "loss": 0.1,
+      "step": 403
+    },
+    {
+      "epoch": 9.619047619047619,
+      "grad_norm": 0.29565364122390747,
+      "learning_rate": 4.0761904761904765e-05,
+      "loss": 0.1024,
+      "step": 404
+    },
+    {
+      "epoch": 9.642857142857142,
+      "grad_norm": 0.27408862113952637,
+      "learning_rate": 4.0714285714285717e-05,
+      "loss": 0.0998,
+      "step": 405
+    },
+    {
+      "epoch": 9.666666666666666,
+      "grad_norm": 0.28693994879722595,
+      "learning_rate": 4.066666666666667e-05,
+      "loss": 0.0887,
+      "step": 406
+    },
+    {
+      "epoch": 9.69047619047619,
+      "grad_norm": 0.2059289962053299,
+      "learning_rate": 4.061904761904762e-05,
+      "loss": 0.1142,
+      "step": 407
+    },
+    {
+      "epoch": 9.714285714285714,
+      "grad_norm": 0.2377048283815384,
+      "learning_rate": 4.057142857142858e-05,
+      "loss": 0.0896,
+      "step": 408
+    },
+    {
+      "epoch": 9.738095238095237,
+      "grad_norm": 0.5583423376083374,
+      "learning_rate": 4.052380952380953e-05,
+      "loss": 0.1117,
+      "step": 409
+    },
+    {
+      "epoch": 9.761904761904763,
+      "grad_norm": 0.2231013923883438,
+      "learning_rate": 4.047619047619048e-05,
+      "loss": 0.0923,
+      "step": 410
+    },
+    {
+      "epoch": 9.785714285714286,
+      "grad_norm": 0.37593528628349304,
+      "learning_rate": 4.042857142857143e-05,
+      "loss": 0.1224,
+      "step": 411
+    },
+    {
+      "epoch": 9.80952380952381,
+      "grad_norm": 0.2936907410621643,
+      "learning_rate": 4.0380952380952384e-05,
+      "loss": 0.0954,
+      "step": 412
+    },
+    {
+      "epoch": 9.833333333333334,
+      "grad_norm": 0.2800084054470062,
+      "learning_rate": 4.0333333333333336e-05,
+      "loss": 0.1021,
+      "step": 413
+    },
+    {
+      "epoch": 9.857142857142858,
+      "grad_norm": 0.2564726173877716,
+      "learning_rate": 4.028571428571428e-05,
+      "loss": 0.1039,
+      "step": 414
+    },
+    {
+      "epoch": 9.880952380952381,
+      "grad_norm": 0.3026561439037323,
+      "learning_rate": 4.023809523809524e-05,
+      "loss": 0.0859,
+      "step": 415
+    },
+    {
+      "epoch": 9.904761904761905,
+      "grad_norm": 0.3142635226249695,
+      "learning_rate": 4.019047619047619e-05,
+      "loss": 0.0934,
+      "step": 416
+    },
+    {
+      "epoch": 9.928571428571429,
+      "grad_norm": 0.3444440960884094,
+      "learning_rate": 4.014285714285714e-05,
+      "loss": 0.0981,
+      "step": 417
+    },
+    {
+      "epoch": 9.952380952380953,
+      "grad_norm": 0.3670935332775116,
+      "learning_rate": 4.0095238095238093e-05,
+      "loss": 0.1012,
+      "step": 418
+    },
+    {
+      "epoch": 9.976190476190476,
+      "grad_norm": 0.508635938167572,
+      "learning_rate": 4.0047619047619045e-05,
+      "loss": 0.096,
+      "step": 419
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.5292103886604309,
+      "learning_rate": 3.9999999999999996e-05,
+      "loss": 0.1119,
+      "step": 420
+    },
+    {
+      "epoch": 10.0,
+      "eval_dice_score": 0.918242837228066,
+      "eval_loss": 0.09436747431755066,
+      "eval_runtime": 8.8686,
+      "eval_samples_per_second": 10.035,
+      "eval_steps_per_second": 1.015,
+      "step": 420
+    },
+    {
+      "epoch": 10.023809523809524,
+      "grad_norm": 0.24404850602149963,
+      "learning_rate": 3.995238095238095e-05,
+      "loss": 0.0966,
+      "step": 421
+    },
+    {
+      "epoch": 10.047619047619047,
+      "grad_norm": 0.21881720423698425,
+      "learning_rate": 3.9904761904761906e-05,
+      "loss": 0.0868,
+      "step": 422
+    },
+    {
+      "epoch": 10.071428571428571,
+      "grad_norm": 0.2544311583042145,
+      "learning_rate": 3.985714285714286e-05,
+      "loss": 0.0981,
+      "step": 423
+    },
+    {
+      "epoch": 10.095238095238095,
+      "grad_norm": 0.3842509686946869,
+      "learning_rate": 3.980952380952381e-05,
+      "loss": 0.102,
+      "step": 424
+    },
+    {
+      "epoch": 10.119047619047619,
+      "grad_norm": 0.2687145471572876,
+      "learning_rate": 3.976190476190476e-05,
+      "loss": 0.0896,
+      "step": 425
+    },
+    {
+      "epoch": 10.142857142857142,
+      "grad_norm": 0.23764175176620483,
+      "learning_rate": 3.971428571428571e-05,
+      "loss": 0.1104,
+      "step": 426
+    },
+    {
+      "epoch": 10.166666666666666,
+      "grad_norm": 0.36458808183670044,
+      "learning_rate": 3.9666666666666664e-05,
+      "loss": 0.1026,
+      "step": 427
+    },
+    {
+      "epoch": 10.19047619047619,
+      "grad_norm": 0.29205796122550964,
+      "learning_rate": 3.961904761904762e-05,
+      "loss": 0.0931,
+      "step": 428
+    },
+    {
+      "epoch": 10.214285714285714,
+      "grad_norm": 0.30714672803878784,
+      "learning_rate": 3.9571428571428574e-05,
+      "loss": 0.0899,
+      "step": 429
+    },
+    {
+      "epoch": 10.238095238095237,
+      "grad_norm": 0.4927786588668823,
+      "learning_rate": 3.9523809523809526e-05,
+      "loss": 0.1027,
+      "step": 430
+    },
+    {
+      "epoch": 10.261904761904763,
+      "grad_norm": 0.2786138951778412,
+      "learning_rate": 3.947619047619048e-05,
+      "loss": 0.1026,
+      "step": 431
+    },
+    {
+      "epoch": 10.285714285714286,
+      "grad_norm": 0.318553626537323,
+      "learning_rate": 3.942857142857143e-05,
+      "loss": 0.1101,
+      "step": 432
+    },
+    {
+      "epoch": 10.30952380952381,
+      "grad_norm": 0.3124142289161682,
+      "learning_rate": 3.938095238095238e-05,
+      "loss": 0.1258,
+      "step": 433
+    },
+    {
+      "epoch": 10.333333333333334,
+      "grad_norm": 0.3818247318267822,
+      "learning_rate": 3.933333333333333e-05,
+      "loss": 0.0995,
+      "step": 434
+    },
+    {
+      "epoch": 10.357142857142858,
+      "grad_norm": 0.281456857919693,
+      "learning_rate": 3.928571428571429e-05,
+      "loss": 0.0874,
+      "step": 435
+    },
+    {
+      "epoch": 10.380952380952381,
+      "grad_norm": 0.3376539349555969,
+      "learning_rate": 3.923809523809524e-05,
+      "loss": 0.0799,
+      "step": 436
+    },
+    {
+      "epoch": 10.404761904761905,
+      "grad_norm": 0.2396553009748459,
+      "learning_rate": 3.919047619047619e-05,
+      "loss": 0.1047,
+      "step": 437
+    },
+    {
+      "epoch": 10.428571428571429,
+      "grad_norm": 0.2931698262691498,
+      "learning_rate": 3.9142857142857145e-05,
+      "loss": 0.0972,
+      "step": 438
+    },
+    {
+      "epoch": 10.452380952380953,
+      "grad_norm": 0.30064502358436584,
+      "learning_rate": 3.9095238095238096e-05,
+      "loss": 0.0975,
+      "step": 439
+    },
+    {
+      "epoch": 10.476190476190476,
+      "grad_norm": 0.4932076036930084,
+      "learning_rate": 3.904761904761905e-05,
+      "loss": 0.095,
+      "step": 440
+    },
+    {
+      "epoch": 10.476190476190476,
+      "eval_dice_score": 0.9177181678377411,
+      "eval_loss": 0.0928926020860672,
+      "eval_runtime": 8.9369,
+      "eval_samples_per_second": 9.959,
+      "eval_steps_per_second": 1.007,
+      "step": 440
+    },
+    {
+      "epoch": 10.5,
+      "grad_norm": 0.26641082763671875,
+      "learning_rate": 3.9e-05,
+      "loss": 0.0955,
+      "step": 441
+    },
+    {
+      "epoch": 10.523809523809524,
+      "grad_norm": 0.32811999320983887,
+      "learning_rate": 3.895238095238096e-05,
+      "loss": 0.1066,
+      "step": 442
+    },
+    {
+      "epoch": 10.547619047619047,
+      "grad_norm": 0.46325215697288513,
+      "learning_rate": 3.890476190476191e-05,
+      "loss": 0.1004,
+      "step": 443
+    },
+    {
+      "epoch": 10.571428571428571,
+      "grad_norm": 0.28526756167411804,
+      "learning_rate": 3.885714285714286e-05,
+      "loss": 0.0977,
+      "step": 444
+    },
+    {
+      "epoch": 10.595238095238095,
+      "grad_norm": 0.27527350187301636,
+      "learning_rate": 3.880952380952381e-05,
+      "loss": 0.0879,
+      "step": 445
+    },
+    {
+      "epoch": 10.619047619047619,
+      "grad_norm": 0.36892035603523254,
+      "learning_rate": 3.8761904761904764e-05,
+      "loss": 0.1032,
+      "step": 446
+    },
+    {
+      "epoch": 10.642857142857142,
+      "grad_norm": 0.3519724905490875,
+      "learning_rate": 3.8714285714285715e-05,
+      "loss": 0.1044,
+      "step": 447
+    },
+    {
+      "epoch": 10.666666666666666,
+      "grad_norm": 1.0675806999206543,
+      "learning_rate": 3.866666666666667e-05,
+      "loss": 0.1241,
+      "step": 448
+    },
+    {
+      "epoch": 10.69047619047619,
+      "grad_norm": 0.3300955891609192,
+      "learning_rate": 3.8619047619047625e-05,
+      "loss": 0.1002,
+      "step": 449
+    },
+    {
+      "epoch": 10.714285714285714,
+      "grad_norm": 0.3200720250606537,
+      "learning_rate": 3.857142857142858e-05,
+      "loss": 0.092,
+      "step": 450
+    },
+    {
+      "epoch": 10.738095238095237,
+      "grad_norm": 0.32000890374183655,
+      "learning_rate": 3.852380952380953e-05,
+      "loss": 0.0894,
+      "step": 451
+    },
+    {
+      "epoch": 10.761904761904763,
+      "grad_norm": 0.5081201195716858,
+      "learning_rate": 3.847619047619048e-05,
+      "loss": 0.116,
+      "step": 452
+    },
+    {
+      "epoch": 10.785714285714286,
+      "grad_norm": 0.4235091507434845,
+      "learning_rate": 3.8428571428571425e-05,
+      "loss": 0.1047,
+      "step": 453
+    },
+    {
+      "epoch": 10.80952380952381,
+      "grad_norm": 0.24375148117542267,
+      "learning_rate": 3.8380952380952376e-05,
+      "loss": 0.0878,
+      "step": 454
+    },
+    {
+      "epoch": 10.833333333333334,
+      "grad_norm": 0.32877597212791443,
+      "learning_rate": 3.8333333333333334e-05,
+      "loss": 0.0963,
+      "step": 455
+    },
+    {
+      "epoch": 10.857142857142858,
+      "grad_norm": 0.31234005093574524,
+      "learning_rate": 3.8285714285714286e-05,
+      "loss": 0.1155,
+      "step": 456
+    },
+    {
+      "epoch": 10.880952380952381,
+      "grad_norm": 0.2086906135082245,
+      "learning_rate": 3.823809523809524e-05,
+      "loss": 0.0933,
+      "step": 457
+    },
+    {
+      "epoch": 10.904761904761905,
+      "grad_norm": 0.22270095348358154,
+      "learning_rate": 3.819047619047619e-05,
+      "loss": 0.0858,
+      "step": 458
+    },
+    {
+      "epoch": 10.928571428571429,
+      "grad_norm": 0.2681645154953003,
+      "learning_rate": 3.814285714285714e-05,
+      "loss": 0.1056,
+      "step": 459
+    },
+    {
+      "epoch": 10.952380952380953,
+      "grad_norm": 0.5809670686721802,
+      "learning_rate": 3.809523809523809e-05,
+      "loss": 0.1205,
+      "step": 460
+    },
+    {
+      "epoch": 10.952380952380953,
+      "eval_dice_score": 0.9194055696314088,
+      "eval_loss": 0.09057500213384628,
+      "eval_runtime": 8.8575,
+      "eval_samples_per_second": 10.048,
+      "eval_steps_per_second": 1.016,
+      "step": 460
+    },
+    {
+      "epoch": 10.976190476190476,
+      "grad_norm": 0.31106501817703247,
+      "learning_rate": 3.8047619047619044e-05,
+      "loss": 0.1124,
+      "step": 461
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.23329056799411774,
+      "learning_rate": 3.8e-05,
+      "loss": 0.1003,
+      "step": 462
+    },
+    {
+      "epoch": 11.023809523809524,
+      "grad_norm": 0.3269498944282532,
+      "learning_rate": 3.7952380952380954e-05,
+      "loss": 0.1102,
+      "step": 463
+    },
+    {
+      "epoch": 11.047619047619047,
+      "grad_norm": 0.41568589210510254,
+      "learning_rate": 3.7904761904761905e-05,
+      "loss": 0.0839,
+      "step": 464
+    },
+    {
+      "epoch": 11.071428571428571,
+      "grad_norm": 0.34682050347328186,
+      "learning_rate": 3.785714285714286e-05,
+      "loss": 0.0886,
+      "step": 465
+    },
+    {
+      "epoch": 11.095238095238095,
+      "grad_norm": 0.4217880070209503,
+      "learning_rate": 3.780952380952381e-05,
+      "loss": 0.0871,
+      "step": 466
+    },
+    {
+      "epoch": 11.119047619047619,
+      "grad_norm": 0.2800375819206238,
+      "learning_rate": 3.776190476190476e-05,
+      "loss": 0.1045,
+      "step": 467
+    },
+    {
+      "epoch": 11.142857142857142,
+      "grad_norm": 0.25883740186691284,
+      "learning_rate": 3.771428571428571e-05,
+      "loss": 0.0879,
+      "step": 468
+    },
+    {
+      "epoch": 11.166666666666666,
+      "grad_norm": 0.26584306359291077,
+      "learning_rate": 3.766666666666667e-05,
+      "loss": 0.1041,
+      "step": 469
+    },
+    {
+      "epoch": 11.19047619047619,
+      "grad_norm": 0.3690215051174164,
+      "learning_rate": 3.761904761904762e-05,
+      "loss": 0.1049,
+      "step": 470
+    },
+    {
+      "epoch": 11.214285714285714,
+      "grad_norm": 0.23434555530548096,
+      "learning_rate": 3.757142857142857e-05,
+      "loss": 0.0929,
+      "step": 471
+    },
+    {
+      "epoch": 11.238095238095237,
+      "grad_norm": 0.23841549456119537,
+      "learning_rate": 3.7523809523809524e-05,
+      "loss": 0.0853,
+      "step": 472
+    },
+    {
+      "epoch": 11.261904761904763,
+      "grad_norm": 0.21769563853740692,
+      "learning_rate": 3.7476190476190476e-05,
+      "loss": 0.0873,
+      "step": 473
+    },
+    {
+      "epoch": 11.285714285714286,
+      "grad_norm": 0.2575165033340454,
+      "learning_rate": 3.742857142857143e-05,
+      "loss": 0.0969,
+      "step": 474
+    },
+    {
+      "epoch": 11.30952380952381,
+      "grad_norm": 0.7376023530960083,
+      "learning_rate": 3.738095238095238e-05,
+      "loss": 0.107,
+      "step": 475
+    },
+    {
+      "epoch": 11.333333333333334,
+      "grad_norm": 0.6353978514671326,
+      "learning_rate": 3.733333333333334e-05,
+      "loss": 0.117,
+      "step": 476
+    },
+    {
+      "epoch": 11.357142857142858,
+      "grad_norm": 0.3576112985610962,
+      "learning_rate": 3.728571428571429e-05,
+      "loss": 0.0776,
+      "step": 477
+    },
+    {
+      "epoch": 11.380952380952381,
+      "grad_norm": 0.3014508783817291,
+      "learning_rate": 3.723809523809524e-05,
+      "loss": 0.1066,
+      "step": 478
+    },
+    {
+      "epoch": 11.404761904761905,
+      "grad_norm": 0.3285028040409088,
+      "learning_rate": 3.719047619047619e-05,
+      "loss": 0.1115,
+      "step": 479
+    },
+    {
+      "epoch": 11.428571428571429,
+      "grad_norm": 0.4150606393814087,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1011,
+      "step": 480
+    },
+    {
+      "epoch": 11.428571428571429,
+      "eval_dice_score": 0.9186469055451411,
+      "eval_loss": 0.09131216257810593,
+      "eval_runtime": 8.8331,
+      "eval_samples_per_second": 10.076,
+      "eval_steps_per_second": 1.019,
+      "step": 480
+    },
+    {
+      "epoch": 11.452380952380953,
+      "grad_norm": 0.3601835072040558,
+      "learning_rate": 3.7095238095238095e-05,
+      "loss": 0.1069,
+      "step": 481
+    },
+    {
+      "epoch": 11.476190476190476,
+      "grad_norm": 0.30209657549858093,
+      "learning_rate": 3.704761904761905e-05,
+      "loss": 0.1357,
+      "step": 482
+    },
+    {
+      "epoch": 11.5,
+      "grad_norm": 0.3924671709537506,
+      "learning_rate": 3.7000000000000005e-05,
