End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9863,
-    "eval_loss": 0.09992218017578125,
-    "eval_runtime": 87.3311,
-    "eval_samples_per_second": 114.507,
-    "eval_steps_per_second": 3.584,
+    "eval_accuracy": 0.9855,
+    "eval_loss": 0.10110656172037125,
+    "eval_runtime": 153.2781,
+    "eval_samples_per_second": 65.241,
+    "eval_steps_per_second": 2.042,
     "total_flos": 1.161843208194687e+19,
-    "train_loss": 0.5011348225112654,
-    "train_runtime": 3321.3479,
-    "train_samples_per_second": 45.162,
-    "train_steps_per_second": 0.352
+    "train_loss": 0.5034837816515539,
+    "train_runtime": 6176.4988,
+    "train_samples_per_second": 24.286,
+    "train_steps_per_second": 0.189
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 0.9863,
-    "eval_loss": 0.09992218017578125,
-    "eval_runtime": 87.3311,
-    "eval_samples_per_second": 114.507,
-    "eval_steps_per_second": 3.584
+    "eval_accuracy": 0.9855,
+    "eval_loss": 0.10110656172037125,
+    "eval_runtime": 153.2781,
+    "eval_samples_per_second": 65.241,
+    "eval_steps_per_second": 2.042
 }

runs/Feb14_15-28-49_f9ff78a7b58d/events.out.tfevents.1644859402.f9ff78a7b58d.82.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb36ac69f21bd4591d1a9404d1e7eae2769a696718bfb744c1ca7cf931622a7
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "total_flos": 1.161843208194687e+19,
-    "train_loss": 0.5011348225112654,
-    "train_runtime": 3321.3479,
-    "train_samples_per_second": 45.162,
-    "train_steps_per_second": 0.352
+    "train_loss": 0.5034837816515539,
+    "train_runtime": 6176.4988,
+    "train_samples_per_second": 24.286,
+    "train_steps_per_second": 0.189
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9863,
+  "best_metric": 0.9855,
   "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-cifar10/checkpoint-1170",
   "epoch": 2.998080614203455,
   "global_step": 1170,
@@ -10,145 +10,145 @@
     {
       "epoch": 0.03,
       "learning_rate": 4.273504273504274e-06,
-      "loss": 2.3003,
+      "loss": 2.3074,
       "step": 10
     },
     {
       "epoch": 0.05,
       "learning_rate": 8.547008547008548e-06,
-      "loss": 2.2805,
+      "loss": 2.2947,
       "step": 20
     },
     {
       "epoch": 0.08,
       "learning_rate": 1.282051282051282e-05,
-      "loss": 2.2424,
+      "loss": 2.2588,
       "step": 30
     },
     {
       "epoch": 0.1,
       "learning_rate": 1.7094017094017095e-05,
-      "loss": 2.1851,
+      "loss": 2.2041,
       "step": 40
     },
     {
       "epoch": 0.13,
       "learning_rate": 2.1367521367521368e-05,
-      "loss": 2.0954,
+      "loss": 2.1221,
       "step": 50
     },
     {
       "epoch": 0.15,
       "learning_rate": 2.564102564102564e-05,
-      "loss": 1.9605,
+      "loss": 2.0014,
       "step": 60
     },
     {
       "epoch": 0.18,
       "learning_rate": 2.9914529914529915e-05,
-      "loss": 1.8082,
+      "loss": 1.8365,
       "step": 70
     },
     {
       "epoch": 0.2,
       "learning_rate": 3.418803418803419e-05,
-      "loss": 1.6045,
+      "loss": 1.6374,
       "step": 80
     },
     {
       "epoch": 0.23,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 1.416,
+      "loss": 1.438,
       "step": 90
     },
     {
       "epoch": 0.26,
       "learning_rate": 4.2735042735042735e-05,
-      "loss": 1.2521,
+      "loss": 1.2628,
       "step": 100
     },
     {
       "epoch": 0.28,
       "learning_rate": 4.700854700854701e-05,
