End of training

all_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 11.99,
-    "train_loss": 0.6240079896062867,
-    "train_runtime": 9692.0216,
-    "train_samples_per_second": 61.907,
-    "train_steps_per_second": 0.483
+    "epoch": 15.99,
+    "train_loss": 0.4757408973498222,
+    "train_runtime": 9951.3836,
+    "train_samples_per_second": 80.391,
+    "train_steps_per_second": 0.627
 }

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 11.99,
-    "train_loss": 0.6240079896062867,
-    "train_runtime": 9692.0216,
-    "train_samples_per_second": 61.907,
-    "train_steps_per_second": 0.483
+    "epoch": 15.99,
+    "train_loss": 0.4757408973498222,
+    "train_runtime": 9951.3836,
+    "train_samples_per_second": 80.391,
+    "train_steps_per_second": 0.627
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 11.990403071017274,
+  "epoch": 15.990403071017274,
   "eval_steps": 500,
-  "global_step": 4680,
+  "global_step": 6240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -239,37 +239,119 @@
       "step": 4500
     },
     {
-      "epoch": 11.99,
+      "epoch": 12.0,
       "eval_f1": 0.9092,
-      "eval_gen_len": 26.54581818181818,
-      "eval_loss": 1.654082179069519,
-      "eval_precision": 0.9101,
-      "eval_recall": 0.9085,
-      "eval_rouge1": 0.4665,
-      "eval_rouge2": 0.2182,
-      "eval_rougeL": 0.3824,
-      "eval_rougeLsum": 0.3824,
-      "eval_runtime": 506.4111,
-      "eval_samples_per_second": 5.43,
-      "eval_steps_per_second": 0.34,
+      "eval_gen_len": 26.439272727272726,
+      "eval_loss": 1.6508440971374512,
+      "eval_precision": 0.9103,
+      "eval_recall": 0.9084,
+      "eval_rouge1": 0.4668,
+      "eval_rouge2": 0.2175,
+      "eval_rougeL": 0.3823,
+      "eval_rougeLsum": 0.3822,
+      "eval_runtime": 506.6299,
+      "eval_samples_per_second": 5.428,
+      "eval_steps_per_second": 0.339,
       "step": 4680
     },
     {
-      "epoch": 11.99,
-      "step": 4680,
-      "total_flos": 8.651981084751299e+17,
-      "train_loss": 0.6240079896062867,
-      "train_runtime": 9692.0216,
-      "train_samples_per_second": 61.907,
-      "train_steps_per_second": 0.483
+      "epoch": 12.82,
+      "learning_rate": 3.974358974358974e-06,
+      "loss": 1.7165,
+      "step": 5000
+    },
+    {
+      "epoch": 13.0,
+      "eval_f1": 0.9094,
+      "eval_gen_len": 26.638545454545454,
+      "eval_loss": 1.6451035737991333,
+      "eval_precision": 0.9103,
+      "eval_recall": 0.9089,
+      "eval_rouge1": 0.4687,
+      "eval_rouge2": 0.2191,
+      "eval_rougeL": 0.3834,
+      "eval_rougeLsum": 0.3834,
+      "eval_runtime": 509.0115,
+      "eval_samples_per_second": 5.403,
+      "eval_steps_per_second": 0.338,
+      "step": 5071
+    },
+    {
+      "epoch": 14.0,
+      "eval_f1": 0.9095,
+      "eval_gen_len": 26.415636363636363,
+      "eval_loss": 1.6405242681503296,
+      "eval_precision": 0.9106,
+      "eval_recall": 0.9087,
+      "eval_rouge1": 0.4691,
+      "eval_rouge2": 0.2193,
+      "eval_rougeL": 0.3845,
+      "eval_rougeLsum": 0.3844,
+      "eval_runtime": 502.7756,
+      "eval_samples_per_second": 5.47,
+      "eval_steps_per_second": 0.342,
+      "step": 5462
+    },
+    {
+      "epoch": 14.1,
+      "learning_rate": 2.371794871794872e-06,
+      "loss": 1.7068,
+      "step": 5500
+    },
+    {
+      "epoch": 15.0,
+      "eval_f1": 0.9097,
+      "eval_gen_len": 26.45709090909091,
+      "eval_loss": 1.6383482217788696,
+      "eval_precision": 0.9108,
+      "eval_recall": 0.9089,
+      "eval_rouge1": 0.4699,
+      "eval_rouge2": 0.2204,
+      "eval_rougeL": 0.3853,
+      "eval_rougeLsum": 0.3853,
+      "eval_runtime": 506.608,
+      "eval_samples_per_second": 5.428,
+      "eval_steps_per_second": 0.34,
+      "step": 5853
+    },
+    {
+      "epoch": 15.38,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 1.7004,
+      "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,
+      "eval_rougeL": 0.385,
+      "eval_rougeLsum": 0.3849,
+      "eval_runtime": 501.725,
+      "eval_samples_per_second": 5.481,
+      "eval_steps_per_second": 0.343,
+      "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
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4680,
+  "max_steps": 6240,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 12,
+  "num_train_epochs": 16,
   "save_steps": 500,
-  "total_flos": 8.651981084751299e+17,
+  "total_flos": 1.153589772728402e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null