+      "loss": 0.0962,
+      "step": 483
+    },
+    {
+      "epoch": 11.523809523809524,
+      "grad_norm": 0.28674888610839844,
+      "learning_rate": 3.6952380952380956e-05,
+      "loss": 0.0863,
+      "step": 484
+    },
+    {
+      "epoch": 11.547619047619047,
+      "grad_norm": 0.25611767172813416,
+      "learning_rate": 3.690476190476191e-05,
+      "loss": 0.1034,
+      "step": 485
+    },
+    {
+      "epoch": 11.571428571428571,
+      "grad_norm": 0.41438841819763184,
+      "learning_rate": 3.685714285714286e-05,
+      "loss": 0.0873,
+      "step": 486
+    },
+    {
+      "epoch": 11.595238095238095,
+      "grad_norm": 0.5831720232963562,
+      "learning_rate": 3.680952380952381e-05,
+      "loss": 0.0965,
+      "step": 487
+    },
+    {
+      "epoch": 11.619047619047619,
+      "grad_norm": 0.2311333268880844,
+      "learning_rate": 3.676190476190476e-05,
+      "loss": 0.0888,
+      "step": 488
+    },
+    {
+      "epoch": 11.642857142857142,
+      "grad_norm": 0.2140185832977295,
+      "learning_rate": 3.671428571428572e-05,
+      "loss": 0.1114,
+      "step": 489
+    },
+    {
+      "epoch": 11.666666666666666,
+      "grad_norm": 0.2992950975894928,
+      "learning_rate": 3.666666666666667e-05,
+      "loss": 0.0936,
+      "step": 490
+    },
+    {
+      "epoch": 11.69047619047619,
+      "grad_norm": 0.4659542143344879,
+      "learning_rate": 3.6619047619047624e-05,
+      "loss": 0.1018,
+      "step": 491
+    },
+    {
+      "epoch": 11.714285714285714,
+      "grad_norm": 0.32879874110221863,
+      "learning_rate": 3.6571428571428576e-05,
+      "loss": 0.0946,
+      "step": 492
+    },
+    {
+      "epoch": 11.738095238095237,
+      "grad_norm": 0.3425431251525879,
+      "learning_rate": 3.652380952380952e-05,
+      "loss": 0.0885,
+      "step": 493
+    },
+    {
+      "epoch": 11.761904761904763,
+      "grad_norm": 0.24904826283454895,
+      "learning_rate": 3.647619047619047e-05,
+      "loss": 0.0913,
+      "step": 494
+    },
+    {
+      "epoch": 11.785714285714286,
+      "grad_norm": 0.26887717843055725,
+      "learning_rate": 3.6428571428571423e-05,
+      "loss": 0.0827,
+      "step": 495
+    },
+    {
+      "epoch": 11.80952380952381,
+      "grad_norm": 0.2696100175380707,
+      "learning_rate": 3.638095238095238e-05,
+      "loss": 0.0879,
+      "step": 496
+    },
+    {
+      "epoch": 11.833333333333334,
+      "grad_norm": 0.28169044852256775,
+      "learning_rate": 3.633333333333333e-05,
+      "loss": 0.1033,
+      "step": 497
+    },
+    {
+      "epoch": 11.857142857142858,
+      "grad_norm": 0.35210949182510376,
+      "learning_rate": 3.6285714285714285e-05,
+      "loss": 0.0972,
+      "step": 498
+    },
+    {
+      "epoch": 11.880952380952381,
+      "grad_norm": 0.2619428336620331,
+      "learning_rate": 3.6238095238095236e-05,
+      "loss": 0.1012,
+      "step": 499
+    },
+    {
+      "epoch": 11.904761904761905,
+      "grad_norm": 0.2678943872451782,
+      "learning_rate": 3.619047619047619e-05,
+      "loss": 0.0886,
+      "step": 500
+    },
+    {
+      "epoch": 11.904761904761905,
+      "eval_dice_score": 0.9205903160321725,
+      "eval_loss": 0.09119950979948044,
+      "eval_runtime": 8.8433,
+      "eval_samples_per_second": 10.064,
+      "eval_steps_per_second": 1.018,
+      "step": 500
+    },
+    {
+      "epoch": 11.928571428571429,
+      "grad_norm": 0.26405709981918335,
+      "learning_rate": 3.614285714285714e-05,
+      "loss": 0.1017,
+      "step": 501
+    },
+    {
+      "epoch": 11.952380952380953,
+      "grad_norm": 0.8981893062591553,
+      "learning_rate": 3.60952380952381e-05,
+      "loss": 0.1172,
+      "step": 502
+    },
+    {
+      "epoch": 11.976190476190476,
+      "grad_norm": 0.504848837852478,
+      "learning_rate": 3.604761904761905e-05,
+      "loss": 0.0996,
+      "step": 503
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 0.2666759490966797,
+      "learning_rate": 3.6e-05,
+      "loss": 0.0847,
+      "step": 504
+    },
+    {
+      "epoch": 12.023809523809524,
+      "grad_norm": 0.2892642021179199,
+      "learning_rate": 3.595238095238095e-05,
+      "loss": 0.085,
+      "step": 505
+    },
+    {
+      "epoch": 12.047619047619047,
+      "grad_norm": 0.5016778707504272,
+      "learning_rate": 3.5904761904761904e-05,
+      "loss": 0.0999,
+      "step": 506
+    },
+    {
+      "epoch": 12.071428571428571,
+      "grad_norm": 0.302896648645401,
+      "learning_rate": 3.5857142857142856e-05,
+      "loss": 0.1005,
+      "step": 507
+    },
+    {
+      "epoch": 12.095238095238095,
+      "grad_norm": 1.1246132850646973,
+      "learning_rate": 3.580952380952381e-05,
+      "loss": 0.1163,
+      "step": 508
+    },
+    {
+      "epoch": 12.119047619047619,
+      "grad_norm": 0.33260878920555115,
+      "learning_rate": 3.5761904761904765e-05,
+      "loss": 0.1062,
+      "step": 509
+    },
+    {
+      "epoch": 12.142857142857142,
+      "grad_norm": 0.2548132538795471,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.1041,
+      "step": 510
+    },
+    {
+      "epoch": 12.166666666666666,
+      "grad_norm": 0.3263733983039856,
+      "learning_rate": 3.566666666666667e-05,
+      "loss": 0.076,
+      "step": 511
+    },
+    {
+      "epoch": 12.19047619047619,
+      "grad_norm": 0.4536820948123932,
+      "learning_rate": 3.561904761904762e-05,
+      "loss": 0.1004,
+      "step": 512
+    },
+    {
+      "epoch": 12.214285714285714,
+      "grad_norm": 0.3253687024116516,
+      "learning_rate": 3.557142857142857e-05,
+      "loss": 0.0959,
+      "step": 513
+    },
+    {
+      "epoch": 12.238095238095237,
+      "grad_norm": 0.4816119968891144,
+      "learning_rate": 3.552380952380952e-05,
+      "loss": 0.1107,
+      "step": 514
+    },
+    {
+      "epoch": 12.261904761904763,
+      "grad_norm": 0.2999951243400574,
+      "learning_rate": 3.5476190476190475e-05,
+      "loss": 0.1103,
+      "step": 515
+    },
+    {
+      "epoch": 12.285714285714286,
+      "grad_norm": 0.2874181866645813,
+      "learning_rate": 3.542857142857143e-05,
+      "loss": 0.0994,
+      "step": 516
+    },
+    {
+      "epoch": 12.30952380952381,
+      "grad_norm": 0.22572314739227295,
+      "learning_rate": 3.5380952380952385e-05,
+      "loss": 0.1032,
+      "step": 517
+    },
+    {
+      "epoch": 12.333333333333334,
+      "grad_norm": 0.24886523187160492,
+      "learning_rate": 3.5333333333333336e-05,
+      "loss": 0.0892,
+      "step": 518
+    },
+    {
+      "epoch": 12.357142857142858,
+      "grad_norm": 0.34990599751472473,
+      "learning_rate": 3.528571428571429e-05,
+      "loss": 0.0866,
+      "step": 519
+    },
+    {
+      "epoch": 12.380952380952381,
+      "grad_norm": 0.4522719383239746,
+      "learning_rate": 3.523809523809524e-05,
+      "loss": 0.1032,
+      "step": 520
+    },
+    {
+      "epoch": 12.380952380952381,
+      "eval_dice_score": 0.9187127339740486,
+      "eval_loss": 0.0883309617638588,
+      "eval_runtime": 8.8468,
+      "eval_samples_per_second": 10.06,
+      "eval_steps_per_second": 1.017,
+      "step": 520
+    },
+    {
+      "epoch": 12.404761904761905,
+      "grad_norm": 0.4739639461040497,
+      "learning_rate": 3.519047619047619e-05,
+      "loss": 0.0934,
+      "step": 521
+    },
+    {
+      "epoch": 12.428571428571429,
+      "grad_norm": 0.3023685812950134,
+      "learning_rate": 3.514285714285714e-05,
+      "loss": 0.0798,
+      "step": 522
+    },
+    {
+      "epoch": 12.452380952380953,
+      "grad_norm": 0.49961069226264954,
+      "learning_rate": 3.50952380952381e-05,
+      "loss": 0.1129,
+      "step": 523
+    },
+    {
+      "epoch": 12.476190476190476,
+      "grad_norm": 0.2742755711078644,
+      "learning_rate": 3.504761904761905e-05,
+      "loss": 0.0854,
+      "step": 524
+    },
+    {
+      "epoch": 12.5,
+      "grad_norm": 0.3225800096988678,
+      "learning_rate": 3.5000000000000004e-05,
+      "loss": 0.1067,
+      "step": 525
+    },
+    {
+      "epoch": 12.523809523809524,
+      "grad_norm": 0.18689511716365814,
+      "learning_rate": 3.4952380952380955e-05,
+      "loss": 0.0837,
+      "step": 526
+    },
+    {
+      "epoch": 12.547619047619047,
+      "grad_norm": 0.4479936957359314,
+      "learning_rate": 3.490476190476191e-05,
+      "loss": 0.1043,
+      "step": 527
+    },
+    {
+      "epoch": 12.571428571428571,
+      "grad_norm": 0.229170560836792,
+      "learning_rate": 3.485714285714286e-05,
+      "loss": 0.0899,
+      "step": 528
+    },
+    {
+      "epoch": 12.595238095238095,
+      "grad_norm": 0.28260865807533264,
+      "learning_rate": 3.480952380952382e-05,
+      "loss": 0.0812,
+      "step": 529
+    },
+    {
+      "epoch": 12.619047619047619,
+      "grad_norm": 0.22655904293060303,
+      "learning_rate": 3.476190476190477e-05,
+      "loss": 0.0781,
+      "step": 530
+    },
+    {
+      "epoch": 12.642857142857142,
+      "grad_norm": 0.23291556537151337,
+      "learning_rate": 3.471428571428572e-05,
+      "loss": 0.0821,
+      "step": 531
+    },
+    {
+      "epoch": 12.666666666666666,
+      "grad_norm": 0.3438037931919098,
+      "learning_rate": 3.4666666666666665e-05,
+      "loss": 0.1208,
+      "step": 532
+    },
+    {
+      "epoch": 12.69047619047619,
+      "grad_norm": 0.4007132947444916,
+      "learning_rate": 3.4619047619047616e-05,
+      "loss": 0.1209,
+      "step": 533
+    },
+    {
+      "epoch": 12.714285714285714,
+      "grad_norm": 0.22857758402824402,
+      "learning_rate": 3.457142857142857e-05,
+      "loss": 0.0906,
+      "step": 534
+    },
+    {
+      "epoch": 12.738095238095237,
+      "grad_norm": 0.33069509267807007,
+      "learning_rate": 3.452380952380952e-05,
+      "loss": 0.0862,
+      "step": 535
+    },
+    {
+      "epoch": 12.761904761904763,
+      "grad_norm": 0.2187006175518036,
+      "learning_rate": 3.447619047619048e-05,
+      "loss": 0.0881,
+      "step": 536
+    },
+    {
+      "epoch": 12.785714285714286,
+      "grad_norm": 0.3787616193294525,
+      "learning_rate": 3.442857142857143e-05,
+      "loss": 0.1055,
+      "step": 537
+    },
+    {
+      "epoch": 12.80952380952381,
+      "grad_norm": 0.3690526485443115,
+      "learning_rate": 3.438095238095238e-05,
+      "loss": 0.0939,
+      "step": 538
+    },
+    {
+      "epoch": 12.833333333333334,
+      "grad_norm": 0.2734881341457367,
+      "learning_rate": 3.433333333333333e-05,
+      "loss": 0.0799,
+      "step": 539
+    },
+    {
+      "epoch": 12.857142857142858,
+      "grad_norm": 0.32879072427749634,
+      "learning_rate": 3.4285714285714284e-05,
+      "loss": 0.094,
+      "step": 540
+    },
+    {
+      "epoch": 12.857142857142858,
+      "eval_dice_score": 0.9214406627422178,
+      "eval_loss": 0.08812714368104935,
+      "eval_runtime": 8.9703,
+      "eval_samples_per_second": 9.922,
+      "eval_steps_per_second": 1.003,
+      "step": 540
+    },
+    {
+      "epoch": 12.880952380952381,
+      "grad_norm": 0.2516557276248932,
+      "learning_rate": 3.4238095238095235e-05,
+      "loss": 0.1029,
+      "step": 541
+    },
+    {
+      "epoch": 12.904761904761905,
+      "grad_norm": 0.28719547390937805,
+      "learning_rate": 3.419047619047619e-05,
+      "loss": 0.0961,
+      "step": 542
+    },
+    {
+      "epoch": 12.928571428571429,
+      "grad_norm": 0.2147558480501175,
+      "learning_rate": 3.4142857142857145e-05,
+      "loss": 0.0948,
+      "step": 543
+    },
+    {
+      "epoch": 12.952380952380953,
+      "grad_norm": 0.36119335889816284,
+      "learning_rate": 3.40952380952381e-05,
+      "loss": 0.0835,
+      "step": 544
+    },
+    {
+      "epoch": 12.976190476190476,
+      "grad_norm": 0.4117376208305359,
+      "learning_rate": 3.404761904761905e-05,
+      "loss": 0.0906,
+      "step": 545
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 0.27086928486824036,
+      "learning_rate": 3.4e-05,
+      "loss": 0.0933,
+      "step": 546
+    },
+    {
+      "epoch": 13.023809523809524,
+      "grad_norm": 0.17893396317958832,
+      "learning_rate": 3.395238095238095e-05,
+      "loss": 0.088,
+      "step": 547
+    },
+    {
+      "epoch": 13.047619047619047,
+      "grad_norm": 0.22478432953357697,
+      "learning_rate": 3.39047619047619e-05,
+      "loss": 0.0761,
+      "step": 548
+    },
+    {
+      "epoch": 13.071428571428571,
+      "grad_norm": 0.6636024713516235,
+      "learning_rate": 3.3857142857142854e-05,
+      "loss": 0.0998,
+      "step": 549
+    },
+    {
+      "epoch": 13.095238095238095,
+      "grad_norm": 0.2301420271396637,
+      "learning_rate": 3.380952380952381e-05,
+      "loss": 0.0837,
+      "step": 550
+    },
+    {
+      "epoch": 13.119047619047619,
+      "grad_norm": 0.32941392064094543,
+      "learning_rate": 3.3761904761904764e-05,
+      "loss": 0.1032,
+      "step": 551
+    },
+    {
+      "epoch": 13.142857142857142,
+      "grad_norm": 0.47822511196136475,
+      "learning_rate": 3.3714285714285716e-05,
+      "loss": 0.1213,
+      "step": 552
+    },
+    {
+      "epoch": 13.166666666666666,
+      "grad_norm": 0.38559892773628235,
+      "learning_rate": 3.366666666666667e-05,
+      "loss": 0.0967,
+      "step": 553
+    },
+    {
+      "epoch": 13.19047619047619,
+      "grad_norm": 0.3795962333679199,
+      "learning_rate": 3.361904761904762e-05,
+      "loss": 0.0848,
+      "step": 554
+    },
+    {
+      "epoch": 13.214285714285714,
+      "grad_norm": 0.33221033215522766,
+      "learning_rate": 3.357142857142857e-05,
+      "loss": 0.0829,
+      "step": 555
+    },
+    {
+      "epoch": 13.238095238095237,
+      "grad_norm": 0.2630172371864319,
+      "learning_rate": 3.352380952380953e-05,
+      "loss": 0.1104,
+      "step": 556
+    },
+    {
+      "epoch": 13.261904761904763,
+      "grad_norm": 0.2620716989040375,
+      "learning_rate": 3.347619047619048e-05,
+      "loss": 0.0883,
+      "step": 557
+    },
+    {
+      "epoch": 13.285714285714286,
+      "grad_norm": 0.4213337004184723,
+      "learning_rate": 3.342857142857143e-05,
+      "loss": 0.0923,
+      "step": 558
+    },
+    {
+      "epoch": 13.30952380952381,
+      "grad_norm": 0.17743177711963654,
+      "learning_rate": 3.338095238095238e-05,
+      "loss": 0.0811,
+      "step": 559
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 0.27336469292640686,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0801,
+      "step": 560
+    },
+    {
+      "epoch": 13.333333333333334,
+      "eval_dice_score": 0.921020052010366,
+      "eval_loss": 0.08740828186273575,
+      "eval_runtime": 8.8356,
+      "eval_samples_per_second": 10.073,
+      "eval_steps_per_second": 1.019,
+      "step": 560
+    },
+    {
+      "epoch": 13.357142857142858,
+      "grad_norm": 0.1683669090270996,
+      "learning_rate": 3.3285714285714286e-05,
+      "loss": 0.0831,
+      "step": 561
+    },
+    {
+      "epoch": 13.380952380952381,
+      "grad_norm": 0.2313413918018341,
+      "learning_rate": 3.323809523809524e-05,
+      "loss": 0.0945,
+      "step": 562
+    },
+    {
+      "epoch": 13.404761904761905,
+      "grad_norm": 0.2842002511024475,
+      "learning_rate": 3.3190476190476196e-05,
