End of training

all_results.json

@@ -1,17 +1,17 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.6093189964157706,
+    "eval_accuracy": 0.4444444444444444,
     "eval_gen_len": 1.0,
-    "eval_loss": 1.8930710554122925,
+    "eval_loss": 3.502009630203247,
     "eval_rouge1": 0.0,
     "eval_rouge2": 0.0,
     "eval_rougeL": 0.0,
     "eval_rougeLsum": 0.0,
-    "eval_runtime": 0.1102,
-    "eval_samples": 3,
-    "eval_samples_per_second": 27.23,
-    "eval_steps_per_second": 9.077,
-    "perplexity": 6.639728375930875,
+    "eval_runtime": 0.2824,
+    "eval_samples": 176,
+    "eval_samples_per_second": 623.144,
+    "eval_steps_per_second": 77.893,
+    "perplexity": 33.18206868621658,
     "predict_gen_len": 1.0,
     "predict_loss": 5.266044616699219,
     "predict_rouge1": 0.0,
@@ -22,9 +22,9 @@
     "predict_samples": 3527,
     "predict_samples_per_second": 237.858,
     "predict_steps_per_second": 14.904,
-    "train_loss": 2.381487710135324,
-    "train_runtime": 18.165,
-    "train_samples": 49,
-    "train_samples_per_second": 26.975,
-    "train_steps_per_second": 3.854
+    "train_loss": 3.5042898788946366,
+    "train_runtime": 613.1863,
+    "train_samples": 3507,
+    "train_samples_per_second": 57.193,
+    "train_steps_per_second": 7.159
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.6093189964157706,
-    "eval_loss": 1.8930710554122925,
-    "eval_runtime": 0.1102,
-    "eval_samples": 3,
-    "eval_samples_per_second": 27.23,
-    "eval_steps_per_second": 9.077,
-    "perplexity": 6.639728375930875
+    "eval_accuracy": 0.4444444444444444,
+    "eval_loss": 3.502009630203247,
+    "eval_runtime": 0.2824,
+    "eval_samples": 176,
+    "eval_samples_per_second": 623.144,
+    "eval_steps_per_second": 77.893,
+    "perplexity": 33.18206868621658
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 2.381487710135324,
-    "train_runtime": 18.165,
-    "train_samples": 49,
-    "train_samples_per_second": 26.975,
-    "train_steps_per_second": 3.854
+    "train_loss": 3.5042898788946366,
+    "train_runtime": 613.1863,
+    "train_samples": 3507,
+    "train_samples_per_second": 57.193,
+    "train_steps_per_second": 7.159
 }