Training complete

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 717876559872000.0,
-    "train_loss": 0.18599607192145454,
-    "train_runtime": 356.9797,
-    "train_samples_per_second": 50.423,
-    "train_steps_per_second": 6.303
+    "epoch": 1.0,
+    "total_flos": 358938279936000.0,
+    "train_loss": 0.19838053215874565,
+    "train_runtime": 189.1751,
+    "train_samples_per_second": 47.575,
+    "train_steps_per_second": 5.947
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 717876559872000.0,
-    "train_loss": 0.18599607192145454,
-    "train_runtime": 356.9797,
-    "train_samples_per_second": 50.423,
-    "train_steps_per_second": 6.303
+    "epoch": 1.0,
+    "total_flos": 358938279936000.0,
+    "train_loss": 0.19838053215874565,
+    "train_runtime": 189.1751,
+    "train_samples_per_second": 47.575,
+    "train_steps_per_second": 5.947
 }

trainer_state.json

@@ -1,131 +1,75 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
-  "global_step": 2250,
+  "epoch": 1.0,
+  "global_step": 1125,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.18,
-      "learning_rate": 0.0009111111111111111,
-      "loss": 0.2592,
+      "learning_rate": 0.0008222222222222222,
+      "loss": 0.2605,
       "step": 200
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.0008222222222222222,
-      "loss": 0.2127,
+      "learning_rate": 0.0006444444444444444,
+      "loss": 0.2056,
       "step": 400
     },
     {
       "epoch": 0.44,
-      "eval_loss": 0.3103792071342468,
-      "eval_macro-f1": 0.3107897037355991,
-      "eval_micro-f1": 0.44178454842219805,
-      "eval_runtime": 16.3051,
-      "eval_samples_per_second": 61.331,
-      "eval_steps_per_second": 7.666,
+      "eval_loss": 0.2846032381057739,
+      "eval_macro-f1": 0.3334545959979333,
+      "eval_micro-f1": 0.47625951431678143,
+      "eval_runtime": 15.3528,
+      "eval_samples_per_second": 65.135,
+      "eval_steps_per_second": 8.142,
       "step": 500
     },
     {
       "epoch": 0.53,
-      "learning_rate": 0.0007333333333333333,
-      "loss": 0.201,
+      "learning_rate": 0.00046666666666666666,
+      "loss": 0.1915,
       "step": 600
     },
     {
       "epoch": 0.71,
-      "learning_rate": 0.0006444444444444444,
-      "loss": 0.1913,
+      "learning_rate": 0.0002888888888888889,
+      "loss": 0.1815,
       "step": 800
     },
     {
       "epoch": 0.89,
-      "learning_rate": 0.0005555555555555556,
-      "loss": 0.1808,
+      "learning_rate": 0.0001111111111111111,
+      "loss": 0.1698,
       "step": 1000
     },
     {
       "epoch": 0.89,
-      "eval_loss": 0.2961287200450897,
-      "eval_macro-f1": 0.3452123032410471,
-      "eval_micro-f1": 0.5021520803443329,
-      "eval_runtime": 17.2348,
-      "eval_samples_per_second": 58.022,
-      "eval_steps_per_second": 7.253,
+      "eval_loss": 0.24874328076839447,
+      "eval_macro-f1": 0.40516740460648787,
+      "eval_micro-f1": 0.5660096690219413,
+      "eval_runtime": 15.0266,
+      "eval_samples_per_second": 66.549,
+      "eval_steps_per_second": 8.319,
       "step": 1000
     },
     {
-      "epoch": 1.07,
-      "learning_rate": 0.00046666666666666666,
-      "loss": 0.1758,
-      "step": 1200
-    },
-    {
-      "epoch": 1.24,
-      "learning_rate": 0.00037777777777777777,
-      "loss": 0.1762,
-      "step": 1400
-    },
-    {
-      "epoch": 1.33,
-      "eval_loss": 0.2807115316390991,
-      "eval_macro-f1": 0.3945425420350508,
-      "eval_micro-f1": 0.527331189710611,
-      "eval_runtime": 14.5084,
-      "eval_samples_per_second": 68.925,
-      "eval_steps_per_second": 8.616,
-      "step": 1500
-    },
-    {
-      "epoch": 1.42,
-      "learning_rate": 0.0002888888888888889,
-      "loss": 0.1698,
-      "step": 1600
-    },
-    {
-      "epoch": 1.6,
-      "learning_rate": 0.0002,
-      "loss": 0.1622,
-      "step": 1800
-    },
-    {
-      "epoch": 1.78,
-      "learning_rate": 0.0001111111111111111,
-      "loss": 0.16,
-      "step": 2000
-    },
-    {
-      "epoch": 1.78,
-      "eval_loss": 0.2669467031955719,
-      "eval_macro-f1": 0.4193019540132512,
-      "eval_micro-f1": 0.5572842998585573,
-      "eval_runtime": 14.4979,
-      "eval_samples_per_second": 68.976,
-      "eval_steps_per_second": 8.622,
-      "step": 2000
-    },
-    {
-      "epoch": 1.96,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.1648,
-      "step": 2200
-    },
-    {
-      "epoch": 2.0,
-      "step": 2250,
-      "total_flos": 717876559872000.0,
-      "train_loss": 0.18599607192145454,
-      "train_runtime": 356.9797,
-      "train_samples_per_second": 50.423,
-      "train_steps_per_second": 6.303
+      "epoch": 1.0,
+      "step": 1125,
+      "total_flos": 358938279936000.0,
+      "train_loss": 0.19838053215874565,
+      "train_runtime": 189.1751,
+      "train_samples_per_second": 47.575,
+      "train_steps_per_second": 5.947
     }
   ],
-  "max_steps": 2250,
-  "num_train_epochs": 2,
-  "total_flos": 717876559872000.0,
+  "max_steps": 1125,
+  "num_train_epochs": 1,
+  "total_flos": 358938279936000.0,
   "trial_name": null,
   "trial_params": null
 }