End of training

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 9.96,
-    "eval_accuracy": 0.9988649262202043,
-    "eval_loss": 0.00955167319625616,
-    "eval_runtime": 8.0129,
-    "eval_samples_per_second": 109.948,
-    "eval_steps_per_second": 13.853,
-    "train_loss": 0.0672107882744877,
-    "train_runtime": 1935.8008,
-    "train_samples_per_second": 40.929,
-    "train_steps_per_second": 0.728
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.005081297364085913,
+    "eval_runtime": 7.7113,
+    "eval_samples_per_second": 114.248,
+    "eval_steps_per_second": 14.395,
+    "train_loss": 0.0649210772510116,
+    "train_runtime": 2044.7769,
+    "train_samples_per_second": 38.748,
+    "train_steps_per_second": 0.69
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.96,
-    "eval_accuracy": 0.9988649262202043,
-    "eval_loss": 0.00955167319625616,
-    "eval_runtime": 8.0129,
-    "eval_samples_per_second": 109.948,
-    "eval_steps_per_second": 13.853
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.005081297364085913,
+    "eval_runtime": 7.7113,
+    "eval_samples_per_second": 114.248,
+    "eval_steps_per_second": 14.395
 }

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 9.96,
-    "train_loss": 0.0672107882744877,
-    "train_runtime": 1935.8008,
-    "train_samples_per_second": 40.929,
-    "train_steps_per_second": 0.728
+    "train_loss": 0.0649210772510116,
+    "train_runtime": 2044.7769,
+    "train_samples_per_second": 38.748,
+    "train_steps_per_second": 0.69
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9988649262202043,
-  "best_model_checkpoint": "convnext-large-224-22k-1k-BottomSportsCasual/checkpoint-849",
+  "best_metric": 1.0,
+  "best_model_checkpoint": "convnext-large-224-22k-1k-BottomSportsCasual/checkpoint-283",
   "epoch": 9.959636730575177,
   "global_step": 1410,
   "is_hyper_param_search": false,
@@ -10,947 +10,947 @@
     {
       "epoch": 0.07,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.6501,
+      "loss": 0.6592,
       "step": 10
     },
     {
       "epoch": 0.14,
       "learning_rate": 4.982078853046595e-05,
-      "loss": 0.2875,
+      "loss": 0.2847,
       "step": 20
     },
     {
       "epoch": 0.21,
       "learning_rate": 4.9462365591397855e-05,
-      "loss": 0.1896,
+      "loss": 0.1765,
       "step": 30
     },
     {
       "epoch": 0.28,
       "learning_rate": 4.910394265232976e-05,
-      "loss": 0.1512,
+      "loss": 0.1438,
       "step": 40
     },
     {
       "epoch": 0.35,
       "learning_rate": 4.874551971326165e-05,
-      "loss": 0.1433,
+      "loss": 0.0968,
       "step": 50
     },
     {
       "epoch": 0.42,
       "learning_rate": 4.8387096774193554e-05,
-      "loss": 0.1148,
+      "loss": 0.1287,
       "step": 60
     },
     {
       "epoch": 0.49,
       "learning_rate": 4.802867383512545e-05,
-      "loss": 0.1106,
+      "loss": 0.1062,
       "step": 70
     },
     {
       "epoch": 0.57,
       "learning_rate": 4.767025089605735e-05,
-      "loss": 0.1079,
+      "loss": 0.1034,
       "step": 80
     },
     {
       "epoch": 0.64,
       "learning_rate": 4.731182795698925e-05,
-      "loss": 0.1282,
+      "loss": 0.1068,
       "step": 90
     },
     {
       "epoch": 0.71,
       "learning_rate": 4.695340501792115e-05,
-      "loss": 0.1306,
+      "loss": 0.0912,
       "step": 100
     },
     {
       "epoch": 0.78,
       "learning_rate": 4.659498207885305e-05,
