End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- id
 license: mit
 base_model: indolem/indobert-base-uncased
 tags:

all_results.json

@@ -1,21 +1,21 @@
 {
-    "accuracy": 0.9233067729083665,
+    "accuracy": 0.9193227091633466,
     "epoch": 20.0,
-    "eval_accuracy": 0.9223057644110275,
-    "eval_f1": 0.9035367518034705,
-    "eval_loss": 0.717867374420166,
-    "eval_precision": 0.9169940112048426,
-    "eval_recall": 0.8925259138025095,
-    "eval_runtime": 4.7405,
+    "eval_accuracy": 0.8922305764411027,
+    "eval_f1": 0.8703223612108386,
+    "eval_loss": 0.8053188323974609,
+    "eval_precision": 0.8694131129742446,
+    "eval_recall": 0.8712493180578287,
+    "eval_runtime": 1.6567,
     "eval_samples": 399,
-    "eval_samples_per_second": 84.169,
-    "eval_steps_per_second": 10.547,
-    "f1": 0.9051778803991193,
-    "precision": 0.9160780898395615,
-    "recall": 0.8959519018875156,
-    "train_loss": 0.05354873166098947,
-    "train_runtime": 2701.55,
+    "eval_samples_per_second": 240.841,
+    "eval_steps_per_second": 30.181,
+    "f1": 0.9021106797869459,
+    "precision": 0.9041861252933917,
+    "recall": 0.900110184919038,
+    "train_loss": 0.05254678238855034,
+    "train_runtime": 869.5662,
     "train_samples": 3645,
-    "train_samples_per_second": 26.985,
-    "train_steps_per_second": 0.903
+    "train_samples_per_second": 83.835,
+    "train_steps_per_second": 2.806
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.9223057644110275,
-    "eval_f1": 0.9035367518034705,
-    "eval_loss": 0.717867374420166,
-    "eval_precision": 0.9169940112048426,
-    "eval_recall": 0.8925259138025095,
-    "eval_runtime": 4.7405,
+    "eval_accuracy": 0.8922305764411027,
+    "eval_f1": 0.8703223612108386,
+    "eval_loss": 0.8053188323974609,
+    "eval_precision": 0.8694131129742446,
+    "eval_recall": 0.8712493180578287,
+    "eval_runtime": 1.6567,
     "eval_samples": 399,
-    "eval_samples_per_second": 84.169,
-    "eval_steps_per_second": 10.547
+    "eval_samples_per_second": 240.841,
+    "eval_steps_per_second": 30.181
 }

predict_results.json

@@ -1,6 +1,6 @@
 {
-    "accuracy": 0.9233067729083665,
-    "f1": 0.9051778803991193,
-    "precision": 0.9160780898395615,
-    "recall": 0.8959519018875156
+    "accuracy": 0.9193227091633466,
+    "f1": 0.9021106797869459,
+    "precision": 0.9041861252933917,
+    "recall": 0.900110184919038
 }

predict_results.txt

@@ -1,6 +1,6 @@
 index	prediction
-0	1
-1	0
+0	0
+1	1
 2	1
 3	1
 4	1
@@ -16,8 +16,8 @@ index	prediction
 14	1
 15	1
 16	0
-17	1
-18	0
+17	0
+18	1
 19	1
 20	1
 21	0
@@ -28,7 +28,7 @@ index	prediction
 26	1
 27	1
 28	1
-29	0
+29	1
 30	0
 31	1
 32	1
@@ -51,14 +51,14 @@ index	prediction
 49	1
 50	1
 51	1
-52	1
+52	0
 53	1
-54	0
+54	1
 55	1
-56	1
+56	0
 57	1
-58	0
-59	0
+58	1
+59	1
 60	1
 61	1
 62	1
@@ -77,10 +77,10 @@ index	prediction
 75	0
 76	1
 77	1
-78	0
+78	1
 79	1
 80	1
-81	1
+81	0
 82	1
 83	1
 84	1
@@ -93,12 +93,12 @@ index	prediction
 91	1
 92	0
 93	1
-94	0
+94	1
 95	1
 96	0
-97	1
+97	0
 98	1
-99	1
+99	0
 100	1
 101	1
 102	1
@@ -127,7 +127,7 @@ index	prediction
 125	0
 126	1
 127	1
-128	0
+128	1
 129	1
 130	1
 131	1
@@ -142,7 +142,7 @@ index	prediction
 140	1
 141	1
 142	1
-143	0
+143	1
 144	1
 145	1
 146	1
@@ -152,11 +152,11 @@ index	prediction
 150	1
 151	1
 152	1
-153	0
+153	1
 154	1
 155	1
 156	1
-157	0
+157	1
 158	1
 159	1
 160	1
@@ -184,12 +184,12 @@ index	prediction
 182	1
 183	1
 184	1
-185	1
+185	0
 186	1
 187	0
 188	1
 189	1
-190	0
+190	1
 191	1
 192	1
