End of training

all_results.json

@@ -5,9 +5,9 @@
     "eval_runtime": 0.2994,
     "eval_samples_per_second": 73.482,
     "eval_steps_per_second": 6.68,
-    "total_flos": 2.348890959163392e+16,
-    "train_loss": 0.2989959716796875,
-    "train_runtime": 40.2987,
-    "train_samples_per_second": 23.45,
-    "train_steps_per_second": 0.372
+    "total_flos": 7.322993018578944e+16,
+    "train_loss": 0.5733050982157389,
+    "train_runtime": 83.7822,
+    "train_samples_per_second": 11.279,
+    "train_steps_per_second": 0.179
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "total_flos": 2.348890959163392e+16,
-    "train_loss": 0.2989959716796875,
-    "train_runtime": 40.2987,
-    "train_samples_per_second": 23.45,
-    "train_steps_per_second": 0.372
+    "total_flos": 7.322993018578944e+16,
+    "train_loss": 0.5733050982157389,
+    "train_runtime": 83.7822,
+    "train_samples_per_second": 11.279,
+    "train_steps_per_second": 0.179
 }

trainer_state.json

@@ -10,70 +10,70 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5454545454545454,
-      "eval_loss": 1.666408658027649,
-      "eval_runtime": 0.2367,
-      "eval_samples_per_second": 92.955,
-      "eval_steps_per_second": 8.45,
+      "eval_accuracy": 0.6818181818181818,
+      "eval_loss": 0.62430739402771,
+      "eval_runtime": 0.4086,
+      "eval_samples_per_second": 53.848,
+      "eval_steps_per_second": 4.895,
       "step": 3
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.7272727272727273,
-      "eval_loss": 0.7899118065834045,
-      "eval_runtime": 0.3107,
-      "eval_samples_per_second": 70.812,
-      "eval_steps_per_second": 6.437,
+      "eval_loss": 0.5459609031677246,
+      "eval_runtime": 0.4897,
+      "eval_samples_per_second": 44.927,
+      "eval_steps_per_second": 4.084,
       "step": 6
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.7272727272727273,
-      "eval_loss": 0.770358145236969,
-      "eval_runtime": 0.2398,
-      "eval_samples_per_second": 91.755,
-      "eval_steps_per_second": 8.341,
+      "eval_loss": 0.5539888143539429,
+      "eval_runtime": 0.3962,
+      "eval_samples_per_second": 55.526,
+      "eval_steps_per_second": 5.048,
       "step": 9
     },
     {
       "epoch": 3.33,
       "learning_rate": 1.923076923076923e-05,
-      "loss": 0.3108,
+      "loss": 0.6502,
       "step": 10
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7272727272727273,
-      "eval_loss": 0.6269326210021973,
-      "eval_runtime": 0.2432,
-      "eval_samples_per_second": 90.452,
-      "eval_steps_per_second": 8.223,
+      "eval_accuracy": 0.6818181818181818,
+      "eval_loss": 0.5746610164642334,
+      "eval_runtime": 0.3881,
+      "eval_samples_per_second": 56.691,
+      "eval_steps_per_second": 5.154,
       "step": 12
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6818181818181818,
-      "eval_loss": 0.644472062587738,
-      "eval_runtime": 0.2431,
-      "eval_samples_per_second": 90.505,
-      "eval_steps_per_second": 8.228,
+      "eval_accuracy": 0.5909090909090909,
+      "eval_loss": 0.5886295437812805,
+      "eval_runtime": 0.391,
+      "eval_samples_per_second": 56.271,
+      "eval_steps_per_second": 5.116,
       "step": 15
     },
     {
       "epoch": 5.0,
       "step": 15,
-      "total_flos": 2.348890959163392e+16,
-      "train_loss": 0.2989959716796875,
-      "train_runtime": 40.2987,
-      "train_samples_per_second": 23.45,
-      "train_steps_per_second": 0.372
+      "total_flos": 7.322993018578944e+16,
+      "train_loss": 0.5733050982157389,
+      "train_runtime": 83.7822,
+      "train_samples_per_second": 11.279,
+      "train_steps_per_second": 0.179
     }
   ],
   "logging_steps": 10,
   "max_steps": 15,
   "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 2.348890959163392e+16,
+  "total_flos": 7.322993018578944e+16,
   "trial_name": null,
   "trial_params": null
 }