End of training

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 40.0,
-    "eval_f1": 0.9746835443037974,
-    "eval_loss": 0.09489229321479797,
-    "eval_runtime": 9.9057,
-    "eval_samples_per_second": 7.975,
-    "eval_steps_per_second": 0.101,
-    "train_loss": 0.17715629466942379,
-    "train_runtime": 2286.0002,
-    "train_samples_per_second": 7.804,
-    "train_steps_per_second": 0.122
+    "eval_f1": 0.9873417721518988,
+    "eval_loss": 0.06278952211141586,
+    "eval_runtime": 8.9053,
+    "eval_samples_per_second": 8.871,
+    "eval_steps_per_second": 0.112,
+    "train_loss": 0.09473916946777276,
+    "train_runtime": 2533.8499,
+    "train_samples_per_second": 7.041,
+    "train_steps_per_second": 0.111
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 40.0,
-    "eval_f1": 0.9746835443037974,
-    "eval_loss": 0.09489229321479797,
-    "eval_runtime": 9.9057,
-    "eval_samples_per_second": 7.975,
-    "eval_steps_per_second": 0.101
+    "eval_f1": 0.9873417721518988,
+    "eval_loss": 0.06278952211141586,
+    "eval_runtime": 8.9053,
+    "eval_samples_per_second": 8.871,
+    "eval_steps_per_second": 0.112
 }

runs/Dec07_14-05-13_4c8c9451709b/events.out.tfevents.1670424940.4c8c9451709b.490.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b8451272354caee02c3e6a72fdbb1f11ec5b8e5815bc8e25ab2189699e6229b
+size 357

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 40.0,
-    "train_loss": 0.17715629466942379,
-    "train_runtime": 2286.0002,
-    "train_samples_per_second": 7.804,
-    "train_steps_per_second": 0.122
+    "train_loss": 0.09473916946777276,
+    "train_runtime": 2533.8499,
+    "train_samples_per_second": 7.041,
+    "train_steps_per_second": 0.111
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.09489229321479797,
-  "best_model_checkpoint": "/vit-base-patch32-224-in21k/checkpoint-273",
+  "best_metric": 0.06278952211141586,
+  "best_model_checkpoint": "/vit-base-patch32-224-in21k/checkpoint-217",
   "epoch": 40.0,
   "global_step": 280,
   "is_hyper_param_search": false,
@@ -10,539 +10,539 @@
     {
       "epoch": 1.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.45760300755500793,
-      "eval_runtime": 9.802,
-      "eval_samples_per_second": 8.06,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.4529457688331604,
+      "eval_runtime": 8.6926,
+      "eval_samples_per_second": 9.088,
+      "eval_steps_per_second": 0.115,
       "step": 7
     },
     {
       "epoch": 1.43,
       "learning_rate": 1.928571428571429e-05,
-      "loss": 0.5021,
+      "loss": 0.5024,
       "step": 10
     },
     {
       "epoch": 2.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.39525437355041504,
-      "eval_runtime": 9.734,
-      "eval_samples_per_second": 8.116,
-      "eval_steps_per_second": 0.103,
+      "eval_loss": 0.39888718724250793,
+      "eval_runtime": 9.0793,
+      "eval_samples_per_second": 8.701,
+      "eval_steps_per_second": 0.11,
       "step": 14
     },
     {
       "epoch": 2.86,
       "learning_rate": 1.8571428571428575e-05,
-      "loss": 0.3595,
+      "loss": 0.3533,
       "step": 20
     },
     {
       "epoch": 3.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.38087165355682373,
-      "eval_runtime": 9.8033,
-      "eval_samples_per_second": 8.059,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.37408992648124695,
+      "eval_runtime": 9.0097,
+      "eval_samples_per_second": 8.768,
+      "eval_steps_per_second": 0.111,
       "step": 21
     },
     {
       "epoch": 4.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.32862257957458496,
-      "eval_runtime": 9.7791,
-      "eval_samples_per_second": 8.078,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.3160648047924042,
+      "eval_runtime": 9.1428,
