End of training

all_results.json

@@ -4,30 +4,30 @@
     "eval_NG_number": 3198,
     "eval_NG_precision": 0.7309048178613397,
     "eval_NG_recall": 0.7779862414008756,
-    "eval_loss": 0.29878920316696167,
+    "eval_loss": 0.29878926277160645,
     "eval_overall_accuracy": 0.9019303173369988,
     "eval_overall_f1": 0.7537109966676764,
     "eval_overall_precision": 0.7309048178613397,
     "eval_overall_recall": 0.7779862414008756,
-    "eval_runtime": 2.4235,
+    "eval_runtime": 2.4602,
     "eval_samples": 391,
-    "eval_samples_per_second": 161.337,
-    "eval_steps_per_second": 10.316,
+    "eval_samples_per_second": 158.929,
+    "eval_steps_per_second": 10.162,
     "predict_NG_f1": 0.7636470935379168,
     "predict_NG_number": 3337,
     "predict_NG_precision": 0.7521069456553328,
     "predict_NG_recall": 0.7755468984117471,
-    "predict_loss": 0.2863216698169708,
+    "predict_loss": 0.28632164001464844,
     "predict_overall_accuracy": 0.9084193245778611,
     "predict_overall_f1": 0.7636470935379168,
     "predict_overall_precision": 0.7521069456553328,
     "predict_overall_recall": 0.7755468984117471,
-    "predict_runtime": 2.281,
-    "predict_samples_per_second": 171.414,
-    "predict_steps_per_second": 10.96,
-    "train_loss": 0.23679824628328022,
-    "train_runtime": 150.2286,
+    "predict_runtime": 2.2477,
+    "predict_samples_per_second": 173.954,
+    "predict_steps_per_second": 11.122,
+    "train_loss": 0.23679824461016738,
+    "train_runtime": 149.9061,
     "train_samples": 1824,
-    "train_samples_per_second": 60.707,
-    "train_steps_per_second": 3.794
+    "train_samples_per_second": 60.838,
+    "train_steps_per_second": 3.802
 }

eval_results.json

@@ -4,13 +4,13 @@
     "eval_NG_number": 3198,
     "eval_NG_precision": 0.7309048178613397,
     "eval_NG_recall": 0.7779862414008756,
-    "eval_loss": 0.29878920316696167,
+    "eval_loss": 0.29878926277160645,
     "eval_overall_accuracy": 0.9019303173369988,
     "eval_overall_f1": 0.7537109966676764,
     "eval_overall_precision": 0.7309048178613397,
     "eval_overall_recall": 0.7779862414008756,
-    "eval_runtime": 2.4235,
+    "eval_runtime": 2.4602,
     "eval_samples": 391,
-    "eval_samples_per_second": 161.337,
-    "eval_steps_per_second": 10.316
+    "eval_samples_per_second": 158.929,
+    "eval_steps_per_second": 10.162
 }

predict_results.json

@@ -3,12 +3,12 @@
     "predict_NG_number": 3337,
     "predict_NG_precision": 0.7521069456553328,
     "predict_NG_recall": 0.7755468984117471,
-    "predict_loss": 0.2863216698169708,
+    "predict_loss": 0.28632164001464844,
     "predict_overall_accuracy": 0.9084193245778611,
     "predict_overall_f1": 0.7636470935379168,
     "predict_overall_precision": 0.7521069456553328,
     "predict_overall_recall": 0.7755468984117471,
-    "predict_runtime": 2.281,
-    "predict_samples_per_second": 171.414,
-    "predict_steps_per_second": 10.96
+    "predict_runtime": 2.2477,
+    "predict_samples_per_second": 173.954,
+    "predict_steps_per_second": 11.122
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 0.23679824628328022,
-    "train_runtime": 150.2286,
+    "train_loss": 0.23679824461016738,
+    "train_runtime": 149.9061,
     "train_samples": 1824,
-    "train_samples_per_second": 60.707,
-    "train_steps_per_second": 3.794
+    "train_samples_per_second": 60.838,
+    "train_steps_per_second": 3.802
 }

