Training in progress, step 3000

{checkpoint-2600 → checkpoint-2900}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:446a2b6be86378473522da6f8340e8995c1f93f317dbf12dd45b200fe2a7cc10
+oid sha256:459fd9cc8b47164c5e619165015eaa256572f1c2c4cf53b6619361a7675d46a1
 size 2490337809

{checkpoint-2700 → checkpoint-2900}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7dbbb4678046f14503aa95af2d02c0cc6a2fb7071e468967b463cf71594e2914
+oid sha256:d39abe947b6927f79b7ffad55a6a3ba07cdd2ab995efc1159bb1ca188d440e7f
 size 1262063089

{checkpoint-2600 → checkpoint-2900}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9452e603549f04f4903c5fb527b7bfc83e1fbaca4162bb7fdd612e22dd24235
+oid sha256:4d40993892ed72701230daaea4a79ca45b84c615b387b7bc61e6e9982f8358e4
 size 14567

{checkpoint-2700 → checkpoint-2900}/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7970626d52baf6e9c965f8503c1a6318b3c55b1f5ba73fdd8b8209e28d5d529e
+oid sha256:daedbf6df7d4c1927bb965444b3e2d13f98980e3a59e64cf90683cbd1e80bdf5
 size 559

{checkpoint-2600 → checkpoint-2900}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e6505acc346b949554e5cbd939e1a9cf4a757843e611b8b1dc1e36690134723
+oid sha256:da5b41dc4267a4eb8832f6b51e0de784a14e4fa5985ff5c1c6135876b04560d7
 size 623

{checkpoint-2700 → checkpoint-2900}/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 29.666666666666668,
-  "global_step": 2700,
+  "epoch": 31.863387978142075,
+  "global_step": 2900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1059,11 +1059,89 @@
       "eval_steps_per_second": 0.803,
       "eval_wer": 0.14774340051092819,
       "step": 2700
+    },
+    {
+      "epoch": 29.89,
+      "learning_rate": 0.0001796078431372549,
+      "loss": 0.8573,
+      "step": 2720
+    },
+    {
+      "epoch": 30.11,
+      "learning_rate": 0.00017764705882352942,
+      "loss": 0.8686,
+      "step": 2740
+    },
+    {
+      "epoch": 30.33,
+      "learning_rate": 0.00017568627450980392,
+      "loss": 0.8462,
+      "step": 2760
+    },
+    {
+      "epoch": 30.55,
+      "learning_rate": 0.00017372549019607845,
+      "loss": 0.8419,
+      "step": 2780
+    },
+    {
+      "epoch": 30.77,
+      "learning_rate": 0.00017176470588235293,
+      "loss": 0.8659,
+      "step": 2800
+    },
+    {
+      "epoch": 30.77,
+      "eval_loss": 0.1636602133512497,
+      "eval_runtime": 190.245,
+      "eval_samples_per_second": 25.457,
+      "eval_steps_per_second": 0.799,
+      "eval_wer": 0.1422367300596083,
+      "step": 2800
+    },
+    {
+      "epoch": 30.98,
+      "learning_rate": 0.00016980392156862745,
+      "loss": 0.8607,
+      "step": 2820
+    },
+    {
+      "epoch": 31.21,
+      "learning_rate": 0.00016784313725490196,
+      "loss": 0.8484,
+      "step": 2840
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 0.00016588235294117648,
+      "loss": 0.8335,
+      "step": 2860
+    },
+    {
+      "epoch": 31.64,
+      "learning_rate": 0.00016392156862745098,
+      "loss": 0.8321,
+      "step": 2880
+    },
+    {
+      "epoch": 31.86,
+      "learning_rate": 0.00016196078431372549,
+      "loss": 0.8419,
+      "step": 2900
+    },
+    {
+      "epoch": 31.86,
+      "eval_loss": 0.16137854754924774,
+      "eval_runtime": 190.2053,
+      "eval_samples_per_second": 25.462,
+      "eval_steps_per_second": 0.799,
+      "eval_wer": 0.1397104740278172,
+      "step": 2900
     }
   ],
   "max_steps": 4550,
   "num_train_epochs": 50,
-  "total_flos": 4.176183820424104e+19,
+  "total_flos": 4.48632216197919e+19,
   "trial_name": null,
   "trial_params": null
 }

