End of training

all_results.json

@@ -3,13 +3,13 @@
     "eval_loss": NaN,
     "eval_mean_acc": 0.0,
     "eval_median_acc": 0.0,
-    "eval_runtime": 17.3158,
+    "eval_runtime": 16.7609,
     "eval_samples": 18477,
-    "eval_samples_per_second": 1067.062,
-    "eval_steps_per_second": 1.444,
+    "eval_samples_per_second": 1102.386,
+    "eval_steps_per_second": 0.597,
     "train_loss": 0.0,
-    "train_runtime": 127.9233,
+    "train_runtime": 109.5006,
     "train_samples": 100000,
-    "train_samples_per_second": 781.718,
-    "train_steps_per_second": 2.447
+    "train_samples_per_second": 913.237,
+    "train_steps_per_second": 0.895
 }

eval_results.json

@@ -3,8 +3,8 @@
     "eval_loss": NaN,
     "eval_mean_acc": 0.0,
     "eval_median_acc": 0.0,
-    "eval_runtime": 17.3158,
+    "eval_runtime": 16.7609,
     "eval_samples": 18477,
-    "eval_samples_per_second": 1067.062,
-    "eval_steps_per_second": 1.444
+    "eval_samples_per_second": 1102.386,
+    "eval_steps_per_second": 0.597
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
     "train_loss": 0.0,
-    "train_runtime": 127.9233,
+    "train_runtime": 109.5006,
     "train_samples": 100000,
-    "train_samples_per_second": 781.718,
-    "train_steps_per_second": 2.447
+    "train_samples_per_second": 913.237,
+    "train_steps_per_second": 0.895
 }

trainer_state.json

@@ -2,32 +2,22 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 1.0,
-  "global_step": 313,
+  "global_step": 98,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 0.51,
-      "eval_loss": NaN,
-      "eval_mean_acc": 0.0,
-      "eval_median_acc": 0.0,
-      "eval_runtime": 16.7694,
-      "eval_samples_per_second": 1101.83,
-      "eval_steps_per_second": 1.491,
-      "step": 160
-    },
     {
       "epoch": 1.0,
-      "step": 313,
+      "step": 98,
       "total_flos": 1.038017592e+16,
       "train_loss": 0.0,
-      "train_runtime": 127.9233,
-      "train_samples_per_second": 781.718,
-      "train_steps_per_second": 2.447
+      "train_runtime": 109.5006,
+      "train_samples_per_second": 913.237,
+      "train_steps_per_second": 0.895
     }
   ],
-  "max_steps": 313,
+  "max_steps": 98,
   "num_train_epochs": 1,
   "total_flos": 1.038017592e+16,
   "trial_name": null,