trainer_state.json

@@ -24,9 +24,9 @@
       "eval_overall_f1": 0.7225241545893719,
       "eval_overall_precision": 0.698482194979568,
       "eval_overall_recall": 0.7482801751094434,
-      "eval_runtime": 2.2675,
-      "eval_samples_per_second": 172.436,
-      "eval_steps_per_second": 11.025,
+      "eval_runtime": 2.2533,
+      "eval_samples_per_second": 173.524,
+      "eval_steps_per_second": 11.095,
       "step": 114
     },
     {
@@ -41,14 +41,14 @@
       "eval_NG_number": 3198,
       "eval_NG_precision": 0.7293508076805851,
       "eval_NG_recall": 0.7482801751094434,
-      "eval_loss": 0.29169175028800964,
+      "eval_loss": 0.29169172048568726,
       "eval_overall_accuracy": 0.8932050439251777,
       "eval_overall_f1": 0.7386942429387251,
       "eval_overall_precision": 0.7293508076805851,
       "eval_overall_recall": 0.7482801751094434,
-      "eval_runtime": 2.3333,
-      "eval_samples_per_second": 167.577,
-      "eval_steps_per_second": 10.715,
+      "eval_runtime": 2.3222,
+      "eval_samples_per_second": 168.377,
+      "eval_steps_per_second": 10.766,
       "step": 228
     },
     {
@@ -63,14 +63,14 @@
       "eval_NG_number": 3198,
       "eval_NG_precision": 0.7306327616794795,
       "eval_NG_recall": 0.7726704190118824,
-      "eval_loss": 0.2864079773426056,
+      "eval_loss": 0.286408007144928,
       "eval_overall_accuracy": 0.9000179286439969,
       "eval_overall_f1": 0.751063829787234,
       "eval_overall_precision": 0.7306327616794795,
       "eval_overall_recall": 0.7726704190118824,
-      "eval_runtime": 2.3686,
-      "eval_samples_per_second": 165.074,
-      "eval_steps_per_second": 10.555,
+      "eval_runtime": 2.3542,
+      "eval_samples_per_second": 166.089,
+      "eval_steps_per_second": 10.62,
       "step": 342
     },
     {
@@ -85,14 +85,14 @@
       "eval_NG_number": 3198,
       "eval_NG_precision": 0.7309048178613397,
       "eval_NG_recall": 0.7779862414008756,
-      "eval_loss": 0.29878920316696167,
+      "eval_loss": 0.29878926277160645,
       "eval_overall_accuracy": 0.9019303173369988,
       "eval_overall_f1": 0.7537109966676764,
       "eval_overall_precision": 0.7309048178613397,
       "eval_overall_recall": 0.7779862414008756,
-      "eval_runtime": 2.3061,
-      "eval_samples_per_second": 169.554,
-      "eval_steps_per_second": 10.841,
+      "eval_runtime": 2.2946,
+      "eval_samples_per_second": 170.401,
+      "eval_steps_per_second": 10.895,
       "step": 456
     },
     {
@@ -112,19 +112,19 @@
       "eval_overall_f1": 0.7514363471424251,
       "eval_overall_precision": 0.7274590163934426,
       "eval_overall_recall": 0.7770481550969356,
-      "eval_runtime": 2.3108,
-      "eval_samples_per_second": 169.206,
-      "eval_steps_per_second": 10.819,
+      "eval_runtime": 2.2998,
+      "eval_samples_per_second": 170.014,
+      "eval_steps_per_second": 10.87,
       "step": 570
     },
     {
       "epoch": 5.0,
       "step": 570,
       "total_flos": 2383047966228480.0,
-      "train_loss": 0.23679824628328022,
-      "train_runtime": 150.2286,
-      "train_samples_per_second": 60.707,
-      "train_steps_per_second": 3.794
+      "train_loss": 0.23679824461016738,
+      "train_runtime": 149.9061,
+      "train_samples_per_second": 60.838,
+      "train_steps_per_second": 3.802
     }
   ],
   "max_steps": 570,