End of training

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.9638339795334647,
-    "eval_f1": 0.6566383257030739,
-    "eval_loss": 0.21393825113773346,
-    "eval_precision": 0.7243867243867244,
-    "eval_recall": 0.6004784688995215,
-    "eval_runtime": 1.5301,
-    "eval_samples": 1009,
-    "eval_samples_per_second": 659.428,
-    "eval_steps_per_second": 0.654,
-    "train_loss": 0.06530485324285648,
-    "train_runtime": 166.7998,
-    "train_samples": 3394,
-    "train_samples_per_second": 203.477,
-    "train_steps_per_second": 3.237
+    "eval_accuracy": 0.9925820645613489,
+    "eval_f1": 0.9571847507331379,
+    "eval_loss": 0.04227477312088013,
+    "eval_precision": 0.953111963957951,
+    "eval_recall": 0.9612924941097274,
+    "eval_runtime": 6.0007,
+    "eval_samples": 3251,
+    "eval_samples_per_second": 541.768,
+    "eval_steps_per_second": 0.667,
+    "train_loss": 0.021953743980689482,
+    "train_runtime": 1324.4932,
+    "train_samples": 14042,
+    "train_samples_per_second": 106.018,
+    "train_steps_per_second": 1.661
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.9638339795334647,
-    "eval_f1": 0.6566383257030739,
-    "eval_loss": 0.21393825113773346,
-    "eval_precision": 0.7243867243867244,
-    "eval_recall": 0.6004784688995215,
-    "eval_runtime": 1.5301,
-    "eval_samples": 1009,
-    "eval_samples_per_second": 659.428,
-    "eval_steps_per_second": 0.654
+    "eval_accuracy": 0.9925820645613489,
+    "eval_f1": 0.9571847507331379,
+    "eval_loss": 0.04227477312088013,
+    "eval_precision": 0.953111963957951,
+    "eval_recall": 0.9612924941097274,
+    "eval_runtime": 6.0007,
+    "eval_samples": 3251,
+    "eval_samples_per_second": 541.768,
+    "eval_steps_per_second": 0.667
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.06530485324285648,
-    "train_runtime": 166.7998,
-    "train_samples": 3394,
-    "train_samples_per_second": 203.477,
-    "train_steps_per_second": 3.237
+    "train_loss": 0.021953743980689482,
+    "train_runtime": 1324.4932,
+    "train_samples": 14042,
+    "train_samples_per_second": 106.018,
+    "train_steps_per_second": 1.661
 }

trainer_state.json

@@ -1,283 +1,1105 @@
 {
-  "best_metric": 0.6566383257030739,
-  "best_model_checkpoint": "twitter-roberta-base-CoNLL/checkpoint-500",
+  "best_metric": 0.9571847507331379,
+  "best_model_checkpoint": "twitter-roberta-base-CoNLL/checkpoint-2000",
   "epoch": 10.0,
-  "global_step": 540,
+  "global_step": 2200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.46,
-      "eval_accuracy": 0.9230280302548782,
-      "eval_f1": 0.03744493392070484,
-      "eval_loss": 0.28825512528419495,
-      "eval_precision": 0.2361111111111111,
-      "eval_recall": 0.02033492822966507,
-      "eval_runtime": 1.8273,
-      "eval_samples_per_second": 552.196,
-      "eval_steps_per_second": 0.547,
+      "epoch": 0.11,
+      "eval_accuracy": 0.9385927339278065,
+      "eval_f1": 0.6587314799400699,
+      "eval_loss": 0.2062799483537674,
+      "eval_precision": 0.6516798418972332,
+      "eval_recall": 0.6659373948165601,
+      "eval_runtime": 5.976,
+      "eval_samples_per_second": 544.011,
+      "eval_steps_per_second": 0.669,
       "step": 25
     },
     {
-      "epoch": 0.93,
-      "eval_accuracy": 0.9474988876882985,
-      "eval_f1": 0.5355029585798816,
-      "eval_loss": 0.20930655300617218,
-      "eval_precision": 0.7015503875968992,
-      "eval_recall": 0.43301435406698563,
-      "eval_runtime": 1.5316,
-      "eval_samples_per_second": 658.776,
-      "eval_steps_per_second": 0.653,
+      "epoch": 0.23,
+      "eval_accuracy": 0.9771231649857871,
+      "eval_f1": 0.8565321055660609,
+      "eval_loss": 0.08096875995397568,
+      "eval_precision": 0.8373251888763864,
+      "eval_recall": 0.8766408616627398,
+      "eval_runtime": 6.2407,
+      "eval_samples_per_second": 520.938,
+      "eval_steps_per_second": 0.641,
       "step": 50
     },
     {
-      "epoch": 1.39,
-      "eval_accuracy": 0.9579228373482489,
-      "eval_f1": 0.6288323548597522,
-      "eval_loss": 0.1807074397802353,
-      "eval_precision": 0.6915351506456241,
-      "eval_recall": 0.5765550239234449,
-      "eval_runtime": 1.5064,
-      "eval_samples_per_second": 669.792,
-      "eval_steps_per_second": 0.664,
+      "epoch": 0.34,
+      "eval_accuracy": 0.9826914839764807,
+      "eval_f1": 0.8996990972918756,
+      "eval_loss": 0.06511491537094116,
+      "eval_precision": 0.8937230156094321,
+      "eval_recall": 0.9057556378323797,
+      "eval_runtime": 6.0069,
+      "eval_samples_per_second": 541.208,
+      "eval_steps_per_second": 0.666,
       "step": 75
     },
     {
-      "epoch": 1.85,
-      "eval_accuracy": 0.9608466281065277,
-      "eval_f1": 0.6203411244472521,
-      "eval_loss": 0.1702219694852829,
-      "eval_precision": 0.6572958500669344,
-      "eval_recall": 0.5873205741626795,
-      "eval_runtime": 1.5233,
-      "eval_samples_per_second": 662.362,
-      "eval_steps_per_second": 0.656,
+      "epoch": 0.45,
+      "eval_accuracy": 0.9849110237140298,
+      "eval_f1": 0.907388833166165,
+      "eval_loss": 0.05366199463605881,
+      "eval_precision": 0.9013616738625041,
+      "eval_recall": 0.9134971390104342,
+      "eval_runtime": 6.3828,
+      "eval_samples_per_second": 509.335,
+      "eval_steps_per_second": 0.627,
       "step": 100
     },
     {
-      "epoch": 2.31,
-      "eval_accuracy": 0.9616093561304265,
-      "eval_f1": 0.6293800539083557,
