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": 0.9109792284866469,
+    "accuracy": 0.9119683481701286,
     "epoch": 20.0,
-    "eval_accuracy": 0.8972431077694235,
-    "eval_f1": 0.8744522298370696,
-    "eval_loss": 0.7891051769256592,
-    "eval_precision": 0.8795731707317074,
-    "eval_recall": 0.8697945080923805,
-    "eval_runtime": 4.7492,
+    "eval_accuracy": 0.9022556390977443,
+    "eval_f1": 0.8799463033398397,
+    "eval_loss": 0.790817379951477,
+    "eval_precision": 0.8874803397294746,
+    "eval_recall": 0.8733406073831607,
+    "eval_runtime": 1.6569,
     "eval_samples": 399,
-    "eval_samples_per_second": 84.015,
-    "eval_steps_per_second": 10.528,
-    "f1": 0.8941622310727313,
-    "precision": 0.8899245317748838,
-    "recall": 0.8987993825126839,
-    "train_loss": 0.05962105130807298,
-    "train_runtime": 2691.1784,
+    "eval_samples_per_second": 240.805,
+    "eval_steps_per_second": 30.176,
+    "f1": 0.8952398693685564,
+    "precision": 0.8913160733549084,
+    "recall": 0.8995006447847737,
+    "train_loss": 0.0588726386183598,
+    "train_runtime": 864.0501,
     "train_samples": 3638,
-    "train_samples_per_second": 27.036,
-    "train_steps_per_second": 0.907
+    "train_samples_per_second": 84.208,
+    "train_steps_per_second": 2.824
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.8972431077694235,
-    "eval_f1": 0.8744522298370696,
-    "eval_loss": 0.7891051769256592,
-    "eval_precision": 0.8795731707317074,
-    "eval_recall": 0.8697945080923805,
-    "eval_runtime": 4.7492,
+    "eval_accuracy": 0.9022556390977443,
+    "eval_f1": 0.8799463033398397,
+    "eval_loss": 0.790817379951477,
+    "eval_precision": 0.8874803397294746,
+    "eval_recall": 0.8733406073831607,
+    "eval_runtime": 1.6569,
     "eval_samples": 399,
-    "eval_samples_per_second": 84.015,
-    "eval_steps_per_second": 10.528
+    "eval_samples_per_second": 240.805,
+    "eval_steps_per_second": 30.176
 }

predict_results.json

@@ -1,6 +1,6 @@
 {
-    "accuracy": 0.9109792284866469,
-    "f1": 0.8941622310727313,
-    "precision": 0.8899245317748838,
-    "recall": 0.8987993825126839
+    "accuracy": 0.9119683481701286,
+    "f1": 0.8952398693685564,
+    "precision": 0.8913160733549084,
+    "recall": 0.8995006447847737
 }

predict_results.txt

@@ -3,7 +3,7 @@ index	prediction
 1	1
 2	1
 3	1
-4	1
+4	0
 5	0
 6	1
 7	1
@@ -20,12 +20,12 @@ index	prediction
 18	1
 19	1
 20	1
-21	0
-22	1
+21	1
+22	0
 23	1
 24	1
-25	0
-26	0
+25	1
+26	1
 27	1
 28	1
 29	1
@@ -35,7 +35,7 @@ index	prediction
 33	1
 34	1
 35	1
-36	0
+36	1
 37	1
 38	1
 39	1
@@ -50,13 +50,13 @@ index	prediction
 48	1
 49	1
 50	1
-51	1
+51	0
 52	1
 53	1
 54	0
 55	1
 56	1
-57	1
+57	0
 58	1
 59	1
 60	0
@@ -66,7 +66,7 @@ index	prediction
 64	1
 65	1
 66	1
-67	0
+67	1
 68	0
 69	1
 70	1
@@ -86,7 +86,7 @@ index	prediction
 84	1
 85	1
 86	1
-87	0
+87	1
 88	1
 89	1
 90	0
@@ -109,17 +109,17 @@ index	prediction
 107	1
 108	1
 109	1
-110	1
+110	0
 111	1
 112	1
 113	1
 114	1
-115	0
+115	1
 116	1
 117	1
 118	1
 119	1
-120	0
+120	1
 121	0
 122	1
 123	1
@@ -130,7 +130,7 @@ index	prediction
 128	1
 129	1
 130	0
-131	1
+131	0