 193	0
@@ -237,7 +237,7 @@ index	prediction
 235	1
 236	1
 237	1
-238	1
+238	0
 239	1
 240	1
 241	0
@@ -267,7 +267,7 @@ index	prediction
 265	1
 266	1
 267	1
-268	1
+268	0
 269	0
 270	1
 271	1
@@ -279,15 +279,15 @@ index	prediction
 277	1
 278	1
 279	1
-280	0
-281	0
-282	0
+280	1
+281	1
+282	1
 283	1
 284	1
 285	1
 286	1
 287	0
-288	0
+288	1
 289	1
 290	1
 291	1
@@ -311,7 +311,7 @@ index	prediction
 309	0
 310	0
 311	0
-312	1
+312	0
 313	0
 314	0
 315	0
@@ -359,7 +359,7 @@ index	prediction
 357	0
 358	0
 359	0
-360	0
+360	1
 361	0
 362	0
 363	0
@@ -383,8 +383,8 @@ index	prediction
 381	0
 382	0
 383	0
-384	0
-385	0
+384	1
+385	1
 386	0
 387	0
 388	0
@@ -429,7 +429,7 @@ index	prediction
 427	0
 428	0
 429	0
-430	1
+430	0
 431	0
 432	0
 433	0
@@ -463,7 +463,7 @@ index	prediction
 461	0
 462	0
 463	0
-464	0
+464	1
 465	0
 466	0
 467	0
@@ -485,7 +485,7 @@ index	prediction
 483	0
 484	0
 485	0
-486	0
+486	1
 487	0
 488	0
 489	0
@@ -542,7 +542,7 @@ index	prediction
 540	0
 541	0
 542	0
-543	0
+543	1
 544	0
 545	0
 546	0
@@ -558,7 +558,7 @@ index	prediction
 556	0
 557	0
 558	0
-559	1
+559	0
 560	0
 561	0
 562	0
@@ -648,7 +648,7 @@ index	prediction
 646	0
 647	0
 648	0
-649	1
+649	0
 650	0
 651	0
 652	0
@@ -661,7 +661,7 @@ index	prediction
 659	0
 660	0
 661	0
-662	0
+662	1
 663	0
 664	0
 665	0
@@ -737,7 +737,7 @@ index	prediction
 735	0
 736	0
 737	0
-738	0
+738	1
 739	0
 740	0
 741	0
@@ -747,8 +747,8 @@ index	prediction
 745	0
 746	0
 747	0
-748	0
-749	0
+748	1
+749	1
 750	0
 751	0
 752	0
@@ -811,12 +811,12 @@ index	prediction
 809	0
 810	0
 811	0
-812	1
+812	0
 813	0
 814	0
 815	0
 816	0
-817	0
+817	1
 818	0
 819	0
 820	0
@@ -833,7 +833,7 @@ index	prediction
 831	0
 832	0
 833	0
-834	0
+834	1
 835	1
 836	0
 837	0
@@ -903,7 +903,7 @@ index	prediction
 901	0
 902	0
 903	0
-904	0
+904	1
 905	0
 906	1
 907	0
@@ -921,7 +921,7 @@ index	prediction
 919	0
 920	0
 921	0
-922	0
+922	1
 923	0
 924	0
 925	0
@@ -976,9 +976,9 @@ index	prediction
 974	0
 975	0
 976	0
-977	0
+977	1
 978	0
-979	0
+979	1
 980	0
 981	0
 982	0

runs/Jun03_10-43-30_a358b85c7679/events.out.tfevents.1717412308.a358b85c7679.35934.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:554144f589ad48e3ad0a0ca4be8515e378169031ed793e69c00649b94988fee3
+size 560

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.05354873166098947,
-    "train_runtime": 2701.55,
+    "train_loss": 0.05254678238855034,
+    "train_runtime": 869.5662,
     "train_samples": 3645,
-    "train_samples_per_second": 26.985,
-    "train_steps_per_second": 0.903
+    "train_samples_per_second": 83.835,
+    "train_steps_per_second": 2.806
 }

trainer_state.json

@@ -10,392 +10,392 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 24.956905364990234,
+      "grad_norm": 6.880457878112793,
       "learning_rate": 4.75e-05,
-      "loss": 0.3833,
+      "loss": 0.371,
       "step": 122
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8696741854636592,
-      "eval_f1": 0.8395012067578439,
-      "eval_loss": 0.31313076615333557,
-      "eval_precision": 0.8473584308763049,
-      "eval_recall": 0.8327877795962902,
-      "eval_runtime": 5.0044,
-      "eval_samples_per_second": 79.73,
-      "eval_steps_per_second": 9.991,
+      "eval_accuracy": 0.8847117794486216,
+      "eval_f1": 0.8556621579112929,
+      "eval_loss": 0.2788735032081604,
+      "eval_precision": 0.871654421411703,
+      "eval_recall": 0.8434260774686306,
+      "eval_runtime": 1.6426,
+      "eval_samples_per_second": 242.907,
+      "eval_steps_per_second": 30.439,