-      "eval_loss": 0.1766974776983261,
-      "eval_precision": 0.720679012345679,
-      "eval_recall": 0.5586124401913876,
-      "eval_runtime": 1.5236,
-      "eval_samples_per_second": 662.237,
-      "eval_steps_per_second": 0.656,
+      "epoch": 0.57,
+      "eval_accuracy": 0.986682761574705,
+      "eval_f1": 0.9170283806343906,
+      "eval_loss": 0.04637985676527023,
+      "eval_precision": 0.9097383239483272,
+      "eval_recall": 0.924436216762033,
+      "eval_runtime": 6.0475,
+      "eval_samples_per_second": 537.576,
+      "eval_steps_per_second": 0.661,
       "step": 125
     },
     {
-      "epoch": 2.78,
-      "eval_accuracy": 0.9617364774677429,
-      "eval_f1": 0.6370860927152318,
-      "eval_loss": 0.17372390627861023,
-      "eval_precision": 0.7136498516320475,
-      "eval_recall": 0.5753588516746412,
-      "eval_runtime": 1.5265,
-      "eval_samples_per_second": 660.978,
-      "eval_steps_per_second": 0.655,
+      "epoch": 0.68,
+      "eval_accuracy": 0.9884544994353802,
+      "eval_f1": 0.9296410942859533,
+      "eval_loss": 0.042301442474126816,
+      "eval_precision": 0.9243054400266179,
+      "eval_recall": 0.9350387075058902,
+      "eval_runtime": 6.0733,
+      "eval_samples_per_second": 535.298,
+      "eval_steps_per_second": 0.659,
       "step": 150
     },
     {
-      "epoch": 3.24,
-      "eval_accuracy": 0.9635161761901735,
-      "eval_f1": 0.6610279765777489,
-      "eval_loss": 0.1732863336801529,
-      "eval_precision": 0.724679029957204,
-      "eval_recall": 0.6076555023923444,
-      "eval_runtime": 1.6272,
-      "eval_samples_per_second": 620.089,
-      "eval_steps_per_second": 0.615,
+      "epoch": 0.8,
+      "eval_accuracy": 0.9899731318873876,
+      "eval_f1": 0.934277384423157,
+      "eval_loss": 0.038059305399656296,
+      "eval_precision": 0.9249546429160481,
+      "eval_recall": 0.9437899697071693,
+      "eval_runtime": 6.3028,
+      "eval_samples_per_second": 515.805,
+      "eval_steps_per_second": 0.635,
       "step": 175
     },
     {
-      "epoch": 3.7,
-      "eval_accuracy": 0.9644695862200471,
-      "eval_f1": 0.6645728643216081,
-      "eval_loss": 0.16830594837665558,
-      "eval_precision": 0.6997354497354498,
-      "eval_recall": 0.6327751196172249,
-      "eval_runtime": 1.5645,
-      "eval_samples_per_second": 644.943,
-      "eval_steps_per_second": 0.639,
+      "epoch": 0.91,
+      "eval_accuracy": 0.9895642693041549,
+      "eval_f1": 0.9354220481626531,
+      "eval_loss": 0.038840554654598236,
+      "eval_precision": 0.9263904934807724,
+      "eval_recall": 0.9446314372265231,
+      "eval_runtime": 6.0497,
+      "eval_samples_per_second": 537.384,
+      "eval_steps_per_second": 0.661,
       "step": 200
     },
     {
-      "epoch": 4.17,
-      "eval_accuracy": 0.9647873895633382,
-      "eval_f1": 0.6692160611854684,
-      "eval_loss": 0.1790447235107422,
-      "eval_precision": 0.7162346521145976,
-      "eval_recall": 0.6279904306220095,
-      "eval_runtime": 1.5303,
-      "eval_samples_per_second": 659.327,
-      "eval_steps_per_second": 0.653,
+      "epoch": 1.02,
+      "eval_accuracy": 0.9898368443596434,
+      "eval_f1": 0.9384409501505521,
+      "eval_loss": 0.03943933546543121,
+      "eval_precision": 0.9328234120385767,
+      "eval_recall": 0.9441265567149109,
+      "eval_runtime": 6.4439,
+      "eval_samples_per_second": 504.505,
+      "eval_steps_per_second": 0.621,
       "step": 225
     },
     {
-      "epoch": 4.63,
-      "eval_accuracy": 0.9645331468887053,
-      "eval_f1": 0.6596858638743455,
-      "eval_loss": 0.1847098022699356,
-      "eval_precision": 0.7283236994219653,
-      "eval_recall": 0.6028708133971292,
-      "eval_runtime": 1.5507,
-      "eval_samples_per_second": 650.675,
-      "eval_steps_per_second": 0.645,
+      "epoch": 1.14,
+      "eval_accuracy": 0.9895837389509754,
+      "eval_f1": 0.9402710389827672,
+      "eval_loss": 0.042259786278009415,
+      "eval_precision": 0.9347970725216235,
+      "eval_recall": 0.9458094917536183,
+      "eval_runtime": 6.16,
+      "eval_samples_per_second": 527.761,
+      "eval_steps_per_second": 0.649,
       "step": 250
     },
     {
-      "epoch": 5.09,
-      "eval_accuracy": 0.9638339795334647,
-      "eval_f1": 0.6632057105775471,
-      "eval_loss": 0.1948511302471161,
-      "eval_precision": 0.724822695035461,
-      "eval_recall": 0.611244019138756,
-      "eval_runtime": 1.7607,
-      "eval_samples_per_second": 573.054,
-      "eval_steps_per_second": 0.568,
+      "epoch": 1.25,
+      "eval_accuracy": 0.9892138156613839,
+      "eval_f1": 0.9354757720311323,
+      "eval_loss": 0.04323225095868111,
+      "eval_precision": 0.9304145163975363,
+      "eval_recall": 0.9405923931336251,
+      "eval_runtime": 6.3592,
+      "eval_samples_per_second": 511.225,
+      "eval_steps_per_second": 0.629,
       "step": 275
     },
     {
-      "epoch": 5.56,
-      "eval_accuracy": 0.9647873895633382,
-      "eval_f1": 0.6649649010848755,
-      "eval_loss": 0.18528707325458527,
-      "eval_precision": 0.7127222982216143,
-      "eval_recall": 0.6232057416267942,
-      "eval_runtime": 1.5232,
-      "eval_samples_per_second": 662.408,
-      "eval_steps_per_second": 0.656,
+      "epoch": 1.36,
+      "eval_accuracy": 0.9900899497683112,
+      "eval_f1": 0.9432760787599497,
+      "eval_loss": 0.03824295476078987,
+      "eval_precision": 0.9392624728850325,
+      "eval_recall": 0.9473241332884551,