+      "eval_samples_per_second": 8.641,
+      "eval_steps_per_second": 0.109,
       "step": 28
     },
     {
       "epoch": 4.29,
       "learning_rate": 1.785714285714286e-05,
-      "loss": 0.3009,
+      "loss": 0.285,
       "step": 30
     },
     {
       "epoch": 5.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.29453742504119873,
-      "eval_runtime": 9.9491,
-      "eval_samples_per_second": 7.94,
-      "eval_steps_per_second": 0.101,
+      "eval_loss": 0.282362163066864,
+      "eval_runtime": 9.1103,
+      "eval_samples_per_second": 8.671,
+      "eval_steps_per_second": 0.11,
       "step": 35
     },
     {
       "epoch": 5.71,
       "learning_rate": 1.7142857142857142e-05,
-      "loss": 0.2843,
+      "loss": 0.2491,
       "step": 40
     },
     {
       "epoch": 6.0,
       "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.35278087854385376,
-      "eval_runtime": 9.8405,
-      "eval_samples_per_second": 8.028,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.2700817883014679,
+      "eval_runtime": 8.8127,
+      "eval_samples_per_second": 8.964,
+      "eval_steps_per_second": 0.113,
       "step": 42
     },
     {
       "epoch": 7.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.23452825844287872,
-      "eval_runtime": 9.7839,
-      "eval_samples_per_second": 8.074,
-      "eval_steps_per_second": 0.102,
+      "eval_f1": 0.9113924050632911,
+      "eval_loss": 0.2062235176563263,
+      "eval_runtime": 8.5646,
+      "eval_samples_per_second": 9.224,
+      "eval_steps_per_second": 0.117,
       "step": 49
     },
     {
       "epoch": 7.14,
       "learning_rate": 1.642857142857143e-05,
-      "loss": 0.266,
+      "loss": 0.2032,
       "step": 50
     },
     {
       "epoch": 8.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.24986670911312103,
-      "eval_runtime": 9.8907,
-      "eval_samples_per_second": 7.987,
-      "eval_steps_per_second": 0.101,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.20497918128967285,
+      "eval_runtime": 8.6022,
+      "eval_samples_per_second": 9.184,
+      "eval_steps_per_second": 0.116,
       "step": 56
     },
     {
       "epoch": 8.57,
       "learning_rate": 1.5714285714285715e-05,
-      "loss": 0.222,
+      "loss": 0.157,
       "step": 60
     },
     {
       "epoch": 9.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.25441065430641174,
-      "eval_runtime": 9.8794,
-      "eval_samples_per_second": 7.996,
-      "eval_steps_per_second": 0.101,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.20131482183933258,
+      "eval_runtime": 8.5286,
+      "eval_samples_per_second": 9.263,
+      "eval_steps_per_second": 0.117,
       "step": 63
     },
     {
       "epoch": 10.0,
       "learning_rate": 1.5000000000000002e-05,
-      "loss": 0.2018,
+      "loss": 0.1127,
       "step": 70
     },
     {
       "epoch": 10.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.19540712237358093,
-      "eval_runtime": 10.3358,
-      "eval_samples_per_second": 7.643,
-      "eval_steps_per_second": 0.097,
+      "eval_f1": 0.9367088607594937,
+      "eval_loss": 0.19601519405841827,
+      "eval_runtime": 8.52,
+      "eval_samples_per_second": 9.272,
+      "eval_steps_per_second": 0.117,
       "step": 70
     },
     {
       "epoch": 11.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.23508581519126892,
-      "eval_runtime": 9.7761,
-      "eval_samples_per_second": 8.081,
-      "eval_steps_per_second": 0.102,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.14171478152275085,
+      "eval_runtime": 8.4584,
+      "eval_samples_per_second": 9.34,
+      "eval_steps_per_second": 0.118,
       "step": 77
     },
     {
       "epoch": 11.43,
       "learning_rate": 1.4285714285714287e-05,
-      "loss": 0.1948,