       "step": 122
     },
     {
       "epoch": 2.0,
-      "grad_norm": 5.629110813140869,
+      "grad_norm": 6.694629192352295,
       "learning_rate": 4.5e-05,
-      "loss": 0.2232,
+      "loss": 0.214,
       "step": 244
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8872180451127819,
-      "eval_f1": 0.8629148629148629,
-      "eval_loss": 0.26969170570373535,
-      "eval_precision": 0.8657894736842104,
-      "eval_recall": 0.860201854882706,
-      "eval_runtime": 4.9679,
-      "eval_samples_per_second": 80.315,
-      "eval_steps_per_second": 10.065,
+      "eval_accuracy": 0.8947368421052632,
+      "eval_f1": 0.8760282890453928,
+      "eval_loss": 0.2702818810939789,
+      "eval_precision": 0.8682260305697083,
+      "eval_recall": 0.8855246408437898,
+      "eval_runtime": 1.6739,
+      "eval_samples_per_second": 238.362,
+      "eval_steps_per_second": 29.87,
       "step": 244
     },
     {
       "epoch": 3.0,
-      "grad_norm": 20.295835494995117,
+      "grad_norm": 3.0087978839874268,
       "learning_rate": 4.25e-05,
-      "loss": 0.1574,
+      "loss": 0.141,
       "step": 366
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8596491228070176,
-      "eval_f1": 0.8079620462046204,
-      "eval_loss": 0.5306064486503601,
-      "eval_precision": 0.8777160493827161,
-      "eval_recall": 0.7781869430805601,
-      "eval_runtime": 4.8971,
-      "eval_samples_per_second": 81.477,
-      "eval_steps_per_second": 10.21,
+      "eval_accuracy": 0.87468671679198,
+      "eval_f1": 0.843111041207927,
+      "eval_loss": 0.4446346163749695,
+      "eval_precision": 0.8585673051692468,
+      "eval_recall": 0.831332969630842,
+      "eval_runtime": 1.6711,
+      "eval_samples_per_second": 238.76,
+      "eval_steps_per_second": 29.92,
       "step": 366
     },
     {
       "epoch": 4.0,
-      "grad_norm": 2.211064100265503,
+      "grad_norm": 15.80390739440918,
       "learning_rate": 4e-05,
-      "loss": 0.0861,
+      "loss": 0.093,
       "step": 488
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9147869674185464,
-      "eval_f1": 0.8991765265473572,
-      "eval_loss": 0.39748647809028625,
-      "eval_precision": 0.8922773722627737,
-      "eval_recall": 0.9072104018912529,
-      "eval_runtime": 4.9778,
-      "eval_samples_per_second": 80.157,
-      "eval_steps_per_second": 10.045,
+      "eval_accuracy": 0.8696741854636592,
+      "eval_f1": 0.8368354828562441,
+      "eval_loss": 0.5896394848823547,
+      "eval_precision": 0.8520237470480189,
+      "eval_recall": 0.8252864157119476,
+      "eval_runtime": 1.6746,
+      "eval_samples_per_second": 238.269,
+      "eval_steps_per_second": 29.858,
       "step": 488
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.07590118050575256,
+      "grad_norm": 0.24515673518180847,
       "learning_rate": 3.7500000000000003e-05,
-      "loss": 0.0444,
+      "loss": 0.0469,
       "step": 610
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8757339815412664,
-      "eval_loss": 0.4915539622306824,
-      "eval_precision": 0.8766906299500427,
-      "eval_recall": 0.8747954173486088,
-      "eval_runtime": 4.9432,
-      "eval_samples_per_second": 80.717,
-      "eval_steps_per_second": 10.115,
+      "eval_accuracy": 0.8796992481203008,
+      "eval_f1": 0.8563025210084034,
+      "eval_loss": 0.6098523736000061,
+      "eval_precision": 0.8529936381473334,
+      "eval_recall": 0.8598836152027641,
+      "eval_runtime": 1.6766,