+      "eval_runtime": 6.1034,
+      "eval_samples_per_second": 532.654,
+      "eval_steps_per_second": 0.655,
       "step": 300
     },
     {
-      "epoch": 6.02,
-      "eval_accuracy": 0.9644695862200471,
-      "eval_f1": 0.669172932330827,
-      "eval_loss": 0.18068380653858185,
-      "eval_precision": 0.7026315789473684,
-      "eval_recall": 0.638755980861244,
-      "eval_runtime": 1.5586,
-      "eval_samples_per_second": 647.372,
-      "eval_steps_per_second": 0.642,
+      "epoch": 1.48,
+      "eval_accuracy": 0.99005101047467,
+      "eval_f1": 0.9414020171709594,
+      "eval_loss": 0.0380987711250782,
+      "eval_precision": 0.932617671345995,
+      "eval_recall": 0.9503534163581285,
+      "eval_runtime": 6.1204,
+      "eval_samples_per_second": 531.176,
+      "eval_steps_per_second": 0.654,
       "step": 325
     },
     {
-      "epoch": 6.48,
-      "eval_accuracy": 0.9641517828767558,
-      "eval_f1": 0.664951768488746,
-      "eval_loss": 0.20004822313785553,
-      "eval_precision": 0.7190542420027817,
-      "eval_recall": 0.618421052631579,
-      "eval_runtime": 1.5707,
-      "eval_samples_per_second": 642.403,
-      "eval_steps_per_second": 0.637,
+      "epoch": 1.59,
+      "eval_accuracy": 0.9902262372960554,
+      "eval_f1": 0.9429309339331834,
+      "eval_loss": 0.03868376836180687,
+      "eval_precision": 0.933674311169774,
+      "eval_recall": 0.9523729384045776,
+      "eval_runtime": 6.3273,
+      "eval_samples_per_second": 513.801,
+      "eval_steps_per_second": 0.632,
       "step": 350
     },
     {
-      "epoch": 6.94,
-      "eval_accuracy": 0.9638339795334647,
-      "eval_f1": 0.6614987080103358,
-      "eval_loss": 0.19970329105854034,
-      "eval_precision": 0.7191011235955056,
-      "eval_recall": 0.6124401913875598,
-      "eval_runtime": 1.5546,
-      "eval_samples_per_second": 649.03,
-      "eval_steps_per_second": 0.643,
+      "epoch": 1.7,
+      "eval_accuracy": 0.990148358708773,
+      "eval_f1": 0.9439181825802665,
+      "eval_loss": 0.03645554929971695,
+      "eval_precision": 0.9403708034073827,
+      "eval_recall": 0.9474924267923258,
+      "eval_runtime": 6.2492,
+      "eval_samples_per_second": 520.224,
+      "eval_steps_per_second": 0.64,
       "step": 375
     },
     {
-      "epoch": 7.41,
-      "eval_accuracy": 0.9640246615394394,
-      "eval_f1": 0.663659793814433,
-      "eval_loss": 0.2035072147846222,
-      "eval_precision": 0.7192737430167597,
-      "eval_recall": 0.6160287081339713,
-      "eval_runtime": 1.5893,
-      "eval_samples_per_second": 634.88,
-      "eval_steps_per_second": 0.629,
+      "epoch": 1.82,
+      "eval_accuracy": 0.9905182819983646,
+      "eval_f1": 0.9473948735131512,
+      "eval_loss": 0.0382225401699543,
+      "eval_precision": 0.9431287525016678,
+      "eval_recall": 0.9516997643890945,
+      "eval_runtime": 6.4588,
+      "eval_samples_per_second": 503.348,
+      "eval_steps_per_second": 0.619,
       "step": 400
     },
     {
-      "epoch": 7.87,
-      "eval_accuracy": 0.9643424648827306,
-      "eval_f1": 0.6557169930511686,
-      "eval_loss": 0.19765512645244598,
-      "eval_precision": 0.6947791164658634,
-      "eval_recall": 0.6208133971291866,
-      "eval_runtime": 1.57,
-      "eval_samples_per_second": 642.691,
-      "eval_steps_per_second": 0.637,
+      "epoch": 1.93,
+      "eval_accuracy": 0.9903235855301585,
+      "eval_f1": 0.9460837582546184,
+      "eval_loss": 0.03733981028199196,
+      "eval_precision": 0.9398770968277694,
+      "eval_recall": 0.9523729384045776,
+      "eval_runtime": 6.1594,
+      "eval_samples_per_second": 527.813,
+      "eval_steps_per_second": 0.649,
       "step": 425
     },
     {
-      "epoch": 8.33,
-      "eval_accuracy": 0.9645331468887053,
-      "eval_f1": 0.6683706070287541,
-      "eval_loss": 0.20478704571723938,
-      "eval_precision": 0.7174211248285323,
-      "eval_recall": 0.6255980861244019,
-      "eval_runtime": 1.58,
-      "eval_samples_per_second": 638.623,
-      "eval_steps_per_second": 0.633,
+      "epoch": 2.05,
+      "eval_accuracy": 0.9910439624625209,
+      "eval_f1": 0.9497365559923059,
+      "eval_loss": 0.03666611388325691,
+      "eval_precision": 0.943973399833749,
+      "eval_recall": 0.9555705149781218,
+      "eval_runtime": 6.4014,
+      "eval_samples_per_second": 507.859,
+      "eval_steps_per_second": 0.625,
       "step": 450
     },
     {
-      "epoch": 8.8,
-      "eval_accuracy": 0.964215343545414,
-      "eval_f1": 0.6610608020698577,
-      "eval_loss": 0.2073480784893036,
-      "eval_precision": 0.719718309859155,
-      "eval_recall": 0.611244019138756,
-      "eval_runtime": 1.6091,
-      "eval_samples_per_second": 627.07,
-      "eval_steps_per_second": 0.621,
+      "epoch": 2.16,
+      "eval_accuracy": 0.99069350881975,
+      "eval_f1": 0.9474914433592119,
+      "eval_loss": 0.03960287570953369,
+      "eval_precision": 0.9400364419413616,
+      "eval_recall": 0.9550656344665096,
+      "eval_runtime": 6.2344,
+      "eval_samples_per_second": 521.465,
+      "eval_steps_per_second": 0.642,
       "step": 475
     },
     {
-      "epoch": 9.26,
-      "learning_rate": 3.7037037037037037e-06,
-      "loss": 0.07,
+      "epoch": 2.27,
+      "learning_rate": 4.636363636363636e-05,
+      "loss": 0.0771,
       "step": 500
     },
     {
-      "epoch": 9.26,
-      "eval_accuracy": 0.9638339795334647,
-      "eval_f1": 0.6566383257030739,