+      "loss": 0.1027,
+      "step": 563
+    },
+    {
+      "epoch": 13.428571428571429,
+      "grad_norm": 0.39948076009750366,
+      "learning_rate": 3.314285714285715e-05,
+      "loss": 0.1064,
+      "step": 564
+    },
+    {
+      "epoch": 13.452380952380953,
+      "grad_norm": 0.2080223709344864,
+      "learning_rate": 3.30952380952381e-05,
+      "loss": 0.0781,
+      "step": 565
+    },
+    {
+      "epoch": 13.476190476190476,
+      "grad_norm": 0.39122873544692993,
+      "learning_rate": 3.304761904761905e-05,
+      "loss": 0.1042,
+      "step": 566
+    },
+    {
+      "epoch": 13.5,
+      "grad_norm": 0.1916409730911255,
+      "learning_rate": 3.3e-05,
+      "loss": 0.087,
+      "step": 567
+    },
+    {
+      "epoch": 13.523809523809524,
+      "grad_norm": 0.21745239198207855,
+      "learning_rate": 3.2952380952380954e-05,
+      "loss": 0.0863,
+      "step": 568
+    },
+    {
+      "epoch": 13.547619047619047,
+      "grad_norm": 0.2460186928510666,
+      "learning_rate": 3.2904761904761906e-05,
+      "loss": 0.1136,
+      "step": 569
+    },
+    {
+      "epoch": 13.571428571428571,
+      "grad_norm": 0.2653687596321106,
+      "learning_rate": 3.2857142857142864e-05,
+      "loss": 0.0737,
+      "step": 570
+    },
+    {
+      "epoch": 13.595238095238095,
+      "grad_norm": 0.261576384305954,
+      "learning_rate": 3.280952380952381e-05,
+      "loss": 0.0931,
+      "step": 571
+    },
+    {
+      "epoch": 13.619047619047619,
+      "grad_norm": 0.2636834383010864,
+      "learning_rate": 3.276190476190476e-05,
+      "loss": 0.0826,
+      "step": 572
+    },
+    {
+      "epoch": 13.642857142857142,
+      "grad_norm": 0.23599408566951752,
+      "learning_rate": 3.271428571428571e-05,
+      "loss": 0.0951,
+      "step": 573
+    },
+    {
+      "epoch": 13.666666666666666,
+      "grad_norm": 0.2800326347351074,
+      "learning_rate": 3.266666666666666e-05,
+      "loss": 0.103,
+      "step": 574
+    },
+    {
+      "epoch": 13.69047619047619,
+      "grad_norm": 0.5657874345779419,
+      "learning_rate": 3.2619047619047615e-05,
+      "loss": 0.1053,
+      "step": 575
+    },
+    {
+      "epoch": 13.714285714285714,
+      "grad_norm": 0.23854580521583557,
+      "learning_rate": 3.2571428571428566e-05,
+      "loss": 0.0909,
+      "step": 576
+    },
+    {
+      "epoch": 13.738095238095237,
+      "grad_norm": 0.21590976417064667,
+      "learning_rate": 3.2523809523809525e-05,
+      "loss": 0.1133,
+      "step": 577
+    },
+    {
+      "epoch": 13.761904761904763,
+      "grad_norm": 0.2769882082939148,
+      "learning_rate": 3.2476190476190476e-05,
+      "loss": 0.0795,
+      "step": 578
+    },
+    {
+      "epoch": 13.785714285714286,
+      "grad_norm": 0.47914767265319824,
+      "learning_rate": 3.242857142857143e-05,
+      "loss": 0.1,
+      "step": 579
+    },
+    {
+      "epoch": 13.80952380952381,
+      "grad_norm": 0.3476030230522156,
+      "learning_rate": 3.238095238095238e-05,
+      "loss": 0.0953,
+      "step": 580
+    },
+    {
+      "epoch": 13.80952380952381,
+      "eval_dice_score": 0.919850237804958,
+      "eval_loss": 0.08700626343488693,
+      "eval_runtime": 8.7245,
+      "eval_samples_per_second": 10.201,
+      "eval_steps_per_second": 1.032,
+      "step": 580
+    },
+    {
+      "epoch": 13.833333333333334,
+      "grad_norm": 0.7643526792526245,
+      "learning_rate": 3.233333333333333e-05,
+      "loss": 0.0931,
+      "step": 581
+    },
+    {
+      "epoch": 13.857142857142858,
+      "grad_norm": 0.25581708550453186,
+      "learning_rate": 3.228571428571428e-05,
+      "loss": 0.0783,
+      "step": 582
+    },
+    {
+      "epoch": 13.880952380952381,
+      "grad_norm": 0.17205297946929932,
+      "learning_rate": 3.223809523809524e-05,
+      "loss": 0.0924,
+      "step": 583
+    },
+    {
+      "epoch": 13.904761904761905,
+      "grad_norm": 0.21472246944904327,
+      "learning_rate": 3.219047619047619e-05,
+      "loss": 0.0827,
+      "step": 584
+    },
+    {
+      "epoch": 13.928571428571429,
+      "grad_norm": 0.16439926624298096,
+      "learning_rate": 3.2142857142857144e-05,
+      "loss": 0.0802,
+      "step": 585
+    },
+    {
+      "epoch": 13.952380952380953,
+      "grad_norm": 0.7240442037582397,
+      "learning_rate": 3.2095238095238095e-05,
+      "loss": 0.1241,
+      "step": 586
+    },
+    {
+      "epoch": 13.976190476190476,
+      "grad_norm": 0.20413804054260254,
+      "learning_rate": 3.204761904761905e-05,
+      "loss": 0.083,
+      "step": 587
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 0.5824400186538696,
+      "learning_rate": 3.2e-05,
+      "loss": 0.0981,
+      "step": 588
+    },
+    {
+      "epoch": 14.023809523809524,
+      "grad_norm": 0.7712008953094482,
+      "learning_rate": 3.195238095238095e-05,
+      "loss": 0.1011,
+      "step": 589
+    },
+    {
+      "epoch": 14.047619047619047,
+      "grad_norm": 0.24950990080833435,
+      "learning_rate": 3.190476190476191e-05,
+      "loss": 0.0825,
+      "step": 590
+    },
+    {
+      "epoch": 14.071428571428571,
+      "grad_norm": 0.9160091876983643,
+      "learning_rate": 3.185714285714286e-05,
+      "loss": 0.1299,
+      "step": 591
+    },
+    {
+      "epoch": 14.095238095238095,
+      "grad_norm": 0.4385213255882263,
+      "learning_rate": 3.180952380952381e-05,
+      "loss": 0.1004,
+      "step": 592
+    },
+    {
+      "epoch": 14.119047619047619,
+      "grad_norm": 0.27217090129852295,
+      "learning_rate": 3.176190476190476e-05,
+      "loss": 0.0853,
+      "step": 593
+    },
+    {
+      "epoch": 14.142857142857142,
+      "grad_norm": 0.3268568813800812,
+      "learning_rate": 3.1714285714285715e-05,
+      "loss": 0.0902,
+      "step": 594
+    },
+    {
+      "epoch": 14.166666666666666,
+      "grad_norm": 0.2754589021205902,
+      "learning_rate": 3.1666666666666666e-05,
+      "loss": 0.0906,
+      "step": 595
+    },
+    {
+      "epoch": 14.19047619047619,
+      "grad_norm": 0.16892971098423004,
+      "learning_rate": 3.161904761904762e-05,
+      "loss": 0.09,
+      "step": 596
+    },
+    {
+      "epoch": 14.214285714285714,
+      "grad_norm": 0.16637921333312988,
+      "learning_rate": 3.1571428571428576e-05,
+      "loss": 0.0899,
+      "step": 597
+    },
+    {
+      "epoch": 14.238095238095237,
+      "grad_norm": 0.22213107347488403,
+      "learning_rate": 3.152380952380953e-05,
+      "loss": 0.0772,
+      "step": 598
+    },
+    {
+      "epoch": 14.261904761904763,
+      "grad_norm": 0.556175172328949,
+      "learning_rate": 3.147619047619048e-05,
+      "loss": 0.0973,
+      "step": 599
+    },
+    {
+      "epoch": 14.285714285714286,
+      "grad_norm": 0.2054072469472885,
+      "learning_rate": 3.142857142857143e-05,
+      "loss": 0.0841,
+      "step": 600
+    },
+    {
+      "epoch": 14.285714285714286,
+      "eval_dice_score": 0.9219252570440926,
+      "eval_loss": 0.08547207713127136,
+      "eval_runtime": 8.7966,
+      "eval_samples_per_second": 10.117,
+      "eval_steps_per_second": 1.023,
+      "step": 600
+    },
+    {
+      "epoch": 14.30952380952381,
+      "grad_norm": 0.2162121832370758,
+      "learning_rate": 3.138095238095238e-05,
+      "loss": 0.0864,
+      "step": 601
+    },
+    {
+      "epoch": 14.333333333333334,
+      "grad_norm": 0.19158528745174408,
+      "learning_rate": 3.1333333333333334e-05,
+      "loss": 0.0969,
+      "step": 602
+    },
+    {
+      "epoch": 14.357142857142858,
+      "grad_norm": 0.20269101858139038,
+      "learning_rate": 3.1285714285714285e-05,
+      "loss": 0.0871,
+      "step": 603
+    },
+    {
+      "epoch": 14.380952380952381,
+      "grad_norm": 0.4201182425022125,
+      "learning_rate": 3.1238095238095244e-05,
+      "loss": 0.0966,
+      "step": 604
+    },
+    {
+      "epoch": 14.404761904761905,
+      "grad_norm": 0.2613025903701782,
+      "learning_rate": 3.1190476190476195e-05,
+      "loss": 0.0884,
+      "step": 605
+    },
+    {
+      "epoch": 14.428571428571429,
+      "grad_norm": 0.19924290478229523,
+      "learning_rate": 3.114285714285715e-05,
+      "loss": 0.0896,
+      "step": 606
+    },
+    {
+      "epoch": 14.452380952380953,
+      "grad_norm": 0.17426665127277374,
+      "learning_rate": 3.10952380952381e-05,
+      "loss": 0.098,
+      "step": 607
+    },
+    {
+      "epoch": 14.476190476190476,
+      "grad_norm": 0.25578776001930237,
+      "learning_rate": 3.104761904761905e-05,
+      "loss": 0.082,
+      "step": 608
+    },
+    {
+      "epoch": 14.5,
+      "grad_norm": 0.7310316562652588,
+      "learning_rate": 3.1e-05,
+      "loss": 0.1199,
+      "step": 609
+    },
+    {
+      "epoch": 14.523809523809524,
+      "grad_norm": 0.223856583237648,
+      "learning_rate": 3.095238095238096e-05,
+      "loss": 0.0942,
+      "step": 610
+    },
+    {
+      "epoch": 14.547619047619047,
+      "grad_norm": 0.22037817537784576,
+      "learning_rate": 3.0904761904761904e-05,
+      "loss": 0.1062,
+      "step": 611
+    },
+    {
+      "epoch": 14.571428571428571,
+      "grad_norm": 0.2451525330543518,
+      "learning_rate": 3.0857142857142856e-05,
+      "loss": 0.108,
+      "step": 612
+    },
+    {
+      "epoch": 14.595238095238095,
+      "grad_norm": 0.2040538340806961,
+      "learning_rate": 3.080952380952381e-05,
+      "loss": 0.0892,
+      "step": 613
+    },
+    {
+      "epoch": 14.619047619047619,
+      "grad_norm": 0.2738243043422699,
+      "learning_rate": 3.076190476190476e-05,
+      "loss": 0.098,
+      "step": 614
+    },
+    {
+      "epoch": 14.642857142857142,
+      "grad_norm": 0.47061553597450256,
+      "learning_rate": 3.071428571428571e-05,
+      "loss": 0.0919,
+      "step": 615
+    },
+    {
+      "epoch": 14.666666666666666,
+      "grad_norm": 0.590526819229126,
+      "learning_rate": 3.066666666666666e-05,
+      "loss": 0.0851,
+      "step": 616
+    },
+    {
+      "epoch": 14.69047619047619,
+      "grad_norm": 0.2558976113796234,
+      "learning_rate": 3.061904761904762e-05,
+      "loss": 0.0781,
+      "step": 617
+    },
+    {
+      "epoch": 14.714285714285714,
+      "grad_norm": 0.406343936920166,
+      "learning_rate": 3.057142857142857e-05,
+      "loss": 0.0798,
+      "step": 618
+    },
+    {
+      "epoch": 14.738095238095237,
+      "grad_norm": 0.5376405119895935,
+      "learning_rate": 3.0523809523809524e-05,
+      "loss": 0.0992,
+      "step": 619
+    },
+    {
+      "epoch": 14.761904761904763,
+      "grad_norm": 0.359608918428421,
+      "learning_rate": 3.0476190476190475e-05,
+      "loss": 0.0821,
+      "step": 620
+    },
+    {
+      "epoch": 14.761904761904763,
+      "eval_dice_score": 0.921945237846471,
+      "eval_loss": 0.08672548085451126,
+      "eval_runtime": 8.8578,
+      "eval_samples_per_second": 10.048,
+      "eval_steps_per_second": 1.016,
+      "step": 620
+    },
+    {
+      "epoch": 14.785714285714286,
+      "grad_norm": 0.4258129894733429,
+      "learning_rate": 3.0428571428571427e-05,
+      "loss": 0.086,
+      "step": 621
+    },
+    {
+      "epoch": 14.80952380952381,
+      "grad_norm": 0.2852553129196167,
+      "learning_rate": 3.038095238095238e-05,
+      "loss": 0.1004,
+      "step": 622
+    },
+    {
+      "epoch": 14.833333333333334,
+      "grad_norm": 0.43911564350128174,
+      "learning_rate": 3.0333333333333333e-05,
+      "loss": 0.0763,
+      "step": 623
+    },
+    {
+      "epoch": 14.857142857142858,
+      "grad_norm": 0.26557502150535583,
+      "learning_rate": 3.0285714285714285e-05,
+      "loss": 0.0822,
+      "step": 624
+    },
+    {
+      "epoch": 14.880952380952381,
+      "grad_norm": 0.27375146746635437,
+      "learning_rate": 3.023809523809524e-05,
+      "loss": 0.0931,
+      "step": 625
+    },
+    {
+      "epoch": 14.904761904761905,
+      "grad_norm": 0.279346764087677,
+      "learning_rate": 3.019047619047619e-05,
+      "loss": 0.072,
+      "step": 626
+    },
+    {
+      "epoch": 14.928571428571429,
+      "grad_norm": 0.23838214576244354,
+      "learning_rate": 3.0142857142857143e-05,
+      "loss": 0.0806,
+      "step": 627
+    },
+    {
+      "epoch": 14.952380952380953,
+      "grad_norm": 0.16995024681091309,
+      "learning_rate": 3.0095238095238094e-05,
+      "loss": 0.0919,
+      "step": 628
+    },
+    {
+      "epoch": 14.976190476190476,
+      "grad_norm": 0.2560848593711853,
+      "learning_rate": 3.004761904761905e-05,
+      "loss": 0.0957,
+      "step": 629
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 0.3053649067878723,
+      "learning_rate": 3e-05,
+      "loss": 0.1034,
+      "step": 630
+    },
+    {
+      "epoch": 15.023809523809524,
+      "grad_norm": 0.1794043630361557,
+      "learning_rate": 2.9952380952380952e-05,
+      "loss": 0.0822,
+      "step": 631
+    },
+    {
+      "epoch": 15.047619047619047,
+      "grad_norm": 0.21695193648338318,
+      "learning_rate": 2.9904761904761907e-05,
+      "loss": 0.106,
+      "step": 632
+    },
+    {
+      "epoch": 15.071428571428571,
+      "grad_norm": 0.38391971588134766,
+      "learning_rate": 2.985714285714286e-05,
+      "loss": 0.0961,
+      "step": 633
+    },
+    {
+      "epoch": 15.095238095238095,
+      "grad_norm": 0.5132598280906677,
+      "learning_rate": 2.980952380952381e-05,
+      "loss": 0.1142,
+      "step": 634
+    },
+    {
+      "epoch": 15.119047619047619,
+      "grad_norm": 0.2727733850479126,
+      "learning_rate": 2.9761904761904762e-05,
+      "loss": 0.092,
+      "step": 635
+    },
+    {
+      "epoch": 15.142857142857142,
+      "grad_norm": 0.2456737458705902,
+      "learning_rate": 2.9714285714285717e-05,
+      "loss": 0.0793,
+      "step": 636
+    },
+    {
+      "epoch": 15.166666666666666,
+      "grad_norm": 0.2777925133705139,
+      "learning_rate": 2.966666666666667e-05,
+      "loss": 0.0848,
+      "step": 637
+    },
+    {
+      "epoch": 15.19047619047619,
+      "grad_norm": 0.5437929034233093,
+      "learning_rate": 2.961904761904762e-05,
+      "loss": 0.1062,
+      "step": 638
+    },
+    {
+      "epoch": 15.214285714285714,
+      "grad_norm": 0.2751214802265167,
+      "learning_rate": 2.9571428571428575e-05,
+      "loss": 0.0968,
+      "step": 639
+    },
+    {
+      "epoch": 15.238095238095237,
+      "grad_norm": 0.21941329538822174,
+      "learning_rate": 2.9523809523809523e-05,
+      "loss": 0.0828,
+      "step": 640
+    },
+    {
+      "epoch": 15.238095238095237,
+      "eval_dice_score": 0.92234169468477,
+      "eval_loss": 0.08497842401266098,
+      "eval_runtime": 8.8201,
+      "eval_samples_per_second": 10.091,
+      "eval_steps_per_second": 1.02,
+      "step": 640
+    },
+    {
+      "epoch": 15.261904761904763,
+      "grad_norm": 0.24691537022590637,
+      "learning_rate": 2.9476190476190475e-05,
+      "loss": 0.0884,
+      "step": 641
+    },
+    {
+      "epoch": 15.285714285714286,
+      "grad_norm": 0.2982950806617737,
+      "learning_rate": 2.942857142857143e-05,
+      "loss": 0.0957,
+      "step": 642
+    },
+    {
+      "epoch": 15.30952380952381,
+      "grad_norm": 0.2718714475631714,
+      "learning_rate": 2.938095238095238e-05,