+      "eval_samples_per_second": 237.984,
+      "eval_steps_per_second": 29.823,
       "step": 610
     },
     {
       "epoch": 6.0,
-      "grad_norm": 6.3078179359436035,
+      "grad_norm": 0.06188611686229706,
       "learning_rate": 3.5e-05,
-      "loss": 0.0323,
+      "loss": 0.0498,
       "step": 732
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9047619047619048,
-      "eval_f1": 0.8851154755410074,
-      "eval_loss": 0.4762665033340454,
-      "eval_precision": 0.8851154755410074,
-      "eval_recall": 0.8851154755410074,
-      "eval_runtime": 4.9518,
-      "eval_samples_per_second": 80.576,
-      "eval_steps_per_second": 10.097,
+      "eval_accuracy": 0.899749373433584,
+      "eval_f1": 0.871530684525726,
+      "eval_loss": 0.6609992980957031,
+      "eval_precision": 0.9016447368421052,
+      "eval_recall": 0.8515639207128569,
+      "eval_runtime": 1.6777,
+      "eval_samples_per_second": 237.819,
+      "eval_steps_per_second": 29.802,
       "step": 732
     },
     {
       "epoch": 7.0,
-      "grad_norm": 12.544130325317383,
+      "grad_norm": 1.2003283500671387,
       "learning_rate": 3.2500000000000004e-05,
-      "loss": 0.0343,
+      "loss": 0.0257,
       "step": 854
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.87468671679198,
-      "eval_f1": 0.8353634383046148,
-      "eval_loss": 0.850492000579834,
-      "eval_precision": 0.8784261715296198,
-      "eval_recall": 0.8113293326059283,
-      "eval_runtime": 4.9345,
-      "eval_samples_per_second": 80.859,
-      "eval_steps_per_second": 10.133,
+      "eval_accuracy": 0.8872180451127819,
+      "eval_f1": 0.8532332818362393,
+      "eval_loss": 0.6781011819839478,
+      "eval_precision": 0.891747572815534,
+      "eval_recall": 0.8301963993453355,
+      "eval_runtime": 1.6733,
+      "eval_samples_per_second": 238.446,
+      "eval_steps_per_second": 29.88,
       "step": 854
     },
     {
       "epoch": 8.0,
-      "grad_norm": 0.023242522031068802,
+      "grad_norm": 1.8772869110107422,
       "learning_rate": 3e-05,
-      "loss": 0.021,
+      "loss": 0.0267,
       "step": 976
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9072681704260651,
-      "eval_f1": 0.8878574955372402,
-      "eval_loss": 0.5842882990837097,
-      "eval_precision": 0.8888448885098087,
-      "eval_recall": 0.8868885251863976,
-      "eval_runtime": 4.9396,
-      "eval_samples_per_second": 80.775,
-      "eval_steps_per_second": 10.122,
+      "eval_accuracy": 0.8872180451127819,
+      "eval_f1": 0.8523012380208119,
+      "eval_loss": 0.8199848532676697,
+      "eval_precision": 0.8950599239988307,
+      "eval_recall": 0.8276959447172213,
+      "eval_runtime": 1.6789,
+      "eval_samples_per_second": 237.65,
+      "eval_steps_per_second": 29.781,
       "step": 976
     },
     {
       "epoch": 9.0,
-      "grad_norm": 4.315354824066162,
+      "grad_norm": 2.237112522125244,
       "learning_rate": 2.7500000000000004e-05,
-      "loss": 0.0171,
+      "loss": 0.016,
       "step": 1098
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8694882125334078,
-      "eval_loss": 0.7051700353622437,
-      "eval_precision": 0.8940436639772188,
-      "eval_recall": 0.8522913256955811,
-      "eval_runtime": 4.9356,
-      "eval_samples_per_second": 80.841,
-      "eval_steps_per_second": 10.13,
+      "eval_accuracy": 0.899749373433584,
+      "eval_f1": 0.881931703852755,
+      "eval_loss": 0.5965662002563477,
+      "eval_precision": 0.8740012737378415,
+      "eval_recall": 0.8915711947626841,