-      "eval_loss": 0.21393825113773346,
-      "eval_precision": 0.7243867243867244,
-      "eval_recall": 0.6004784688995215,
-      "eval_runtime": 1.5553,
-      "eval_samples_per_second": 648.763,
-      "eval_steps_per_second": 0.643,
+      "epoch": 2.27,
+      "eval_accuracy": 0.991238658930727,
+      "eval_f1": 0.9507813152836967,
+      "eval_loss": 0.035283163189888,
+      "eval_precision": 0.9442323651452282,
+      "eval_recall": 0.9574217435207001,
+      "eval_runtime": 6.1631,
+      "eval_samples_per_second": 527.493,
+      "eval_steps_per_second": 0.649,
       "step": 500
     },
     {
-      "epoch": 9.72,
-      "eval_accuracy": 0.96472382889468,
-      "eval_f1": 0.663284717818643,
-      "eval_loss": 0.20485757291316986,
-      "eval_precision": 0.7058029689608637,
-      "eval_recall": 0.6255980861244019,
-      "eval_runtime": 1.544,
-      "eval_samples_per_second": 653.493,
-      "eval_steps_per_second": 0.648,
+      "epoch": 2.39,
+      "eval_accuracy": 0.9905572212920057,
+      "eval_f1": 0.9453602209020165,
+      "eval_loss": 0.03937030956149101,
+      "eval_precision": 0.9400898652021967,
+      "eval_recall": 0.9506900033658701,
+      "eval_runtime": 6.243,
+      "eval_samples_per_second": 520.742,
+      "eval_steps_per_second": 0.641,
       "step": 525
     },
+    {
+      "epoch": 2.5,
+      "eval_accuracy": 0.9909660838752385,
+      "eval_f1": 0.9484536082474228,
+      "eval_loss": 0.036998968571424484,
+      "eval_precision": 0.9447320086825848,
+      "eval_recall": 0.9522046449007069,
+      "eval_runtime": 6.0773,
+      "eval_samples_per_second": 534.945,
+      "eval_steps_per_second": 0.658,
+      "step": 550
+    },
+    {
+      "epoch": 2.61,
+      "eval_accuracy": 0.9908492659943149,
+      "eval_f1": 0.9472013366750209,
+      "eval_loss": 0.035233963280916214,
+      "eval_precision": 0.940444591904446,
+      "eval_recall": 0.9540558734432851,
+      "eval_runtime": 6.3038,
+      "eval_samples_per_second": 515.72,
+      "eval_steps_per_second": 0.635,
+      "step": 575
+    },
+    {
+      "epoch": 2.73,
+      "eval_accuracy": 0.9908103267006737,
+      "eval_f1": 0.9448281601065159,
+      "eval_loss": 0.0385683998465538,
+      "eval_precision": 0.9344855967078189,
+      "eval_recall": 0.9554022214742511,
+      "eval_runtime": 6.1499,
+      "eval_samples_per_second": 528.629,
+      "eval_steps_per_second": 0.65,
+      "step": 600
+    },
+    {
+      "epoch": 2.84,
+      "eval_accuracy": 0.9916280518671391,
+      "eval_f1": 0.9501544627202138,
+      "eval_loss": 0.036591168493032455,
+      "eval_precision": 0.9428334714167357,
+      "eval_recall": 0.9575900370245709,
+      "eval_runtime": 6.1996,
+      "eval_samples_per_second": 524.389,
+      "eval_steps_per_second": 0.645,
+      "step": 625
+    },
+    {
+      "epoch": 2.95,
+      "eval_accuracy": 0.9913360071648301,
+      "eval_f1": 0.948574295509658,
+      "eval_loss": 0.035347189754247665,
+      "eval_precision": 0.9426624563736081,
+      "eval_recall": 0.9545607539548974,
+      "eval_runtime": 6.4287,
+      "eval_samples_per_second": 505.702,
+      "eval_steps_per_second": 0.622,
+      "step": 650
+    },
+    {
+      "epoch": 3.07,
+      "eval_accuracy": 0.9911023714029827,
+      "eval_f1": 0.9477730425336343,
+      "eval_loss": 0.035899706184864044,
+      "eval_precision": 0.9412448132780084,
+      "eval_recall": 0.9543924604510265,
+      "eval_runtime": 6.1591,
+      "eval_samples_per_second": 527.837,
+      "eval_steps_per_second": 0.649,
+      "step": 675
+    },
+    {
+      "epoch": 3.18,
+      "eval_accuracy": 0.9919979751567306,
+      "eval_f1": 0.9534164087981936,
+      "eval_loss": 0.0355631485581398,
+      "eval_precision": 0.9476309226932669,
+      "eval_recall": 0.9592729720632783,
+      "eval_runtime": 6.1923,
+      "eval_samples_per_second": 525.005,
+      "eval_steps_per_second": 0.646,
+      "step": 700
+    },
+    {
+      "epoch": 3.3,
+      "eval_accuracy": 0.9918422179821658,
+      "eval_f1": 0.953465015065283,
+      "eval_loss": 0.03445196524262428,
+      "eval_precision": 0.9483849483849484,
+      "eval_recall": 0.9585997980477954,
+      "eval_runtime": 6.3335,
+      "eval_samples_per_second": 513.302,
+      "eval_steps_per_second": 0.632,
+      "step": 725
+    },
+    {
+      "epoch": 3.41,
+      "eval_accuracy": 0.9915696429266773,
+      "eval_f1": 0.9491893698813304,
+      "eval_loss": 0.03445504605770111,
+      "eval_precision": 0.9427290836653387,
+      "eval_recall": 0.9557388084819926,
+      "eval_runtime": 6.1661,
+      "eval_samples_per_second": 527.234,
+      "eval_steps_per_second": 0.649,
+      "step": 750
+    },
+    {
+      "epoch": 3.52,
+      "eval_accuracy": 0.9913944161052919,
+      "eval_f1": 0.9478246374395733,
+      "eval_loss": 0.03635535016655922,
+      "eval_precision": 0.9389035667107001,
+      "eval_recall": 0.9569168630090878,
+      "eval_runtime": 6.4373,
+      "eval_samples_per_second": 505.028,
+      "eval_steps_per_second": 0.621,
+      "step": 775
+    },
+    {
+      "epoch": 3.64,
+      "eval_accuracy": 0.9914917643393949,
+      "eval_f1": 0.95067189717052,
+      "eval_loss": 0.03599384054541588,
+      "eval_precision": 0.943036926643484,
+      "eval_recall": 0.9584315045439246,
+      "eval_runtime": 6.1546,
+      "eval_samples_per_second": 528.219,
+      "eval_steps_per_second": 0.65,
+      "step": 800
+    },
+    {
+      "epoch": 3.75,