-      "loss": 1.1004,
+      "loss": 1.0934,
       "step": 110
     },
     {
       "epoch": 0.31,
       "learning_rate": 4.985754985754986e-05,
-      "loss": 0.9802,
+      "loss": 0.9678,
       "step": 120
     },
     {
       "epoch": 0.33,
       "learning_rate": 4.938271604938271e-05,
-      "loss": 0.8694,
+      "loss": 0.8619,
       "step": 130
     },
     {
       "epoch": 0.36,
       "learning_rate": 4.890788224121557e-05,
-      "loss": 0.8239,
+      "loss": 0.837,
       "step": 140
     },
     {
       "epoch": 0.38,
       "learning_rate": 4.8433048433048433e-05,
-      "loss": 0.726,
+      "loss": 0.7366,
       "step": 150
     },
     {
       "epoch": 0.41,
       "learning_rate": 4.7958214624881294e-05,
-      "loss": 0.6769,
+      "loss": 0.6799,
       "step": 160
     },
     {
       "epoch": 0.44,
       "learning_rate": 4.7483380816714154e-05,
-      "loss": 0.653,
+      "loss": 0.6741,
       "step": 170
     },
     {
       "epoch": 0.46,
       "learning_rate": 4.700854700854701e-05,
-      "loss": 0.6043,
+      "loss": 0.6262,
       "step": 180
     },
     {
       "epoch": 0.49,
       "learning_rate": 4.653371320037987e-05,
-      "loss": 0.6188,
+      "loss": 0.6355,
       "step": 190
     },
     {
       "epoch": 0.51,
       "learning_rate": 4.605887939221273e-05,
-      "loss": 0.5447,
+      "loss": 0.5781,
       "step": 200
     },
     {
       "epoch": 0.54,
       "learning_rate": 4.558404558404559e-05,
-      "loss": 0.5582,
+      "loss": 0.5671,
       "step": 210
     },
     {
       "epoch": 0.56,
       "learning_rate": 4.510921177587845e-05,
-      "loss": 0.525,
+      "loss": 0.5192,
       "step": 220
     },
     {
       "epoch": 0.59,
       "learning_rate": 4.463437796771131e-05,
-      "loss": 0.5545,
+      "loss": 0.5498,
       "step": 230
     },
     {
       "epoch": 0.61,
       "learning_rate": 4.415954415954416e-05,
-      "loss": 0.5203,
+      "loss": 0.507,
       "step": 240
     },
     {
@@ -160,590 +160,590 @@
     {
       "epoch": 0.67,
       "learning_rate": 4.3209876543209875e-05,
-      "loss": 0.4704,
+      "loss": 0.4586,
       "step": 260
     },
     {
       "epoch": 0.69,
       "learning_rate": 4.2735042735042735e-05,
-      "loss": 0.5016,
+      "loss": 0.4815,
       "step": 270
     },
     {
       "epoch": 0.72,
       "learning_rate": 4.2260208926875595e-05,
-      "loss": 0.4651,
+      "loss": 0.4656,
       "step": 280
     },
     {
       "epoch": 0.74,
       "learning_rate": 4.1785375118708455e-05,
-      "loss": 0.4805,
+      "loss": 0.4937,
       "step": 290
     },
     {
       "epoch": 0.77,
       "learning_rate": 4.131054131054131e-05,
-      "loss": 0.4609,
+      "loss": 0.4649,
       "step": 300
     },
     {
       "epoch": 0.79,
       "learning_rate": 4.083570750237417e-05,
-      "loss": 0.4392,
+      "loss": 0.4279,
       "step": 310
     },
     {
       "epoch": 0.82,
       "learning_rate": 4.036087369420703e-05,
-      "loss": 0.4761,
+      "loss": 0.4736,
       "step": 320
     },
     {
       "epoch": 0.84,
       "learning_rate": 3.988603988603989e-05,
-      "loss": 0.4517,
+      "loss": 0.4434,
       "step": 330
     },
     {
       "epoch": 0.87,
       "learning_rate": 3.941120607787275e-05,
-      "loss": 0.4361,
+      "loss": 0.4459,
       "step": 340
     },
     {
       "epoch": 0.9,
       "learning_rate": 3.893637226970561e-05,
