End of training

all_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 15.99,
-    "train_loss": 0.4757408973498222,
-    "train_runtime": 9951.3836,
-    "train_samples_per_second": 80.391,
-    "train_steps_per_second": 0.627
+    "epoch": 19.99,
+    "train_loss": 0.3883497776129307,
+    "train_runtime": 10238.8851,
+    "train_samples_per_second": 97.667,
+    "train_steps_per_second": 0.762
 }

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 15.99,
-    "train_loss": 0.4757408973498222,
-    "train_runtime": 9951.3836,
-    "train_samples_per_second": 80.391,
-    "train_steps_per_second": 0.627
+    "epoch": 19.99,
+    "train_loss": 0.3883497776129307,
+    "train_runtime": 10238.8851,
+    "train_samples_per_second": 97.667,
+    "train_steps_per_second": 0.762
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 15.990403071017274,
+  "epoch": 19.990403071017276,
   "eval_steps": 500,
-  "global_step": 6240,
+  "global_step": 7800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -321,37 +321,119 @@
       "step": 6000
     },
     {
-      "epoch": 15.99,
-      "eval_f1": 0.9096,
-      "eval_gen_len": 26.52509090909091,
-      "eval_loss": 1.6378074884414673,
-      "eval_precision": 0.9107,
-      "eval_recall": 0.909,
-      "eval_rouge1": 0.4698,
-      "eval_rouge2": 0.2197,
+      "epoch": 16.0,
+      "eval_f1": 0.9097,
+      "eval_gen_len": 26.424727272727274,
+      "eval_loss": 1.6346381902694702,
+      "eval_precision": 0.9108,
+      "eval_recall": 0.9089,
+      "eval_rouge1": 0.4703,
+      "eval_rouge2": 0.2204,
       "eval_rougeL": 0.385,
-      "eval_rougeLsum": 0.3849,
-      "eval_runtime": 501.725,
-      "eval_samples_per_second": 5.481,
-      "eval_steps_per_second": 0.343,
+      "eval_rougeLsum": 0.385,
+      "eval_runtime": 506.5023,
+      "eval_samples_per_second": 5.429,
+      "eval_steps_per_second": 0.34,
       "step": 6240
     },
     {
-      "epoch": 15.99,
-      "step": 6240,
-      "total_flos": 1.153589772728402e+18,
-      "train_loss": 0.4757408973498222,
-      "train_runtime": 9951.3836,
-      "train_samples_per_second": 80.391,
-      "train_steps_per_second": 0.627
+      "epoch": 16.66,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.6923,
+      "step": 6500
+    },
+    {
+      "epoch": 17.0,
+      "eval_f1": 0.9099,
+      "eval_gen_len": 26.443636363636365,
+      "eval_loss": 1.6305155754089355,
+      "eval_precision": 0.911,
+      "eval_recall": 0.9091,
+      "eval_rouge1": 0.4706,
+      "eval_rouge2": 0.221,
+      "eval_rougeL": 0.3855,
+      "eval_rougeLsum": 0.3856,
+      "eval_runtime": 506.0984,
+      "eval_samples_per_second": 5.434,
+      "eval_steps_per_second": 0.34,
+      "step": 6631
+    },
+    {
+      "epoch": 17.94,
+      "learning_rate": 2.0512820512820513e-06,
+      "loss": 1.6839,
+      "step": 7000
+    },
+    {
+      "epoch": 18.0,
+      "eval_f1": 0.9098,
+      "eval_gen_len": 26.612,
+      "eval_loss": 1.6285477876663208,
+      "eval_precision": 0.9106,
+      "eval_recall": 0.9094,
+      "eval_rouge1": 0.4712,
+      "eval_rouge2": 0.2215,
+      "eval_rougeL": 0.3862,
+      "eval_rougeLsum": 0.3864,
+      "eval_runtime": 509.5093,
+      "eval_samples_per_second": 5.397,
+      "eval_steps_per_second": 0.338,
+      "step": 7022
+    },
+    {
+      "epoch": 19.0,
+      "eval_f1": 0.9099,
+      "eval_gen_len": 26.529090909090908,
+      "eval_loss": 1.6263303756713867,
+      "eval_precision": 0.9108,
+      "eval_recall": 0.9093,
+      "eval_rouge1": 0.4709,
+      "eval_rouge2": 0.2217,
+      "eval_rougeL": 0.3862,
+      "eval_rougeLsum": 0.3864,
+      "eval_runtime": 510.6771,
+      "eval_samples_per_second": 5.385,
+      "eval_steps_per_second": 0.337,
+      "step": 7413
+    },
+    {
+      "epoch": 19.22,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 1.6743,
+      "step": 7500
+    },
+    {
+      "epoch": 19.99,
+      "eval_f1": 0.9099,
+      "eval_gen_len": 26.541090909090908,
+      "eval_loss": 1.6258454322814941,
+      "eval_precision": 0.9108,
+      "eval_recall": 0.9093,
+      "eval_rouge1": 0.4708,
+      "eval_rouge2": 0.2214,
+      "eval_rougeL": 0.3861,
+      "eval_rougeLsum": 0.3863,
+      "eval_runtime": 510.6936,
+      "eval_samples_per_second": 5.385,
+      "eval_steps_per_second": 0.337,
+      "step": 7800
+    },
+    {
+      "epoch": 19.99,
+      "step": 7800,
+      "total_flos": 1.441981436981674e+18,
+      "train_loss": 0.3883497776129307,
+      "train_runtime": 10238.8851,
+      "train_samples_per_second": 97.667,
+      "train_steps_per_second": 0.762
     }
   ],
   "logging_steps": 500,
-  "max_steps": 6240,
+  "max_steps": 7800,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 16,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 1.153589772728402e+18,
+  "total_flos": 1.441981436981674e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null