-      "loss": 0.0917,
+      "loss": 0.0704,
       "step": 110
     },
     {
       "epoch": 0.85,
       "learning_rate": 4.6236559139784944e-05,
-      "loss": 0.0934,
+      "loss": 0.1192,
       "step": 120
     },
     {
       "epoch": 0.92,
       "learning_rate": 4.5878136200716846e-05,
-      "loss": 0.068,
+      "loss": 0.0991,
       "step": 130
     },
     {
       "epoch": 0.99,
       "learning_rate": 4.551971326164875e-05,
-      "loss": 0.0752,
+      "loss": 0.0732,
       "step": 140
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9920544835414302,
-      "eval_loss": 0.028408855199813843,
-      "eval_runtime": 13.4966,
-      "eval_samples_per_second": 65.276,
-      "eval_steps_per_second": 8.224,
+      "eval_accuracy": 0.9943246311010215,
+      "eval_loss": 0.011665264144539833,
+      "eval_runtime": 9.6566,
+      "eval_samples_per_second": 91.233,
+      "eval_steps_per_second": 11.495,
       "step": 141
     },
     {
       "epoch": 1.06,
       "learning_rate": 4.516129032258064e-05,
-      "loss": 0.0647,
+      "loss": 0.0578,
       "step": 150
     },
     {
       "epoch": 1.13,
       "learning_rate": 4.4802867383512545e-05,
-      "loss": 0.1433,
+      "loss": 0.1351,
       "step": 160
     },
     {
       "epoch": 1.2,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.0918,
+      "loss": 0.1143,
       "step": 170
     },
     {
       "epoch": 1.27,
       "learning_rate": 4.408602150537635e-05,
-      "loss": 0.0898,
+      "loss": 0.1396,
       "step": 180
     },
     {
       "epoch": 1.34,
       "learning_rate": 4.372759856630825e-05,
-      "loss": 0.0849,
+      "loss": 0.0934,
       "step": 190
     },
     {
       "epoch": 1.41,
       "learning_rate": 4.3369175627240145e-05,
-      "loss": 0.0816,
+      "loss": 0.1086,
       "step": 200
     },
     {
       "epoch": 1.48,
       "learning_rate": 4.301075268817205e-05,
-      "loss": 0.0793,
+      "loss": 0.0862,
       "step": 210
     },
     {
       "epoch": 1.55,
       "learning_rate": 4.265232974910394e-05,
-      "loss": 0.1098,
+      "loss": 0.0786,
       "step": 220
     },
     {
       "epoch": 1.62,
       "learning_rate": 4.2293906810035844e-05,
-      "loss": 0.0801,
+      "loss": 0.1058,
       "step": 230
     },
     {
       "epoch": 1.7,
       "learning_rate": 4.1935483870967746e-05,
-      "loss": 0.096,
+      "loss": 0.0606,
       "step": 240
     },
     {
       "epoch": 1.77,
       "learning_rate": 4.157706093189964e-05,
-      "loss": 0.0685,
+      "loss": 0.0776,
       "step": 250
     },
     {
       "epoch": 1.84,
       "learning_rate": 4.121863799283154e-05,
-      "loss": 0.069,
+      "loss": 0.0909,
       "step": 260
     },
     {
       "epoch": 1.91,
       "learning_rate": 4.0860215053763444e-05,
-      "loss": 0.0788,
+      "loss": 0.0704,
       "step": 270
     },
     {
       "epoch": 1.98,
       "learning_rate": 4.050179211469534e-05,
-      "loss": 0.0711,
+      "loss": 0.0664,
       "step": 280
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.996594778660613,
-      "eval_loss": 0.007982436567544937,
-      "eval_runtime": 10.1876,
-      "eval_samples_per_second": 86.477,
-      "eval_steps_per_second": 10.896,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.005081297364085913,
+      "eval_runtime": 9.5071,
+      "eval_samples_per_second": 92.667,
+      "eval_steps_per_second": 11.675,
       "step": 283