-      "loss": 0.4287,
+      "loss": 0.4307,
       "step": 350
     },
     {
       "epoch": 0.92,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.4151,
+      "loss": 0.4029,
       "step": 360
     },
     {
       "epoch": 0.95,
       "learning_rate": 3.798670465337132e-05,
-      "loss": 0.4098,
+      "loss": 0.4007,
       "step": 370
     },
     {
       "epoch": 0.97,
       "learning_rate": 3.7511870845204176e-05,
-      "loss": 0.4236,
+      "loss": 0.4279,
       "step": 380
     },
     {
       "epoch": 1.0,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.3897,
+      "loss": 0.3831,
       "step": 390
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9757,
-      "eval_loss": 0.20828521251678467,
-      "eval_runtime": 85.3525,
-      "eval_samples_per_second": 117.161,
-      "eval_steps_per_second": 3.667,
+      "eval_accuracy": 0.978,
+      "eval_loss": 0.20566777884960175,
+      "eval_runtime": 152.329,
+      "eval_samples_per_second": 65.647,
+      "eval_steps_per_second": 2.055,
       "step": 390
     },
     {
       "epoch": 1.03,
       "learning_rate": 3.65622032288699e-05,
-      "loss": 0.4293,
+      "loss": 0.4258,
       "step": 400
     },
     {
       "epoch": 1.05,
       "learning_rate": 3.608736942070276e-05,
-      "loss": 0.3485,
+      "loss": 0.3416,
       "step": 410
     },
     {
       "epoch": 1.08,
       "learning_rate": 3.561253561253561e-05,
-      "loss": 0.387,
+      "loss": 0.3881,
       "step": 420
     },
     {
       "epoch": 1.1,
       "learning_rate": 3.513770180436847e-05,
-      "loss": 0.3417,
+      "loss": 0.3453,
       "step": 430
     },
     {
       "epoch": 1.13,
       "learning_rate": 3.466286799620133e-05,
-      "loss": 0.3836,
+      "loss": 0.3806,
       "step": 440
     },
     {
       "epoch": 1.15,
       "learning_rate": 3.418803418803419e-05,
-      "loss": 0.3623,
+      "loss": 0.361,
       "step": 450
     },
     {
       "epoch": 1.18,
       "learning_rate": 3.371320037986705e-05,
-      "loss": 0.3662,
+      "loss": 0.3834,
       "step": 460
     },
     {
       "epoch": 1.2,
       "learning_rate": 3.323836657169991e-05,
-      "loss": 0.3724,
+      "loss": 0.3696,
       "step": 470
     },
     {
       "epoch": 1.23,
       "learning_rate": 3.2763532763532764e-05,
-      "loss": 0.3278,
+      "loss": 0.3389,
       "step": 480
     },
     {
       "epoch": 1.26,
       "learning_rate": 3.2288698955365625e-05,
-      "loss": 0.3703,
+      "loss": 0.3671,
       "step": 490
     },
     {
       "epoch": 1.28,
       "learning_rate": 3.181386514719848e-05,
-      "loss": 0.3691,
+      "loss": 0.3685,
       "step": 500
     },
     {
       "epoch": 1.31,
       "learning_rate": 3.133903133903134e-05,
-      "loss": 0.3644,
+      "loss": 0.3533,
       "step": 510
     },
     {
       "epoch": 1.33,
       "learning_rate": 3.08641975308642e-05,
-      "loss": 0.3536,
+      "loss": 0.3539,
       "step": 520
     },
     {
       "epoch": 1.36,
       "learning_rate": 3.0389363722697055e-05,
-      "loss": 0.3383,
+      "loss": 0.3499,
       "step": 530
     },
     {
       "epoch": 1.38,
       "learning_rate": 2.9914529914529915e-05,
-      "loss": 0.3239,
+      "loss": 0.3246,
       "step": 540
     },
     {
       "epoch": 1.41,
       "learning_rate": 2.9439696106362775e-05,
-      "loss": 0.3155,
+      "loss": 0.3134,
       "step": 550
     },
     {
       "epoch": 1.44,