+      "loss": 0.0883,
+      "step": 643
+    },
+    {
+      "epoch": 15.333333333333334,
+      "grad_norm": 0.29031896591186523,
+      "learning_rate": 2.9333333333333333e-05,
+      "loss": 0.0712,
+      "step": 644
+    },
+    {
+      "epoch": 15.357142857142858,
+      "grad_norm": 0.2872285842895508,
+      "learning_rate": 2.9285714285714284e-05,
+      "loss": 0.1028,
+      "step": 645
+    },
+    {
+      "epoch": 15.380952380952381,
+      "grad_norm": 0.3029137849807739,
+      "learning_rate": 2.923809523809524e-05,
+      "loss": 0.0778,
+      "step": 646
+    },
+    {
+      "epoch": 15.404761904761905,
+      "grad_norm": 0.3122120499610901,
+      "learning_rate": 2.919047619047619e-05,
+      "loss": 0.0728,
+      "step": 647
+    },
+    {
+      "epoch": 15.428571428571429,
+      "grad_norm": 0.28851887583732605,
+      "learning_rate": 2.9142857142857142e-05,
+      "loss": 0.0641,
+      "step": 648
+    },
+    {
+      "epoch": 15.452380952380953,
+      "grad_norm": 0.4654085338115692,
+      "learning_rate": 2.9095238095238097e-05,
+      "loss": 0.1011,
+      "step": 649
+    },
+    {
+      "epoch": 15.476190476190476,
+      "grad_norm": 0.32005026936531067,
+      "learning_rate": 2.904761904761905e-05,
+      "loss": 0.0953,
+      "step": 650
+    },
+    {
+      "epoch": 15.5,
+      "grad_norm": 0.2644088864326477,
+      "learning_rate": 2.9e-05,
+      "loss": 0.0854,
+      "step": 651
+    },
+    {
+      "epoch": 15.523809523809524,
+      "grad_norm": 0.37003859877586365,
+      "learning_rate": 2.8952380952380955e-05,
+      "loss": 0.0975,
+      "step": 652
+    },
+    {
+      "epoch": 15.547619047619047,
+      "grad_norm": 0.30068716406822205,
+      "learning_rate": 2.8904761904761907e-05,
+      "loss": 0.0806,
+      "step": 653
+    },
+    {
+      "epoch": 15.571428571428571,
+      "grad_norm": 0.2711564600467682,
+      "learning_rate": 2.8857142857142858e-05,
+      "loss": 0.0765,
+      "step": 654
+    },
+    {
+      "epoch": 15.595238095238095,
+      "grad_norm": 0.2577073574066162,
+      "learning_rate": 2.880952380952381e-05,
+      "loss": 0.0898,
+      "step": 655
+    },
+    {
+      "epoch": 15.619047619047619,
+      "grad_norm": 0.24594584107398987,
+      "learning_rate": 2.8761904761904765e-05,
+      "loss": 0.0844,
+      "step": 656
+    },
+    {
+      "epoch": 15.642857142857142,
+      "grad_norm": 0.22383585572242737,
+      "learning_rate": 2.8714285714285716e-05,
+      "loss": 0.094,
+      "step": 657
+    },
+    {
+      "epoch": 15.666666666666666,
+      "grad_norm": 0.3904590904712677,
+      "learning_rate": 2.8666666666666668e-05,
+      "loss": 0.0976,
+      "step": 658
+    },
+    {
+      "epoch": 15.69047619047619,
+      "grad_norm": 0.2078755795955658,
+      "learning_rate": 2.8619047619047623e-05,
+      "loss": 0.1011,
+      "step": 659
+    },
+    {
+      "epoch": 15.714285714285714,
+      "grad_norm": 0.35800865292549133,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 0.0914,
+      "step": 660
+    },
+    {
+      "epoch": 15.714285714285714,
+      "eval_dice_score": 0.9226904064180195,
+      "eval_loss": 0.08527851849794388,
+      "eval_runtime": 8.7169,
+      "eval_samples_per_second": 10.21,
+      "eval_steps_per_second": 1.032,
+      "step": 660
+    },
+    {
+      "epoch": 15.738095238095237,
+      "grad_norm": 0.30371710658073425,
+      "learning_rate": 2.8523809523809522e-05,
+      "loss": 0.0848,
+      "step": 661
+    },
+    {
+      "epoch": 15.761904761904763,
+      "grad_norm": 0.17579156160354614,
+      "learning_rate": 2.8476190476190477e-05,
+      "loss": 0.0881,
+      "step": 662
+    },
+    {
+      "epoch": 15.785714285714286,
+      "grad_norm": 0.2887127697467804,
+      "learning_rate": 2.842857142857143e-05,
+      "loss": 0.1119,
+      "step": 663
+    },
+    {
+      "epoch": 15.80952380952381,
+      "grad_norm": 0.20994120836257935,
+      "learning_rate": 2.838095238095238e-05,
+      "loss": 0.1133,
+      "step": 664
+    },
+    {
+      "epoch": 15.833333333333334,
+      "grad_norm": 0.23240219056606293,
+      "learning_rate": 2.8333333333333332e-05,
+      "loss": 0.0946,
+      "step": 665
+    },
+    {
+      "epoch": 15.857142857142858,
+      "grad_norm": 0.2476823329925537,
+      "learning_rate": 2.8285714285714287e-05,
+      "loss": 0.0881,
+      "step": 666
+    },
+    {
+      "epoch": 15.880952380952381,
+      "grad_norm": 0.5070866942405701,
+      "learning_rate": 2.823809523809524e-05,
+      "loss": 0.0869,
+      "step": 667
+    },
+    {
+      "epoch": 15.904761904761905,
+      "grad_norm": 0.2617017328739166,
+      "learning_rate": 2.819047619047619e-05,
+      "loss": 0.0823,
+      "step": 668
+    },
+    {
+      "epoch": 15.928571428571429,
+      "grad_norm": 0.2422521710395813,
+      "learning_rate": 2.8142857142857145e-05,
+      "loss": 0.0909,
+      "step": 669
+    },
+    {
+      "epoch": 15.952380952380953,
+      "grad_norm": 0.24192140996456146,
+      "learning_rate": 2.8095238095238096e-05,
+      "loss": 0.0836,
+      "step": 670
+    },
+    {
+      "epoch": 15.976190476190476,
+      "grad_norm": 0.32048043608665466,
+      "learning_rate": 2.8047619047619048e-05,
+      "loss": 0.0906,
+      "step": 671
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 0.6783073544502258,
+      "learning_rate": 2.8e-05,
+      "loss": 0.0956,
+      "step": 672
+    },
+    {
+      "epoch": 16.023809523809526,
+      "grad_norm": 0.21156437695026398,
+      "learning_rate": 2.7952380952380955e-05,
+      "loss": 0.0833,
+      "step": 673
+    },
+    {
+      "epoch": 16.047619047619047,
+      "grad_norm": 0.28836673498153687,
+      "learning_rate": 2.7904761904761906e-05,
+      "loss": 0.0775,
+      "step": 674
+    },
+    {
+      "epoch": 16.071428571428573,
+      "grad_norm": 0.2659715712070465,
+      "learning_rate": 2.7857142857142858e-05,
+      "loss": 0.0855,
+      "step": 675
+    },
+    {
+      "epoch": 16.095238095238095,
+      "grad_norm": 0.2857109010219574,
+      "learning_rate": 2.7809523809523813e-05,
+      "loss": 0.0847,
+      "step": 676
+    },
+    {
+      "epoch": 16.11904761904762,
+      "grad_norm": 0.24819140136241913,
+      "learning_rate": 2.7761904761904764e-05,
+      "loss": 0.0854,
+      "step": 677
+    },
+    {
+      "epoch": 16.142857142857142,
+      "grad_norm": 0.22153520584106445,
+      "learning_rate": 2.7714285714285716e-05,
+      "loss": 0.0821,
+      "step": 678
+    },
+    {
+      "epoch": 16.166666666666668,
+      "grad_norm": 0.2610468566417694,
+      "learning_rate": 2.766666666666667e-05,
+      "loss": 0.115,
+      "step": 679
+    },
+    {
+      "epoch": 16.19047619047619,
+      "grad_norm": 0.19806088507175446,
+      "learning_rate": 2.761904761904762e-05,
+      "loss": 0.0746,
+      "step": 680
+    },
+    {
+      "epoch": 16.19047619047619,
+      "eval_dice_score": 0.9234079552722338,
+      "eval_loss": 0.08446983993053436,
+      "eval_runtime": 8.6982,
+      "eval_samples_per_second": 10.232,
+      "eval_steps_per_second": 1.035,
+      "step": 680
+    },
+    {
+      "epoch": 16.214285714285715,
+      "grad_norm": 0.30673784017562866,
+      "learning_rate": 2.757142857142857e-05,
+      "loss": 0.0779,
+      "step": 681
+    },
+    {
+      "epoch": 16.238095238095237,
+      "grad_norm": 0.2549842596054077,
+      "learning_rate": 2.7523809523809522e-05,
+      "loss": 0.0823,
+      "step": 682
+    },
+    {
+      "epoch": 16.261904761904763,
+      "grad_norm": 0.24637040495872498,
+      "learning_rate": 2.7476190476190477e-05,
+      "loss": 0.0739,
+      "step": 683
+    },
+    {
+      "epoch": 16.285714285714285,
+      "grad_norm": 0.2508900761604309,
+      "learning_rate": 2.7428571428571428e-05,
+      "loss": 0.0928,
+      "step": 684
+    },
+    {
+      "epoch": 16.30952380952381,
+      "grad_norm": 0.2124183028936386,
+      "learning_rate": 2.738095238095238e-05,
+      "loss": 0.0795,
+      "step": 685
+    },
+    {
+      "epoch": 16.333333333333332,
+      "grad_norm": 0.29679664969444275,
+      "learning_rate": 2.7333333333333335e-05,
+      "loss": 0.0809,
+      "step": 686
+    },
+    {
+      "epoch": 16.357142857142858,
+      "grad_norm": 0.193775936961174,
+      "learning_rate": 2.7285714285714286e-05,
+      "loss": 0.0813,
+      "step": 687
+    },
+    {
+      "epoch": 16.38095238095238,
+      "grad_norm": 0.17316065728664398,
+      "learning_rate": 2.7238095238095238e-05,
+      "loss": 0.0796,
+      "step": 688
+    },
+    {
+      "epoch": 16.404761904761905,
+      "grad_norm": 0.39464396238327026,
+      "learning_rate": 2.7190476190476193e-05,
+      "loss": 0.1009,
+      "step": 689
+    },
+    {
+      "epoch": 16.428571428571427,
+      "grad_norm": 0.2255675494670868,
+      "learning_rate": 2.7142857142857144e-05,
+      "loss": 0.0858,
+      "step": 690
+    },
+    {
+      "epoch": 16.452380952380953,
+      "grad_norm": 0.20199432969093323,
+      "learning_rate": 2.7095238095238096e-05,
+      "loss": 0.079,
+      "step": 691
+    },
+    {
+      "epoch": 16.476190476190474,
+      "grad_norm": 0.3646969199180603,
+      "learning_rate": 2.7047619047619047e-05,
+      "loss": 0.1239,
+      "step": 692
+    },
+    {
+      "epoch": 16.5,
+      "grad_norm": 0.20324058830738068,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.0983,
+      "step": 693
+    },
+    {
+      "epoch": 16.523809523809526,
+      "grad_norm": 0.21907328069210052,
+      "learning_rate": 2.6952380952380954e-05,
+      "loss": 0.0836,
+      "step": 694
+    },
+    {
+      "epoch": 16.547619047619047,
+      "grad_norm": 0.27519315481185913,
+      "learning_rate": 2.6904761904761905e-05,
+      "loss": 0.0873,
+      "step": 695
+    },
+    {
+      "epoch": 16.571428571428573,
+      "grad_norm": 0.3098788261413574,
+      "learning_rate": 2.685714285714286e-05,
+      "loss": 0.1047,
+      "step": 696
+    },
+    {
+      "epoch": 16.595238095238095,
+      "grad_norm": 0.21530640125274658,
+      "learning_rate": 2.6809523809523812e-05,
+      "loss": 0.0742,
+      "step": 697
+    },
+    {
+      "epoch": 16.61904761904762,
+      "grad_norm": 0.19972284138202667,
+      "learning_rate": 2.6761904761904763e-05,
+      "loss": 0.0989,
+      "step": 698
+    },
+    {
+      "epoch": 16.642857142857142,
+      "grad_norm": 0.25694721937179565,
+      "learning_rate": 2.671428571428571e-05,
+      "loss": 0.0837,
+      "step": 699
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 0.4011896252632141,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.1062,
+      "step": 700
+    },
+    {
+      "epoch": 16.666666666666668,
+      "eval_dice_score": 0.9230252870594667,
+      "eval_loss": 0.08297789841890335,
+      "eval_runtime": 8.732,
+      "eval_samples_per_second": 10.192,
+      "eval_steps_per_second": 1.031,
+      "step": 700
+    },
+    {
+      "epoch": 16.69047619047619,
+      "grad_norm": 0.26972779631614685,
+      "learning_rate": 2.6619047619047618e-05,
+      "loss": 0.0878,
+      "step": 701
+    },
+    {
+      "epoch": 16.714285714285715,
+      "grad_norm": 0.3032207190990448,
+      "learning_rate": 2.657142857142857e-05,
+      "loss": 0.0933,
+      "step": 702
+    },
+    {
+      "epoch": 16.738095238095237,
+      "grad_norm": 0.18521499633789062,
+      "learning_rate": 2.6523809523809525e-05,
+      "loss": 0.0846,
+      "step": 703
+    },
+    {
+      "epoch": 16.761904761904763,
+      "grad_norm": 0.27679115533828735,
+      "learning_rate": 2.6476190476190476e-05,
+      "loss": 0.0716,
+      "step": 704
+    },
+    {
+      "epoch": 16.785714285714285,
+      "grad_norm": 0.6110713481903076,
+      "learning_rate": 2.6428571428571428e-05,
+      "loss": 0.1121,
+      "step": 705
+    },
+    {
+      "epoch": 16.80952380952381,
+      "grad_norm": 0.1876603215932846,
+      "learning_rate": 2.6380952380952383e-05,
+      "loss": 0.0844,
+      "step": 706
+    },
+    {
+      "epoch": 16.833333333333332,
+      "grad_norm": 0.3138282001018524,
+      "learning_rate": 2.6333333333333334e-05,
+      "loss": 0.0904,
+      "step": 707
+    },
+    {
+      "epoch": 16.857142857142858,
+      "grad_norm": 0.20735274255275726,
+      "learning_rate": 2.6285714285714286e-05,
+      "loss": 0.0968,
+      "step": 708
+    },
+    {
+      "epoch": 16.88095238095238,
+      "grad_norm": 0.31236979365348816,
+      "learning_rate": 2.6238095238095237e-05,
+      "loss": 0.1186,
+      "step": 709
+    },
+    {
+      "epoch": 16.904761904761905,
+      "grad_norm": 0.295156866312027,
+      "learning_rate": 2.6190476190476192e-05,
+      "loss": 0.082,
+      "step": 710
+    },
+    {
+      "epoch": 16.928571428571427,
+      "grad_norm": 0.24397049844264984,
+      "learning_rate": 2.6142857142857144e-05,
+      "loss": 0.0965,
+      "step": 711
+    },
+    {
+      "epoch": 16.952380952380953,
+      "grad_norm": 0.2571556568145752,
+      "learning_rate": 2.6095238095238095e-05,
+      "loss": 0.0704,
+      "step": 712
+    },
+    {
+      "epoch": 16.976190476190474,
+      "grad_norm": 0.24409866333007812,
+      "learning_rate": 2.604761904761905e-05,
+      "loss": 0.1001,
+      "step": 713
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 0.20111119747161865,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 0.0824,
+      "step": 714
+    },
+    {
+      "epoch": 17.023809523809526,
+      "grad_norm": 0.7431234121322632,
+      "learning_rate": 2.5952380952380953e-05,
+      "loss": 0.0929,
+      "step": 715
+    },
+    {
+      "epoch": 17.047619047619047,
+      "grad_norm": 0.27233293652534485,
+      "learning_rate": 2.5904761904761908e-05,
+      "loss": 0.0917,
+      "step": 716
+    },
+    {
+      "epoch": 17.071428571428573,
+      "grad_norm": 0.24907705187797546,
+      "learning_rate": 2.585714285714286e-05,
+      "loss": 0.0951,
+      "step": 717
+    },
+    {
+      "epoch": 17.095238095238095,
+      "grad_norm": 0.23256130516529083,
+      "learning_rate": 2.580952380952381e-05,
+      "loss": 0.0925,
+      "step": 718
+    },
+    {
+      "epoch": 17.11904761904762,
+      "grad_norm": 0.5666278004646301,
+      "learning_rate": 2.576190476190476e-05,
+      "loss": 0.1042,
+      "step": 719
+    },
+    {
+      "epoch": 17.142857142857142,
+      "grad_norm": 0.2410479336977005,
+      "learning_rate": 2.5714285714285714e-05,
+      "loss": 0.0666,
+      "step": 720
+    },
+    {
+      "epoch": 17.142857142857142,
+      "eval_dice_score": 0.9229571361874431,
+      "eval_loss": 0.08340851962566376,
+      "eval_runtime": 8.7189,
+      "eval_samples_per_second": 10.208,
+      "eval_steps_per_second": 1.032,
+      "step": 720
+    },
+    {
+      "epoch": 17.166666666666668,
+      "grad_norm": 0.20309841632843018,
+      "learning_rate": 2.5666666666666666e-05,
+      "loss": 0.067,
+      "step": 721
+    },
+    {
+      "epoch": 17.19047619047619,
+      "grad_norm": 0.3129255473613739,
+      "learning_rate": 2.5619047619047618e-05,
+      "loss": 0.0894,
+      "step": 722
+    },
+    {
+      "epoch": 17.214285714285715,
+      "grad_norm": 0.27786898612976074,