     },
     {
       "epoch": 2.05,
       "learning_rate": 4.014336917562724e-05,
-      "loss": 0.0824,
+      "loss": 0.0667,
       "step": 290
     },
     {
       "epoch": 2.12,
       "learning_rate": 3.978494623655914e-05,
-      "loss": 0.0357,
+      "loss": 0.072,
       "step": 300
     },
     {
       "epoch": 2.19,
       "learning_rate": 3.9426523297491045e-05,
-      "loss": 0.0834,
+      "loss": 0.078,
       "step": 310
     },
     {
       "epoch": 2.26,
       "learning_rate": 3.906810035842295e-05,
-      "loss": 0.0925,
+      "loss": 0.0978,
       "step": 320
     },
     {
       "epoch": 2.33,
       "learning_rate": 3.870967741935484e-05,
-      "loss": 0.1001,
+      "loss": 0.0667,
       "step": 330
     },
     {
       "epoch": 2.4,
       "learning_rate": 3.8351254480286743e-05,
-      "loss": 0.0533,
+      "loss": 0.0646,
       "step": 340
     },
     {
       "epoch": 2.47,
       "learning_rate": 3.799283154121864e-05,
-      "loss": 0.0652,
+      "loss": 0.0509,
       "step": 350
     },
     {
       "epoch": 2.54,
       "learning_rate": 3.763440860215054e-05,
-      "loss": 0.0738,
+      "loss": 0.0676,
       "step": 360
     },
     {
       "epoch": 2.61,
       "learning_rate": 3.727598566308244e-05,
-      "loss": 0.1049,
+      "loss": 0.0474,
       "step": 370
     },
     {
       "epoch": 2.68,
       "learning_rate": 3.691756272401434e-05,
-      "loss": 0.0918,
+      "loss": 0.084,
       "step": 380
     },
     {
       "epoch": 2.75,
       "learning_rate": 3.655913978494624e-05,
-      "loss": 0.0882,
+      "loss": 0.0636,
       "step": 390
     },
     {
       "epoch": 2.83,
       "learning_rate": 3.6200716845878134e-05,
-      "loss": 0.0636,
+      "loss": 0.102,
       "step": 400
     },
     {
       "epoch": 2.9,
       "learning_rate": 3.5842293906810036e-05,
-      "loss": 0.094,
+      "loss": 0.0863,
       "step": 410
     },
     {
       "epoch": 2.97,
       "learning_rate": 3.548387096774194e-05,
-      "loss": 0.074,
+      "loss": 0.065,
       "step": 420
     },
     {
       "epoch": 2.99,
-      "eval_accuracy": 0.996594778660613,
-      "eval_loss": 0.007469804957509041,
-      "eval_runtime": 10.2147,
-      "eval_samples_per_second": 86.248,
-      "eval_steps_per_second": 10.867,
+      "eval_accuracy": 0.9988649262202043,
+      "eval_loss": 0.004680068697780371,
+      "eval_runtime": 9.346,
+      "eval_samples_per_second": 94.265,
+      "eval_steps_per_second": 11.877,
       "step": 424
     },
     {
       "epoch": 3.04,
       "learning_rate": 3.512544802867383e-05,
-      "loss": 0.0748,
+      "loss": 0.0655,
       "step": 430
     },
     {
       "epoch": 3.11,
       "learning_rate": 3.4767025089605734e-05,
-      "loss": 0.0573,
+      "loss": 0.0511,
       "step": 440
     },
     {
       "epoch": 3.18,
       "learning_rate": 3.4408602150537636e-05,
-      "loss": 0.101,
+      "loss": 0.0829,
       "step": 450
     },
     {
       "epoch": 3.25,
       "learning_rate": 3.405017921146954e-05,
-      "loss": 0.0578,
+      "loss": 0.0413,
       "step": 460
     },
     {
       "epoch": 3.32,
       "learning_rate": 3.369175627240144e-05,
-      "loss": 0.0723,
+      "loss": 0.0587,
       "step": 470
     },
     {
       "epoch": 3.39,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0327,
+      "loss": 0.077,
       "step": 480
     },
     {
       "epoch": 3.46,