       "learning_rate": 2.8964862298195632e-05,
-      "loss": 0.3401,
+      "loss": 0.3425,
       "step": 560
     },
     {
       "epoch": 1.46,
       "learning_rate": 2.8490028490028492e-05,
-      "loss": 0.3221,
+      "loss": 0.3185,
       "step": 570
     },
     {
       "epoch": 1.49,
       "learning_rate": 2.8015194681861352e-05,
-      "loss": 0.3183,
+      "loss": 0.3215,
       "step": 580
     },
     {
       "epoch": 1.51,
       "learning_rate": 2.754036087369421e-05,
-      "loss": 0.3256,
+      "loss": 0.3268,
       "step": 590
     },
     {
       "epoch": 1.54,
       "learning_rate": 2.706552706552707e-05,
-      "loss": 0.3407,
+      "loss": 0.3439,
       "step": 600
     },
     {
       "epoch": 1.56,
       "learning_rate": 2.6590693257359926e-05,
-      "loss": 0.2972,
+      "loss": 0.3049,
       "step": 610
     },
     {
       "epoch": 1.59,
       "learning_rate": 2.611585944919278e-05,
-      "loss": 0.3366,
+      "loss": 0.3306,
       "step": 620
     },
     {
       "epoch": 1.61,
       "learning_rate": 2.564102564102564e-05,
-      "loss": 0.2942,
+      "loss": 0.3158,
       "step": 630
     },
     {
       "epoch": 1.64,
       "learning_rate": 2.51661918328585e-05,
-      "loss": 0.3066,
+      "loss": 0.3118,
       "step": 640
     },
     {
       "epoch": 1.67,
       "learning_rate": 2.4691358024691357e-05,
-      "loss": 0.3247,
+      "loss": 0.3312,
       "step": 650
     },
     {
       "epoch": 1.69,
       "learning_rate": 2.4216524216524217e-05,
-      "loss": 0.3191,
+      "loss": 0.3094,
       "step": 660
     },
     {
       "epoch": 1.72,
       "learning_rate": 2.3741690408357077e-05,
-      "loss": 0.3076,
+      "loss": 0.3064,
       "step": 670
     },
     {
       "epoch": 1.74,
       "learning_rate": 2.3266856600189934e-05,
-      "loss": 0.3037,
+      "loss": 0.3144,
       "step": 680
     },
     {
       "epoch": 1.77,
       "learning_rate": 2.2792022792022794e-05,
-      "loss": 0.2986,
+      "loss": 0.2858,
       "step": 690
     },
     {
       "epoch": 1.79,
       "learning_rate": 2.2317188983855654e-05,
-      "loss": 0.2659,
+      "loss": 0.2634,
       "step": 700
     },
     {
       "epoch": 1.82,
       "learning_rate": 2.184235517568851e-05,
-      "loss": 0.3203,
+      "loss": 0.31,
       "step": 710
     },
     {
       "epoch": 1.84,
       "learning_rate": 2.1367521367521368e-05,
-      "loss": 0.2952,
+      "loss": 0.3035,
       "step": 720
     },
     {
       "epoch": 1.87,
       "learning_rate": 2.0892687559354228e-05,
-      "loss": 0.3053,
+      "loss": 0.2942,
       "step": 730
     },
     {
       "epoch": 1.9,
       "learning_rate": 2.0417853751187084e-05,
-      "loss": 0.2889,
+      "loss": 0.2951,
       "step": 740
     },
     {
       "epoch": 1.92,
       "learning_rate": 1.9943019943019945e-05,
-      "loss": 0.2518,
+      "loss": 0.26,
       "step": 750
     },
     {
       "epoch": 1.95,
       "learning_rate": 1.9468186134852805e-05,
-      "loss": 0.2816,
+      "loss": 0.2861,
       "step": 760
     },
     {
       "epoch": 1.97,
       "learning_rate": 1.899335232668566e-05,
-      "loss": 0.2813,
+      "loss": 0.2839,
       "step": 770
     },
     {
       "epoch": 2.0,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.3045,
+      "loss": 0.3007,
       "step": 780
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.986,
-      "eval_loss": 0.11745991557836533,