+      "learning_rate": 2.5571428571428572e-05,
+      "loss": 0.0839,
+      "step": 723
+    },
+    {
+      "epoch": 17.238095238095237,
+      "grad_norm": 0.267407089471817,
+      "learning_rate": 2.5523809523809524e-05,
+      "loss": 0.1054,
+      "step": 724
+    },
+    {
+      "epoch": 17.261904761904763,
+      "grad_norm": 0.4915502965450287,
+      "learning_rate": 2.5476190476190476e-05,
+      "loss": 0.0938,
+      "step": 725
+    },
+    {
+      "epoch": 17.285714285714285,
+      "grad_norm": 0.21978172659873962,
+      "learning_rate": 2.5428571428571427e-05,
+      "loss": 0.0857,
+      "step": 726
+    },
+    {
+      "epoch": 17.30952380952381,
+      "grad_norm": 0.35925644636154175,
+      "learning_rate": 2.5380952380952382e-05,
+      "loss": 0.1016,
+      "step": 727
+    },
+    {
+      "epoch": 17.333333333333332,
+      "grad_norm": 0.2245437353849411,
+      "learning_rate": 2.5333333333333334e-05,
+      "loss": 0.0848,
+      "step": 728
+    },
+    {
+      "epoch": 17.357142857142858,
+      "grad_norm": 0.24263395369052887,
+      "learning_rate": 2.5285714285714285e-05,
+      "loss": 0.0948,
+      "step": 729
+    },
+    {
+      "epoch": 17.38095238095238,
+      "grad_norm": 0.21132177114486694,
+      "learning_rate": 2.523809523809524e-05,
+      "loss": 0.0953,
+      "step": 730
+    },
+    {
+      "epoch": 17.404761904761905,
+      "grad_norm": 0.6164664030075073,
+      "learning_rate": 2.519047619047619e-05,
+      "loss": 0.0985,
+      "step": 731
+    },
+    {
+      "epoch": 17.428571428571427,
+      "grad_norm": 0.34183213114738464,
+      "learning_rate": 2.5142857142857143e-05,
+      "loss": 0.1059,
+      "step": 732
+    },
+    {
+      "epoch": 17.452380952380953,
+      "grad_norm": 0.27018675208091736,
+      "learning_rate": 2.5095238095238098e-05,
+      "loss": 0.0955,
+      "step": 733
+    },
+    {
+      "epoch": 17.476190476190474,
+      "grad_norm": 0.291178435087204,
+      "learning_rate": 2.504761904761905e-05,
+      "loss": 0.0796,
+      "step": 734
+    },
+    {
+      "epoch": 17.5,
+      "grad_norm": 0.23403896391391754,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0787,
+      "step": 735
+    },
+    {
+      "epoch": 17.523809523809526,
+      "grad_norm": 0.45682311058044434,
+      "learning_rate": 2.4952380952380953e-05,
+      "loss": 0.0884,
+      "step": 736
+    },
+    {
+      "epoch": 17.547619047619047,
+      "grad_norm": 0.23823651671409607,
+      "learning_rate": 2.4904761904761908e-05,
+      "loss": 0.0977,
+      "step": 737
+    },
+    {
+      "epoch": 17.571428571428573,
+      "grad_norm": 0.22397005558013916,
+      "learning_rate": 2.485714285714286e-05,
+      "loss": 0.0775,
+      "step": 738
+    },
+    {
+      "epoch": 17.595238095238095,
+      "grad_norm": 0.2246365249156952,
+      "learning_rate": 2.4809523809523807e-05,
+      "loss": 0.0685,
+      "step": 739
+    },
+    {
+      "epoch": 17.61904761904762,
+      "grad_norm": 0.2863618731498718,
+      "learning_rate": 2.4761904761904762e-05,
+      "loss": 0.0904,
+      "step": 740
+    },
+    {
+      "epoch": 17.61904761904762,
+      "eval_dice_score": 0.9211544390867972,
+      "eval_loss": 0.0836789682507515,
+      "eval_runtime": 8.7029,
+      "eval_samples_per_second": 10.226,
+      "eval_steps_per_second": 1.034,
+      "step": 740
+    },
+    {
+      "epoch": 17.642857142857142,
+      "grad_norm": 0.39199358224868774,
+      "learning_rate": 2.4714285714285714e-05,
+      "loss": 0.1074,
+      "step": 741
+    },
+    {
+      "epoch": 17.666666666666668,
+      "grad_norm": 0.29316845536231995,
+      "learning_rate": 2.4666666666666665e-05,
+      "loss": 0.0857,
+      "step": 742
+    },
+    {
+      "epoch": 17.69047619047619,
+      "grad_norm": 0.2797645926475525,
+      "learning_rate": 2.461904761904762e-05,
+      "loss": 0.0803,
+      "step": 743
+    },
+    {
+      "epoch": 17.714285714285715,
+      "grad_norm": 0.2353767305612564,
+      "learning_rate": 2.4571428571428572e-05,
+      "loss": 0.0861,
+      "step": 744
+    },
+    {
+      "epoch": 17.738095238095237,
+      "grad_norm": 0.24464334547519684,
+      "learning_rate": 2.4523809523809523e-05,
+      "loss": 0.0785,
+      "step": 745
+    },
+    {
+      "epoch": 17.761904761904763,
+      "grad_norm": 0.3035442531108856,
+      "learning_rate": 2.4476190476190475e-05,
+      "loss": 0.0896,
+      "step": 746
+    },
+    {
+      "epoch": 17.785714285714285,
+      "grad_norm": 0.20400097966194153,
+      "learning_rate": 2.442857142857143e-05,
+      "loss": 0.0898,
+      "step": 747
+    },
+    {
+      "epoch": 17.80952380952381,
+      "grad_norm": 0.23570357263088226,
+      "learning_rate": 2.438095238095238e-05,
+      "loss": 0.0836,
+      "step": 748
+    },
+    {
+      "epoch": 17.833333333333332,
+      "grad_norm": 0.321524053812027,
+      "learning_rate": 2.4333333333333333e-05,
+      "loss": 0.0744,
+      "step": 749
+    },
+    {
+      "epoch": 17.857142857142858,
+      "grad_norm": 0.22331471741199493,
+      "learning_rate": 2.4285714285714288e-05,
+      "loss": 0.0872,
+      "step": 750
+    },
+    {
+      "epoch": 17.88095238095238,
+      "grad_norm": 0.31584519147872925,
+      "learning_rate": 2.423809523809524e-05,
+      "loss": 0.0746,
+      "step": 751
+    },
+    {
+      "epoch": 17.904761904761905,
+      "grad_norm": 0.32943403720855713,
+      "learning_rate": 2.419047619047619e-05,
+      "loss": 0.0896,
+      "step": 752
+    },
+    {
+      "epoch": 17.928571428571427,
+      "grad_norm": 0.21786193549633026,
+      "learning_rate": 2.4142857142857146e-05,
+      "loss": 0.0828,
+      "step": 753
+    },
+    {
+      "epoch": 17.952380952380953,
+      "grad_norm": 0.17888040840625763,
+      "learning_rate": 2.4095238095238098e-05,
+      "loss": 0.0775,
+      "step": 754
+    },
+    {
+      "epoch": 17.976190476190474,
+      "grad_norm": 0.250286728143692,
+      "learning_rate": 2.404761904761905e-05,
+      "loss": 0.0844,
+      "step": 755
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 0.3801717162132263,
+      "learning_rate": 2.4e-05,
+      "loss": 0.0883,
+      "step": 756
+    },
+    {
+      "epoch": 18.023809523809526,
+      "grad_norm": 0.2086687684059143,
+      "learning_rate": 2.3952380952380956e-05,
+      "loss": 0.0784,
+      "step": 757
+    },
+    {
+      "epoch": 18.047619047619047,
+      "grad_norm": 0.23379236459732056,
+      "learning_rate": 2.3904761904761904e-05,
+      "loss": 0.0947,
+      "step": 758
+    },
+    {
+      "epoch": 18.071428571428573,
+      "grad_norm": 0.2616595923900604,
+      "learning_rate": 2.3857142857142855e-05,
+      "loss": 0.0869,
+      "step": 759
+    },
+    {
+      "epoch": 18.095238095238095,
+      "grad_norm": 0.18408074975013733,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 0.0791,
+      "step": 760
+    },
+    {
+      "epoch": 18.095238095238095,
+      "eval_dice_score": 0.9236365296024307,
+      "eval_loss": 0.0821303203701973,
+      "eval_runtime": 8.7731,
+      "eval_samples_per_second": 10.145,
+      "eval_steps_per_second": 1.026,
+      "step": 760
+    },
+    {
+      "epoch": 18.11904761904762,
+      "grad_norm": 0.17643745243549347,
+      "learning_rate": 2.3761904761904762e-05,
+      "loss": 0.0809,
+      "step": 761
+    },
+    {
+      "epoch": 18.142857142857142,
+      "grad_norm": 0.28116345405578613,
+      "learning_rate": 2.3714285714285713e-05,
+      "loss": 0.085,
+      "step": 762
+    },
+    {
+      "epoch": 18.166666666666668,
+      "grad_norm": 0.18162474036216736,
+      "learning_rate": 2.3666666666666665e-05,
+      "loss": 0.078,
+      "step": 763
+    },
+    {
+      "epoch": 18.19047619047619,
+      "grad_norm": 0.27260643243789673,
+      "learning_rate": 2.361904761904762e-05,
+      "loss": 0.0914,
+      "step": 764
+    },
+    {
+      "epoch": 18.214285714285715,
+      "grad_norm": 0.26358166337013245,
+      "learning_rate": 2.357142857142857e-05,
+      "loss": 0.0857,
+      "step": 765
+    },
+    {
+      "epoch": 18.238095238095237,
+      "grad_norm": 0.32413822412490845,
+      "learning_rate": 2.3523809523809523e-05,
+      "loss": 0.0894,
+      "step": 766
+    },
+    {
+      "epoch": 18.261904761904763,
+      "grad_norm": 0.23978236317634583,
+      "learning_rate": 2.3476190476190478e-05,
+      "loss": 0.091,
+      "step": 767
+    },
+    {
+      "epoch": 18.285714285714285,
+      "grad_norm": 0.3409174978733063,
+      "learning_rate": 2.342857142857143e-05,
+      "loss": 0.1105,
+      "step": 768
+    },
+    {
+      "epoch": 18.30952380952381,
+      "grad_norm": 0.1719333976507187,
+      "learning_rate": 2.338095238095238e-05,
+      "loss": 0.0799,
+      "step": 769
+    },
+    {
+      "epoch": 18.333333333333332,
+      "grad_norm": 0.2759392559528351,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 0.0943,
+      "step": 770
+    },
+    {
+      "epoch": 18.357142857142858,
+      "grad_norm": 0.2284536361694336,
+      "learning_rate": 2.3285714285714287e-05,
+      "loss": 0.0899,
+      "step": 771
+    },
+    {
+      "epoch": 18.38095238095238,
+      "grad_norm": 0.2436296045780182,
+      "learning_rate": 2.323809523809524e-05,
+      "loss": 0.0662,
+      "step": 772
+    },
+    {
+      "epoch": 18.404761904761905,
+      "grad_norm": 0.16819937527179718,
+      "learning_rate": 2.319047619047619e-05,
+      "loss": 0.0865,
+      "step": 773
+    },
+    {
+      "epoch": 18.428571428571427,
+      "grad_norm": 0.22437210381031036,
+      "learning_rate": 2.3142857142857145e-05,
+      "loss": 0.0849,
+      "step": 774
+    },
+    {
+      "epoch": 18.452380952380953,
+      "grad_norm": 0.28795716166496277,
+      "learning_rate": 2.3095238095238097e-05,
+      "loss": 0.0997,
+      "step": 775
+    },
+    {
+      "epoch": 18.476190476190474,
+      "grad_norm": 0.40768590569496155,
+      "learning_rate": 2.304761904761905e-05,
+      "loss": 0.1003,
+      "step": 776
+    },
+    {
+      "epoch": 18.5,
+      "grad_norm": 0.2193804681301117,
+      "learning_rate": 2.3000000000000003e-05,
+      "loss": 0.0896,
+      "step": 777
+    },
+    {
+      "epoch": 18.523809523809526,
+      "grad_norm": 0.29064154624938965,
+      "learning_rate": 2.295238095238095e-05,
+      "loss": 0.0778,
+      "step": 778
+    },
+    {
+      "epoch": 18.547619047619047,
+      "grad_norm": 0.39261043071746826,
+      "learning_rate": 2.2904761904761903e-05,
+      "loss": 0.0924,
+      "step": 779
+    },
+    {
+      "epoch": 18.571428571428573,
+      "grad_norm": 0.37924641370773315,
+      "learning_rate": 2.2857142857142858e-05,
+      "loss": 0.1048,
+      "step": 780
+    },
+    {
+      "epoch": 18.571428571428573,
+      "eval_dice_score": 0.9232294350220722,
+      "eval_loss": 0.08147627115249634,
+      "eval_runtime": 8.7455,
+      "eval_samples_per_second": 10.177,
+      "eval_steps_per_second": 1.029,
+      "step": 780
+    },
+    {
+      "epoch": 18.595238095238095,
+      "grad_norm": 0.24612796306610107,
+      "learning_rate": 2.280952380952381e-05,
+      "loss": 0.0774,
+      "step": 781
+    },
+    {
+      "epoch": 18.61904761904762,
+      "grad_norm": 0.18419839441776276,
+      "learning_rate": 2.276190476190476e-05,
+      "loss": 0.0829,
+      "step": 782
+    },
+    {
+      "epoch": 18.642857142857142,
+      "grad_norm": 0.19330959022045135,
+      "learning_rate": 2.2714285714285713e-05,
+      "loss": 0.0729,
+      "step": 783
+    },
+    {
+      "epoch": 18.666666666666668,
+      "grad_norm": 0.21217748522758484,
+      "learning_rate": 2.2666666666666668e-05,
+      "loss": 0.0729,
+      "step": 784
+    },
+    {
+      "epoch": 18.69047619047619,
+      "grad_norm": 0.22022300958633423,
+      "learning_rate": 2.261904761904762e-05,
+      "loss": 0.075,
+      "step": 785
+    },
+    {
+      "epoch": 18.714285714285715,
+      "grad_norm": 0.2307768315076828,
+      "learning_rate": 2.257142857142857e-05,
+      "loss": 0.0995,
+      "step": 786
+    },
+    {
+      "epoch": 18.738095238095237,
+      "grad_norm": 0.499103307723999,
+      "learning_rate": 2.2523809523809526e-05,
+      "loss": 0.0927,
+      "step": 787
+    },
+    {
+      "epoch": 18.761904761904763,
+      "grad_norm": 0.1646272838115692,
+      "learning_rate": 2.2476190476190477e-05,
+      "loss": 0.0728,
+      "step": 788
+    },
+    {
+      "epoch": 18.785714285714285,
+      "grad_norm": 0.19184449315071106,
+      "learning_rate": 2.242857142857143e-05,
+      "loss": 0.0768,
+      "step": 789
+    },
+    {
+      "epoch": 18.80952380952381,
+      "grad_norm": 0.3876766562461853,
+      "learning_rate": 2.238095238095238e-05,
+      "loss": 0.1064,
+      "step": 790
+    },
+    {
+      "epoch": 18.833333333333332,
+      "grad_norm": 0.18872225284576416,
+      "learning_rate": 2.2333333333333335e-05,
+      "loss": 0.0783,
+      "step": 791
+    },
+    {
+      "epoch": 18.857142857142858,
+      "grad_norm": 0.4279574453830719,
+      "learning_rate": 2.2285714285714287e-05,
+      "loss": 0.0971,
+      "step": 792
+    },
+    {
+      "epoch": 18.88095238095238,
+      "grad_norm": 0.1798621416091919,
+      "learning_rate": 2.223809523809524e-05,
+      "loss": 0.0862,
+      "step": 793
+    },
+    {
+      "epoch": 18.904761904761905,
+      "grad_norm": 0.4626399874687195,
+      "learning_rate": 2.2190476190476193e-05,
+      "loss": 0.0922,
+      "step": 794
+    },
+    {
+      "epoch": 18.928571428571427,
+      "grad_norm": 0.2156091034412384,
+      "learning_rate": 2.2142857142857145e-05,
+      "loss": 0.0826,
+      "step": 795
+    },
+    {
+      "epoch": 18.952380952380953,
+      "grad_norm": 0.30277004837989807,
+      "learning_rate": 2.2095238095238096e-05,
+      "loss": 0.0881,
+      "step": 796
+    },
+    {
+      "epoch": 18.976190476190474,
+      "grad_norm": 0.20252549648284912,
+      "learning_rate": 2.204761904761905e-05,
+      "loss": 0.0814,
+      "step": 797
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 0.3133484423160553,
+      "learning_rate": 2.2e-05,
+      "loss": 0.091,
+      "step": 798
+    },
+    {
+      "epoch": 19.023809523809526,
+      "grad_norm": 0.2910543978214264,
+      "learning_rate": 2.195238095238095e-05,
+      "loss": 0.0761,
+      "step": 799
+    },
+    {
+      "epoch": 19.047619047619047,
+      "grad_norm": 0.19995278120040894,
+      "learning_rate": 2.1904761904761903e-05,
+      "loss": 0.0935,
+      "step": 800
+    },
+    {
+      "epoch": 19.047619047619047,
+      "eval_dice_score": 0.9242091310096217,
+      "eval_loss": 0.08286911994218826,
+      "eval_runtime": 9.0935,
+      "eval_samples_per_second": 9.787,
+      "eval_steps_per_second": 0.99,
+      "step": 800
+    },
+    {
+      "epoch": 19.071428571428573,
+      "grad_norm": 0.29460981488227844,
+      "learning_rate": 2.1857142857142857e-05,
+      "loss": 0.0838,
+      "step": 801
+    },
+    {
+      "epoch": 19.095238095238095,
+      "grad_norm": 0.18889278173446655,
+      "learning_rate": 2.180952380952381e-05,
+      "loss": 0.1012,
+      "step": 802
+    },
+    {
+      "epoch": 19.11904761904762,