+      "eval_accuracy": 0.9914528250457537,
+      "eval_f1": 0.950514945993469,
+      "eval_loss": 0.03866244852542877,
+      "eval_precision": 0.9458423596067322,
+      "eval_recall": 0.9552339279703803,
+      "eval_runtime": 6.3798,
+      "eval_samples_per_second": 509.575,
+      "eval_steps_per_second": 0.627,
+      "step": 825
+    },
+    {
+      "epoch": 3.86,
+      "eval_accuracy": 0.9917059304544216,
+      "eval_f1": 0.9521419009370816,
+      "eval_loss": 0.034695982933044434,
+      "eval_precision": 0.9467554076539102,
+      "eval_recall": 0.9575900370245709,
+      "eval_runtime": 6.175,
+      "eval_samples_per_second": 526.482,
+      "eval_steps_per_second": 0.648,
+      "step": 850
+    },
+    {
+      "epoch": 3.98,
+      "eval_accuracy": 0.9915307036330361,
+      "eval_f1": 0.9509402423735897,
+      "eval_loss": 0.03567422926425934,
+      "eval_precision": 0.9445459073551387,
+      "eval_recall": 0.9574217435207001,
+      "eval_runtime": 6.2028,
+      "eval_samples_per_second": 524.119,
+      "eval_steps_per_second": 0.645,
+      "step": 875
+    },
+    {
+      "epoch": 4.09,
+      "eval_accuracy": 0.9918032786885246,
+      "eval_f1": 0.9520702634880803,
+      "eval_loss": 0.03815401718020439,
+      "eval_precision": 0.9464493597206054,
+      "eval_recall": 0.9577583305284416,
+      "eval_runtime": 6.4111,
+      "eval_samples_per_second": 507.091,
+      "eval_steps_per_second": 0.624,
+      "step": 900
+    },
+    {
+      "epoch": 4.2,
+      "eval_accuracy": 0.9917643393948834,
+      "eval_f1": 0.9518385124382277,
+      "eval_loss": 0.03912338241934776,
+      "eval_precision": 0.9474737368684342,
+      "eval_recall": 0.9562436889936049,
+      "eval_runtime": 6.1023,
+      "eval_samples_per_second": 532.748,
+      "eval_steps_per_second": 0.655,
+      "step": 925
+    },
+    {
+      "epoch": 4.32,
+      "eval_accuracy": 0.9911802499902652,
+      "eval_f1": 0.9506493506493505,
+      "eval_loss": 0.0428401380777359,
+      "eval_precision": 0.9466043717670616,
+      "eval_recall": 0.9547290474587681,
+      "eval_runtime": 6.3492,
+      "eval_samples_per_second": 512.03,
+      "eval_steps_per_second": 0.63,
+      "step": 950
+    },
+    {
+      "epoch": 4.43,
+      "eval_accuracy": 0.9913360071648301,
+      "eval_f1": 0.9506028131279304,
+      "eval_loss": 0.04035865515470505,
+      "eval_precision": 0.9458513828723759,
+      "eval_recall": 0.9554022214742511,
+      "eval_runtime": 6.1625,
+      "eval_samples_per_second": 527.542,
+      "eval_steps_per_second": 0.649,
+      "step": 975
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.2727272727272725e-05,
+      "loss": 0.0118,
+      "step": 1000
+    },
+    {
+      "epoch": 4.55,
+      "eval_accuracy": 0.9908687356411354,
+      "eval_f1": 0.9461397365349341,
+      "eval_loss": 0.04028761386871338,
+      "eval_precision": 0.9375413086582948,
+      "eval_recall": 0.9548973409626389,
+      "eval_runtime": 6.261,
+      "eval_samples_per_second": 519.243,
+      "eval_steps_per_second": 0.639,
+      "step": 1000
+    },
+    {
+      "epoch": 4.66,
+      "eval_accuracy": 0.9919200965694482,
+      "eval_f1": 0.9533852205205458,
+      "eval_loss": 0.036938250064849854,
+      "eval_precision": 0.9482270684201765,
+      "eval_recall": 0.9585997980477954,
+      "eval_runtime": 6.1962,
+      "eval_samples_per_second": 524.674,
+      "eval_steps_per_second": 0.646,
+      "step": 1025
+    },
+    {
+      "epoch": 4.77,
+      "eval_accuracy": 0.9918032786885246,
+      "eval_f1": 0.9520227348712804,
+      "eval_loss": 0.037420712411403656,
+      "eval_precision": 0.9456991032879442,
+      "eval_recall": 0.9584315045439246,
+      "eval_runtime": 5.9766,
+      "eval_samples_per_second": 543.958,
+      "eval_steps_per_second": 0.669,
+      "step": 1050
+    },
+    {
+      "epoch": 4.89,
+      "eval_accuracy": 0.9922900198590398,
+      "eval_f1": 0.9538745387453874,
+      "eval_loss": 0.035940222442150116,
+      "eval_precision": 0.9506853895018389,
+      "eval_recall": 0.9570851565129586,
+      "eval_runtime": 6.2768,
+      "eval_samples_per_second": 517.943,
+      "eval_steps_per_second": 0.637,
+      "step": 1075
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9919006269226276,
+      "eval_f1": 0.9523093627328155,
+      "eval_loss": 0.03730743005871773,
+      "eval_precision": 0.9452827060189023,
+      "eval_recall": 0.9594412655671491,
+      "eval_runtime": 6.0853,
+      "eval_samples_per_second": 534.236,
+      "eval_steps_per_second": 0.657,
+      "step": 1100
+    },
+    {
+      "epoch": 5.11,
+      "eval_accuracy": 0.9923873680931428,
+      "eval_f1": 0.9546215673141327,
+      "eval_loss": 0.037039875984191895,
+      "eval_precision": 0.9498500499833389,
+      "eval_recall": 0.9594412655671491,
+      "eval_runtime": 6.3128,
+      "eval_samples_per_second": 514.986,
+      "eval_steps_per_second": 0.634,
+      "step": 1125
+    },
+    {
+      "epoch": 5.23,
+      "eval_accuracy": 0.9922316109185779,
+      "eval_f1": 0.9555313625324512,
+      "eval_loss": 0.038839343935251236,
+      "eval_precision": 0.9509918319719953,
+      "eval_recall": 0.9601144395826321,
+      "eval_runtime": 6.1584,
+      "eval_samples_per_second": 527.897,
+      "eval_steps_per_second": 0.65,
+      "step": 1150
+    },
+    {
+      "epoch": 5.34,
+      "eval_accuracy": 0.9919785055099101,
+      "eval_f1": 0.9522212908633696,
+      "eval_loss": 0.03950090706348419,
