End of training

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 14.81,
-    "eval_loss": 1.0876343250274658,
-    "eval_runtime": 430.2181,
-    "eval_samples_per_second": 1.19,
-    "eval_steps_per_second": 0.149,
-    "eval_wer": 57.99031476997578,
-    "train_loss": 0.027764302641153336,
-    "train_runtime": 533.3671,
-    "train_samples_per_second": 11.999,
-    "train_steps_per_second": 0.75
+    "epoch": 18.52,
+    "eval_loss": 1.1797882318496704,
+    "eval_runtime": 450.6771,
+    "eval_samples_per_second": 1.136,
+    "eval_steps_per_second": 0.142,
+    "eval_wer": 57.87681598062954,
+    "train_loss": 0.0056979965269565586,
+    "train_runtime": 579.6036,
+    "train_samples_per_second": 13.803,
+    "train_steps_per_second": 0.863
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 14.81,
-    "eval_loss": 1.0876343250274658,
-    "eval_runtime": 430.2181,
-    "eval_samples_per_second": 1.19,
-    "eval_steps_per_second": 0.149,
-    "eval_wer": 57.99031476997578
+    "epoch": 18.52,
+    "eval_loss": 1.1797882318496704,
+    "eval_runtime": 450.6771,
+    "eval_samples_per_second": 1.136,
+    "eval_steps_per_second": 0.142,
+    "eval_wer": 57.87681598062954
 }

runs/Dec16_17-39-22_129-146-104-29/events.out.tfevents.1671213470.129-146-104-29.136366.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0e41f9f380a847cd5940de28e629b05b10d7854b06fc01af4b0b868a9f4c782
+size 358

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 14.81,
-    "train_loss": 0.027764302641153336,
-    "train_runtime": 533.3671,
-    "train_samples_per_second": 11.999,
-    "train_steps_per_second": 0.75
+    "epoch": 18.52,
+    "train_loss": 0.0056979965269565586,
+    "train_runtime": 579.6036,
+    "train_samples_per_second": 13.803,
+    "train_steps_per_second": 0.863
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 57.99031476997578,
-  "best_model_checkpoint": "./checkpoint-400",
-  "epoch": 14.814814814814815,
-  "global_step": 400,
+  "best_metric": 57.87681598062954,
+  "best_model_checkpoint": "./checkpoint-500",
+  "epoch": 18.51851851851852,
+  "global_step": 500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -176,18 +176,87 @@
       "step": 400
     },
     {
-      "epoch": 14.81,
-      "step": 400,
-      "total_flos": 1.84290636644352e+18,
-      "train_loss": 0.027764302641153336,
-      "train_runtime": 533.3671,
-      "train_samples_per_second": 11.999,
-      "train_steps_per_second": 0.75
+      "epoch": 15.19,
+      "learning_rate": 8.120000000000002e-06,
+      "loss": 0.0465,
+      "step": 410
+    },
+    {
+      "epoch": 15.56,
+      "learning_rate": 8.32e-06,
+      "loss": 0.0392,
+      "step": 420
+    },
+    {
+      "epoch": 15.93,
+      "learning_rate": 8.52e-06,
+      "loss": 0.0354,
+      "step": 430
+    },
+    {
+      "epoch": 16.3,
+      "learning_rate": 8.720000000000001e-06,
+      "loss": 0.0296,
+      "step": 440
+    },
+    {
+      "epoch": 16.67,
+      "learning_rate": 8.920000000000001e-06,
+      "loss": 0.0255,
+      "step": 450
+    },
+    {
+      "epoch": 17.04,
+      "learning_rate": 9.12e-06,
+      "loss": 0.0274,
+      "step": 460
+    },
+    {
+      "epoch": 17.41,
+      "learning_rate": 9.32e-06,
+      "loss": 0.0216,
+      "step": 470
+    },
+    {
+      "epoch": 17.78,
+      "learning_rate": 9.52e-06,
+      "loss": 0.022,
+      "step": 480
+    },
+    {
+      "epoch": 18.15,
+      "learning_rate": 9.72e-06,
+      "loss": 0.0219,
+      "step": 490
+    },
+    {
+      "epoch": 18.52,
+      "learning_rate": 9.920000000000002e-06,
+      "loss": 0.0159,
+      "step": 500
+    },
+    {
+      "epoch": 18.52,
+      "eval_loss": 1.1797882318496704,
+      "eval_runtime": 450.9533,
+      "eval_samples_per_second": 1.135,
+      "eval_steps_per_second": 0.142,
+      "eval_wer": 57.87681598062954,
+      "step": 500
+    },
+    {
+      "epoch": 18.52,
+      "step": 500,
+      "total_flos": 2.30348866535424e+18,
+      "train_loss": 0.0056979965269565586,
+      "train_runtime": 579.6036,
+      "train_samples_per_second": 13.803,
+      "train_steps_per_second": 0.863
     }
   ],
-  "max_steps": 400,
-  "num_train_epochs": 15,
-  "total_flos": 1.84290636644352e+18,
+  "max_steps": 500,
+  "num_train_epochs": 19,
+  "total_flos": 2.30348866535424e+18,
   "trial_name": null,
   "trial_params": null
 }