 132	1
 133	1
 134	1
@@ -148,9 +148,9 @@ index	prediction
 146	1
 147	1
 148	1
-149	1
+149	0
 150	1
-151	0
+151	1
 152	1
 153	1
 154	1
@@ -180,7 +180,7 @@ index	prediction
 178	1
 179	1
 180	1
-181	1
+181	0
 182	1
 183	1
 184	1
@@ -239,7 +239,7 @@ index	prediction
 237	0
 238	1
 239	1
-240	0
+240	1
 241	1
 242	1
 243	1
@@ -253,16 +253,16 @@ index	prediction
 251	1
 252	1
 253	1
-254	0
+254	1
 255	1
-256	1
-257	0
-258	1
+256	0
+257	1
+258	0
 259	1
-260	1
+260	0
 261	1
 262	1
-263	0
+263	1
 264	1
 265	1
 266	1
@@ -285,10 +285,10 @@ index	prediction
 283	1
 284	1
 285	1
-286	1
+286	0
 287	1
 288	1
-289	1
+289	0
 290	1
 291	1
 292	1
@@ -322,7 +322,7 @@ index	prediction
 320	0
 321	0
 322	0
-323	0
+323	1
 324	0
 325	0
 326	0
@@ -339,9 +339,9 @@ index	prediction
 337	0
 338	1
 339	0
-340	0
+340	1
 341	0
-342	1
+342	0
 343	0
 344	0
 345	0
@@ -368,7 +368,7 @@ index	prediction
 366	0
 367	0
 368	0
-369	0
+369	1
 370	0
 371	0
 372	0
@@ -434,7 +434,7 @@ index	prediction
 432	0
 433	0
 434	1
-435	1
+435	0
 436	0
 437	0
 438	0
@@ -456,7 +456,7 @@ index	prediction
 454	0
 455	1
 456	0
-457	0
+457	1
 458	0
 459	0
 460	0
@@ -486,7 +486,7 @@ index	prediction
 484	0
 485	0
 486	0
-487	0
+487	1
 488	0
 489	0
 490	0
@@ -497,7 +497,7 @@ index	prediction
 495	0
 496	0
 497	0
-498	0
+498	1
 499	0
 500	0
 501	0
@@ -508,7 +508,7 @@ index	prediction
 506	0
 507	0
 508	1
-509	0
+509	1
 510	0
 511	0
 512	0
@@ -521,8 +521,8 @@ index	prediction
 519	0
 520	0
 521	0
-522	0
-523	1
+522	1
+523	0
 524	0
 525	0
 526	0
@@ -580,7 +580,7 @@ index	prediction
 578	0
 579	1
 580	0
-581	1
+581	0
 582	1
 583	0
 584	0
@@ -590,7 +590,7 @@ index	prediction
 588	1
 589	0
 590	0
-591	1
+591	0
 592	0
 593	0
 594	0
@@ -605,7 +605,7 @@ index	prediction
 603	0
 604	1
 605	0
-606	1
+606	0
 607	0
 608	0
 609	0
@@ -664,7 +664,7 @@ index	prediction
 662	0
 663	0
 664	0
-665	1
+665	0
 666	0
 667	1
 668	0
@@ -709,7 +709,7 @@ index	prediction
 707	0
 708	0
 709	0
-710	0
+710	1
 711	0
 712	1
 713	0
@@ -723,7 +723,7 @@ index	prediction
 721	0
 722	0
 723	0
-724	1
+724	0
 725	0
 726	0
 727	0
@@ -736,7 +736,7 @@ index	prediction
 734	1
 735	0
 736	0
-737	0
+737	1
 738	0
 739	0
 740	0
@@ -812,7 +812,7 @@ index	prediction
 810	0
 811	0
 812	1
-813	1
+813	0
 814	0
 815	0
 816	0
@@ -838,19 +838,19 @@ index	prediction
 836	0
 837	0
 838	0
-839	1
+839	0
 840	0
 841	0
 842	0
 843	0
 844	0
 845	0
-846	0
+846	1
 847	0
 848	0
 849	0
-850	0
-851	0
+850	1
+851	1
 852	1
 853	0
 854	1
@@ -862,7 +862,7 @@ index	prediction
 860	0
 861	0
 862	0
-863	1
+863	0
 864	0
 865	1
 866	0
@@ -892,13 +892,13 @@ index	prediction
 890	0
 891	0
 892	0
-893	0
+893	1
 894	0
 895	0
 896	1
 897	0
 898	0
-899	1
+899	0
 900	0
 901	0
 902	0
@@ -916,7 +916,7 @@ index	prediction
 914	0
 915	0
 916	0
-917	1
+917	0
 918	0
 919	0
 920	0
@@ -953,7 +953,7 @@ index	prediction
 951	0
 952	0
 953	0
-954	1
+954	0
 955	0
 956	0
 957	0
@@ -981,7 +981,7 @@ index	prediction
 979	0
 980	0
 981	0
-982	1
+982	0
 983	0
 984	0
 985	0

runs/Jun03_09-57-46_a358b85c7679/events.out.tfevents.1717409561.a358b85c7679.18986.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d48601fe52b0be1678a007861adb9d43e2e9a7da888092d086e7a47c15375e71