+      "loss": 0.0903,
       "step": 80
     },
     {
       "epoch": 12.0,
-      "eval_f1": 0.8607594936708861,
-      "eval_loss": 0.17053687572479248,
-      "eval_runtime": 9.7433,
-      "eval_samples_per_second": 8.108,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.1306915283203125,
+      "eval_runtime": 8.4328,
+      "eval_samples_per_second": 9.368,
+      "eval_steps_per_second": 0.119,
       "step": 84
     },
     {
       "epoch": 12.86,
       "learning_rate": 1.3571428571428574e-05,
-      "loss": 0.2053,
+      "loss": 0.0922,
       "step": 90
     },
     {
       "epoch": 13.0,
-      "eval_f1": 0.8734177215189873,
-      "eval_loss": 0.16247014701366425,
-      "eval_runtime": 9.7199,
-      "eval_samples_per_second": 8.128,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9873417721518988,
+      "eval_loss": 0.08702569454908371,
+      "eval_runtime": 8.8303,
+      "eval_samples_per_second": 8.947,
+      "eval_steps_per_second": 0.113,
       "step": 91
     },
     {
       "epoch": 14.0,
-      "eval_f1": 0.9367088607594937,
-      "eval_loss": 0.17189449071884155,
-      "eval_runtime": 9.7734,
-      "eval_samples_per_second": 8.083,
-      "eval_steps_per_second": 0.102,
+      "eval_f1": 0.9240506329113924,
+      "eval_loss": 0.20480988919734955,
+      "eval_runtime": 8.6551,
+      "eval_samples_per_second": 9.128,
+      "eval_steps_per_second": 0.116,
       "step": 98
     },
     {
       "epoch": 14.29,
       "learning_rate": 1.2857142857142859e-05,
-      "loss": 0.1729,
+      "loss": 0.0595,
       "step": 100
     },
     {
       "epoch": 15.0,
-      "eval_f1": 0.9367088607594937,
-      "eval_loss": 0.1488722860813141,
-      "eval_runtime": 9.7192,
-      "eval_samples_per_second": 8.128,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9620253164556962,
+      "eval_loss": 0.12036000937223434,
+      "eval_runtime": 8.6389,
+      "eval_samples_per_second": 9.145,
+      "eval_steps_per_second": 0.116,
       "step": 105
     },
     {
       "epoch": 15.71,
       "learning_rate": 1.2142857142857142e-05,
-      "loss": 0.1535,
+      "loss": 0.0527,
       "step": 110
     },
     {
       "epoch": 16.0,
-      "eval_f1": 0.9493670886075949,
-      "eval_loss": 0.14498455822467804,
-      "eval_runtime": 9.7322,
-      "eval_samples_per_second": 8.117,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9367088607594937,
+      "eval_loss": 0.2552852928638458,
+      "eval_runtime": 8.4875,
+      "eval_samples_per_second": 9.308,
+      "eval_steps_per_second": 0.118,
       "step": 112
     },
     {
       "epoch": 17.0,
-      "eval_f1": 0.9493670886075949,
-      "eval_loss": 0.1749649941921234,
-      "eval_runtime": 9.718,
-      "eval_samples_per_second": 8.129,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9367088607594937,
+      "eval_loss": 0.16753825545310974,
+      "eval_runtime": 8.4252,
+      "eval_samples_per_second": 9.377,
+      "eval_steps_per_second": 0.119,
       "step": 119
     },
     {
       "epoch": 17.14,
       "learning_rate": 1.1428571428571429e-05,
-      "loss": 0.1492,
+      "loss": 0.0477,
       "step": 120
     },
     {
       "epoch": 18.0,
-      "eval_f1": 0.9493670886075949,
-      "eval_loss": 0.15143541991710663,
-      "eval_runtime": 9.7273,
-      "eval_samples_per_second": 8.121,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9240506329113924,
+      "eval_loss": 0.22650040686130524,
+      "eval_runtime": 8.392,
+      "eval_samples_per_second": 9.414,
+      "eval_steps_per_second": 0.119,
       "step": 126
     },
     {
       "epoch": 18.57,
       "learning_rate": 1.0714285714285714e-05,
-      "loss": 0.1349,
+      "loss": 0.0411,
       "step": 130
     },
     {