+      "grad_norm": 0.2679443061351776,
+      "learning_rate": 2.176190476190476e-05,
+      "loss": 0.0893,
+      "step": 803
+    },
+    {
+      "epoch": 19.142857142857142,
+      "grad_norm": 0.2455090433359146,
+      "learning_rate": 2.1714285714285715e-05,
+      "loss": 0.0715,
+      "step": 804
+    },
+    {
+      "epoch": 19.166666666666668,
+      "grad_norm": 0.17029725015163422,
+      "learning_rate": 2.1666666666666667e-05,
+      "loss": 0.0729,
+      "step": 805
+    },
+    {
+      "epoch": 19.19047619047619,
+      "grad_norm": 0.3500533699989319,
+      "learning_rate": 2.161904761904762e-05,
+      "loss": 0.0991,
+      "step": 806
+    },
+    {
+      "epoch": 19.214285714285715,
+      "grad_norm": 0.17729905247688293,
+      "learning_rate": 2.1571428571428574e-05,
+      "loss": 0.0822,
+      "step": 807
+    },
+    {
+      "epoch": 19.238095238095237,
+      "grad_norm": 0.24052225053310394,
+      "learning_rate": 2.1523809523809525e-05,
+      "loss": 0.0736,
+      "step": 808
+    },
+    {
+      "epoch": 19.261904761904763,
+      "grad_norm": 0.24907103180885315,
+      "learning_rate": 2.1476190476190477e-05,
+      "loss": 0.0874,
+      "step": 809
+    },
+    {
+      "epoch": 19.285714285714285,
+      "grad_norm": 0.2517390251159668,
+      "learning_rate": 2.1428571428571428e-05,
+      "loss": 0.087,
+      "step": 810
+    },
+    {
+      "epoch": 19.30952380952381,
+      "grad_norm": 0.21364130079746246,
+      "learning_rate": 2.1380952380952383e-05,
+      "loss": 0.0926,
+      "step": 811
+    },
+    {
+      "epoch": 19.333333333333332,
+      "grad_norm": 0.16167882084846497,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 0.087,
+      "step": 812
+    },
+    {
+      "epoch": 19.357142857142858,
+      "grad_norm": 0.21866513788700104,
+      "learning_rate": 2.1285714285714286e-05,
+      "loss": 0.0732,
+      "step": 813
+    },
+    {
+      "epoch": 19.38095238095238,
+      "grad_norm": 0.30814534425735474,
+      "learning_rate": 2.123809523809524e-05,
+      "loss": 0.0951,
+      "step": 814
+    },
+    {
+      "epoch": 19.404761904761905,
+      "grad_norm": 0.20811501145362854,
+      "learning_rate": 2.1190476190476193e-05,
+      "loss": 0.0677,
+      "step": 815
+    },
+    {
+      "epoch": 19.428571428571427,
+      "grad_norm": 0.3216896057128906,
+      "learning_rate": 2.1142857142857144e-05,
+      "loss": 0.0872,
+      "step": 816
+    },
+    {
+      "epoch": 19.452380952380953,
+      "grad_norm": 0.3017369210720062,
+      "learning_rate": 2.1095238095238096e-05,
+      "loss": 0.0696,
+      "step": 817
+    },
+    {
+      "epoch": 19.476190476190474,
+      "grad_norm": 0.22521443665027618,
+      "learning_rate": 2.1047619047619047e-05,
+      "loss": 0.1094,
+      "step": 818
+    },
+    {
+      "epoch": 19.5,
+      "grad_norm": 0.1566617488861084,
+      "learning_rate": 2.1e-05,
+      "loss": 0.0851,
+      "step": 819
+    },
+    {
+      "epoch": 19.523809523809526,
+      "grad_norm": 0.2676796019077301,
+      "learning_rate": 2.095238095238095e-05,
+      "loss": 0.0745,
+      "step": 820
+    },
+    {
+      "epoch": 19.523809523809526,
+      "eval_dice_score": 0.9234152837764387,
+      "eval_loss": 0.08154100179672241,
+      "eval_runtime": 8.8155,
+      "eval_samples_per_second": 10.096,
+      "eval_steps_per_second": 1.021,
+      "step": 820
+    },
+    {
+      "epoch": 19.547619047619047,
+      "grad_norm": 0.2537257969379425,
+      "learning_rate": 2.0904761904761905e-05,
+      "loss": 0.0889,
+      "step": 821
+    },
+    {
+      "epoch": 19.571428571428573,
+      "grad_norm": 0.16671201586723328,
+      "learning_rate": 2.0857142857142857e-05,
+      "loss": 0.084,
+      "step": 822
+    },
+    {
+      "epoch": 19.595238095238095,
+      "grad_norm": 0.21541264653205872,
+      "learning_rate": 2.080952380952381e-05,
+      "loss": 0.0993,
+      "step": 823
+    },
+    {
+      "epoch": 19.61904761904762,
+      "grad_norm": 0.20101720094680786,
+      "learning_rate": 2.0761904761904763e-05,
+      "loss": 0.0906,
+      "step": 824
+    },
+    {
+      "epoch": 19.642857142857142,
+      "grad_norm": 0.155084490776062,
+      "learning_rate": 2.0714285714285715e-05,
+      "loss": 0.0703,
+      "step": 825
+    },
+    {
+      "epoch": 19.666666666666668,
+      "grad_norm": 0.19437827169895172,
+      "learning_rate": 2.0666666666666666e-05,
+      "loss": 0.0827,
+      "step": 826
+    },
+    {
+      "epoch": 19.69047619047619,
+      "grad_norm": 0.2180471420288086,
+      "learning_rate": 2.0619047619047618e-05,
+      "loss": 0.088,
+      "step": 827
+    },
+    {
+      "epoch": 19.714285714285715,
+      "grad_norm": 0.30691877007484436,
+      "learning_rate": 2.0571428571428573e-05,
+      "loss": 0.1097,
+      "step": 828
+    },
+    {
+      "epoch": 19.738095238095237,
+      "grad_norm": 0.16602082550525665,
+      "learning_rate": 2.0523809523809524e-05,
+      "loss": 0.0706,
+      "step": 829
+    },
+    {
+      "epoch": 19.761904761904763,
+      "grad_norm": 0.19758841395378113,
+      "learning_rate": 2.0476190476190476e-05,
+      "loss": 0.1042,
+      "step": 830
+    },
+    {
+      "epoch": 19.785714285714285,
+      "grad_norm": 0.2860798239707947,
+      "learning_rate": 2.042857142857143e-05,
+      "loss": 0.0872,
+      "step": 831
+    },
+    {
+      "epoch": 19.80952380952381,
+      "grad_norm": 0.175958052277565,
+      "learning_rate": 2.0380952380952382e-05,
+      "loss": 0.0812,
+      "step": 832
+    },
+    {
+      "epoch": 19.833333333333332,
+      "grad_norm": 0.2437559813261032,
+      "learning_rate": 2.0333333333333334e-05,
+      "loss": 0.0811,
+      "step": 833
+    },
+    {
+      "epoch": 19.857142857142858,
+      "grad_norm": 0.16146957874298096,
+      "learning_rate": 2.028571428571429e-05,
+      "loss": 0.0726,
+      "step": 834
+    },
+    {
+      "epoch": 19.88095238095238,
+      "grad_norm": 0.18311820924282074,
+      "learning_rate": 2.023809523809524e-05,
+      "loss": 0.0936,
+      "step": 835
+    },
+    {
+      "epoch": 19.904761904761905,
+      "grad_norm": 0.23156321048736572,
+      "learning_rate": 2.0190476190476192e-05,
+      "loss": 0.0959,
+      "step": 836
+    },
+    {
+      "epoch": 19.928571428571427,
+      "grad_norm": 0.6132747530937195,
+      "learning_rate": 2.014285714285714e-05,
+      "loss": 0.1083,
+      "step": 837
+    },
+    {
+      "epoch": 19.952380952380953,
+      "grad_norm": 0.2138746976852417,
+      "learning_rate": 2.0095238095238095e-05,
+      "loss": 0.0686,
+      "step": 838
+    },
+    {
+      "epoch": 19.976190476190474,
+      "grad_norm": 0.1906508058309555,
+      "learning_rate": 2.0047619047619047e-05,
+      "loss": 0.0819,
+      "step": 839
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 0.19612105190753937,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 0.0811,
+      "step": 840
+    },
+    {
+      "epoch": 20.0,
+      "eval_dice_score": 0.9237751835199779,
+      "eval_loss": 0.08120287954807281,
+      "eval_runtime": 8.9191,
+      "eval_samples_per_second": 9.979,
+      "eval_steps_per_second": 1.009,
+      "step": 840
+    },
+    {
+      "epoch": 20.023809523809526,
+      "grad_norm": 0.20686964690685272,
+      "learning_rate": 1.9952380952380953e-05,
+      "loss": 0.0828,
+      "step": 841
+    },
+    {
+      "epoch": 20.047619047619047,
+      "grad_norm": 0.19367481768131256,
+      "learning_rate": 1.9904761904761905e-05,
+      "loss": 0.077,
+      "step": 842
+    },
+    {
+      "epoch": 20.071428571428573,
+      "grad_norm": 0.24994997680187225,
+      "learning_rate": 1.9857142857142856e-05,
+      "loss": 0.0735,
+      "step": 843
+    },
+    {
+      "epoch": 20.095238095238095,
+      "grad_norm": 0.22139909863471985,
+      "learning_rate": 1.980952380952381e-05,
+      "loss": 0.0784,
+      "step": 844
+    },
+    {
+      "epoch": 20.11904761904762,
+      "grad_norm": 0.30801865458488464,
+      "learning_rate": 1.9761904761904763e-05,
+      "loss": 0.103,
+      "step": 845
+    },
+    {
+      "epoch": 20.142857142857142,
+      "grad_norm": 0.33742642402648926,
+      "learning_rate": 1.9714285714285714e-05,
+      "loss": 0.1045,
+      "step": 846
+    },
+    {
+      "epoch": 20.166666666666668,
+      "grad_norm": 0.22293123602867126,
+      "learning_rate": 1.9666666666666666e-05,
+      "loss": 0.0856,
+      "step": 847
+    },
+    {
+      "epoch": 20.19047619047619,
+      "grad_norm": 0.3034045398235321,
+      "learning_rate": 1.961904761904762e-05,
+      "loss": 0.0824,
+      "step": 848
+    },
+    {
+      "epoch": 20.214285714285715,
+      "grad_norm": 0.20768654346466064,
+      "learning_rate": 1.9571428571428572e-05,
+      "loss": 0.0766,
+      "step": 849
+    },
+    {
+      "epoch": 20.238095238095237,
+      "grad_norm": 0.22169268131256104,
+      "learning_rate": 1.9523809523809524e-05,
+      "loss": 0.0778,
+      "step": 850
+    },
+    {
+      "epoch": 20.261904761904763,
+      "grad_norm": 0.1683131903409958,
+      "learning_rate": 1.947619047619048e-05,
+      "loss": 0.0758,
+      "step": 851
+    },
+    {
+      "epoch": 20.285714285714285,
+      "grad_norm": 0.3005770742893219,
+      "learning_rate": 1.942857142857143e-05,
+      "loss": 0.1013,
+      "step": 852
+    },
+    {
+      "epoch": 20.30952380952381,
+      "grad_norm": 0.48266512155532837,
+      "learning_rate": 1.9380952380952382e-05,
+      "loss": 0.0957,
+      "step": 853
+    },
+    {
+      "epoch": 20.333333333333332,
+      "grad_norm": 0.37430715560913086,
+      "learning_rate": 1.9333333333333333e-05,
+      "loss": 0.0702,
+      "step": 854
+    },
+    {
+      "epoch": 20.357142857142858,
+      "grad_norm": 0.3011479675769806,
+      "learning_rate": 1.928571428571429e-05,
+      "loss": 0.0862,
+      "step": 855
+    },
+    {
+      "epoch": 20.38095238095238,
+      "grad_norm": 0.3303626477718353,
+      "learning_rate": 1.923809523809524e-05,
+      "loss": 0.081,
+      "step": 856
+    },
+    {
+      "epoch": 20.404761904761905,
+      "grad_norm": 0.1781713366508484,
+      "learning_rate": 1.9190476190476188e-05,
+      "loss": 0.0755,
+      "step": 857
+    },
+    {
+      "epoch": 20.428571428571427,
+      "grad_norm": 0.45133450627326965,
+      "learning_rate": 1.9142857142857143e-05,
+      "loss": 0.1125,
+      "step": 858
+    },
+    {
+      "epoch": 20.452380952380953,
+      "grad_norm": 0.3273387551307678,
+      "learning_rate": 1.9095238095238095e-05,
+      "loss": 0.1023,
+      "step": 859
+    },
+    {
+      "epoch": 20.476190476190474,
+      "grad_norm": 0.2446010559797287,
+      "learning_rate": 1.9047619047619046e-05,
+      "loss": 0.0905,
+      "step": 860
+    },
+    {
+      "epoch": 20.476190476190474,
+      "eval_dice_score": 0.9241981800769812,
+      "eval_loss": 0.08037485182285309,
+      "eval_runtime": 8.7085,
+      "eval_samples_per_second": 10.22,
+      "eval_steps_per_second": 1.033,
+      "step": 860
+    },
+    {
+      "epoch": 20.5,
+      "grad_norm": 0.3432914912700653,
+      "learning_rate": 1.9e-05,
+      "loss": 0.0834,
+      "step": 861
+    },
+    {
+      "epoch": 20.523809523809526,
+      "grad_norm": 0.3280867040157318,
+      "learning_rate": 1.8952380952380953e-05,
+      "loss": 0.091,
+      "step": 862
+    },
+    {
+      "epoch": 20.547619047619047,
+      "grad_norm": 0.21562229096889496,
+      "learning_rate": 1.8904761904761904e-05,
+      "loss": 0.0803,
+      "step": 863
+    },
+    {
+      "epoch": 20.571428571428573,
+      "grad_norm": 0.22910885512828827,
+      "learning_rate": 1.8857142857142856e-05,
+      "loss": 0.0763,
+      "step": 864
+    },
+    {
+      "epoch": 20.595238095238095,
+      "grad_norm": 0.20336787402629852,
+      "learning_rate": 1.880952380952381e-05,
+      "loss": 0.0839,
+      "step": 865
+    },
+    {
+      "epoch": 20.61904761904762,
+      "grad_norm": 0.22064198553562164,
+      "learning_rate": 1.8761904761904762e-05,
+      "loss": 0.081,
+      "step": 866
+    },
+    {
+      "epoch": 20.642857142857142,
+      "grad_norm": 0.2525925040245056,
+      "learning_rate": 1.8714285714285714e-05,
+      "loss": 0.0853,
+      "step": 867
+    },
+    {
+      "epoch": 20.666666666666668,
+      "grad_norm": 0.21353179216384888,
+      "learning_rate": 1.866666666666667e-05,
+      "loss": 0.0862,
+      "step": 868
+    },
+    {
+      "epoch": 20.69047619047619,
+      "grad_norm": 0.17838771641254425,
+      "learning_rate": 1.861904761904762e-05,
+      "loss": 0.0809,
+      "step": 869
+    },
+    {
+      "epoch": 20.714285714285715,
+      "grad_norm": 0.2589264512062073,
+      "learning_rate": 1.8571428571428572e-05,
+      "loss": 0.1112,
+      "step": 870
+    },
+    {
+      "epoch": 20.738095238095237,
+      "grad_norm": 0.2731986939907074,
+      "learning_rate": 1.8523809523809527e-05,
+      "loss": 0.0861,
+      "step": 871
+    },
+    {
+      "epoch": 20.761904761904763,
+      "grad_norm": 0.18752317130565643,
+      "learning_rate": 1.8476190476190478e-05,
+      "loss": 0.068,
+      "step": 872
+    },
+    {
+      "epoch": 20.785714285714285,
+      "grad_norm": 0.18171530961990356,
+      "learning_rate": 1.842857142857143e-05,
+      "loss": 0.0722,
+      "step": 873
+    },
+    {
+      "epoch": 20.80952380952381,
+      "grad_norm": 0.20566533505916595,
+      "learning_rate": 1.838095238095238e-05,
+      "loss": 0.0669,
+      "step": 874
+    },
+    {
+      "epoch": 20.833333333333332,
+      "grad_norm": 0.2387693226337433,
+      "learning_rate": 1.8333333333333336e-05,
+      "loss": 0.0868,
+      "step": 875
+    },
+    {
+      "epoch": 20.857142857142858,
+      "grad_norm": 0.2408524453639984,
+      "learning_rate": 1.8285714285714288e-05,
+      "loss": 0.0716,
+      "step": 876
+    },
+    {
+      "epoch": 20.88095238095238,
+      "grad_norm": 0.35495293140411377,
+      "learning_rate": 1.8238095238095236e-05,
+      "loss": 0.0902,
+      "step": 877
+    },
+    {
+      "epoch": 20.904761904761905,
+      "grad_norm": 0.19206169247627258,
+      "learning_rate": 1.819047619047619e-05,
+      "loss": 0.0771,
+      "step": 878
+    },
+    {
+      "epoch": 20.928571428571427,
+      "grad_norm": 0.22178664803504944,
+      "learning_rate": 1.8142857142857142e-05,
+      "loss": 0.1119,
+      "step": 879
+    },
+    {
+      "epoch": 20.952380952380953,
+      "grad_norm": 0.1991724669933319,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.069,
+      "step": 880
+    },
+    {
+      "epoch": 20.952380952380953,
+      "eval_dice_score": 0.9220529684642296,
+      "eval_loss": 0.08112800121307373,
+      "eval_runtime": 8.7757,
+      "eval_samples_per_second": 10.142,
+      "eval_steps_per_second": 1.026,
+      "step": 880
+    },
+    {
+      "epoch": 20.976190476190474,
+      "grad_norm": 0.29708757996559143,
+      "learning_rate": 1.804761904761905e-05,
+      "loss": 0.0901,
+      "step": 881
+    },
+    {
+      "epoch": 21.0,
+      "grad_norm": 0.28050467371940613,
+      "learning_rate": 1.8e-05,
+      "loss": 0.1043,
+      "step": 882
+    },
+    {
+      "epoch": 21.023809523809526,
