End of training

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 10.0,
-    "eval_gen_len": 52.717333333333336,
-    "eval_loss": 2.173374652862549,
-    "eval_rouge1": 26.8011,
-    "eval_rouge2": 8.2533,
-    "eval_rougeL": 18.7776,
-    "eval_rougeLsum": 23.2859,
-    "eval_runtime": 369.7996,
+    "eval_gen_len": 52.897333333333336,
+    "eval_loss": 2.143275499343872,
+    "eval_rouge1": 27.2333,
+    "eval_rouge2": 8.451,
+    "eval_rougeL": 18.8946,
+    "eval_rougeLsum": 23.575,
+    "eval_runtime": 250.9824,
     "eval_samples": 1500,
-    "eval_samples_per_second": 4.056,
-    "eval_steps_per_second": 0.508,
-    "train_loss": 2.152468052455357,
-    "train_runtime": 10217.243,
+    "eval_samples_per_second": 5.977,
+    "eval_steps_per_second": 0.749,
+    "train_loss": 2.0453601527622767,
+    "train_runtime": 6220.6473,
     "train_samples": 7000,
-    "train_samples_per_second": 6.851,
-    "train_steps_per_second": 0.856
+    "train_samples_per_second": 11.253,
+    "train_steps_per_second": 1.407
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_gen_len": 52.717333333333336,
-    "eval_loss": 2.173374652862549,
-    "eval_rouge1": 26.8011,
-    "eval_rouge2": 8.2533,
-    "eval_rougeL": 18.7776,
-    "eval_rougeLsum": 23.2859,
-    "eval_runtime": 369.7996,
+    "eval_gen_len": 52.897333333333336,
+    "eval_loss": 2.143275499343872,
+    "eval_rouge1": 27.2333,
+    "eval_rouge2": 8.451,
+    "eval_rougeL": 18.8946,
+    "eval_rougeLsum": 23.575,
+    "eval_runtime": 250.9824,
     "eval_samples": 1500,
-    "eval_samples_per_second": 4.056,
-    "eval_steps_per_second": 0.508
+    "eval_samples_per_second": 5.977,
+    "eval_steps_per_second": 0.749
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 2.152468052455357,
-    "train_runtime": 10217.243,
+    "train_loss": 2.0453601527622767,
+    "train_runtime": 6220.6473,
     "train_samples": 7000,
-    "train_samples_per_second": 6.851,
-    "train_steps_per_second": 0.856
+    "train_samples_per_second": 11.253,
+    "train_steps_per_second": 1.407
 }

trainer_state.json

@@ -10,118 +10,118 @@
     {
       "epoch": 0.57,
       "learning_rate": 4.714285714285714e-05,
-      "loss": 2.5639,
+      "loss": 2.3409,
       "step": 500
     },
     {
       "epoch": 1.14,
       "learning_rate": 4.428571428571428e-05,
-      "loss": 2.4439,
+      "loss": 2.3049,
       "step": 1000
     },
     {
       "epoch": 1.71,
       "learning_rate": 4.1428571428571437e-05,
-      "loss": 2.3715,
+      "loss": 2.2384,
       "step": 1500
     },
     {
       "epoch": 2.29,
       "learning_rate": 3.857142857142858e-05,
-      "loss": 2.2965,
+      "loss": 2.1886,
       "step": 2000
     },
     {
       "epoch": 2.86,
       "learning_rate": 3.571428571428572e-05,
-      "loss": 2.2701,
+      "loss": 2.1368,
       "step": 2500
     },
     {
       "epoch": 3.43,
       "learning_rate": 3.285714285714286e-05,
-      "loss": 2.2025,
+      "loss": 2.1029,
       "step": 3000
     },
     {
       "epoch": 4.0,
       "learning_rate": 3e-05,
-      "loss": 2.1752,
+      "loss": 2.0797,
       "step": 3500
     },
     {
       "epoch": 4.57,
       "learning_rate": 2.714285714285714e-05,
-      "loss": 2.1169,
+      "loss": 2.0263,
       "step": 4000
     },
     {
       "epoch": 5.14,
       "learning_rate": 2.4285714285714288e-05,
-      "loss": 2.1151,
+      "loss": 2.0152,
       "step": 4500
     },
     {
       "epoch": 5.71,
       "learning_rate": 2.1428571428571428e-05,
-      "loss": 2.0713,
+      "loss": 1.9843,
       "step": 5000
     },
     {
       "epoch": 6.29,
       "learning_rate": 1.8571428571428572e-05,
-      "loss": 2.0616,
+      "loss": 1.9578,
       "step": 5500
     },
     {
       "epoch": 6.86,
       "learning_rate": 1.5714285714285715e-05,
-      "loss": 2.0385,
+      "loss": 1.9462,
       "step": 6000
     },
     {
       "epoch": 7.43,
       "learning_rate": 1.2857142857142857e-05,
-      "loss": 2.0265,
+      "loss": 1.9366,
       "step": 6500
     },
     {
       "epoch": 8.0,
       "learning_rate": 1e-05,
-      "loss": 1.9998,
+      "loss": 1.9146,
       "step": 7000
     },
     {
       "epoch": 8.57,
       "learning_rate": 7.142857142857143e-06,
-      "loss": 1.9831,
+      "loss": 1.8964,
       "step": 7500
     },
     {
       "epoch": 9.14,
       "learning_rate": 4.285714285714286e-06,
-      "loss": 1.9831,
+      "loss": 1.8958,
       "step": 8000
     },
     {
       "epoch": 9.71,
       "learning_rate": 1.4285714285714286e-06,
-      "loss": 1.9695,
+      "loss": 1.8854,
       "step": 8500
     },
     {
       "epoch": 10.0,
       "step": 8750,
-      "total_flos": 4.262140044877824e+16,
-      "train_loss": 2.152468052455357,
-      "train_runtime": 10217.243,
-      "train_samples_per_second": 6.851,
-      "train_steps_per_second": 0.856
+      "total_flos": 4.261889801244672e+16,
+      "train_loss": 2.0453601527622767,
+      "train_runtime": 6220.6473,
+      "train_samples_per_second": 11.253,
+      "train_steps_per_second": 1.407
     }
   ],
   "max_steps": 8750,
   "num_train_epochs": 10,
-  "total_flos": 4.262140044877824e+16,
+  "total_flos": 4.261889801244672e+16,
   "trial_name": null,
   "trial_params": null
 }