+size 560

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "train_loss": 0.05962105130807298,
-    "train_runtime": 2691.1784,
+    "train_loss": 0.0588726386183598,
+    "train_runtime": 864.0501,
     "train_samples": 3638,
-    "train_samples_per_second": 27.036,
-    "train_steps_per_second": 0.907
+    "train_samples_per_second": 84.208,
+    "train_steps_per_second": 2.824
 }

trainer_state.json

@@ -10,392 +10,392 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 21.05661964416504,
+      "grad_norm": 20.136756896972656,
       "learning_rate": 4.75e-05,
-      "loss": 0.411,
+      "loss": 0.3889,
       "step": 122
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8721804511278195,
-      "eval_f1": 0.8446368446368446,
-      "eval_loss": 0.275076687335968,
-      "eval_precision": 0.8473684210526315,
-      "eval_recall": 0.8420621931260229,
-      "eval_runtime": 4.9569,
-      "eval_samples_per_second": 80.493,
-      "eval_steps_per_second": 10.087,
+      "eval_accuracy": 0.8045112781954887,
+      "eval_f1": 0.7109554944646705,
+      "eval_loss": 0.4199941158294678,
+      "eval_precision": 0.8255285412262157,
+      "eval_recall": 0.6866703036915802,
+      "eval_runtime": 1.6394,
+      "eval_samples_per_second": 243.375,
+      "eval_steps_per_second": 30.498,
       "step": 122
     },
     {
       "epoch": 2.0,
-      "grad_norm": 21.74665069580078,
+      "grad_norm": 24.683944702148438,
       "learning_rate": 4.5e-05,
-      "loss": 0.2264,
+      "loss": 0.2335,
       "step": 244
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8872180451127819,
-      "eval_f1": 0.8718936621074937,
-      "eval_loss": 0.3036659359931946,
-      "eval_precision": 0.8573529411764707,
-      "eval_recall": 0.897708674304419,
-      "eval_runtime": 4.9408,
-      "eval_samples_per_second": 80.756,
-      "eval_steps_per_second": 10.12,
+      "eval_accuracy": 0.8922305764411027,
+      "eval_f1": 0.8739355018846853,
+      "eval_loss": 0.3136064410209656,
+      "eval_precision": 0.864426651415499,
+      "eval_recall": 0.886252045826514,
+      "eval_runtime": 1.6497,
+      "eval_samples_per_second": 241.866,
+      "eval_steps_per_second": 30.309,
       "step": 244
     },
     {
       "epoch": 3.0,
-      "grad_norm": 2.0689539909362793,
+      "grad_norm": 66.46725463867188,
       "learning_rate": 4.25e-05,
-      "loss": 0.1467,
+      "loss": 0.1411,
       "step": 366
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8771929824561403,
-      "eval_f1": 0.8582079268956014,
-      "eval_loss": 0.3442274332046509,
-      "eval_precision": 0.8464828897338403,
-      "eval_recall": 0.8756137479541735,
-      "eval_runtime": 4.9579,
-      "eval_samples_per_second": 80.478,
-      "eval_steps_per_second": 10.085,
+      "eval_accuracy": 0.8972431077694235,
+      "eval_f1": 0.8751002084335417,
+      "eval_loss": 0.35689812898635864,
+      "eval_precision": 0.8780701754385964,
+      "eval_recall": 0.8722949627204946,
+      "eval_runtime": 1.6606,
+      "eval_samples_per_second": 240.275,
+      "eval_steps_per_second": 30.11,
       "step": 366
     },