       "learning_rate": 3.297491039426524e-05,
-      "loss": 0.0735,
+      "loss": 0.0637,
       "step": 490
     },
     {
       "epoch": 3.53,
       "learning_rate": 3.261648745519714e-05,
-      "loss": 0.1024,
+      "loss": 0.0822,
       "step": 500
     },
     {
       "epoch": 3.6,
       "learning_rate": 3.2258064516129034e-05,
-      "loss": 0.0746,
+      "loss": 0.0491,
       "step": 510
     },
     {
       "epoch": 3.67,
       "learning_rate": 3.1899641577060935e-05,
-      "loss": 0.0762,
+      "loss": 0.0638,
       "step": 520
     },
     {
       "epoch": 3.74,
       "learning_rate": 3.154121863799283e-05,
-      "loss": 0.0392,
+      "loss": 0.0866,
       "step": 530
     },
     {
       "epoch": 3.81,
       "learning_rate": 3.118279569892473e-05,
-      "loss": 0.0389,
+      "loss": 0.0516,
       "step": 540
     },
     {
       "epoch": 3.88,
       "learning_rate": 3.0824372759856634e-05,
-      "loss": 0.091,
+      "loss": 0.0443,
       "step": 550
     },
     {
       "epoch": 3.96,
       "learning_rate": 3.046594982078853e-05,
-      "loss": 0.0696,
+      "loss": 0.0635,
       "step": 560
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.996594778660613,
-      "eval_loss": 0.0062737599946558475,
-      "eval_runtime": 9.8804,
-      "eval_samples_per_second": 89.166,
-      "eval_steps_per_second": 11.234,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0028926546219736338,
+      "eval_runtime": 9.7296,
+      "eval_samples_per_second": 90.549,
+      "eval_steps_per_second": 11.409,
       "step": 566
     },
     {
       "epoch": 4.03,
       "learning_rate": 3.010752688172043e-05,
-      "loss": 0.0496,
+      "loss": 0.0762,
       "step": 570
     },
     {
       "epoch": 4.1,
       "learning_rate": 2.974910394265233e-05,
-      "loss": 0.0279,
+      "loss": 0.0427,
       "step": 580
     },
     {
       "epoch": 4.17,
       "learning_rate": 2.939068100358423e-05,
-      "loss": 0.0647,
+      "loss": 0.0484,
       "step": 590
     },
     {
       "epoch": 4.24,
       "learning_rate": 2.9032258064516133e-05,
-      "loss": 0.052,
+      "loss": 0.0608,
       "step": 600
     },
     {
       "epoch": 4.31,
       "learning_rate": 2.8673835125448028e-05,
-      "loss": 0.0606,
+      "loss": 0.04,
       "step": 610
     },
     {
       "epoch": 4.38,
       "learning_rate": 2.831541218637993e-05,
-      "loss": 0.0843,
+      "loss": 0.0436,
       "step": 620
     },
     {
       "epoch": 4.45,
       "learning_rate": 2.7956989247311828e-05,
-      "loss": 0.0687,
+      "loss": 0.0418,
       "step": 630
     },
     {
       "epoch": 4.52,
       "learning_rate": 2.759856630824373e-05,
-      "loss": 0.0473,
+      "loss": 0.0385,
       "step": 640
     },
     {
       "epoch": 4.59,
       "learning_rate": 2.7240143369175632e-05,
-      "loss": 0.0448,
+      "loss": 0.0511,
       "step": 650
     },
     {
       "epoch": 4.66,
       "learning_rate": 2.6881720430107527e-05,
-      "loss": 0.0504,
+      "loss": 0.061,
       "step": 660
     },
     {
       "epoch": 4.73,
       "learning_rate": 2.652329749103943e-05,
-      "loss": 0.0595,
+      "loss": 0.0256,
       "step": 670
     },
     {
       "epoch": 4.8,
       "learning_rate": 2.616487455197133e-05,
-      "loss": 0.0721,
+      "loss": 0.0679,
       "step": 680
     },
     {
       "epoch": 4.87,
       "learning_rate": 2.5806451612903226e-05,