-      "eval_runtime": 84.0012,
-      "eval_samples_per_second": 119.046,
-      "eval_steps_per_second": 3.726,
+      "eval_accuracy": 0.9845,
+      "eval_loss": 0.11994641274213791,
+      "eval_runtime": 153.4534,
+      "eval_samples_per_second": 65.166,
+      "eval_steps_per_second": 2.04,
       "step": 780
     },
     {
       "epoch": 2.03,
       "learning_rate": 1.804368471035138e-05,
-      "loss": 0.2973,
+      "loss": 0.3043,
       "step": 790
     },
     {
       "epoch": 2.05,
       "learning_rate": 1.7568850902184235e-05,
-      "loss": 0.27,
+      "loss": 0.2769,
       "step": 800
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.7094017094017095e-05,
-      "loss": 0.2932,
+      "loss": 0.2983,
       "step": 810
     },
     {
       "epoch": 2.1,
       "learning_rate": 1.6619183285849956e-05,
-      "loss": 0.3179,
+      "loss": 0.3186,
       "step": 820
     },
     {
       "epoch": 2.13,
       "learning_rate": 1.6144349477682812e-05,
-      "loss": 0.2581,
+      "loss": 0.2619,
       "step": 830
     },
     {
       "epoch": 2.15,
       "learning_rate": 1.566951566951567e-05,
-      "loss": 0.275,
+      "loss": 0.283,
       "step": 840
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.5194681861348528e-05,
-      "loss": 0.2882,
+      "loss": 0.287,
       "step": 850
     },
     {
       "epoch": 2.2,
       "learning_rate": 1.4719848053181388e-05,
-      "loss": 0.26,
+      "loss": 0.2631,
       "step": 860
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.4245014245014246e-05,
-      "loss": 0.2748,
+      "loss": 0.263,
       "step": 870
     },
     {
       "epoch": 2.26,
       "learning_rate": 1.3770180436847105e-05,
-      "loss": 0.2553,
+      "loss": 0.2554,
       "step": 880
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.3295346628679963e-05,
-      "loss": 0.2578,
+      "loss": 0.2702,
       "step": 890
     },
     {
       "epoch": 2.31,
       "learning_rate": 1.282051282051282e-05,
-      "loss": 0.3364,
+      "loss": 0.3224,
       "step": 900
     },
     {
       "epoch": 2.33,
       "learning_rate": 1.2345679012345678e-05,
-      "loss": 0.2927,
+      "loss": 0.2801,
       "step": 910
     },
     {
       "epoch": 2.36,
       "learning_rate": 1.1870845204178538e-05,
-      "loss": 0.2481,
+      "loss": 0.2521,
       "step": 920
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.1396011396011397e-05,
-      "loss": 0.2591,
+      "loss": 0.2613,
       "step": 930
     },
     {
       "epoch": 2.41,
       "learning_rate": 1.0921177587844255e-05,
-      "loss": 0.2648,
+      "loss": 0.2485,
       "step": 940
     },
     {
       "epoch": 2.44,
       "learning_rate": 1.0446343779677114e-05,
-      "loss": 0.2691,
+      "loss": 0.2672,
       "step": 950
     },
     {
       "epoch": 2.46,
       "learning_rate": 9.971509971509972e-06,
-      "loss": 0.2627,
+      "loss": 0.2695,
       "step": 960
     },
     {
       "epoch": 2.49,
       "learning_rate": 9.49667616334283e-06,
-      "loss": 0.2612,
+      "loss": 0.2593,
       "step": 970
     },
     {
       "epoch": 2.51,
       "learning_rate": 9.02184235517569e-06,
-      "loss": 0.2389,
+      "loss": 0.2402,
       "step": 980
     },
     {
       "epoch": 2.54,
       "learning_rate": 8.547008547008548e-06,
-      "loss": 0.2792,
+      "loss": 0.2759,
       "step": 990
     },
     {
       "epoch": 2.56,
       "learning_rate": 8.072174738841406e-06,