     {
       "epoch": 4.0,
-      "grad_norm": 1.8201183080673218,
+      "grad_norm": 42.06414031982422,
       "learning_rate": 4e-05,
-      "loss": 0.0961,
+      "loss": 0.1078,
       "step": 488
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.899749373433584,
-      "eval_f1": 0.8778322106552358,
-      "eval_loss": 0.3736657500267029,
-      "eval_precision": 0.8818924438393465,
-      "eval_recall": 0.8740680123658847,
-      "eval_runtime": 4.9699,
-      "eval_samples_per_second": 80.283,
-      "eval_steps_per_second": 10.061,
+      "eval_accuracy": 0.9147869674185464,
+      "eval_f1": 0.8991765265473572,
+      "eval_loss": 0.35370269417762756,
+      "eval_precision": 0.8922773722627737,
+      "eval_recall": 0.9072104018912529,
+      "eval_runtime": 1.653,
+      "eval_samples_per_second": 241.378,
+      "eval_steps_per_second": 30.248,
       "step": 488
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.0896231159567833,
+      "grad_norm": 88.54315185546875,
       "learning_rate": 3.7500000000000003e-05,
-      "loss": 0.0726,
+      "loss": 0.0822,
       "step": 610
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.899749373433584,
-      "eval_f1": 0.8771929824561404,
-      "eval_loss": 0.43064403533935547,
-      "eval_precision": 0.8834928678678678,
-      "eval_recall": 0.8715675577377705,
-      "eval_runtime": 4.9329,
-      "eval_samples_per_second": 80.886,
-      "eval_steps_per_second": 10.136,
+      "eval_accuracy": 0.8796992481203008,
+      "eval_f1": 0.8439374185136896,
+      "eval_loss": 0.5069139003753662,
+      "eval_precision": 0.8794955044955045,
+      "eval_recall": 0.822376795781051,
+      "eval_runtime": 1.6524,
+      "eval_samples_per_second": 241.466,
+      "eval_steps_per_second": 30.259,
       "step": 610
     },
     {
       "epoch": 6.0,
-      "grad_norm": 53.643978118896484,
+      "grad_norm": 114.8245849609375,
       "learning_rate": 3.5e-05,
-      "loss": 0.0514,
+      "loss": 0.0529,
       "step": 732
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8847117794486216,
-      "eval_f1": 0.8677208256457565,
-      "eval_loss": 0.6448621153831482,
-      "eval_precision": 0.8546209186496956,
-      "eval_recall": 0.8884342607746863,
-      "eval_runtime": 4.9504,
-      "eval_samples_per_second": 80.6,
-      "eval_steps_per_second": 10.1,
+      "eval_accuracy": 0.9072681704260651,
+      "eval_f1": 0.888964101175568,
+      "eval_loss": 0.42624175548553467,
+      "eval_precision": 0.8862007168458781,
+      "eval_recall": 0.8918894344426259,
+      "eval_runtime": 1.6561,
+      "eval_samples_per_second": 240.934,
+      "eval_steps_per_second": 30.192,
       "step": 732
     },
     {
       "epoch": 7.0,
-      "grad_norm": 0.03521590679883957,
+      "grad_norm": 0.022069375962018967,
       "learning_rate": 3.2500000000000004e-05,
-      "loss": 0.0532,
+      "loss": 0.0365,
       "step": 854
     },
     {
       "epoch": 7.0,
       "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8763538792940554,
-      "eval_loss": 0.5595228672027588,
-      "eval_precision": 0.8754297605404427,