-      "loss": 0.0709,
+      "loss": 0.0461,
       "step": 690
     },
     {
       "epoch": 4.94,
       "learning_rate": 2.5448028673835127e-05,
-      "loss": 0.0628,
+      "loss": 0.0357,
       "step": 700
     },
     {
       "epoch": 4.99,
-      "eval_accuracy": 0.9954597048808173,
-      "eval_loss": 0.013796533457934856,
-      "eval_runtime": 10.1728,
-      "eval_samples_per_second": 86.603,
-      "eval_steps_per_second": 10.911,
+      "eval_accuracy": 0.996594778660613,
+      "eval_loss": 0.0047698114067316055,
+      "eval_runtime": 9.8714,
+      "eval_samples_per_second": 89.248,
+      "eval_steps_per_second": 11.245,
       "step": 707
     },
     {
       "epoch": 5.02,
       "learning_rate": 2.5089605734767026e-05,
-      "loss": 0.0601,
+      "loss": 0.0668,
       "step": 710
     },
     {
       "epoch": 5.09,
       "learning_rate": 2.4731182795698928e-05,
-      "loss": 0.0512,
+      "loss": 0.0446,
       "step": 720
     },
     {
       "epoch": 5.16,
       "learning_rate": 2.4372759856630826e-05,
-      "loss": 0.0469,
+      "loss": 0.0561,
       "step": 730
     },
     {
       "epoch": 5.23,
       "learning_rate": 2.4014336917562724e-05,
-      "loss": 0.0343,
+      "loss": 0.0538,
       "step": 740
     },
     {
       "epoch": 5.3,
       "learning_rate": 2.3655913978494626e-05,
-      "loss": 0.0334,
+      "loss": 0.0408,
       "step": 750
     },
     {
       "epoch": 5.37,
       "learning_rate": 2.3297491039426525e-05,
-      "loss": 0.0525,
+      "loss": 0.0373,
       "step": 760
     },
     {
       "epoch": 5.44,
       "learning_rate": 2.2939068100358423e-05,
-      "loss": 0.0639,
+      "loss": 0.0336,
       "step": 770
     },
     {
       "epoch": 5.51,
       "learning_rate": 2.258064516129032e-05,
-      "loss": 0.0408,
+      "loss": 0.0571,
       "step": 780
     },
     {
       "epoch": 5.58,
       "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.0327,
+      "loss": 0.0609,
       "step": 790
     },
     {
       "epoch": 5.65,
       "learning_rate": 2.1863799283154125e-05,
-      "loss": 0.0571,
+      "loss": 0.0689,
       "step": 800
     },
     {
       "epoch": 5.72,
       "learning_rate": 2.1505376344086024e-05,
-      "loss": 0.0681,
+      "loss": 0.0654,
       "step": 810
     },
     {
       "epoch": 5.79,
       "learning_rate": 2.1146953405017922e-05,
-      "loss": 0.0589,
+      "loss": 0.0341,
       "step": 820
     },
     {
       "epoch": 5.86,
       "learning_rate": 2.078853046594982e-05,
-      "loss": 0.0515,
+      "loss": 0.0383,
       "step": 830
     },
     {
       "epoch": 5.93,
       "learning_rate": 2.0430107526881722e-05,
-      "loss": 0.0402,
+      "loss": 0.0826,
       "step": 840
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9988649262202043,
-      "eval_loss": 0.00955167319625616,
-      "eval_runtime": 9.9549,
-      "eval_samples_per_second": 88.499,
-      "eval_steps_per_second": 11.15,
+      "eval_loss": 0.00289411679841578,
+      "eval_runtime": 10.3189,
+      "eval_samples_per_second": 85.378,
+      "eval_steps_per_second": 10.757,
       "step": 849
     },
     {
       "epoch": 6.0,
       "learning_rate": 2.007168458781362e-05,
-      "loss": 0.0401,
+      "loss": 0.072,
       "step": 850
     },
     {
       "epoch": 6.07,
       "learning_rate": 1.9713261648745522e-05,
-      "loss": 0.0595,
+      "loss": 0.043,
       "step": 860
     },
     {