+      "eval_precision": 0.948563794255177,
+      "eval_recall": 0.9559071019858634,
+      "eval_runtime": 6.084,
+      "eval_samples_per_second": 534.349,
+      "eval_steps_per_second": 0.657,
+      "step": 1175
+    },
+    {
+      "epoch": 5.45,
+      "eval_accuracy": 0.9923678984463222,
+      "eval_f1": 0.9542866711319491,
+      "eval_loss": 0.03910110890865326,
+      "eval_precision": 0.949516827724092,
+      "eval_recall": 0.9591046785594076,
+      "eval_runtime": 6.3281,
+      "eval_samples_per_second": 513.744,
+      "eval_steps_per_second": 0.632,
+      "step": 1200
+    },
+    {
+      "epoch": 5.57,
+      "eval_accuracy": 0.9923289591526809,
+      "eval_f1": 0.9552313883299799,
+      "eval_loss": 0.037824589759111404,
+      "eval_precision": 0.9517206815903776,
+      "eval_recall": 0.9587680915516661,
+      "eval_runtime": 6.1901,
+      "eval_samples_per_second": 525.192,
+      "eval_steps_per_second": 0.646,
+      "step": 1225
+    },
+    {
+      "epoch": 5.68,
+      "eval_accuracy": 0.9923678984463222,
+      "eval_f1": 0.956470785200067,
+      "eval_loss": 0.038752481341362,
+      "eval_precision": 0.9515323117921386,
+      "eval_recall": 0.9614607876135981,
+      "eval_runtime": 6.178,
+      "eval_samples_per_second": 526.219,
+      "eval_steps_per_second": 0.647,
+      "step": 1250
+    },
+    {
+      "epoch": 5.8,
+      "eval_accuracy": 0.9924457770336046,
+      "eval_f1": 0.9560485558811219,
+      "eval_loss": 0.03838498517870903,
+      "eval_precision": 0.9511910711311011,
+      "eval_recall": 0.9609559071019859,
+      "eval_runtime": 6.3664,
+      "eval_samples_per_second": 510.65,
+      "eval_steps_per_second": 0.628,
+      "step": 1275
+    },
+    {
+      "epoch": 5.91,
+      "eval_accuracy": 0.9924068377399634,
+      "eval_f1": 0.9571045576407506,
+      "eval_loss": 0.03946718946099281,
+      "eval_precision": 0.9529529529529529,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.1196,
+      "eval_samples_per_second": 531.242,
+      "eval_steps_per_second": 0.654,
+      "step": 1300
+    },
+    {
+      "epoch": 6.02,
+      "eval_accuracy": 0.9919200965694482,
+      "eval_f1": 0.9533869885982562,
+      "eval_loss": 0.04075756296515465,
+      "eval_precision": 0.9498830604744404,
+      "eval_recall": 0.9569168630090878,
+      "eval_runtime": 6.1927,
+      "eval_samples_per_second": 524.974,
+      "eval_steps_per_second": 0.646,
+      "step": 1325
+    },
+    {
+      "epoch": 6.14,
+      "eval_accuracy": 0.9921732019781161,
+      "eval_f1": 0.9547998997409975,
+      "eval_loss": 0.04120802879333496,
+      "eval_precision": 0.948067031690725,
+      "eval_recall": 0.9616290811174689,
+      "eval_runtime": 6.3552,
+      "eval_samples_per_second": 511.548,
+      "eval_steps_per_second": 0.629,
+      "step": 1350
+    },
+    {
+      "epoch": 6.25,
+      "eval_accuracy": 0.9923873680931428,
+      "eval_f1": 0.9555667337357477,
+      "eval_loss": 0.0413355678319931,
+      "eval_precision": 0.952054794520548,
+      "eval_recall": 0.9591046785594076,
+      "eval_runtime": 6.2847,
+      "eval_samples_per_second": 517.288,
+      "eval_steps_per_second": 0.636,
+      "step": 1375
+    },
+    {
+      "epoch": 6.36,
+      "eval_accuracy": 0.9917448697480628,
+      "eval_f1": 0.9525004181301221,
+      "eval_loss": 0.04124680161476135,
+      "eval_precision": 0.9466422872340425,
+      "eval_recall": 0.9584315045439246,
+      "eval_runtime": 6.4524,
+      "eval_samples_per_second": 503.841,
+      "eval_steps_per_second": 0.62,
+      "step": 1400
+    },
+    {
+      "epoch": 6.48,
+      "eval_accuracy": 0.9920953233908337,
+      "eval_f1": 0.9555611348229978,
+      "eval_loss": 0.040540654212236404,
+      "eval_precision": 0.9503912102547029,
+      "eval_recall": 0.9607876135981152,
+      "eval_runtime": 6.1266,
+      "eval_samples_per_second": 530.638,
+      "eval_steps_per_second": 0.653,
+      "step": 1425
+    },
+    {
+      "epoch": 6.59,
+      "eval_accuracy": 0.9925041859740664,
+      "eval_f1": 0.9565508580996233,
+      "eval_loss": 0.04004018381237984,
+      "eval_precision": 0.9516908212560387,
+      "eval_recall": 0.9614607876135981,
+      "eval_runtime": 6.4161,
+      "eval_samples_per_second": 506.697,
+      "eval_steps_per_second": 0.623,
+      "step": 1450
+    },
+    {
+      "epoch": 6.7,
+      "eval_accuracy": 0.9922900198590398,
+      "eval_f1": 0.9551813688531458,
+      "eval_loss": 0.039815619587898254,
+      "eval_precision": 0.9509591326105088,
+      "eval_recall": 0.9594412655671491,
+      "eval_runtime": 6.2578,
+      "eval_samples_per_second": 519.511,
+      "eval_steps_per_second": 0.639,
+      "step": 1475
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 1.909090909090909e-05,
+      "loss": 0.0049,
+      "step": 1500
+    },
+    {
+      "epoch": 6.82,
+      "eval_accuracy": 0.9925431252677076,
+      "eval_f1": 0.956871283812076,
+      "eval_loss": 0.03951007500290871,
+      "eval_precision": 0.9523253875645941,
+      "eval_recall": 0.9614607876135981,
+      "eval_runtime": 6.2854,
+      "eval_samples_per_second": 517.232,
+      "eval_steps_per_second": 0.636,
+      "step": 1500
+    },
+    {
+      "epoch": 6.93,
+      "eval_accuracy": 0.9927183520890931,
+      "eval_f1": 0.9571476397723468,
+      "eval_loss": 0.03918459266424179,
+      "eval_precision": 0.952047952047952,
+      "eval_recall": 0.9623022551329519,