       "epoch": 19.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.1304464191198349,
-      "eval_runtime": 9.7442,
-      "eval_samples_per_second": 8.107,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9367088607594937,
+      "eval_loss": 0.1900627613067627,
+      "eval_runtime": 8.4937,
+      "eval_samples_per_second": 9.301,
+      "eval_steps_per_second": 0.118,
       "step": 133
     },
     {
       "epoch": 20.0,
       "learning_rate": 1e-05,
-      "loss": 0.1538,
+      "loss": 0.0299,
       "step": 140
     },
     {
       "epoch": 20.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.12909165024757385,
-      "eval_runtime": 9.7157,
-      "eval_samples_per_second": 8.131,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9240506329113924,
+      "eval_loss": 0.2422873079776764,
+      "eval_runtime": 8.7448,
+      "eval_samples_per_second": 9.034,
+      "eval_steps_per_second": 0.114,
       "step": 140
     },
     {
       "epoch": 21.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.13061794638633728,
-      "eval_runtime": 9.7515,
-      "eval_samples_per_second": 8.101,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9873417721518988,
+      "eval_loss": 0.06394638121128082,
+      "eval_runtime": 8.7204,
+      "eval_samples_per_second": 9.059,
+      "eval_steps_per_second": 0.115,
       "step": 147
     },
     {
       "epoch": 21.43,
       "learning_rate": 9.285714285714288e-06,
-      "loss": 0.1357,
+      "loss": 0.0487,
       "step": 150
     },
     {
       "epoch": 22.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.12830054759979248,
-      "eval_runtime": 9.7367,
-      "eval_samples_per_second": 8.114,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.12548162043094635,
+      "eval_runtime": 8.9171,
+      "eval_samples_per_second": 8.859,
+      "eval_steps_per_second": 0.112,
       "step": 154
     },
     {
       "epoch": 22.86,
       "learning_rate": 8.571428571428571e-06,
-      "loss": 0.147,
+      "loss": 0.0359,
       "step": 160
     },
     {
       "epoch": 23.0,
       "eval_f1": 0.9493670886075949,
-      "eval_loss": 0.12891362607479095,
-      "eval_runtime": 9.7565,
-      "eval_samples_per_second": 8.097,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.12127607315778732,
+      "eval_runtime": 8.6147,
+      "eval_samples_per_second": 9.17,
+      "eval_steps_per_second": 0.116,
       "step": 161
     },
     {
       "epoch": 24.0,
       "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.1338558942079544,
-      "eval_runtime": 9.7379,
-      "eval_samples_per_second": 8.113,
-      "eval_steps_per_second": 0.103,
+      "eval_loss": 0.07274330407381058,
+      "eval_runtime": 8.5807,
+      "eval_samples_per_second": 9.207,
+      "eval_steps_per_second": 0.117,
       "step": 168
     },
     {
       "epoch": 24.29,
       "learning_rate": 7.857142857142858e-06,
-      "loss": 0.1388,
+      "loss": 0.0302,
       "step": 170
     },
     {
       "epoch": 25.0,
       "eval_f1": 0.9493670886075949,
-      "eval_loss": 0.12436553090810776,
-      "eval_runtime": 9.7614,
-      "eval_samples_per_second": 8.093,
-      "eval_steps_per_second": 0.102,
+      "eval_loss": 0.11162865161895752,
+      "eval_runtime": 8.6735,
+      "eval_samples_per_second": 9.108,
+      "eval_steps_per_second": 0.115,
       "step": 175
     },
     {
       "epoch": 25.71,
       "learning_rate": 7.1428571428571436e-06,
-      "loss": 0.1192,
+      "loss": 0.0304,
       "step": 180
     },
     {
       "epoch": 26.0,
-      "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.11165592074394226,
-      "eval_runtime": 9.7177,
-      "eval_samples_per_second": 8.129,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.10622164607048035,
+      "eval_runtime": 8.8211,