       "epoch": 6.15,
       "learning_rate": 1.935483870967742e-05,
-      "loss": 0.0518,
+      "loss": 0.038,
       "step": 870
     },
     {
       "epoch": 6.22,
       "learning_rate": 1.899641577060932e-05,
-      "loss": 0.0354,
+      "loss": 0.061,
       "step": 880
     },
     {
       "epoch": 6.29,
       "learning_rate": 1.863799283154122e-05,
-      "loss": 0.0325,
+      "loss": 0.0468,
       "step": 890
     },
     {
       "epoch": 6.36,
       "learning_rate": 1.827956989247312e-05,
-      "loss": 0.0453,
+      "loss": 0.0689,
       "step": 900
     },
     {
       "epoch": 6.43,
       "learning_rate": 1.7921146953405018e-05,
-      "loss": 0.0283,
+      "loss": 0.0671,
       "step": 910
     },
     {
       "epoch": 6.5,
       "learning_rate": 1.7562724014336916e-05,
-      "loss": 0.0439,
+      "loss": 0.0648,
       "step": 920
     },
     {
       "epoch": 6.57,
       "learning_rate": 1.7204301075268818e-05,
-      "loss": 0.0368,
+      "loss": 0.0502,
       "step": 930
     },
     {
       "epoch": 6.64,
       "learning_rate": 1.684587813620072e-05,
-      "loss": 0.0561,
+      "loss": 0.052,
       "step": 940
     },
     {
       "epoch": 6.71,
       "learning_rate": 1.648745519713262e-05,
-      "loss": 0.054,
+      "loss": 0.0363,
       "step": 950
     },
     {
       "epoch": 6.78,
       "learning_rate": 1.6129032258064517e-05,
-      "loss": 0.0624,
+      "loss": 0.0261,
       "step": 960
     },
     {
       "epoch": 6.85,
       "learning_rate": 1.5770609318996415e-05,
-      "loss": 0.0315,
+      "loss": 0.0413,
       "step": 970
     },
     {
       "epoch": 6.92,
       "learning_rate": 1.5412186379928317e-05,
-      "loss": 0.0625,
+      "loss": 0.0295,
       "step": 980
     },
     {
       "epoch": 6.99,
       "learning_rate": 1.5053763440860215e-05,
-      "loss": 0.062,
+      "loss": 0.0177,
       "step": 990
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9977298524404086,
-      "eval_loss": 0.00731318723410368,
-      "eval_runtime": 9.9651,
-      "eval_samples_per_second": 88.409,
-      "eval_steps_per_second": 11.139,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0018128241645172238,
+      "eval_runtime": 9.7183,
+      "eval_samples_per_second": 90.654,
+      "eval_steps_per_second": 11.422,
       "step": 991
     },
     {
       "epoch": 7.06,
       "learning_rate": 1.4695340501792116e-05,
-      "loss": 0.0506,
+      "loss": 0.0518,
       "step": 1000
     },
     {
       "epoch": 7.13,
       "learning_rate": 1.4336917562724014e-05,
-      "loss": 0.0394,
+      "loss": 0.0421,
       "step": 1010
     },
     {
       "epoch": 7.2,
       "learning_rate": 1.3978494623655914e-05,
-      "loss": 0.0458,
+      "loss": 0.026,
       "step": 1020
     },
     {
       "epoch": 7.28,
       "learning_rate": 1.3620071684587816e-05,
-      "loss": 0.0257,
+      "loss": 0.0234,
       "step": 1030
     },
     {
       "epoch": 7.35,
       "learning_rate": 1.3261648745519714e-05,
-      "loss": 0.0312,
+      "loss": 0.0433,
       "step": 1040
     },
     {
       "epoch": 7.42,
       "learning_rate": 1.2903225806451613e-05,
-      "loss": 0.0324,
+      "loss": 0.0599,
       "step": 1050
     },
     {
       "epoch": 7.49,
       "learning_rate": 1.2544802867383513e-05,
-      "loss": 0.0416,
+      "loss": 0.0447,
       "step": 1060
     },
     {
       "epoch": 7.56,