+      "eval_runtime": 1.673,
+      "eval_samples_per_second": 238.498,
+      "eval_steps_per_second": 29.887,
       "step": 1098
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.0030266689136624336,
+      "grad_norm": 0.010850044898688793,
       "learning_rate": 2.5e-05,
-      "loss": 0.0125,
+      "loss": 0.0132,
       "step": 1220
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.9047619047619048,
-      "eval_f1": 0.8820775261324042,
-      "eval_loss": 0.7267709970474243,
-      "eval_precision": 0.8934835488413775,
-      "eval_recall": 0.8726132024004365,
-      "eval_runtime": 4.9522,
-      "eval_samples_per_second": 80.57,
-      "eval_steps_per_second": 10.096,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8835263209107715,
+      "eval_loss": 0.6436753273010254,
+      "eval_precision": 0.8791501449961532,
+      "eval_recall": 0.8883433351518457,
+      "eval_runtime": 1.6736,
+      "eval_samples_per_second": 238.404,
+      "eval_steps_per_second": 29.875,
       "step": 1220
     },
     {
       "epoch": 11.0,
-      "grad_norm": 0.10744116455316544,
+      "grad_norm": 0.008115105330944061,
       "learning_rate": 2.25e-05,
-      "loss": 0.0106,
+      "loss": 0.0161,
       "step": 1342
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.9122807017543859,
-      "eval_f1": 0.8904851902099328,
-      "eval_loss": 0.7197956442832947,
-      "eval_precision": 0.9064374185136896,
-      "eval_recall": 0.8779323513366066,
-      "eval_runtime": 4.9857,
-      "eval_samples_per_second": 80.028,
-      "eval_steps_per_second": 10.029,
+      "eval_accuracy": 0.9072681704260651,
+      "eval_f1": 0.8867007927797945,
+      "eval_loss": 0.679655909538269,
+      "eval_precision": 0.89198606271777,
+      "eval_recall": 0.8818876159301692,
+      "eval_runtime": 1.6719,
+      "eval_samples_per_second": 238.648,
+      "eval_steps_per_second": 29.906,
       "step": 1342
     },
     {
       "epoch": 12.0,
-      "grad_norm": 0.00939513836055994,
+      "grad_norm": 0.003001323202624917,
       "learning_rate": 2e-05,
-      "loss": 0.0153,
+      "loss": 0.0091,
       "step": 1464
     },
     {
       "epoch": 12.0,
       "eval_accuracy": 0.9097744360902256,
-      "eval_f1": 0.8917003438084323,
-      "eval_loss": 0.6022568941116333,
-      "eval_precision": 0.8898109243697478,
-      "eval_recall": 0.893662484088016,
-      "eval_runtime": 4.9277,
-      "eval_samples_per_second": 80.97,
-      "eval_steps_per_second": 10.147,
+      "eval_f1": 0.8882839721254355,
+      "eval_loss": 0.6953954696655273,
+      "eval_precision": 0.8998687748047625,
+      "eval_recall": 0.8786597563193308,
+      "eval_runtime": 1.6653,
+      "eval_samples_per_second": 239.59,
+      "eval_steps_per_second": 30.024,
       "step": 1464
     },
     {
       "epoch": 13.0,
-      "grad_norm": 0.0019507030956447124,
+      "grad_norm": 0.0026766008231788874,
       "learning_rate": 1.75e-05,
-      "loss": 0.0047,
+      "loss": 0.0101,
       "step": 1586
     },
     {
       "epoch": 13.0,
-      "eval_accuracy": 0.9147869674185464,
-      "eval_f1": 0.892708003796267,
-      "eval_loss": 0.7098793387413025,
-      "eval_precision": 0.9134992358296232,
-      "eval_recall": 0.8772049463538825,
-      "eval_runtime": 4.9409,
-      "eval_samples_per_second": 80.755,
-      "eval_steps_per_second": 10.12,
+      "eval_accuracy": 0.9122807017543859,
+      "eval_f1": 0.8954723392788977,
+      "eval_loss": 0.6750524044036865,
