End of training

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 30.0,
-    "eval_loss": 0.8710034489631653,
-    "eval_runtime": 279.4192,
-    "eval_samples_per_second": 1.832,
-    "eval_steps_per_second": 0.057,
+    "epoch": 33.33,
+    "eval_loss": 0.8714202046394348,
+    "eval_runtime": 282.6755,
+    "eval_samples_per_second": 1.811,
+    "eval_steps_per_second": 0.113,
     "eval_wer": 60.05599273607748,
-    "train_loss": 0.0,
-    "train_runtime": 0.8373,
-    "train_samples_per_second": 91726.968,
-    "train_steps_per_second": 1433.234
+    "train_loss": 0.018539312619429368,
+    "train_runtime": 504.1118,
+    "train_samples_per_second": 165.043,
+    "train_steps_per_second": 2.579
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 30.0,
-    "eval_loss": 0.8710034489631653,
-    "eval_runtime": 279.4192,
-    "eval_samples_per_second": 1.832,
-    "eval_steps_per_second": 0.057,
+    "epoch": 33.33,
+    "eval_loss": 0.8714202046394348,
+    "eval_runtime": 282.6755,
+    "eval_samples_per_second": 1.811,
+    "eval_steps_per_second": 0.113,
     "eval_wer": 60.05599273607748
 }

runs/Dec20_20-11-43_129-146-32-172/events.out.tfevents.1671567996.129-146-32-172.149339.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f35cb80a34e81a96e6ecea4220ba82e98f79d0b2759d0d7d032926a35627a522
+size 358

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 30.0,
-    "train_loss": 0.0,
-    "train_runtime": 0.8373,
-    "train_samples_per_second": 91726.968,
-    "train_steps_per_second": 1433.234
+    "epoch": 33.33,
+    "train_loss": 0.018539312619429368,
+    "train_runtime": 504.1118,
+    "train_samples_per_second": 165.043,
+    "train_steps_per_second": 2.579
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": 0.8710034489631653,
   "best_model_checkpoint": "./checkpoint-400",
-  "epoch": 30.0,
-  "global_step": 1200,
+  "epoch": 33.32911392405063,
+  "global_step": 1300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -836,18 +836,87 @@
       "step": 1200
     },
     {
-      "epoch": 30.0,
-      "step": 1200,
-      "total_flos": 4.8586623123456e+18,
-      "train_loss": 0.0,
-      "train_runtime": 0.8373,
-      "train_samples_per_second": 91726.968,
-      "train_steps_per_second": 1433.234
+      "epoch": 31.03,
+      "learning_rate": 7.833333333333335e-07,
+      "loss": 0.2489,
+      "step": 1210
+    },
+    {
+      "epoch": 31.28,
+      "learning_rate": 7.000000000000001e-07,
+      "loss": 0.2397,
+      "step": 1220
+    },
+    {
+      "epoch": 31.53,
+      "learning_rate": 6.166666666666668e-07,
+      "loss": 0.2362,
+      "step": 1230
+    },
+    {
+      "epoch": 31.78,
+      "learning_rate": 5.333333333333335e-07,
+      "loss": 0.2419,
+      "step": 1240
+    },
+    {
+      "epoch": 32.05,
+      "learning_rate": 4.5000000000000003e-07,
+      "loss": 0.2533,
+      "step": 1250
+    },
+    {
+      "epoch": 32.3,
+      "learning_rate": 3.666666666666667e-07,
+      "loss": 0.2321,
+      "step": 1260
+    },
+    {
+      "epoch": 32.56,
+      "learning_rate": 2.8333333333333336e-07,
+      "loss": 0.232,
+      "step": 1270
+    },
+    {
+      "epoch": 32.81,
+      "learning_rate": 2.0000000000000002e-07,
+      "loss": 0.2421,
+      "step": 1280
+    },
+    {
+      "epoch": 33.08,
+      "learning_rate": 1.1666666666666668e-07,
+      "loss": 0.2457,
+      "step": 1290
+    },
+    {
+      "epoch": 33.33,
+      "learning_rate": 3.333333333333334e-08,
+      "loss": 0.2383,
+      "step": 1300
+    },
+    {
+      "epoch": 33.33,
+      "eval_loss": 0.9528815746307373,
+      "eval_runtime": 176.305,
+      "eval_samples_per_second": 2.904,
+      "eval_steps_per_second": 0.182,
+      "eval_wer": 59.07990314769975,
+      "step": 1300
+    },
+    {
+      "epoch": 33.33,
+      "step": 1300,
+      "total_flos": 5.27396086775808e+18,
+      "train_loss": 0.018539312619429368,
+      "train_runtime": 504.1118,
+      "train_samples_per_second": 165.043,
+      "train_steps_per_second": 2.579
     }
   ],
-  "max_steps": 1200,
-  "num_train_epochs": 30,
-  "total_flos": 4.8586623123456e+18,
+  "max_steps": 1300,
+  "num_train_epochs": 34,
+  "total_flos": 5.27396086775808e+18,
   "trial_name": null,
   "trial_params": null
 }