-      "eval_recall": 0.877295871976723,
-      "eval_runtime": 4.952,
-      "eval_samples_per_second": 80.574,
-      "eval_steps_per_second": 10.097,
+      "eval_f1": 0.8769602202215754,
+      "eval_loss": 0.5586097836494446,
+      "eval_precision": 0.8742831541218639,
+      "eval_recall": 0.8797963266048372,
+      "eval_runtime": 1.6532,
+      "eval_samples_per_second": 241.352,
+      "eval_steps_per_second": 30.245,
       "step": 854
     },
     {
       "epoch": 8.0,
-      "grad_norm": 0.004897149745374918,
+      "grad_norm": 0.0406961552798748,
       "learning_rate": 3e-05,
-      "loss": 0.0274,
+      "loss": 0.033,
       "step": 976
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8872180451127819,
-      "eval_f1": 0.8614765038536611,
-      "eval_loss": 0.6727907657623291,
-      "eval_precision": 0.8686536646744258,
-      "eval_recall": 0.8552009456264775,
-      "eval_runtime": 4.9909,
-      "eval_samples_per_second": 79.945,
-      "eval_steps_per_second": 10.018,
+      "eval_accuracy": 0.8947368421052632,
+      "eval_f1": 0.8674628282189181,
+      "eval_loss": 0.5012311935424805,
+      "eval_precision": 0.8869858462356303,
+      "eval_recall": 0.8530187306783051,
+      "eval_runtime": 1.6551,
+      "eval_samples_per_second": 241.075,
+      "eval_steps_per_second": 30.21,
       "step": 976
     },
     {
       "epoch": 9.0,
-      "grad_norm": 0.02297130785882473,
+      "grad_norm": 0.6461573243141174,
       "learning_rate": 2.7500000000000004e-05,
-      "loss": 0.0186,
+      "loss": 0.0248,
       "step": 1098
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9072681704260651,
-      "eval_f1": 0.8848664457009163,
-      "eval_loss": 0.6217536926269531,
-      "eval_precision": 0.8977236138837015,
-      "eval_recall": 0.8743862520458265,
-      "eval_runtime": 4.9504,
-      "eval_samples_per_second": 80.6,
-      "eval_steps_per_second": 10.1,
+      "eval_accuracy": 0.8922305764411027,
+      "eval_f1": 0.8631217838765008,
+      "eval_loss": 0.583283007144928,
+      "eval_precision": 0.8872804935927859,
+      "eval_recall": 0.8462447717766868,
+      "eval_runtime": 1.6572,
+      "eval_samples_per_second": 240.772,
+      "eval_steps_per_second": 30.172,
       "step": 1098
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.018355082720518112,
+      "grad_norm": 0.12847253680229187,
       "learning_rate": 2.5e-05,
-      "loss": 0.0121,
+      "loss": 0.0123,
       "step": 1220
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.8922305764411027,
-      "eval_f1": 0.8669226294357184,
-      "eval_loss": 0.6576113104820251,
-      "eval_precision": 0.8766227567773959,
-      "eval_recall": 0.8587470449172576,
-      "eval_runtime": 4.9651,
-      "eval_samples_per_second": 80.361,
-      "eval_steps_per_second": 10.07,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8805765113084321,
+      "eval_loss": 0.6610547304153442,
+      "eval_precision": 0.8857796167247387,
+      "eval_recall": 0.8758410620112748,
+      "eval_runtime": 1.6505,
+      "eval_samples_per_second": 241.744,
+      "eval_steps_per_second": 30.294,
       "step": 1220
     },