+      "grad_norm": 0.25332918763160706,
+      "learning_rate": 1.7952380952380952e-05,
+      "loss": 0.0872,
+      "step": 883
+    },
+    {
+      "epoch": 21.047619047619047,
+      "grad_norm": 0.21389389038085938,
+      "learning_rate": 1.7904761904761904e-05,
+      "loss": 0.1038,
+      "step": 884
+    },
+    {
+      "epoch": 21.071428571428573,
+      "grad_norm": 0.2038821429014206,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.075,
+      "step": 885
+    },
+    {
+      "epoch": 21.095238095238095,
+      "grad_norm": 0.19971756637096405,
+      "learning_rate": 1.780952380952381e-05,
+      "loss": 0.0827,
+      "step": 886
+    },
+    {
+      "epoch": 21.11904761904762,
+      "grad_norm": 0.2837202548980713,
+      "learning_rate": 1.776190476190476e-05,
+      "loss": 0.0781,
+      "step": 887
+    },
+    {
+      "epoch": 21.142857142857142,
+      "grad_norm": 0.24689999222755432,
+      "learning_rate": 1.7714285714285717e-05,
+      "loss": 0.0746,
+      "step": 888
+    },
+    {
+      "epoch": 21.166666666666668,
+      "grad_norm": 0.19299085438251495,
+      "learning_rate": 1.7666666666666668e-05,
+      "loss": 0.0936,
+      "step": 889
+    },
+    {
+      "epoch": 21.19047619047619,
+      "grad_norm": 0.18479064106941223,
+      "learning_rate": 1.761904761904762e-05,
+      "loss": 0.0919,
+      "step": 890
+    },
+    {
+      "epoch": 21.214285714285715,
+      "grad_norm": 0.24078267812728882,
+      "learning_rate": 1.757142857142857e-05,
+      "loss": 0.0836,
+      "step": 891
+    },
+    {
+      "epoch": 21.238095238095237,
+      "grad_norm": 0.33140161633491516,
+      "learning_rate": 1.7523809523809526e-05,
+      "loss": 0.0873,
+      "step": 892
+    },
+    {
+      "epoch": 21.261904761904763,
+      "grad_norm": 0.22419366240501404,
+      "learning_rate": 1.7476190476190478e-05,
+      "loss": 0.0821,
+      "step": 893
+    },
+    {
+      "epoch": 21.285714285714285,
+      "grad_norm": 0.16673584282398224,
+      "learning_rate": 1.742857142857143e-05,
+      "loss": 0.0778,
+      "step": 894
+    },
+    {
+      "epoch": 21.30952380952381,
+      "grad_norm": 0.19760103523731232,
+      "learning_rate": 1.7380952380952384e-05,
+      "loss": 0.0738,
+      "step": 895
+    },
+    {
+      "epoch": 21.333333333333332,
+      "grad_norm": 0.22858861088752747,
+      "learning_rate": 1.7333333333333332e-05,
+      "loss": 0.0628,
+      "step": 896
+    },
+    {
+      "epoch": 21.357142857142858,
+      "grad_norm": 0.27508506178855896,
+      "learning_rate": 1.7285714285714284e-05,
+      "loss": 0.0824,
+      "step": 897
+    },
+    {
+      "epoch": 21.38095238095238,
+      "grad_norm": 0.2053472250699997,
+      "learning_rate": 1.723809523809524e-05,
+      "loss": 0.0791,
+      "step": 898
+    },
+    {
+      "epoch": 21.404761904761905,
+      "grad_norm": 0.28713539242744446,
+      "learning_rate": 1.719047619047619e-05,
+      "loss": 0.1038,
+      "step": 899
+    },
+    {
+      "epoch": 21.428571428571427,
+      "grad_norm": 0.20974591374397278,
+      "learning_rate": 1.7142857142857142e-05,
+      "loss": 0.0816,
+      "step": 900
+    },
+    {
+      "epoch": 21.428571428571427,
+      "eval_dice_score": 0.92420542823889,
+      "eval_loss": 0.0804237425327301,
+      "eval_runtime": 9.1587,
+      "eval_samples_per_second": 9.718,
+      "eval_steps_per_second": 0.983,
+      "step": 900
+    },
+    {
+      "epoch": 21.452380952380953,
+      "grad_norm": 0.18460780382156372,
+      "learning_rate": 1.7095238095238093e-05,
+      "loss": 0.085,
+      "step": 901
+    },
+    {
+      "epoch": 21.476190476190474,
+      "grad_norm": 0.4044642150402069,
+      "learning_rate": 1.704761904761905e-05,
+      "loss": 0.085,
+      "step": 902
+    },
+    {
+      "epoch": 21.5,
+      "grad_norm": 0.32045164704322815,
+      "learning_rate": 1.7e-05,
+      "loss": 0.0902,
+      "step": 903
+    },
+    {
+      "epoch": 21.523809523809526,
+      "grad_norm": 0.19330991804599762,
+      "learning_rate": 1.695238095238095e-05,
+      "loss": 0.0726,
+      "step": 904
+    },
+    {
+      "epoch": 21.547619047619047,
+      "grad_norm": 0.2264699935913086,
+      "learning_rate": 1.6904761904761906e-05,
+      "loss": 0.0879,
+      "step": 905
+    },
+    {
+      "epoch": 21.571428571428573,
+      "grad_norm": 0.25664040446281433,
+      "learning_rate": 1.6857142857142858e-05,
+      "loss": 0.083,
+      "step": 906
+    },
+    {
+      "epoch": 21.595238095238095,
+      "grad_norm": 0.28455498814582825,
+      "learning_rate": 1.680952380952381e-05,
+      "loss": 0.103,
+      "step": 907
+    },
+    {
+      "epoch": 21.61904761904762,
+      "grad_norm": 0.22873304784297943,
+      "learning_rate": 1.6761904761904764e-05,
+      "loss": 0.0927,
+      "step": 908
+    },
+    {
+      "epoch": 21.642857142857142,
+      "grad_norm": 0.20388808846473694,
+      "learning_rate": 1.6714285714285716e-05,
+      "loss": 0.0739,
+      "step": 909
+    },
+    {
+      "epoch": 21.666666666666668,
+      "grad_norm": 0.23918978869915009,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0917,
+      "step": 910
+    },
+    {
+      "epoch": 21.69047619047619,
+      "grad_norm": 0.22957748174667358,
+      "learning_rate": 1.661904761904762e-05,
+      "loss": 0.0695,
+      "step": 911
+    },
+    {
+      "epoch": 21.714285714285715,
+      "grad_norm": 0.26606836915016174,
+      "learning_rate": 1.6571428571428574e-05,
+      "loss": 0.0953,
+      "step": 912
+    },
+    {
+      "epoch": 21.738095238095237,
+      "grad_norm": 0.29274919629096985,
+      "learning_rate": 1.6523809523809525e-05,
+      "loss": 0.0593,
+      "step": 913
+    },
+    {
+      "epoch": 21.761904761904763,
+      "grad_norm": 0.22514043748378754,
+      "learning_rate": 1.6476190476190477e-05,
+      "loss": 0.075,
+      "step": 914
+    },
+    {
+      "epoch": 21.785714285714285,
+      "grad_norm": 0.18376828730106354,
+      "learning_rate": 1.6428571428571432e-05,
+      "loss": 0.0791,
+      "step": 915
+    },
+    {
+      "epoch": 21.80952380952381,
+      "grad_norm": 0.4766561985015869,
+      "learning_rate": 1.638095238095238e-05,
+      "loss": 0.0971,
+      "step": 916
+    },
+    {
+      "epoch": 21.833333333333332,
+      "grad_norm": 0.2249252051115036,
+      "learning_rate": 1.633333333333333e-05,
+      "loss": 0.0792,
+      "step": 917
+    },
+    {
+      "epoch": 21.857142857142858,
+      "grad_norm": 0.3545120656490326,
+      "learning_rate": 1.6285714285714283e-05,
+      "loss": 0.1018,
+      "step": 918
+    },
+    {
+      "epoch": 21.88095238095238,
+      "grad_norm": 0.22790645062923431,
+      "learning_rate": 1.6238095238095238e-05,
+      "loss": 0.0781,
+      "step": 919
+    },
+    {
+      "epoch": 21.904761904761905,
+      "grad_norm": 0.1942400336265564,
+      "learning_rate": 1.619047619047619e-05,
+      "loss": 0.0828,
+      "step": 920
+    },
+    {
+      "epoch": 21.904761904761905,
+      "eval_dice_score": 0.9240529374274217,
+      "eval_loss": 0.08054715394973755,
+      "eval_runtime": 9.145,
+      "eval_samples_per_second": 9.732,
+      "eval_steps_per_second": 0.984,
+      "step": 920
+    },
+    {
+      "epoch": 21.928571428571427,
+      "grad_norm": 0.2710881531238556,
+      "learning_rate": 1.614285714285714e-05,
+      "loss": 0.0849,
+      "step": 921
+    },
+    {
+      "epoch": 21.952380952380953,
+      "grad_norm": 0.20673465728759766,
+      "learning_rate": 1.6095238095238096e-05,
+      "loss": 0.0931,
+      "step": 922
+    },
+    {
+      "epoch": 21.976190476190474,
+      "grad_norm": 0.37557193636894226,
+      "learning_rate": 1.6047619047619048e-05,
+      "loss": 0.0921,
+      "step": 923
+    },
+    {
+      "epoch": 22.0,
+      "grad_norm": 0.29974475502967834,
+      "learning_rate": 1.6e-05,
+      "loss": 0.0946,
+      "step": 924
+    },
+    {
+      "epoch": 22.023809523809526,
+      "grad_norm": 0.2729737162590027,
+      "learning_rate": 1.5952380952380954e-05,
+      "loss": 0.0637,
+      "step": 925
+    },
+    {
+      "epoch": 22.047619047619047,
+      "grad_norm": 0.26678407192230225,
+      "learning_rate": 1.5904761904761906e-05,
+      "loss": 0.0829,
+      "step": 926
+    },
+    {
+      "epoch": 22.071428571428573,
+      "grad_norm": 0.2640640139579773,
+      "learning_rate": 1.5857142857142857e-05,
+      "loss": 0.0854,
+      "step": 927
+    },
+    {
+      "epoch": 22.095238095238095,
+      "grad_norm": 0.27443549036979675,
+      "learning_rate": 1.580952380952381e-05,
+      "loss": 0.0846,
+      "step": 928
+    },
+    {
+      "epoch": 22.11904761904762,
+      "grad_norm": 0.24837443232536316,
+      "learning_rate": 1.5761904761904764e-05,
+      "loss": 0.0705,
+      "step": 929
+    },
+    {
+      "epoch": 22.142857142857142,
+      "grad_norm": 0.17277519404888153,
+      "learning_rate": 1.5714285714285715e-05,
+      "loss": 0.0791,
+      "step": 930
+    },
+    {
+      "epoch": 22.166666666666668,
+      "grad_norm": 0.17575153708457947,
+      "learning_rate": 1.5666666666666667e-05,
+      "loss": 0.0742,
+      "step": 931
+    },
+    {
+      "epoch": 22.19047619047619,
+      "grad_norm": 0.20902031660079956,
+      "learning_rate": 1.5619047619047622e-05,
+      "loss": 0.0801,
+      "step": 932
+    },
+    {
+      "epoch": 22.214285714285715,
+      "grad_norm": 0.2587524652481079,
+      "learning_rate": 1.5571428571428573e-05,
+      "loss": 0.0998,
+      "step": 933
+    },
+    {
+      "epoch": 22.238095238095237,
+      "grad_norm": 0.24326211214065552,
+      "learning_rate": 1.5523809523809525e-05,
+      "loss": 0.09,
+      "step": 934
+    },
+    {
+      "epoch": 22.261904761904763,
+      "grad_norm": 0.15226134657859802,
+      "learning_rate": 1.547619047619048e-05,
+      "loss": 0.0771,
+      "step": 935
+    },
+    {
+      "epoch": 22.285714285714285,
+      "grad_norm": 0.2282165288925171,
+      "learning_rate": 1.5428571428571428e-05,
+      "loss": 0.1008,
+      "step": 936
+    },
+    {
+      "epoch": 22.30952380952381,
+      "grad_norm": 0.6688697934150696,
+      "learning_rate": 1.538095238095238e-05,
+      "loss": 0.1117,
+      "step": 937
+    },
+    {
+      "epoch": 22.333333333333332,
+      "grad_norm": 0.16388161480426788,
+      "learning_rate": 1.533333333333333e-05,
+      "loss": 0.0918,
+      "step": 938
+    },
+    {
+      "epoch": 22.357142857142858,
+      "grad_norm": 0.24473772943019867,
+      "learning_rate": 1.5285714285714286e-05,
+      "loss": 0.0811,
+      "step": 939
+    },
+    {
+      "epoch": 22.38095238095238,
+      "grad_norm": 0.29383912682533264,
+      "learning_rate": 1.5238095238095238e-05,
+      "loss": 0.0964,
+      "step": 940
+    },
+    {
+      "epoch": 22.38095238095238,
+      "eval_dice_score": 0.9245331000875532,
+      "eval_loss": 0.08000662177801132,
+      "eval_runtime": 8.8153,
+      "eval_samples_per_second": 10.096,
+      "eval_steps_per_second": 1.021,
+      "step": 940
+    },
+    {
+      "epoch": 22.404761904761905,
+      "grad_norm": 0.28534939885139465,
+      "learning_rate": 1.519047619047619e-05,
+      "loss": 0.0847,
+      "step": 941
+    },
+    {
+      "epoch": 22.428571428571427,
+      "grad_norm": 0.18751578032970428,
+      "learning_rate": 1.5142857142857142e-05,
+      "loss": 0.0729,
+      "step": 942
+    },
+    {
+      "epoch": 22.452380952380953,
+      "grad_norm": 0.14760446548461914,
+      "learning_rate": 1.5095238095238096e-05,
+      "loss": 0.0786,
+      "step": 943
+    },
+    {
+      "epoch": 22.476190476190474,
+      "grad_norm": 0.40601739287376404,
+      "learning_rate": 1.5047619047619047e-05,
+      "loss": 0.0667,
+      "step": 944
+    },
+    {
+      "epoch": 22.5,
+      "grad_norm": 0.28885194659233093,
+      "learning_rate": 1.5e-05,
+      "loss": 0.0886,
+      "step": 945
+    },
+    {
+      "epoch": 22.523809523809526,
+      "grad_norm": 0.2508743703365326,
+      "learning_rate": 1.4952380952380954e-05,
+      "loss": 0.0827,
+      "step": 946
+    },
+    {
+      "epoch": 22.547619047619047,
+      "grad_norm": 0.2167244702577591,
+      "learning_rate": 1.4904761904761905e-05,
+      "loss": 0.0742,
+      "step": 947
+    },
+    {
+      "epoch": 22.571428571428573,
+      "grad_norm": 0.22318218648433685,
+      "learning_rate": 1.4857142857142858e-05,
+      "loss": 0.0786,
+      "step": 948
+    },
+    {
+      "epoch": 22.595238095238095,
+      "grad_norm": 0.27486732602119446,
+      "learning_rate": 1.480952380952381e-05,
+      "loss": 0.0884,
+      "step": 949
+    },
+    {
+      "epoch": 22.61904761904762,
+      "grad_norm": 0.2715624272823334,
+      "learning_rate": 1.4761904761904761e-05,
+      "loss": 0.0893,
+      "step": 950
+    },
+    {
+      "epoch": 22.642857142857142,
+      "grad_norm": 0.18250462412834167,
+      "learning_rate": 1.4714285714285715e-05,
+      "loss": 0.0962,
+      "step": 951
+    },
+    {
+      "epoch": 22.666666666666668,
+      "grad_norm": 0.42171528935432434,
+      "learning_rate": 1.4666666666666666e-05,
+      "loss": 0.0766,
+      "step": 952
+    },
+    {
+      "epoch": 22.69047619047619,
+      "grad_norm": 0.15866506099700928,
+      "learning_rate": 1.461904761904762e-05,
+      "loss": 0.0734,
+      "step": 953
+    },
+    {
+      "epoch": 22.714285714285715,
+      "grad_norm": 0.19888657331466675,
+      "learning_rate": 1.4571428571428571e-05,
+      "loss": 0.0944,
+      "step": 954
+    },
+    {
+      "epoch": 22.738095238095237,
+      "grad_norm": 0.20909306406974792,
+      "learning_rate": 1.4523809523809524e-05,
+      "loss": 0.0964,
+      "step": 955
+    },
+    {
+      "epoch": 22.761904761904763,
+      "grad_norm": 0.28159016370773315,
+      "learning_rate": 1.4476190476190478e-05,
+      "loss": 0.0901,
+      "step": 956
+    },
+    {
+      "epoch": 22.785714285714285,
+      "grad_norm": 0.2649635672569275,
+      "learning_rate": 1.4428571428571429e-05,
+      "loss": 0.0655,
+      "step": 957
+    },
+    {
+      "epoch": 22.80952380952381,
+      "grad_norm": 0.22923074662685394,
+      "learning_rate": 1.4380952380952382e-05,
+      "loss": 0.1028,
+      "step": 958
+    },
+    {
+      "epoch": 22.833333333333332,
+      "grad_norm": 0.225973978638649,
+      "learning_rate": 1.4333333333333334e-05,
+      "loss": 0.0863,
+      "step": 959
+    },
+    {
+      "epoch": 22.857142857142858,
+      "grad_norm": 0.2754770815372467,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.0893,
+      "step": 960
+    },
+    {
+      "epoch": 22.857142857142858,
+      "eval_dice_score": 0.925071224925649,
+      "eval_loss": 0.08001391589641571,
+      "eval_runtime": 8.8179,
+      "eval_samples_per_second": 10.093,
+      "eval_steps_per_second": 1.021,
+      "step": 960
+    },
+    {
+      "epoch": 22.88095238095238,
+      "grad_norm": 0.18644385039806366,
+      "learning_rate": 1.4238095238095239e-05,
+      "loss": 0.0813,
+      "step": 961
+    },
+    {
+      "epoch": 22.904761904761905,
+      "grad_norm": 0.18705585598945618,
+      "learning_rate": 1.419047619047619e-05,
+      "loss": 0.0642,