+      "eval_samples_per_second": 8.956,
+      "eval_steps_per_second": 0.113,
       "step": 182
     },
     {
       "epoch": 27.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.11046960204839706,
-      "eval_runtime": 9.7417,
-      "eval_samples_per_second": 8.109,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9240506329113924,
+      "eval_loss": 0.20966486632823944,
+      "eval_runtime": 8.7082,
+      "eval_samples_per_second": 9.072,
+      "eval_steps_per_second": 0.115,
       "step": 189
     },
     {
       "epoch": 27.14,
       "learning_rate": 6.4285714285714295e-06,
-      "loss": 0.112,
+      "loss": 0.0274,
       "step": 190
     },
     {
       "epoch": 28.0,
-      "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.1078834980726242,
-      "eval_runtime": 9.7562,
-      "eval_samples_per_second": 8.097,
-      "eval_steps_per_second": 0.102,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.1276017129421234,
+      "eval_runtime": 8.676,
+      "eval_samples_per_second": 9.106,
+      "eval_steps_per_second": 0.115,
       "step": 196
     },
     {
       "epoch": 28.57,
       "learning_rate": 5.7142857142857145e-06,
-      "loss": 0.1215,
+      "loss": 0.0291,
       "step": 200
     },
     {
       "epoch": 29.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.11511888355016708,
-      "eval_runtime": 9.7361,
-      "eval_samples_per_second": 8.114,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.09670199453830719,
+      "eval_runtime": 8.5086,
+      "eval_samples_per_second": 9.285,
+      "eval_steps_per_second": 0.118,
       "step": 203
     },
     {
       "epoch": 30.0,
       "learning_rate": 5e-06,
-      "loss": 0.1139,
+      "loss": 0.0202,
       "step": 210
     },
     {
       "epoch": 30.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.10075395554304123,
-      "eval_runtime": 9.7351,
-      "eval_samples_per_second": 8.115,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9746835443037974,
+      "eval_loss": 0.07649976760149002,
+      "eval_runtime": 8.9676,
+      "eval_samples_per_second": 8.81,
+      "eval_steps_per_second": 0.112,
       "step": 210
     },
     {
       "epoch": 31.0,
-      "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.10330603271722794,
-      "eval_runtime": 9.7699,
-      "eval_samples_per_second": 8.086,
-      "eval_steps_per_second": 0.102,
+      "eval_f1": 0.9873417721518988,
+      "eval_loss": 0.06278952211141586,
+      "eval_runtime": 8.7232,
+      "eval_samples_per_second": 9.056,
+      "eval_steps_per_second": 0.115,
       "step": 217
     },
     {
       "epoch": 31.43,
       "learning_rate": 4.2857142857142855e-06,
-      "loss": 0.1164,
+      "loss": 0.0232,
       "step": 220
     },
     {
       "epoch": 32.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.0984945222735405,
-      "eval_runtime": 9.7232,
-      "eval_samples_per_second": 8.125,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.13882263004779816,
+      "eval_runtime": 8.6516,
+      "eval_samples_per_second": 9.131,
+      "eval_steps_per_second": 0.116,
       "step": 224
     },
     {
       "epoch": 32.86,
       "learning_rate": 3.5714285714285718e-06,
-      "loss": 0.1192,
+      "loss": 0.0264,
       "step": 230
     },
     {
       "epoch": 33.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.09554588049650192,
-      "eval_runtime": 9.7164,
-      "eval_samples_per_second": 8.131,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.10616844147443771,
+      "eval_runtime": 8.6952,
+      "eval_samples_per_second": 9.085,
+      "eval_steps_per_second": 0.115,
       "step": 231
     },
     {
       "epoch": 34.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.10772588849067688,
-      "eval_runtime": 9.7297,