     {
       "epoch": 11.0,
-      "grad_norm": 0.0026700079906731844,
+      "grad_norm": 0.003805552376434207,
       "learning_rate": 2.25e-05,
-      "loss": 0.0244,
+      "loss": 0.0088,
       "step": 1342
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8694882125334078,
-      "eval_loss": 0.7506579160690308,
-      "eval_precision": 0.8940436639772188,
-      "eval_recall": 0.8522913256955811,
-      "eval_runtime": 4.9691,
-      "eval_samples_per_second": 80.296,
-      "eval_steps_per_second": 10.062,
+      "eval_accuracy": 0.8947368421052632,
+      "eval_f1": 0.8682132746146587,
+      "eval_loss": 0.6935672760009766,
+      "eval_precision": 0.884741537654159,
+      "eval_recall": 0.8555191853064193,
+      "eval_runtime": 1.6547,
+      "eval_samples_per_second": 241.138,
+      "eval_steps_per_second": 30.218,
       "step": 1342
     },
     {
       "epoch": 12.0,
-      "grad_norm": 0.002809323836117983,
+      "grad_norm": 0.0037182692904025316,
       "learning_rate": 2e-05,
-      "loss": 0.0062,
+      "loss": 0.0074,
       "step": 1464
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8724195749658803,
-      "eval_loss": 0.6859227418899536,
-      "eval_precision": 0.8848766823362741,
-      "eval_recall": 0.8622931442080378,
-      "eval_runtime": 4.9578,
-      "eval_samples_per_second": 80.48,
-      "eval_steps_per_second": 10.085,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8805765113084321,
+      "eval_loss": 0.6789939403533936,
+      "eval_precision": 0.8857796167247387,
+      "eval_recall": 0.8758410620112748,
+      "eval_runtime": 1.6567,
+      "eval_samples_per_second": 240.838,
+      "eval_steps_per_second": 30.18,
       "step": 1464
     },
     {
       "epoch": 13.0,
-      "grad_norm": 0.004050145391374826,
+      "grad_norm": 0.0025616472121328115,
       "learning_rate": 1.75e-05,
-      "loss": 0.0099,
+      "loss": 0.0141,
       "step": 1586
     },
     {
       "epoch": 13.0,
-      "eval_accuracy": 0.9072681704260651,
-      "eval_f1": 0.8872855539522206,
-      "eval_loss": 0.6514401435852051,
-      "eval_precision": 0.8903508771929824,
-      "eval_recall": 0.8843880705582834,
-      "eval_runtime": 4.9373,
-      "eval_samples_per_second": 80.813,
-      "eval_steps_per_second": 10.127,
+      "eval_accuracy": 0.8972431077694235,
+      "eval_f1": 0.8731122745782431,
+      "eval_loss": 0.6981470584869385,
+      "eval_precision": 0.8829705994654449,
+      "eval_recall": 0.864793598836152,
+      "eval_runtime": 1.6639,
+      "eval_samples_per_second": 239.794,
+      "eval_steps_per_second": 30.049,
       "step": 1586
     },
     {
       "epoch": 14.0,
-      "grad_norm": 0.0045247310772538185,
+      "grad_norm": 0.006673410069197416,
       "learning_rate": 1.5e-05,
-      "loss": 0.0087,
+      "loss": 0.0034,
       "step": 1708
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.899749373433584,
-      "eval_f1": 0.8765393898137261,
-      "eval_loss": 0.7604307532310486,
-      "eval_precision": 0.8852261942423283,
-      "eval_recall": 0.8690671031096563,
-      "eval_runtime": 4.9312,
-      "eval_samples_per_second": 80.913,