+      "eval_runtime": 6.1825,
+      "eval_samples_per_second": 525.837,
+      "eval_steps_per_second": 0.647,
+      "step": 1525
+    },
+    {
+      "epoch": 7.05,
+      "eval_accuracy": 0.9923484287995016,
+      "eval_f1": 0.9551738583996647,
+      "eval_loss": 0.03897989168763161,
+      "eval_precision": 0.9511096278992157,
+      "eval_recall": 0.9592729720632783,
+      "eval_runtime": 6.0468,
+      "eval_samples_per_second": 537.641,
+      "eval_steps_per_second": 0.662,
+      "step": 1550
+    },
+    {
+      "epoch": 7.16,
+      "eval_accuracy": 0.9925431252677076,
+      "eval_f1": 0.9565363034921699,
+      "eval_loss": 0.03931865468621254,
+      "eval_precision": 0.9519919986664445,
+      "eval_recall": 0.9611242006058566,
+      "eval_runtime": 6.2528,
+      "eval_samples_per_second": 519.93,
+      "eval_steps_per_second": 0.64,
+      "step": 1575
+    },
+    {
+      "epoch": 7.27,
+      "eval_accuracy": 0.9925236556208871,
+      "eval_f1": 0.9562233196618398,
+      "eval_loss": 0.03892602398991585,
+      "eval_precision": 0.9512073272273106,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.0818,
+      "eval_samples_per_second": 534.544,
+      "eval_steps_per_second": 0.658,
+      "step": 1600
+    },
+    {
+      "epoch": 7.39,
+      "eval_accuracy": 0.9923678984463222,
+      "eval_f1": 0.9565435820145692,
+      "eval_loss": 0.0404973067343235,
+      "eval_precision": 0.9518413597733711,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.1086,
+      "eval_samples_per_second": 532.197,
+      "eval_steps_per_second": 0.655,
+      "step": 1625
+    },
+    {
+      "epoch": 7.5,
+      "eval_accuracy": 0.9924847163272458,
+      "eval_f1": 0.9558737335677803,
+      "eval_loss": 0.04104885458946228,
+      "eval_precision": 0.9511748041993001,
+      "eval_recall": 0.9606193200942443,
+      "eval_runtime": 6.3603,
+      "eval_samples_per_second": 511.14,
+      "eval_steps_per_second": 0.629,
+      "step": 1650
+    },
+    {
+      "epoch": 7.61,
+      "eval_accuracy": 0.9925431252677076,
+      "eval_f1": 0.9569442117607639,
+      "eval_loss": 0.04079214483499527,
+      "eval_precision": 0.95263509006004,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.0876,
+      "eval_samples_per_second": 534.04,
+      "eval_steps_per_second": 0.657,
+      "step": 1675
+    },
+    {
+      "epoch": 7.73,
+      "eval_accuracy": 0.9921732019781161,
+      "eval_f1": 0.9545302574389837,
+      "eval_loss": 0.043598320335149765,
+      "eval_precision": 0.9481899701095982,
+      "eval_recall": 0.9609559071019859,
+      "eval_runtime": 6.3314,
+      "eval_samples_per_second": 513.469,
+      "eval_steps_per_second": 0.632,
+      "step": 1700
+    },
+    {
+      "epoch": 7.84,
+      "eval_accuracy": 0.9924068377399634,
+      "eval_f1": 0.9559548683660676,
+      "eval_loss": 0.0419035442173481,
+      "eval_precision": 0.9495268138801262,
+      "eval_recall": 0.9624705486368226,
+      "eval_runtime": 6.237,
+      "eval_samples_per_second": 521.246,
+      "eval_steps_per_second": 0.641,
+      "step": 1725
+    },
+    {
+      "epoch": 7.95,
+      "eval_accuracy": 0.9925625949145283,
+      "eval_f1": 0.957126109529392,
+      "eval_loss": 0.04293340817093849,
+      "eval_precision": 0.9525,
+      "eval_recall": 0.9617973746213396,
+      "eval_runtime": 6.1487,
+      "eval_samples_per_second": 528.734,
+      "eval_steps_per_second": 0.651,
+      "step": 1750
+    },
+    {
+      "epoch": 8.07,
+      "eval_accuracy": 0.9924068377399634,
+      "eval_f1": 0.9561506276150629,
+      "eval_loss": 0.041886329650878906,
+      "eval_precision": 0.9508988015978695,
+      "eval_recall": 0.9614607876135981,
+      "eval_runtime": 6.3754,
+      "eval_samples_per_second": 509.93,
+      "eval_steps_per_second": 0.627,
+      "step": 1775
+    },
+    {
+      "epoch": 8.18,
+      "eval_accuracy": 0.9922900198590398,
+      "eval_f1": 0.9555313625324512,
+      "eval_loss": 0.04220963269472122,
+      "eval_precision": 0.9509918319719953,
+      "eval_recall": 0.9601144395826321,
+      "eval_runtime": 6.1129,
+      "eval_samples_per_second": 531.828,
+      "eval_steps_per_second": 0.654,
+      "step": 1800
+    },
+    {
+      "epoch": 8.3,
+      "eval_accuracy": 0.9923873680931428,
+      "eval_f1": 0.9561793045664014,
+      "eval_loss": 0.041702426970005035,
+      "eval_precision": 0.9521107959285834,
+      "eval_recall": 0.9602827330865029,
+      "eval_runtime": 6.3543,
+      "eval_samples_per_second": 511.618,
+      "eval_steps_per_second": 0.629,
+      "step": 1825
+    },
+    {
+      "epoch": 8.41,
+      "eval_accuracy": 0.9924652466804252,
+      "eval_f1": 0.9570171763720151,
+      "eval_loss": 0.04150727763772011,
+      "eval_precision": 0.952945102619723,
+      "eval_recall": 0.9611242006058566,
+      "eval_runtime": 6.1032,
+      "eval_samples_per_second": 532.671,
+      "eval_steps_per_second": 0.655,
+      "step": 1850
+    },
+    {
+      "epoch": 8.52,
+      "eval_accuracy": 0.9924263073867841,
+      "eval_f1": 0.9566965407488064,
+      "eval_loss": 0.04162871465086937,
+      "eval_precision": 0.952309488077372,
+      "eval_recall": 0.9611242006058566,
+      "eval_runtime": 6.1162,
+      "eval_samples_per_second": 531.542,
+      "eval_steps_per_second": 0.654,
+      "step": 1875
+    },
+    {
+      "epoch": 8.64,
+      "eval_accuracy": 0.9922316109185779,
+      "eval_f1": 0.9555611348229978,
+      "eval_loss": 0.04189736768603325,
