End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 0.14093226194381714,
-    "eval_runtime": 180.4366,
+    "epoch": 99.98,
+    "eval_loss": 0.28930261731147766,
+    "eval_runtime": 108.745,
     "eval_samples": 3209,
-    "eval_samples_per_second": 17.785,
-    "eval_steps_per_second": 8.895,
-    "eval_wer": 0.13087634802638468,
-    "train_loss": 1.532381568636213,
-    "train_runtime": 408.5144,
+    "eval_samples_per_second": 29.509,
+    "eval_steps_per_second": 3.697,
+    "eval_wer": 0.27128049418909017,
+    "train_loss": 1.2002243913922992,
+    "train_runtime": 17065.0192,
     "train_samples": 7188,
-    "train_samples_per_second": 17.595,
-    "train_steps_per_second": 0.274
+    "train_samples_per_second": 42.121,
+    "train_steps_per_second": 0.082
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 0.14093226194381714,
-    "eval_runtime": 180.4366,
+    "epoch": 99.98,
+    "eval_loss": 0.28930261731147766,
+    "eval_runtime": 108.745,
     "eval_samples": 3209,
-    "eval_samples_per_second": 17.785,
-    "eval_steps_per_second": 8.895,
-    "eval_wer": 0.13087634802638468
+    "eval_samples_per_second": 29.509,
+    "eval_steps_per_second": 3.697,
+    "eval_wer": 0.27128049418909017
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "train_loss": 1.532381568636213,
-    "train_runtime": 408.5144,
+    "epoch": 99.98,
+    "train_loss": 1.2002243913922992,
+    "train_runtime": 17065.0192,
     "train_samples": 7188,
-    "train_samples_per_second": 17.595,
-    "train_steps_per_second": 0.274
+    "train_samples_per_second": 42.121,
+    "train_steps_per_second": 0.082
 }

trainer_state.json

@@ -1,31 +1,130 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.9955555555555555,
-  "global_step": 112,
+  "epoch": 99.98245614035088,
+  "global_step": 1400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.89,
-      "learning_rate": 1.4999999999999999e-05,
-      "loss": 1.5179,
-      "step": 100
+      "epoch": 14.28,
+      "learning_rate": 0.0002769230769230769,
+      "loss": 1.8647,
+      "step": 200
     },
     {
-      "epoch": 1.0,
-      "step": 112,
-      "total_flos": 2.2375521021350707e+17,
-      "train_loss": 1.532381568636213,
-      "train_runtime": 408.5144,
-      "train_samples_per_second": 17.595,
-      "train_steps_per_second": 0.274
+      "epoch": 14.28,
+      "eval_loss": 0.2757788896560669,
+      "eval_runtime": 104.0875,
+      "eval_samples_per_second": 30.83,
+      "eval_steps_per_second": 3.862,
+      "eval_wer": 0.2568317453669773,
+      "step": 200
+    },
+    {
+      "epoch": 28.56,
+      "learning_rate": 0.00023076923076923076,
+      "loss": 1.3376,
+      "step": 400
+    },
+    {
+      "epoch": 28.56,
+      "eval_loss": 0.2754214107990265,
+      "eval_runtime": 104.1937,
+      "eval_samples_per_second": 30.798,
+      "eval_steps_per_second": 3.858,
+      "eval_wer": 0.2721704533556696,
+      "step": 400
+    },
+    {
+      "epoch": 42.84,
+      "learning_rate": 0.0001846153846153846,
+      "loss": 1.1975,
+      "step": 600
+    },
+    {
+      "epoch": 42.84,
+      "eval_loss": 0.29288050532341003,
+      "eval_runtime": 101.9031,
+      "eval_samples_per_second": 31.491,
+      "eval_steps_per_second": 3.945,
+      "eval_wer": 0.290074337765679,
+      "step": 600
+    },
+    {
+      "epoch": 57.14,
+      "learning_rate": 0.00013846153846153845,
+      "loss": 1.1024,
+      "step": 800
+    },
+    {
+      "epoch": 57.14,
+      "eval_loss": 0.2903825342655182,
+      "eval_runtime": 103.1754,
+      "eval_samples_per_second": 31.102,
+      "eval_steps_per_second": 3.896,
+      "eval_wer": 0.29279656580462776,
+      "step": 800
+    },
+    {
+      "epoch": 71.42,
+      "learning_rate": 9.23076923076923e-05,
+      "loss": 1.0257,
+      "step": 1000
+    },
+    {
+      "epoch": 71.42,
+      "eval_loss": 0.2915154993534088,
+      "eval_runtime": 103.3451,
+      "eval_samples_per_second": 31.051,
+      "eval_steps_per_second": 3.89,
+      "eval_wer": 0.28232645796251704,
+      "step": 1000
+    },
+    {
+      "epoch": 85.7,
+      "learning_rate": 4.615384615384615e-05,
+      "loss": 0.9628,
+      "step": 1200
+    },
+    {
+      "epoch": 85.7,
+      "eval_loss": 0.2936263978481293,
+      "eval_runtime": 102.9525,
+      "eval_samples_per_second": 31.17,
+      "eval_steps_per_second": 3.905,
+      "eval_wer": 0.2749450319338289,
+      "step": 1200
+    },
+    {
+      "epoch": 99.98,
+      "learning_rate": 0.0,
+      "loss": 0.9109,
+      "step": 1400
+    },
+    {
+      "epoch": 99.98,
+      "eval_loss": 0.28930261731147766,
+      "eval_runtime": 104.444,
+      "eval_samples_per_second": 30.725,
+      "eval_steps_per_second": 3.849,
+      "eval_wer": 0.27128049418909017,
+      "step": 1400
+    },
+    {
+      "epoch": 99.98,
+      "step": 1400,
+      "total_flos": 1.9811825337670926e+19,
+      "train_loss": 1.2002243913922992,
+      "train_runtime": 17065.0192,
+      "train_samples_per_second": 42.121,
+      "train_steps_per_second": 0.082
     }
   ],
-  "max_steps": 112,
-  "num_train_epochs": 1,
-  "total_flos": 2.2375521021350707e+17,
+  "max_steps": 1400,
+  "num_train_epochs": 100,
+  "total_flos": 1.9811825337670926e+19,
   "trial_name": null,
   "trial_params": null
 }