+      "eval_precision": 0.8909569746108776,
+      "eval_recall": 0.9004364429896345,
+      "eval_runtime": 1.6681,
+      "eval_samples_per_second": 239.196,
+      "eval_steps_per_second": 29.974,
       "step": 1586
     },
     {
       "epoch": 14.0,
-      "grad_norm": 0.0021610422991216183,
+      "grad_norm": 0.0015771281905472279,
       "learning_rate": 1.5e-05,
-      "loss": 0.0109,
+      "loss": 0.0025,
       "step": 1708
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.9097744360902256,
-      "eval_f1": 0.8882839721254355,
-      "eval_loss": 0.6721974611282349,
-      "eval_precision": 0.8998687748047625,
-      "eval_recall": 0.8786597563193308,
-      "eval_runtime": 4.9095,
-      "eval_samples_per_second": 81.271,
-      "eval_steps_per_second": 10.184,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8779692119482108,
+      "eval_loss": 0.7316891551017761,
+      "eval_precision": 0.8934322033898305,
+      "eval_recall": 0.865839243498818,
+      "eval_runtime": 1.6607,
+      "eval_samples_per_second": 240.257,
+      "eval_steps_per_second": 30.107,
       "step": 1708
     },
     {
       "epoch": 15.0,
-      "grad_norm": 0.0025670749600976706,
+      "grad_norm": 0.003234785981476307,
       "learning_rate": 1.25e-05,
-      "loss": 0.0044,
+      "loss": 0.0088,
       "step": 1830
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.9197994987468672,
-      "eval_f1": 0.8995910663730733,
-      "eval_loss": 0.7060150504112244,
-      "eval_precision": 0.9174593282602992,
-      "eval_recall": 0.8857519549008911,
-      "eval_runtime": 4.9454,
-      "eval_samples_per_second": 80.681,
-      "eval_steps_per_second": 10.11,
+      "eval_accuracy": 0.8897243107769424,
+      "eval_f1": 0.8669758137843244,
+      "eval_loss": 0.6788834929466248,
+      "eval_precision": 0.8669758137843244,
+      "eval_recall": 0.8669758137843244,
+      "eval_runtime": 1.6627,
+      "eval_samples_per_second": 239.966,
+      "eval_steps_per_second": 30.071,
       "step": 1830
     },
     {
       "epoch": 16.0,
-      "grad_norm": 0.0015903374878689647,
+      "grad_norm": 0.0018662125803530216,
       "learning_rate": 1e-05,
-      "loss": 0.0053,
+      "loss": 0.0017,
       "step": 1952
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.9223057644110275,
-      "eval_f1": 0.9030011684716548,
-      "eval_loss": 0.7098459005355835,
-      "eval_precision": 0.9194426336375489,
-      "eval_recall": 0.8900254591743954,
-      "eval_runtime": 4.9282,
-      "eval_samples_per_second": 80.962,
-      "eval_steps_per_second": 10.146,
+      "eval_accuracy": 0.8897243107769424,
+      "eval_f1": 0.8676337535436396,
+      "eval_loss": 0.7504969239234924,
+      "eval_precision": 0.8658613445378152,
+      "eval_recall": 0.8694762684124386,
+      "eval_runtime": 1.6676,
+      "eval_samples_per_second": 239.272,
+      "eval_steps_per_second": 29.984,
       "step": 1952
     },
     {
       "epoch": 17.0,
-      "grad_norm": 0.0019439965253695846,
+      "grad_norm": 0.0024691985454410315,
       "learning_rate": 7.5e-06,
-      "loss": 0.005,
+      "loss": 0.0017,
       "step": 2074
     },
     {
       "epoch": 17.0,
-      "eval_accuracy": 0.9197994987468672,
-      "eval_f1": 0.9012315118509808,
-      "eval_loss": 0.7069065570831299,
-      "eval_precision": 0.9104713698196774,
-      "eval_recall": 0.8932533187852337,
-      "eval_runtime": 5.0714,
-      "eval_samples_per_second": 78.676,
-      "eval_steps_per_second": 9.859,
