End of training

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 3.0,
-    "eval_bleu": 3.1263,
-    "eval_gen_len": 101.3243,
-    "eval_loss": 2.2739031314849854,
-    "eval_runtime": 177.4487,
+    "epoch": 10.0,
+    "eval_bleu": 10.9236,
+    "eval_gen_len": 79.2493,
+    "eval_loss": 1.7087136507034302,
+    "eval_runtime": 177.7595,
     "eval_samples": 12422,
-    "eval_samples_per_second": 70.003,
-    "eval_steps_per_second": 0.552,
-    "train_loss": 2.8308688191229425,
-    "train_runtime": 860.89,
+    "eval_samples_per_second": 69.881,
+    "eval_steps_per_second": 0.551,
+    "train_loss": 2.27133990001286,
+    "train_runtime": 2909.2317,
     "train_samples": 62107,
-    "train_samples_per_second": 216.428,
-    "train_steps_per_second": 0.847
+    "train_samples_per_second": 213.482,
+    "train_steps_per_second": 0.835
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
-    "epoch": 3.0,
-    "eval_bleu": 3.1263,
-    "eval_gen_len": 101.3243,
-    "eval_loss": 2.2739031314849854,
-    "eval_runtime": 177.4487,
+    "epoch": 10.0,
+    "eval_bleu": 10.9236,
+    "eval_gen_len": 79.2493,
+    "eval_loss": 1.7087136507034302,
+    "eval_runtime": 177.7595,
     "eval_samples": 12422,
-    "eval_samples_per_second": 70.003,
-    "eval_steps_per_second": 0.552
+    "eval_samples_per_second": 69.881,
+    "eval_steps_per_second": 0.551
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "train_loss": 2.8308688191229425,
-    "train_runtime": 860.89,
+    "epoch": 10.0,
+    "train_loss": 2.27133990001286,
+    "train_runtime": 2909.2317,
     "train_samples": 62107,
-    "train_samples_per_second": 216.428,
-    "train_steps_per_second": 0.847
+    "train_samples_per_second": 213.482,
+    "train_steps_per_second": 0.835
 }

trainer_state.json

@@ -1,31 +1,49 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
-  "global_step": 729,
+  "epoch": 10.0,
+  "global_step": 2430,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 2.06,
-      "learning_rate": 1.570644718792867e-05,
-      "loss": 2.9439,
+      "learning_rate": 3.971193415637861e-05,
+      "loss": 2.8803,
       "step": 500
     },
     {
-      "epoch": 3.0,
-      "step": 729,
-      "total_flos": 6806063508836352.0,
-      "train_loss": 2.8308688191229425,
-      "train_runtime": 860.89,
-      "train_samples_per_second": 216.428,
-      "train_steps_per_second": 0.847
+      "epoch": 4.12,
+      "learning_rate": 2.9423868312757202e-05,
+      "loss": 2.3089,
+      "step": 1000
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 1.91358024691358e-05,
+      "loss": 2.116,
+      "step": 1500
+    },
+    {
+      "epoch": 8.23,
+      "learning_rate": 8.847736625514404e-06,
+      "loss": 2.0249,
+      "step": 2000
+    },
+    {
+      "epoch": 10.0,
+      "step": 2430,
+      "total_flos": 2.2677509529759744e+16,
+      "train_loss": 2.27133990001286,
+      "train_runtime": 2909.2317,
+      "train_samples_per_second": 213.482,
+      "train_steps_per_second": 0.835
     }
   ],
-  "max_steps": 729,
-  "num_train_epochs": 3,
-  "total_flos": 6806063508836352.0,
+  "max_steps": 2430,
+  "num_train_epochs": 10,
+  "total_flos": 2.2677509529759744e+16,
   "trial_name": null,
   "trial_params": null
 }