-      "eval_samples_per_second": 8.119,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.1320488601922989,
+      "eval_runtime": 8.7896,
+      "eval_samples_per_second": 8.988,
+      "eval_steps_per_second": 0.114,
       "step": 238
     },
     {
       "epoch": 34.29,
       "learning_rate": 2.8571428571428573e-06,
-      "loss": 0.1132,
+      "loss": 0.0219,
       "step": 240
     },
     {
       "epoch": 35.0,
-      "eval_f1": 0.9620253164556962,
-      "eval_loss": 0.11074268072843552,
-      "eval_runtime": 9.7358,
-      "eval_samples_per_second": 8.114,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.1528194099664688,
+      "eval_runtime": 8.8303,
+      "eval_samples_per_second": 8.946,
+      "eval_steps_per_second": 0.113,
       "step": 245
     },
     {
       "epoch": 35.71,
       "learning_rate": 2.1428571428571427e-06,
-      "loss": 0.1021,
+      "loss": 0.0194,
       "step": 250
     },
     {
       "epoch": 36.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.09575933963060379,
-      "eval_runtime": 9.7306,
-      "eval_samples_per_second": 8.119,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.1746273934841156,
+      "eval_runtime": 8.7909,
+      "eval_samples_per_second": 8.987,
+      "eval_steps_per_second": 0.114,
       "step": 252
     },
     {
       "epoch": 37.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.09573517739772797,
-      "eval_runtime": 9.7256,
-      "eval_samples_per_second": 8.123,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.16089513897895813,
+      "eval_runtime": 8.8367,
+      "eval_samples_per_second": 8.94,
+      "eval_steps_per_second": 0.113,
       "step": 259
     },
     {
       "epoch": 37.14,
       "learning_rate": 1.4285714285714286e-06,
-      "loss": 0.0945,
+      "loss": 0.0204,
       "step": 260
     },
     {
       "epoch": 38.0,
-      "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.09509044885635376,
-      "eval_runtime": 9.7255,
-      "eval_samples_per_second": 8.123,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.14817634224891663,
+      "eval_runtime": 8.8819,
+      "eval_samples_per_second": 8.895,
+      "eval_steps_per_second": 0.113,
       "step": 266
     },
     {
       "epoch": 38.57,
       "learning_rate": 7.142857142857143e-07,
-      "loss": 0.1244,
+      "loss": 0.0217,
       "step": 270
     },
     {
       "epoch": 39.0,
-      "eval_f1": 0.9746835443037974,
-      "eval_loss": 0.09489229321479797,
-      "eval_runtime": 9.7321,
-      "eval_samples_per_second": 8.117,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.152223601937294,
+      "eval_runtime": 8.8958,
+      "eval_samples_per_second": 8.881,
+      "eval_steps_per_second": 0.112,
       "step": 273
     },
     {
       "epoch": 40.0,
       "learning_rate": 0.0,
-      "loss": 0.1012,
+      "loss": 0.0216,
       "step": 280
     },
     {
       "epoch": 40.0,
-      "eval_f1": 0.9873417721518988,
-      "eval_loss": 0.09554900228977203,
-      "eval_runtime": 9.7342,
-      "eval_samples_per_second": 8.116,
-      "eval_steps_per_second": 0.103,
+      "eval_f1": 0.9493670886075949,
+      "eval_loss": 0.14991530776023865,
+      "eval_runtime": 8.6557,
+      "eval_samples_per_second": 9.127,
+      "eval_steps_per_second": 0.116,
       "step": 280
     },
     {
       "epoch": 40.0,
       "step": 280,
       "total_flos": 1.4091487038849024e+18,
-      "train_loss": 0.17715629466942379,
-      "train_runtime": 2286.0002,
-      "train_samples_per_second": 7.804,
-      "train_steps_per_second": 0.122
+      "train_loss": 0.09473916946777276,
+      "train_runtime": 2533.8499,
+      "train_samples_per_second": 7.041,
+      "train_steps_per_second": 0.111
     }
   ],
   "max_steps": 280,