-      "eval_steps_per_second": 10.14,
+      "eval_accuracy": 0.8972431077694235,
+      "eval_f1": 0.8751002084335417,
+      "eval_loss": 0.7144644856452942,
+      "eval_precision": 0.8780701754385964,
+      "eval_recall": 0.8722949627204946,
+      "eval_runtime": 1.6531,
+      "eval_samples_per_second": 241.366,
+      "eval_steps_per_second": 30.246,
       "step": 1708
     },
     {
       "epoch": 15.0,
-      "grad_norm": 0.002084016567096114,
+      "grad_norm": 0.0030696168541908264,
       "learning_rate": 1.25e-05,
-      "loss": 0.0056,
+      "loss": 0.0059,
       "step": 1830
     },
     {
       "epoch": 15.0,
-      "eval_accuracy": 0.9022556390977443,
-      "eval_f1": 0.8799463033398397,
-      "eval_loss": 0.7281515598297119,
-      "eval_precision": 0.8874803397294746,
-      "eval_recall": 0.8733406073831607,
-      "eval_runtime": 4.9315,
-      "eval_samples_per_second": 80.909,
-      "eval_steps_per_second": 10.139,
+      "eval_accuracy": 0.899749373433584,
+      "eval_f1": 0.8758710801393728,
+      "eval_loss": 0.7303631901741028,
+      "eval_precision": 0.8870983228779925,
+      "eval_recall": 0.8665666484815421,
+      "eval_runtime": 1.6541,
+      "eval_samples_per_second": 241.215,
+      "eval_steps_per_second": 30.227,
       "step": 1830
     },
     {
       "epoch": 16.0,
-      "grad_norm": 0.0016144708497449756,
+      "grad_norm": 0.0017388605047017336,
       "learning_rate": 1e-05,
-      "loss": 0.0063,
+      "loss": 0.0056,
       "step": 1952
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.9122807017543859,
-      "eval_f1": 0.89337822671156,
-      "eval_loss": 0.6987277269363403,
-      "eval_precision": 0.8964912280701754,
-      "eval_recall": 0.8904346244771777,
-      "eval_runtime": 4.9746,
-      "eval_samples_per_second": 80.207,
-      "eval_steps_per_second": 10.051,
+      "eval_accuracy": 0.899749373433584,
+      "eval_f1": 0.879667048676036,
+      "eval_loss": 0.7517656683921814,
+      "eval_precision": 0.8778361344537815,
+      "eval_recall": 0.8815693762502272,
+      "eval_runtime": 1.6536,
+      "eval_samples_per_second": 241.288,
+      "eval_steps_per_second": 30.237,
       "step": 1952
     },
     {
       "epoch": 17.0,
-      "grad_norm": 0.0015528218355029821,
+      "grad_norm": 0.002333118114620447,
       "learning_rate": 7.5e-06,
-      "loss": 0.0071,
+      "loss": 0.0039,
       "step": 2074
     },
     {
       "epoch": 17.0,
-      "eval_accuracy": 0.9047619047619048,
-      "eval_f1": 0.8833333333333333,
-      "eval_loss": 0.7402216792106628,
-      "eval_precision": 0.8897334834834836,
-      "eval_recall": 0.8776141116566649,
-      "eval_runtime": 4.9544,
-      "eval_samples_per_second": 80.535,
-      "eval_steps_per_second": 10.092,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8793019197207679,
+      "eval_loss": 0.7390431761741638,
+      "eval_precision": 0.8893184421534936,
+      "eval_recall": 0.8708401527550463,
+      "eval_runtime": 1.655,
+      "eval_samples_per_second": 241.08,
+      "eval_steps_per_second": 30.211,
       "step": 2074
     },
     {