+      "eval_precision": 0.9503912102547029,
+      "eval_recall": 0.9607876135981152,
+      "eval_runtime": 6.4808,
+      "eval_samples_per_second": 501.634,
+      "eval_steps_per_second": 0.617,
+      "step": 1900
+    },
+    {
+      "epoch": 8.75,
+      "eval_accuracy": 0.9923873680931428,
+      "eval_f1": 0.9564489112227805,
+      "eval_loss": 0.04165585711598396,
+      "eval_precision": 0.9519839946648883,
+      "eval_recall": 0.9609559071019859,
+      "eval_runtime": 6.2132,
+      "eval_samples_per_second": 523.242,
+      "eval_steps_per_second": 0.644,
+      "step": 1925
+    },
+    {
+      "epoch": 8.86,
+      "eval_accuracy": 0.9926015342081694,
+      "eval_f1": 0.9577818730105546,
+      "eval_loss": 0.041886311024427414,
+      "eval_precision": 0.9534689793195463,
+      "eval_recall": 0.9621339616290812,
+      "eval_runtime": 6.5243,
+      "eval_samples_per_second": 498.291,
+      "eval_steps_per_second": 0.613,
+      "step": 1950
+    },
+    {
+      "epoch": 8.98,
+      "eval_accuracy": 0.9926599431486313,
+      "eval_f1": 0.9575341318368372,
+      "eval_loss": 0.04215257614850998,
+      "eval_precision": 0.9531432382858096,
+      "eval_recall": 0.9619656681252103,
+      "eval_runtime": 6.2104,
+      "eval_samples_per_second": 523.476,
+      "eval_steps_per_second": 0.644,
+      "step": 1975
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 0.0022,
+      "step": 2000
+    },
+    {
+      "epoch": 9.09,
+      "eval_accuracy": 0.9925820645613489,
+      "eval_f1": 0.9571847507331379,
+      "eval_loss": 0.04227477312088013,
+      "eval_precision": 0.953111963957951,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.1893,
+      "eval_samples_per_second": 525.265,
+      "eval_steps_per_second": 0.646,
+      "step": 2000
+    },
+    {
+      "epoch": 9.2,
+      "eval_accuracy": 0.9925431252677076,
+      "eval_f1": 0.9567110441262665,
+      "eval_loss": 0.042577099055051804,
+      "eval_precision": 0.9520079986668889,
+      "eval_recall": 0.9614607876135981,
+      "eval_runtime": 6.2434,
+      "eval_samples_per_second": 520.71,
+      "eval_steps_per_second": 0.641,
+      "step": 2025
+    },
+    {
+      "epoch": 9.32,
+      "eval_accuracy": 0.9924652466804252,
+      "eval_f1": 0.9560338330123105,
+      "eval_loss": 0.04250513017177582,
+      "eval_precision": 0.9514919153192198,
+      "eval_recall": 0.9606193200942443,
+      "eval_runtime": 5.9913,
+      "eval_samples_per_second": 542.619,
+      "eval_steps_per_second": 0.668,
+      "step": 2050
+    },
+    {
+      "epoch": 9.43,
+      "eval_accuracy": 0.9925431252677076,
+      "eval_f1": 0.956463496316142,
+      "eval_loss": 0.0422191396355629,
+      "eval_precision": 0.951682772409197,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.2812,
+      "eval_samples_per_second": 517.58,
+      "eval_steps_per_second": 0.637,
+      "step": 2075
+    },
+    {
+      "epoch": 9.55,
+      "eval_accuracy": 0.9924847163272458,
+      "eval_f1": 0.9559537765868363,
+      "eval_loss": 0.0423436276614666,
+      "eval_precision": 0.9513333333333334,
+      "eval_recall": 0.9606193200942443,
+      "eval_runtime": 6.0529,
+      "eval_samples_per_second": 537.095,
+      "eval_steps_per_second": 0.661,
+      "step": 2100
+    },
+    {
+      "epoch": 9.66,
+      "eval_accuracy": 0.9924652466804252,
+      "eval_f1": 0.9558663428523575,
+      "eval_loss": 0.04235740005970001,
+      "eval_precision": 0.951325220870145,
+      "eval_recall": 0.9604510265903736,
+      "eval_runtime": 6.0731,
+      "eval_samples_per_second": 535.315,
+      "eval_steps_per_second": 0.659,
+      "step": 2125
+    },
+    {
+      "epoch": 9.77,
+      "eval_accuracy": 0.9924652466804252,
+      "eval_f1": 0.9566164154103852,
+      "eval_loss": 0.042295120656490326,
+      "eval_precision": 0.9521507169056352,
+      "eval_recall": 0.9611242006058566,
+      "eval_runtime": 6.3348,
+      "eval_samples_per_second": 513.197,
+      "eval_steps_per_second": 0.631,
+      "step": 2150
+    },
+    {
+      "epoch": 9.89,
+      "eval_accuracy": 0.9924847163272458,
+      "eval_f1": 0.9567037936521229,
+      "eval_loss": 0.04226445034146309,
+      "eval_precision": 0.9521586931155193,
+      "eval_recall": 0.9612924941097274,
+      "eval_runtime": 6.1145,
+      "eval_samples_per_second": 531.688,
+      "eval_steps_per_second": 0.654,
+      "step": 2175
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9925236556208871,
+      "eval_f1": 0.9570387739720292,
+      "eval_loss": 0.04224075376987457,
+      "eval_precision": 0.952492082013669,
+      "eval_recall": 0.9616290811174689,
+      "eval_runtime": 6.3677,
+      "eval_samples_per_second": 510.543,
+      "eval_steps_per_second": 0.628,
+      "step": 2200
+    },
     {
       "epoch": 10.0,
-      "step": 540,
-      "total_flos": 1009327449452316.0,
-      "train_loss": 0.06530485324285648,
-      "train_runtime": 166.7998,
-      "train_samples_per_second": 203.477,
-      "train_steps_per_second": 3.237
+      "step": 2200,
+      "total_flos": 4144211411102244.0,
+      "train_loss": 0.021953743980689482,
+      "train_runtime": 1324.4932,
+      "train_samples_per_second": 106.018,
+      "train_steps_per_second": 1.661
     }
   ],
-  "max_steps": 540,
+  "max_steps": 2200,
   "num_train_epochs": 10,
-  "total_flos": 1009327449452316.0,
+  "total_flos": 4144211411102244.0,
   "trial_name": null,
   "trial_params": null
 }