+      "eval_accuracy": 0.8897243107769424,
+      "eval_f1": 0.8676337535436396,
+      "eval_loss": 0.7755724787712097,
+      "eval_precision": 0.8658613445378152,
+      "eval_recall": 0.8694762684124386,
+      "eval_runtime": 1.6752,
+      "eval_samples_per_second": 238.182,
+      "eval_steps_per_second": 29.847,
       "step": 2074
     },
     {
       "epoch": 18.0,
-      "grad_norm": 0.001250488217920065,
+      "grad_norm": 0.0015471646329388022,
       "learning_rate": 5e-06,
-      "loss": 0.0005,
+      "loss": 0.0011,
       "step": 2196
     },
     {
       "epoch": 18.0,
-      "eval_accuracy": 0.9172932330827067,
-      "eval_f1": 0.8961601249201505,
-      "eval_loss": 0.758560061454773,
-      "eval_precision": 0.9154783125371361,
-      "eval_recall": 0.8814784506273867,
-      "eval_runtime": 4.9396,
-      "eval_samples_per_second": 80.776,
-      "eval_steps_per_second": 10.122,
+      "eval_accuracy": 0.8922305764411027,
+      "eval_f1": 0.8715803025426456,
+      "eval_loss": 0.8041408061981201,
+      "eval_precision": 0.8673433153814287,
+      "eval_recall": 0.8762502273140571,
+      "eval_runtime": 1.6768,
+      "eval_samples_per_second": 237.95,
+      "eval_steps_per_second": 29.818,
       "step": 2196
     },
     {
       "epoch": 19.0,
-      "grad_norm": 0.0011430870508775115,
+      "grad_norm": 0.0015901036094874144,
       "learning_rate": 2.5e-06,
       "loss": 0.0017,
       "step": 2318
     },
     {
       "epoch": 19.0,
-      "eval_accuracy": 0.9223057644110275,
-      "eval_f1": 0.9035367518034705,
-      "eval_loss": 0.7198145985603333,
-      "eval_precision": 0.9169940112048426,
-      "eval_recall": 0.8925259138025095,
-      "eval_runtime": 4.9455,
-      "eval_samples_per_second": 80.68,
-      "eval_steps_per_second": 10.11,
+      "eval_accuracy": 0.8922305764411027,
+      "eval_f1": 0.8703223612108386,
+      "eval_loss": 0.8064257502555847,
+      "eval_precision": 0.8694131129742446,
+      "eval_recall": 0.8712493180578287,
+      "eval_runtime": 1.6795,
+      "eval_samples_per_second": 237.568,
+      "eval_steps_per_second": 29.77,
       "step": 2318
     },
     {
       "epoch": 20.0,
-      "grad_norm": 0.0008615191909484565,
+      "grad_norm": 0.0011909452732652426,
       "learning_rate": 0.0,
-      "loss": 0.0009,
+      "loss": 0.0008,
       "step": 2440
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.9223057644110275,
-      "eval_f1": 0.9035367518034705,
-      "eval_loss": 0.717867374420166,
-      "eval_precision": 0.9169940112048426,
-      "eval_recall": 0.8925259138025095,
-      "eval_runtime": 4.9346,
-      "eval_samples_per_second": 80.858,
-      "eval_steps_per_second": 10.133,
+      "eval_accuracy": 0.8922305764411027,
+      "eval_f1": 0.8703223612108386,
+      "eval_loss": 0.8053188323974609,
+      "eval_precision": 0.8694131129742446,
+      "eval_recall": 0.8712493180578287,
+      "eval_runtime": 1.6806,
+      "eval_samples_per_second": 237.419,
+      "eval_steps_per_second": 29.752,
       "step": 2440
     },
     {
       "epoch": 20.0,
       "step": 2440,
       "total_flos": 7598755382040000.0,
-      "train_loss": 0.05354873166098947,
-      "train_runtime": 2701.55,
-      "train_samples_per_second": 26.985,
-      "train_steps_per_second": 0.903
+      "train_loss": 0.05254678238855034,
+      "train_runtime": 869.5662,
+      "train_samples_per_second": 83.835,
+      "train_steps_per_second": 2.806
     }
   ],
   "logging_steps": 500,