End of training

all_results.json

@@ -1,7 +1,12 @@
 {
     "epoch": 1.03,
-    "train_loss": 0.13366424560546875,
-    "train_runtime": 7967.969,
-    "train_samples_per_second": 4.016,
-    "train_steps_per_second": 0.063
+    "eval_loss": 0.1700439453125,
+    "eval_runtime": 3035.8435,
+    "eval_samples_per_second": 5.112,
+    "eval_steps_per_second": 0.16,
+    "eval_wer": 5.696169637752716,
+    "train_loss": 0.0865213623046875,
+    "train_runtime": 8011.6489,
+    "train_samples_per_second": 3.994,
+    "train_steps_per_second": 0.062
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.03,
+    "eval_loss": 0.1700439453125,
+    "eval_runtime": 3035.8435,
+    "eval_samples_per_second": 5.112,
+    "eval_steps_per_second": 0.16,
+    "eval_wer": 5.696169637752716
+}

runs/Dec16_17-56-05_132-145-140-45/events.out.tfevents.1671224581.132-145-140-45.3273636.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f139750e42de87e1b9e8d503d3b270351c8c4aa7e5d263d51bdf026118d574
+size 358

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 1.03,
-    "train_loss": 0.13366424560546875,
-    "train_runtime": 7967.969,
-    "train_samples_per_second": 4.016,
-    "train_steps_per_second": 0.063
+    "train_loss": 0.0865213623046875,
+    "train_runtime": 8011.6489,
+    "train_samples_per_second": 3.994,
+    "train_steps_per_second": 0.062
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 5.735798555168651,
+  "best_metric": 5.696169637752716,
   "best_model_checkpoint": "./checkpoint-500",
   "epoch": 1.03,
   "global_step": 500,
@@ -9,141 +9,141 @@
   "log_history": [
     {
       "epoch": 0.05,
-      "learning_rate": 6.611096473669595e-07,
-      "loss": 0.6089,
+      "learning_rate": 3.4043195900439816e-07,
+      "loss": 0.0907,
       "step": 25
     },
     {
       "epoch": 0.1,
-      "learning_rate": 8.313789158407869e-07,
-      "loss": 0.1002,
+      "learning_rate": 4.203103093438967e-07,
+      "loss": 0.0802,
       "step": 50
     },
     {
       "epoch": 0.15,
-      "learning_rate": 9.256291743595375e-07,
-      "loss": 0.0973,
+      "learning_rate": 4.6583071503011386e-07,
+      "loss": 0.0774,
       "step": 75
     },
     {
       "epoch": 0.2,
-      "learning_rate": 9.91135616519784e-07,
-      "loss": 0.096,
+      "learning_rate": 4.978065189231237e-07,
+      "loss": 0.0746,
       "step": 100
     },
     {
       "epoch": 0.25,
-      "learning_rate": 1e-06,
-      "loss": 0.1061,
+      "learning_rate": 5e-07,
+      "loss": 0.0834,
       "step": 125
     },
     {
       "epoch": 0.3,
-      "learning_rate": 1e-06,
-      "loss": 0.0993,
+      "learning_rate": 5e-07,
+      "loss": 0.0775,
       "step": 150
     },
     {
       "epoch": 0.35,
-      "learning_rate": 1e-06,
-      "loss": 0.0994,
+      "learning_rate": 5e-07,
+      "loss": 0.0784,
       "step": 175
     },
     {
       "epoch": 0.4,
-      "learning_rate": 1e-06,
-      "loss": 0.0964,
+      "learning_rate": 5e-07,
+      "loss": 0.0756,
       "step": 200
     },
     {
       "epoch": 0.45,
-      "learning_rate": 1e-06,
-      "loss": 0.0959,
+      "learning_rate": 5e-07,
+      "loss": 0.0744,
       "step": 225
     },
     {
       "epoch": 0.5,
-      "learning_rate": 1e-06,
-      "loss": 0.1011,
+      "learning_rate": 5e-07,
+      "loss": 0.0801,
       "step": 250
     },
     {
       "epoch": 0.55,
-      "learning_rate": 1e-06,
-      "loss": 0.0976,
+      "learning_rate": 5e-07,
+      "loss": 0.0768,
       "step": 275
     },
     {
       "epoch": 0.6,
-      "learning_rate": 1e-06,
-      "loss": 0.0995,
+      "learning_rate": 5e-07,
+      "loss": 0.0795,
       "step": 300
     },
     {
       "epoch": 0.65,
-      "learning_rate": 1e-06,
-      "loss": 0.1059,
+      "learning_rate": 5e-07,
+      "loss": 0.0837,
       "step": 325
     },
     {
       "epoch": 0.7,
-      "learning_rate": 1e-06,
-      "loss": 0.1056,
+      "learning_rate": 5e-07,
+      "loss": 0.0827,
       "step": 350
     },
     {
       "epoch": 0.75,
-      "learning_rate": 1e-06,
-      "loss": 0.1078,
+      "learning_rate": 5e-07,
+      "loss": 0.0858,
       "step": 375
     },
     {
       "epoch": 0.8,
-      "learning_rate": 1e-06,
-      "loss": 0.1122,
+      "learning_rate": 5e-07,
+      "loss": 0.088,
       "step": 400
     },
     {
       "epoch": 0.85,
-      "learning_rate": 1e-06,
-      "loss": 0.1098,
+      "learning_rate": 5e-07,
+      "loss": 0.0865,
       "step": 425
     },
     {
       "epoch": 0.9,
-      "learning_rate": 1e-06,
-      "loss": 0.1163,
+      "learning_rate": 5e-07,
+      "loss": 0.0911,
       "step": 450
     },
     {
       "epoch": 0.95,
-      "learning_rate": 1e-06,
-      "loss": 0.1209,
+      "learning_rate": 5e-07,
+      "loss": 0.0976,
       "step": 475
     },
     {
       "epoch": 1.03,
-      "learning_rate": 1e-06,
-      "loss": 0.1972,
+      "learning_rate": 5e-07,
+      "loss": 0.1665,
       "step": 500
     },
     {
       "epoch": 1.03,
-      "eval_loss": 0.1683349609375,
-      "eval_runtime": 3047.5288,
-      "eval_samples_per_second": 5.093,
+      "eval_loss": 0.1700439453125,
+      "eval_runtime": 3050.6887,
+      "eval_samples_per_second": 5.087,
       "eval_steps_per_second": 0.159,
-      "eval_wer": 5.735798555168651,
+      "eval_wer": 5.696169637752716,
       "step": 500
     },
     {
       "epoch": 1.03,
       "step": 500,
       "total_flos": 3.2659357379539763e+19,
-      "train_loss": 0.13366424560546875,
-      "train_runtime": 7967.969,
-      "train_samples_per_second": 4.016,
-      "train_steps_per_second": 0.063
+      "train_loss": 0.0865213623046875,
+      "train_runtime": 8011.6489,
+      "train_samples_per_second": 3.994,
+      "train_steps_per_second": 0.062
     }
   ],
   "max_steps": 500,