-      "loss": 0.2446,
+      "loss": 0.2654,
       "step": 1000
     },
     {
       "epoch": 2.59,
       "learning_rate": 7.597340930674264e-06,
-      "loss": 0.2305,
+      "loss": 0.2363,
       "step": 1010
     },
     {
       "epoch": 2.61,
       "learning_rate": 7.122507122507123e-06,
-      "loss": 0.2527,
+      "loss": 0.2553,
       "step": 1020
     },
     {
       "epoch": 2.64,
       "learning_rate": 6.6476733143399815e-06,
-      "loss": 0.2403,
+      "loss": 0.2384,
       "step": 1030
     },
     {
       "epoch": 2.67,
       "learning_rate": 6.172839506172839e-06,
-      "loss": 0.2254,
+      "loss": 0.2173,
       "step": 1040
     },
     {
       "epoch": 2.69,
       "learning_rate": 5.6980056980056985e-06,
-      "loss": 0.2879,
+      "loss": 0.2794,
       "step": 1050
     },
     {
       "epoch": 2.72,
       "learning_rate": 5.223171889838557e-06,
-      "loss": 0.2233,
+      "loss": 0.2293,
       "step": 1060
     },
     {
       "epoch": 2.74,
       "learning_rate": 4.748338081671415e-06,
-      "loss": 0.2258,
+      "loss": 0.2107,
       "step": 1070
     },
     {
       "epoch": 2.77,
       "learning_rate": 4.273504273504274e-06,
-      "loss": 0.2382,
+      "loss": 0.2387,
       "step": 1080
     },
     {
       "epoch": 2.79,
       "learning_rate": 3.798670465337132e-06,
-      "loss": 0.2564,
+      "loss": 0.2534,
       "step": 1090
     },
     {
       "epoch": 2.82,
       "learning_rate": 3.3238366571699908e-06,
-      "loss": 0.2513,
+      "loss": 0.2449,
       "step": 1100
     },
     {
       "epoch": 2.84,
       "learning_rate": 2.8490028490028492e-06,
-      "loss": 0.2152,
+      "loss": 0.2188,
       "step": 1110
     },
     {
       "epoch": 2.87,
       "learning_rate": 2.3741690408357077e-06,
-      "loss": 0.243,
+      "loss": 0.2444,
       "step": 1120
     },
     {
       "epoch": 2.9,
       "learning_rate": 1.899335232668566e-06,
-      "loss": 0.2701,
+      "loss": 0.2611,
       "step": 1130
     },
     {
       "epoch": 2.92,
       "learning_rate": 1.4245014245014246e-06,
-      "loss": 0.2531,
+      "loss": 0.2607,
       "step": 1140
     },
     {
       "epoch": 2.95,
       "learning_rate": 9.49667616334283e-07,
-      "loss": 0.2637,
+      "loss": 0.2601,
       "step": 1150
     },
     {
       "epoch": 2.97,
       "learning_rate": 4.748338081671415e-07,
-      "loss": 0.2424,
+      "loss": 0.2441,
       "step": 1160
     },
     {
       "epoch": 3.0,
       "learning_rate": 0.0,
-      "loss": 0.2524,
+      "loss": 0.2442,
       "step": 1170
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9863,
-      "eval_loss": 0.09992218017578125,
-      "eval_runtime": 83.7402,
-      "eval_samples_per_second": 119.417,
-      "eval_steps_per_second": 3.738,
+      "eval_accuracy": 0.9855,
+      "eval_loss": 0.10110656172037125,
+      "eval_runtime": 153.3402,
+      "eval_samples_per_second": 65.214,
+      "eval_steps_per_second": 2.041,
       "step": 1170
     },
     {
       "epoch": 3.0,
       "step": 1170,
       "total_flos": 1.161843208194687e+19,
-      "train_loss": 0.5011348225112654,
-      "train_runtime": 3321.3479,
-      "train_samples_per_second": 45.162,
-      "train_steps_per_second": 0.352
+      "train_loss": 0.5034837816515539,
+      "train_runtime": 6176.4988,
+      "train_samples_per_second": 24.286,
+      "train_steps_per_second": 0.189
     }
   ],
   "max_steps": 1170,