       "learning_rate": 1.2186379928315413e-05,
-      "loss": 0.0309,
+      "loss": 0.0178,
       "step": 1070
     },
     {
       "epoch": 7.63,
       "learning_rate": 1.1827956989247313e-05,
-      "loss": 0.0201,
+      "loss": 0.0471,
       "step": 1080
     },
     {
       "epoch": 7.7,
       "learning_rate": 1.1469534050179212e-05,
-      "loss": 0.0531,
+      "loss": 0.0539,
       "step": 1090
     },
     {
       "epoch": 7.77,
       "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.019,
+      "loss": 0.031,
       "step": 1100
     },
     {
       "epoch": 7.84,
       "learning_rate": 1.0752688172043012e-05,
-      "loss": 0.0585,
+      "loss": 0.0202,
       "step": 1110
     },
     {
       "epoch": 7.91,
       "learning_rate": 1.039426523297491e-05,
-      "loss": 0.0464,
+      "loss": 0.0374,
       "step": 1120
     },
     {
       "epoch": 7.98,
       "learning_rate": 1.003584229390681e-05,
-      "loss": 0.021,
+      "loss": 0.0353,
       "step": 1130
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9977298524404086,
-      "eval_loss": 0.007844115607440472,
-      "eval_runtime": 9.6256,
-      "eval_samples_per_second": 91.527,
-      "eval_steps_per_second": 11.532,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0006858365959487855,
+      "eval_runtime": 9.4026,
+      "eval_samples_per_second": 93.698,
+      "eval_steps_per_second": 11.805,
       "step": 1132
     },
     {
       "epoch": 8.05,
       "learning_rate": 9.67741935483871e-06,
-      "loss": 0.0351,
+      "loss": 0.0393,
       "step": 1140
     },
     {
       "epoch": 8.12,
       "learning_rate": 9.31899641577061e-06,
-      "loss": 0.0646,
+      "loss": 0.0318,
       "step": 1150
     },
     {
       "epoch": 8.19,
       "learning_rate": 8.960573476702509e-06,
-      "loss": 0.0377,
+      "loss": 0.0499,
       "step": 1160
     },
     {
       "epoch": 8.26,
       "learning_rate": 8.602150537634409e-06,
-      "loss": 0.0537,
+      "loss": 0.0246,
       "step": 1170
     },
     {
       "epoch": 8.34,
       "learning_rate": 8.24372759856631e-06,
-      "loss": 0.0313,
+      "loss": 0.0689,
       "step": 1180
     },
     {
       "epoch": 8.41,
       "learning_rate": 7.885304659498208e-06,
-      "loss": 0.0477,
+      "loss": 0.0599,
       "step": 1190
     },
     {
       "epoch": 8.48,
       "learning_rate": 7.526881720430108e-06,
-      "loss": 0.0371,
+      "loss": 0.042,
       "step": 1200
     },
     {
       "epoch": 8.55,
       "learning_rate": 7.168458781362007e-06,
-      "loss": 0.0358,
+      "loss": 0.0459,
       "step": 1210
     },
     {
       "epoch": 8.62,
       "learning_rate": 6.810035842293908e-06,
-      "loss": 0.025,
+      "loss": 0.0299,
       "step": 1220
     },
     {
       "epoch": 8.69,
       "learning_rate": 6.451612903225806e-06,
-      "loss": 0.0591,
+      "loss": 0.028,
       "step": 1230
     },
     {
       "epoch": 8.76,
       "learning_rate": 6.0931899641577065e-06,
-      "loss": 0.0385,
+      "loss": 0.0393,
       "step": 1240
     },
     {
       "epoch": 8.83,
       "learning_rate": 5.734767025089606e-06,
-      "loss": 0.0456,
+      "loss": 0.0544,
       "step": 1250
     },
     {
       "epoch": 8.9,
       "learning_rate": 5.376344086021506e-06,
-      "loss": 0.027,
+      "loss": 0.0184,
       "step": 1260
     },
     {
       "epoch": 8.97,
       "learning_rate": 5.017921146953405e-06,
-      "loss": 0.0317,