+      "step": 962
+    },
+    {
+      "epoch": 22.928571428571427,
+      "grad_norm": 0.2833781838417053,
+      "learning_rate": 1.4142857142857143e-05,
+      "loss": 0.0772,
+      "step": 963
+    },
+    {
+      "epoch": 22.952380952380953,
+      "grad_norm": 0.20696187019348145,
+      "learning_rate": 1.4095238095238095e-05,
+      "loss": 0.0883,
+      "step": 964
+    },
+    {
+      "epoch": 22.976190476190474,
+      "grad_norm": 0.24278227984905243,
+      "learning_rate": 1.4047619047619048e-05,
+      "loss": 0.0852,
+      "step": 965
+    },
+    {
+      "epoch": 23.0,
+      "grad_norm": 0.29513198137283325,
+      "learning_rate": 1.4e-05,
+      "loss": 0.0855,
+      "step": 966
+    },
+    {
+      "epoch": 23.023809523809526,
+      "grad_norm": 0.18746696412563324,
+      "learning_rate": 1.3952380952380953e-05,
+      "loss": 0.0707,
+      "step": 967
+    },
+    {
+      "epoch": 23.047619047619047,
+      "grad_norm": 0.21581074595451355,
+      "learning_rate": 1.3904761904761906e-05,
+      "loss": 0.0887,
+      "step": 968
+    },
+    {
+      "epoch": 23.071428571428573,
+      "grad_norm": 0.33836206793785095,
+      "learning_rate": 1.3857142857142858e-05,
+      "loss": 0.118,
+      "step": 969
+    },
+    {
+      "epoch": 23.095238095238095,
+      "grad_norm": 0.2581540048122406,
+      "learning_rate": 1.380952380952381e-05,
+      "loss": 0.0985,
+      "step": 970
+    },
+    {
+      "epoch": 23.11904761904762,
+      "grad_norm": 0.1527651846408844,
+      "learning_rate": 1.3761904761904761e-05,
+      "loss": 0.0695,
+      "step": 971
+    },
+    {
+      "epoch": 23.142857142857142,
+      "grad_norm": 0.17830391228199005,
+      "learning_rate": 1.3714285714285714e-05,
+      "loss": 0.1029,
+      "step": 972
+    },
+    {
+      "epoch": 23.166666666666668,
+      "grad_norm": 0.4796157479286194,
+      "learning_rate": 1.3666666666666667e-05,
+      "loss": 0.093,
+      "step": 973
+    },
+    {
+      "epoch": 23.19047619047619,
+      "grad_norm": 0.21920832991600037,
+      "learning_rate": 1.3619047619047619e-05,
+      "loss": 0.0717,
+      "step": 974
+    },
+    {
+      "epoch": 23.214285714285715,
+      "grad_norm": 0.456992506980896,
+      "learning_rate": 1.3571428571428572e-05,
+      "loss": 0.0958,
+      "step": 975
+    },
+    {
+      "epoch": 23.238095238095237,
+      "grad_norm": 0.44602781534194946,
+      "learning_rate": 1.3523809523809524e-05,
+      "loss": 0.1064,
+      "step": 976
+    },
+    {
+      "epoch": 23.261904761904763,
+      "grad_norm": 0.19102871417999268,
+      "learning_rate": 1.3476190476190477e-05,
+      "loss": 0.0885,
+      "step": 977
+    },
+    {
+      "epoch": 23.285714285714285,
+      "grad_norm": 0.22834639251232147,
+      "learning_rate": 1.342857142857143e-05,
+      "loss": 0.0987,
+      "step": 978
+    },
+    {
+      "epoch": 23.30952380952381,
+      "grad_norm": 0.160812109708786,
+      "learning_rate": 1.3380952380952382e-05,
+      "loss": 0.0712,
+      "step": 979
+    },
+    {
+      "epoch": 23.333333333333332,
+      "grad_norm": 0.25260743498802185,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.0709,
+      "step": 980
+    },
+    {
+      "epoch": 23.333333333333332,
+      "eval_dice_score": 0.9249947230101319,
+      "eval_loss": 0.07975644618272781,
+      "eval_runtime": 9.1651,
+      "eval_samples_per_second": 9.711,
+      "eval_steps_per_second": 0.982,
+      "step": 980
+    },
+    {
+      "epoch": 23.357142857142858,
+      "grad_norm": 0.32486334443092346,
+      "learning_rate": 1.3285714285714285e-05,
+      "loss": 0.0828,
+      "step": 981
+    },
+    {
+      "epoch": 23.38095238095238,
+      "grad_norm": 0.27001824975013733,
+      "learning_rate": 1.3238095238095238e-05,
+      "loss": 0.0746,
+      "step": 982
+    },
+    {
+      "epoch": 23.404761904761905,
+      "grad_norm": 0.20627065002918243,
+      "learning_rate": 1.3190476190476191e-05,
+      "loss": 0.0768,
+      "step": 983
+    },
+    {
+      "epoch": 23.428571428571427,
+      "grad_norm": 0.29352235794067383,
+      "learning_rate": 1.3142857142857143e-05,
+      "loss": 0.0788,
+      "step": 984
+    },
+    {
+      "epoch": 23.452380952380953,
+      "grad_norm": 0.23943500220775604,
+      "learning_rate": 1.3095238095238096e-05,
+      "loss": 0.07,
+      "step": 985
+    },
+    {
+      "epoch": 23.476190476190474,
+      "grad_norm": 0.22379229962825775,
+      "learning_rate": 1.3047619047619048e-05,
+      "loss": 0.0785,
+      "step": 986
+    },
+    {
+      "epoch": 23.5,
+      "grad_norm": 0.3175750970840454,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 0.0999,
+      "step": 987
+    },
+    {
+      "epoch": 23.523809523809526,
+      "grad_norm": 0.24017292261123657,
+      "learning_rate": 1.2952380952380954e-05,
+      "loss": 0.0867,
+      "step": 988
+    },
+    {
+      "epoch": 23.547619047619047,
+      "grad_norm": 0.33831965923309326,
+      "learning_rate": 1.2904761904761906e-05,
+      "loss": 0.0982,
+      "step": 989
+    },
+    {
+      "epoch": 23.571428571428573,
+      "grad_norm": 0.20093227922916412,
+      "learning_rate": 1.2857142857142857e-05,
+      "loss": 0.074,
+      "step": 990
+    },
+    {
+      "epoch": 23.595238095238095,
+      "grad_norm": 0.22127002477645874,
+      "learning_rate": 1.2809523809523809e-05,
+      "loss": 0.0726,
+      "step": 991
+    },
+    {
+      "epoch": 23.61904761904762,
+      "grad_norm": 0.3663376271724701,
+      "learning_rate": 1.2761904761904762e-05,
+      "loss": 0.0784,
+      "step": 992
+    },
+    {
+      "epoch": 23.642857142857142,
+      "grad_norm": 0.5978325009346008,
+      "learning_rate": 1.2714285714285714e-05,
+      "loss": 0.0979,
+      "step": 993
+    },
+    {
+      "epoch": 23.666666666666668,
+      "grad_norm": 0.24766002595424652,
+      "learning_rate": 1.2666666666666667e-05,
+      "loss": 0.0875,
+      "step": 994
+    },
+    {
+      "epoch": 23.69047619047619,
+      "grad_norm": 0.34553080797195435,
+      "learning_rate": 1.261904761904762e-05,
+      "loss": 0.0939,
+      "step": 995
+    },
+    {
+      "epoch": 23.714285714285715,
+      "grad_norm": 0.26458320021629333,
+      "learning_rate": 1.2571428571428572e-05,
+      "loss": 0.0937,
+      "step": 996
+    },
+    {
+      "epoch": 23.738095238095237,
+      "grad_norm": 0.24503840506076813,
+      "learning_rate": 1.2523809523809525e-05,
+      "loss": 0.0836,
+      "step": 997
+    },
+    {
+      "epoch": 23.761904761904763,
+      "grad_norm": 0.2026822566986084,
+      "learning_rate": 1.2476190476190476e-05,
+      "loss": 0.0808,
+      "step": 998
+    },
+    {
+      "epoch": 23.785714285714285,
+      "grad_norm": 0.19072376191616058,
+      "learning_rate": 1.242857142857143e-05,
+      "loss": 0.0735,
+      "step": 999
+    },
+    {
+      "epoch": 23.80952380952381,
+      "grad_norm": 0.39544475078582764,
+      "learning_rate": 1.2380952380952381e-05,
+      "loss": 0.0771,
+      "step": 1000
+    },
+    {
+      "epoch": 23.80952380952381,
+      "eval_dice_score": 0.9249777722999508,
+      "eval_loss": 0.07983684539794922,
+      "eval_runtime": 9.0081,
+      "eval_samples_per_second": 9.88,
+      "eval_steps_per_second": 0.999,
+      "step": 1000
+    },
+    {
+      "epoch": 23.833333333333332,
+      "grad_norm": 0.2170388102531433,
+      "learning_rate": 1.2333333333333333e-05,
+      "loss": 0.0795,
+      "step": 1001
+    },
+    {
+      "epoch": 23.857142857142858,
+      "grad_norm": 0.12621523439884186,
+      "learning_rate": 1.2285714285714286e-05,
+      "loss": 0.0775,
+      "step": 1002
+    },
+    {
+      "epoch": 23.88095238095238,
+      "grad_norm": 0.21630653738975525,
+      "learning_rate": 1.2238095238095237e-05,
+      "loss": 0.0609,
+      "step": 1003
+    },
+    {
+      "epoch": 23.904761904761905,
+      "grad_norm": 0.17754028737545013,
+      "learning_rate": 1.219047619047619e-05,
+      "loss": 0.0806,
+      "step": 1004
+    },
+    {
+      "epoch": 23.928571428571427,
+      "grad_norm": 0.24530617892742157,
+      "learning_rate": 1.2142857142857144e-05,
+      "loss": 0.0696,
+      "step": 1005
+    },
+    {
+      "epoch": 23.952380952380953,
+      "grad_norm": 0.4387490749359131,
+      "learning_rate": 1.2095238095238096e-05,
+      "loss": 0.0929,
+      "step": 1006
+    },
+    {
+      "epoch": 23.976190476190474,
+      "grad_norm": 0.19713687896728516,
+      "learning_rate": 1.2047619047619049e-05,
+      "loss": 0.073,
+      "step": 1007
+    },
+    {
+      "epoch": 24.0,
+      "grad_norm": 0.3298085927963257,
+      "learning_rate": 1.2e-05,
+      "loss": 0.0952,
+      "step": 1008
+    },
+    {
+      "epoch": 24.023809523809526,
+      "grad_norm": 0.13209277391433716,
+      "learning_rate": 1.1952380952380952e-05,
+      "loss": 0.0674,
+      "step": 1009
+    },
+    {
+      "epoch": 24.047619047619047,
+      "grad_norm": 0.25918281078338623,
+      "learning_rate": 1.1904761904761905e-05,
+      "loss": 0.0972,
+      "step": 1010
+    },
+    {
+      "epoch": 24.071428571428573,
+      "grad_norm": 0.18305575847625732,
+      "learning_rate": 1.1857142857142857e-05,
+      "loss": 0.0749,
+      "step": 1011
+    },
+    {
+      "epoch": 24.095238095238095,
+      "grad_norm": 0.22177988290786743,
+      "learning_rate": 1.180952380952381e-05,
+      "loss": 0.0948,
+      "step": 1012
+    },
+    {
+      "epoch": 24.11904761904762,
+      "grad_norm": 0.9742544293403625,
+      "learning_rate": 1.1761904761904761e-05,
+      "loss": 0.0748,
+      "step": 1013
+    },
+    {
+      "epoch": 24.142857142857142,
+      "grad_norm": 0.2539862096309662,
+      "learning_rate": 1.1714285714285715e-05,
+      "loss": 0.0737,
+      "step": 1014
+    },
+    {
+      "epoch": 24.166666666666668,
+      "grad_norm": 0.2597903311252594,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.0694,
+      "step": 1015
+    },
+    {
+      "epoch": 24.19047619047619,
+      "grad_norm": 0.2900591492652893,
+      "learning_rate": 1.161904761904762e-05,
+      "loss": 0.0904,
+      "step": 1016
+    },
+    {
+      "epoch": 24.214285714285715,
+      "grad_norm": 0.30920645594596863,
+      "learning_rate": 1.1571428571428573e-05,
+      "loss": 0.0873,
+      "step": 1017
+    },
+    {
+      "epoch": 24.238095238095237,
+      "grad_norm": 0.2939493954181671,
+      "learning_rate": 1.1523809523809524e-05,
+      "loss": 0.0803,
+      "step": 1018
+    },
+    {
+      "epoch": 24.261904761904763,
+      "grad_norm": 0.323430597782135,
+      "learning_rate": 1.1476190476190476e-05,
+      "loss": 0.0912,
+      "step": 1019
+    },
+    {
+      "epoch": 24.285714285714285,
+      "grad_norm": 0.17873886227607727,
+      "learning_rate": 1.1428571428571429e-05,
+      "loss": 0.0781,
+      "step": 1020
+    },
+    {
+      "epoch": 24.285714285714285,
+      "eval_dice_score": 0.9251687151008064,
+      "eval_loss": 0.07953391969203949,
+      "eval_runtime": 8.8432,
+      "eval_samples_per_second": 10.064,
+      "eval_steps_per_second": 1.018,
+      "step": 1020
+    },
+    {
+      "epoch": 24.30952380952381,
+      "grad_norm": 0.2174515575170517,
+      "learning_rate": 1.138095238095238e-05,
+      "loss": 0.0689,
+      "step": 1021
+    },
+    {
+      "epoch": 24.333333333333332,
+      "grad_norm": 0.25644710659980774,
+      "learning_rate": 1.1333333333333334e-05,
+      "loss": 0.0696,
+      "step": 1022
+    },
+    {
+      "epoch": 24.357142857142858,
+      "grad_norm": 0.3882044851779938,
+      "learning_rate": 1.1285714285714285e-05,
+      "loss": 0.0888,
+      "step": 1023
+    },
+    {
+      "epoch": 24.38095238095238,
+      "grad_norm": 0.23190882802009583,
+      "learning_rate": 1.1238095238095239e-05,
+      "loss": 0.0754,
+      "step": 1024
+    },
+    {
+      "epoch": 24.404761904761905,
+      "grad_norm": 0.17043378949165344,
+      "learning_rate": 1.119047619047619e-05,
+      "loss": 0.0791,
+      "step": 1025
+    },
+    {
+      "epoch": 24.428571428571427,
+      "grad_norm": 0.33094340562820435,
+      "learning_rate": 1.1142857142857143e-05,
+      "loss": 0.0838,
+      "step": 1026
+    },
+    {
+      "epoch": 24.452380952380953,
+      "grad_norm": 0.22381150722503662,
+      "learning_rate": 1.1095238095238097e-05,
+      "loss": 0.1096,
+      "step": 1027
+    },
+    {
+      "epoch": 24.476190476190474,
+      "grad_norm": 0.18582071363925934,
+      "learning_rate": 1.1047619047619048e-05,
+      "loss": 0.0733,
+      "step": 1028
+    },
+    {
+      "epoch": 24.5,
+      "grad_norm": 0.16228701174259186,
+      "learning_rate": 1.1e-05,
+      "loss": 0.0776,
+      "step": 1029
+    },
+    {
+      "epoch": 24.523809523809526,
+      "grad_norm": 0.4295779764652252,
+      "learning_rate": 1.0952380952380951e-05,
+      "loss": 0.0953,
+      "step": 1030
+    },
+    {
+      "epoch": 24.547619047619047,
+      "grad_norm": 0.2032005786895752,
+      "learning_rate": 1.0904761904761905e-05,
+      "loss": 0.0994,
+      "step": 1031
+    },
+    {
+      "epoch": 24.571428571428573,
+      "grad_norm": 0.4408031702041626,
+      "learning_rate": 1.0857142857142858e-05,
+      "loss": 0.0832,
+      "step": 1032
+    },
+    {
+      "epoch": 24.595238095238095,
+      "grad_norm": 0.2662723660469055,
+      "learning_rate": 1.080952380952381e-05,
+      "loss": 0.0718,
+      "step": 1033
+    },
+    {
+      "epoch": 24.61904761904762,
+      "grad_norm": 0.23538383841514587,
+      "learning_rate": 1.0761904761904763e-05,
+      "loss": 0.0834,
+      "step": 1034
+    },
+    {
+      "epoch": 24.642857142857142,
+      "grad_norm": 0.15239837765693665,
+      "learning_rate": 1.0714285714285714e-05,
+      "loss": 0.0809,
+      "step": 1035
+    },
+    {
+      "epoch": 24.666666666666668,
+      "grad_norm": 0.20802651345729828,
+      "learning_rate": 1.0666666666666667e-05,
+      "loss": 0.0951,
+      "step": 1036
+    },
+    {
+      "epoch": 24.69047619047619,
+      "grad_norm": 0.1992301344871521,
+      "learning_rate": 1.061904761904762e-05,
+      "loss": 0.0803,
+      "step": 1037
+    },
+    {
+      "epoch": 24.714285714285715,
+      "grad_norm": 0.24126474559307098,
+      "learning_rate": 1.0571428571428572e-05,
+      "loss": 0.0766,
+      "step": 1038
+    },
+    {
+      "epoch": 24.738095238095237,
+      "grad_norm": 0.2718803286552429,
+      "learning_rate": 1.0523809523809524e-05,
+      "loss": 0.0871,
+      "step": 1039
+    },
+    {
+      "epoch": 24.761904761904763,
+      "grad_norm": 0.27064719796180725,
+      "learning_rate": 1.0476190476190475e-05,
+      "loss": 0.0857,
+      "step": 1040
+    },
+    {
+      "epoch": 24.761904761904763,
+      "eval_dice_score": 0.9250793835928479,
+      "eval_loss": 0.07883252203464508,
+      "eval_runtime": 8.8276,
+      "eval_samples_per_second": 10.082,
+      "eval_steps_per_second": 1.02,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1260,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 30,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.8144939555894067e+17,
+  "train_batch_size": 10,
+  "trial_name": null,
+  "trial_params": null
+}