End of training

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 10.0,
-    "eval_gen_len": 53.322,
-    "eval_loss": 2.0281758308410645,
-    "eval_rouge1": 28.26,
-    "eval_rouge2": 9.2063,
-    "eval_rougeL": 19.7844,
-    "eval_rougeLsum": 24.388,
-    "eval_runtime": 249.008,
+    "eval_gen_len": 54.48733333333333,
+    "eval_loss": 2.0228171348571777,
+    "eval_rouge1": 28.4344,
+    "eval_rouge2": 9.3461,
+    "eval_rougeL": 19.7157,
+    "eval_rougeLsum": 24.6508,
+    "eval_runtime": 254.6251,
     "eval_samples": 1500,
-    "eval_samples_per_second": 6.024,
-    "eval_steps_per_second": 0.755,
-    "train_loss": 1.9120913609095982,
-    "train_runtime": 6207.6309,
+    "eval_samples_per_second": 5.891,
+    "eval_steps_per_second": 0.738,
+    "train_loss": 1.883265816824777,
+    "train_runtime": 6265.8424,
     "train_samples": 7000,
-    "train_samples_per_second": 11.276,
-    "train_steps_per_second": 1.41
+    "train_samples_per_second": 11.172,
+    "train_steps_per_second": 1.396
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_gen_len": 53.322,
-    "eval_loss": 2.0281758308410645,
-    "eval_rouge1": 28.26,
-    "eval_rouge2": 9.2063,
-    "eval_rougeL": 19.7844,
-    "eval_rougeLsum": 24.388,
-    "eval_runtime": 249.008,
+    "eval_gen_len": 54.48733333333333,
+    "eval_loss": 2.0228171348571777,
+    "eval_rouge1": 28.4344,
+    "eval_rouge2": 9.3461,
+    "eval_rougeL": 19.7157,
+    "eval_rougeLsum": 24.6508,
+    "eval_runtime": 254.6251,
     "eval_samples": 1500,
-    "eval_samples_per_second": 6.024,
-    "eval_steps_per_second": 0.755
+    "eval_samples_per_second": 5.891,
+    "eval_steps_per_second": 0.738
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 1.9120913609095982,
-    "train_runtime": 6207.6309,
+    "train_loss": 1.883265816824777,
+    "train_runtime": 6265.8424,
     "train_samples": 7000,
-    "train_samples_per_second": 11.276,
-    "train_steps_per_second": 1.41
+    "train_samples_per_second": 11.172,
+    "train_steps_per_second": 1.396
 }

trainer_state.json

@@ -10,118 +10,118 @@
     {
       "epoch": 0.57,
       "learning_rate": 4.714285714285714e-05,
-      "loss": 2.1935,
+      "loss": 2.1596,
       "step": 500
     },
     {
       "epoch": 1.14,
       "learning_rate": 4.428571428571428e-05,
-      "loss": 2.1618,
+      "loss": 2.1314,
       "step": 1000
     },
     {
       "epoch": 1.71,
       "learning_rate": 4.1428571428571437e-05,
-      "loss": 2.0802,
+      "loss": 2.055,
       "step": 1500
     },
     {
       "epoch": 2.29,
       "learning_rate": 3.857142857142858e-05,
-      "loss": 2.0405,
+      "loss": 2.0122,
       "step": 2000
     },
     {
       "epoch": 2.86,
       "learning_rate": 3.571428571428572e-05,
-      "loss": 2.0094,
+      "loss": 1.9833,
       "step": 2500
     },
     {
       "epoch": 3.43,
       "learning_rate": 3.285714285714286e-05,
-      "loss": 1.9689,
+      "loss": 1.9181,
       "step": 3000
     },
     {
       "epoch": 4.0,
       "learning_rate": 3e-05,
-      "loss": 1.9429,
+      "loss": 1.9276,
       "step": 3500
     },
     {
       "epoch": 4.57,
       "learning_rate": 2.714285714285714e-05,
-      "loss": 1.888,
+      "loss": 1.8599,
       "step": 4000
     },
     {
       "epoch": 5.14,
       "learning_rate": 2.4285714285714288e-05,
-      "loss": 1.8982,
+      "loss": 1.8608,
       "step": 4500
     },
     {
       "epoch": 5.71,
       "learning_rate": 2.1428571428571428e-05,
-      "loss": 1.8509,
+      "loss": 1.8234,
       "step": 5000
     },
     {
       "epoch": 6.29,
       "learning_rate": 1.8571428571428572e-05,
-      "loss": 1.8348,
+      "loss": 1.8171,
       "step": 5500
     },
     {
       "epoch": 6.86,
       "learning_rate": 1.5714285714285715e-05,
-      "loss": 1.8173,
+      "loss": 1.7907,
       "step": 6000
     },
     {
       "epoch": 7.43,
       "learning_rate": 1.2857142857142857e-05,
-      "loss": 1.8017,
+      "loss": 1.7693,
       "step": 6500
     },
     {
       "epoch": 8.0,
       "learning_rate": 1e-05,
-      "loss": 1.7911,
+      "loss": 1.7666,
       "step": 7000
     },
     {
       "epoch": 8.57,
       "learning_rate": 7.142857142857143e-06,
-      "loss": 1.7689,
+      "loss": 1.7393,
       "step": 7500
     },
     {
       "epoch": 9.14,
       "learning_rate": 4.285714285714286e-06,
-      "loss": 1.7795,
+      "loss": 1.7473,
       "step": 8000
     },
     {
       "epoch": 9.71,
       "learning_rate": 1.4285714285714286e-06,
-      "loss": 1.7522,
+      "loss": 1.7235,
       "step": 8500
     },
     {
       "epoch": 10.0,
       "step": 8750,
-      "total_flos": 4.261736610275328e+16,
-      "train_loss": 1.9120913609095982,
-      "train_runtime": 6207.6309,
-      "train_samples_per_second": 11.276,
-      "train_steps_per_second": 1.41
+      "total_flos": 4.261233268518912e+16,
+      "train_loss": 1.883265816824777,
+      "train_runtime": 6265.8424,
+      "train_samples_per_second": 11.172,
+      "train_steps_per_second": 1.396
     }
   ],
   "max_steps": 8750,
   "num_train_epochs": 10,
-  "total_flos": 4.261736610275328e+16,
+  "total_flos": 4.261233268518912e+16,
   "trial_name": null,
   "trial_params": null
 }