+      "loss": 0.0243,
       "step": 1270
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.996594778660613,
-      "eval_loss": 0.008439515717327595,
-      "eval_runtime": 9.5892,
-      "eval_samples_per_second": 91.875,
-      "eval_steps_per_second": 11.576,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0003116827574558556,
+      "eval_runtime": 9.8106,
+      "eval_samples_per_second": 89.8,
+      "eval_steps_per_second": 11.314,
       "step": 1274
     },
     {
       "epoch": 9.04,
       "learning_rate": 4.659498207885305e-06,
-      "loss": 0.0448,
+      "loss": 0.0483,
       "step": 1280
     },
     {
       "epoch": 9.11,
       "learning_rate": 4.3010752688172045e-06,
-      "loss": 0.0307,
+      "loss": 0.0291,
       "step": 1290
     },
     {
       "epoch": 9.18,
       "learning_rate": 3.942652329749104e-06,
-      "loss": 0.0501,
+      "loss": 0.0394,
       "step": 1300
     },
     {
       "epoch": 9.25,
       "learning_rate": 3.5842293906810035e-06,
-      "loss": 0.0339,
+      "loss": 0.0242,
       "step": 1310
     },
     {
       "epoch": 9.32,
       "learning_rate": 3.225806451612903e-06,
-      "loss": 0.0325,
+      "loss": 0.044,
       "step": 1320
     },
     {
       "epoch": 9.39,
       "learning_rate": 2.867383512544803e-06,
-      "loss": 0.0325,
+      "loss": 0.0346,
       "step": 1330
     },
     {
       "epoch": 9.47,
       "learning_rate": 2.5089605734767026e-06,
-      "loss": 0.0321,
+      "loss": 0.0521,
       "step": 1340
     },
     {
       "epoch": 9.54,
       "learning_rate": 2.1505376344086023e-06,
-      "loss": 0.0533,
+      "loss": 0.041,
       "step": 1350
     },
     {
       "epoch": 9.61,
       "learning_rate": 1.7921146953405017e-06,
-      "loss": 0.0308,
+      "loss": 0.0497,
       "step": 1360
     },
     {
       "epoch": 9.68,
       "learning_rate": 1.4336917562724014e-06,
-      "loss": 0.0626,
+      "loss": 0.0324,
       "step": 1370
     },
     {
       "epoch": 9.75,
       "learning_rate": 1.0752688172043011e-06,
-      "loss": 0.0711,
+      "loss": 0.0312,
       "step": 1380
     },
     {
       "epoch": 9.82,
       "learning_rate": 7.168458781362007e-07,
-      "loss": 0.035,
+      "loss": 0.0169,
       "step": 1390
     },
     {
       "epoch": 9.89,
       "learning_rate": 3.5842293906810036e-07,
-      "loss": 0.0239,
+      "loss": 0.0221,
       "step": 1400
     },
     {
       "epoch": 9.96,
       "learning_rate": 0.0,
-      "loss": 0.0339,
+      "loss": 0.0259,
       "step": 1410
     },
     {
       "epoch": 9.96,
-      "eval_accuracy": 0.996594778660613,
-      "eval_loss": 0.008212185464799404,
-      "eval_runtime": 10.1675,
-      "eval_samples_per_second": 86.649,
-      "eval_steps_per_second": 10.917,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0003799856931436807,
+      "eval_runtime": 9.7493,
+      "eval_samples_per_second": 90.366,
+      "eval_steps_per_second": 11.385,
       "step": 1410
     },
     {
       "epoch": 9.96,
       "step": 1410,
       "total_flos": 1.3986242481389875e+19,
-      "train_loss": 0.0672107882744877,
-      "train_runtime": 1935.8008,
-      "train_samples_per_second": 40.929,
-      "train_steps_per_second": 0.728
+      "train_loss": 0.0649210772510116,
+      "train_runtime": 2044.7769,
+      "train_samples_per_second": 38.748,
+      "train_steps_per_second": 0.69
     }
   ],
   "max_steps": 1410,