       "epoch": 18.0,
-      "grad_norm": 0.0025696582160890102,
+      "grad_norm": 0.0018157872837036848,
       "learning_rate": 5e-06,
-      "loss": 0.0023,
+      "loss": 0.004,
       "step": 2196
     },
     {
       "epoch": 18.0,
-      "eval_accuracy": 0.8922305764411027,
-      "eval_f1": 0.8690075356742023,
-      "eval_loss": 0.7846499085426331,
-      "eval_precision": 0.8719298245614036,
-      "eval_recall": 0.8662484088016003,
-      "eval_runtime": 4.9898,
-      "eval_samples_per_second": 79.962,
-      "eval_steps_per_second": 10.02,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8799463033398397,
+      "eval_loss": 0.764133095741272,
+      "eval_precision": 0.8874803397294746,
+      "eval_recall": 0.8733406073831607,
+      "eval_runtime": 1.6667,
+      "eval_samples_per_second": 239.389,
+      "eval_steps_per_second": 29.999,
       "step": 2196
     },
     {
       "epoch": 19.0,
-      "grad_norm": 0.0012161381309852004,
+      "grad_norm": 0.0015570241957902908,
       "learning_rate": 2.5e-06,
-      "loss": 0.0043,
+      "loss": 0.007,
       "step": 2318
     },
     {
       "epoch": 19.0,
-      "eval_accuracy": 0.8922305764411027,
-      "eval_f1": 0.8690075356742023,
-      "eval_loss": 0.7948206067085266,
-      "eval_precision": 0.8719298245614036,
-      "eval_recall": 0.8662484088016003,
-      "eval_runtime": 4.9737,
-      "eval_samples_per_second": 80.222,
-      "eval_steps_per_second": 10.053,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8799463033398397,
+      "eval_loss": 0.7847548723220825,
+      "eval_precision": 0.8874803397294746,
+      "eval_recall": 0.8733406073831607,
+      "eval_runtime": 1.664,
+      "eval_samples_per_second": 239.788,
+      "eval_steps_per_second": 30.049,
       "step": 2318
     },
     {
       "epoch": 20.0,
-      "grad_norm": 0.0017919199308380485,
+      "grad_norm": 0.002853752113878727,
       "learning_rate": 0.0,
-      "loss": 0.0021,
+      "loss": 0.0042,
       "step": 2440
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.8972431077694235,
-      "eval_f1": 0.8744522298370696,
-      "eval_loss": 0.7891051769256592,
-      "eval_precision": 0.8795731707317074,
-      "eval_recall": 0.8697945080923805,
-      "eval_runtime": 4.9669,
-      "eval_samples_per_second": 80.331,
-      "eval_steps_per_second": 10.067,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_f1": 0.8799463033398397,
+      "eval_loss": 0.790817379951477,
+      "eval_precision": 0.8874803397294746,
+      "eval_recall": 0.8733406073831607,
+      "eval_runtime": 1.6678,
+      "eval_samples_per_second": 239.236,
+      "eval_steps_per_second": 29.979,
       "step": 2440
     },
     {
       "epoch": 20.0,
       "step": 2440,
       "total_flos": 7584162436176000.0,
-      "train_loss": 0.05962105130807298,
-      "train_runtime": 2691.1784,
-      "train_samples_per_second": 27.036,
-      "train_steps_per_second": 0.907
+      "train_loss": 0.0588726386183598,
+      "train_runtime": 864.0501,
+      "train_samples_per_second": 84.208,
+      "train_steps_per_second": 2.824
     }
   ],
   "logging_steps": 500,