{checkpoint-2700 → checkpoint-3000}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:865f44abdd5045a8acf626b3b992e8fdc983a2d4a3ba5dd6901b27ee83a91484
+oid sha256:023713187972aa2b714a036a65a7542d7e006ad45109ce7ddc7eafb956260fe3
 size 2490337809

{checkpoint-2600 → checkpoint-3000}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91a8acf504932720b905a27cdf2f6f7fe41c15519686ef918a30514dacbc0cee
+oid sha256:7e2564558a2c8e9677d5526af137102a80a73cbd4be834bcb3fa95a0e5c8ecef
 size 1262063089

{checkpoint-2700 → checkpoint-3000}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9213f1eed5d75326eef26e352d35aa78d567c2885250ab304bce7d59c834157a
-size 14631
+oid sha256:813a474ad7f864479df922c7e373032d552ed88f79371c0f5114bee90fb0f0c6
+size 14567

{checkpoint-2600 → checkpoint-3000}/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:981227375b8c6d2439b9fd2664e9cd784500649faa2e607c97ed38e07fc17be3
+oid sha256:ba461c86794433c4a650b89cfa19a2a74f927a6e243d52bc4fa86b75cc0b4312
 size 559

{checkpoint-2700 → checkpoint-3000}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89b50e7b743fe5caf734a580430810fca62d132f602e2a17e8e075ce20b82717
+oid sha256:fff88e552e894e867e2247a229e76319848f876eaeb56445f8251bd260defd4d
 size 623

{checkpoint-2600 → checkpoint-3000}/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 28.56830601092896,
-  "global_step": 2600,
+  "epoch": 32.96174863387978,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1020,11 +1020,167 @@
       "eval_steps_per_second": 0.774,
       "eval_wer": 0.14459267669599773,
       "step": 2600
+    },
+    {
+      "epoch": 28.79,
+      "learning_rate": 0.00018941176470588236,
+      "loss": 0.8796,
+      "step": 2620
+    },
+    {
+      "epoch": 29.01,
+      "learning_rate": 0.00018745098039215686,
+      "loss": 0.9045,
+      "step": 2640
+    },
+    {
+      "epoch": 29.23,
+      "learning_rate": 0.00018549019607843137,
+      "loss": 0.8515,
+      "step": 2660
+    },
+    {
+      "epoch": 29.45,
+      "learning_rate": 0.0001835294117647059,
+      "loss": 0.861,
+      "step": 2680
+    },
+    {
+      "epoch": 29.67,
+      "learning_rate": 0.0001815686274509804,
+      "loss": 0.8513,
+      "step": 2700
+    },
+    {
+      "epoch": 29.67,
+      "eval_loss": 0.16488835215568542,
+      "eval_runtime": 189.1938,
+      "eval_samples_per_second": 25.598,
+      "eval_steps_per_second": 0.803,
+      "eval_wer": 0.14774340051092819,
+      "step": 2700
+    },
+    {
+      "epoch": 29.89,
+      "learning_rate": 0.0001796078431372549,
+      "loss": 0.8573,
+      "step": 2720
+    },
+    {
+      "epoch": 30.11,
+      "learning_rate": 0.00017764705882352942,
+      "loss": 0.8686,
+      "step": 2740
+    },
+    {
+      "epoch": 30.33,
+      "learning_rate": 0.00017568627450980392,
+      "loss": 0.8462,
+      "step": 2760
+    },
+    {
+      "epoch": 30.55,
+      "learning_rate": 0.00017372549019607845,
+      "loss": 0.8419,
+      "step": 2780
+    },
+    {
+      "epoch": 30.77,
+      "learning_rate": 0.00017176470588235293,
+      "loss": 0.8659,
+      "step": 2800
+    },
+    {
+      "epoch": 30.77,
+      "eval_loss": 0.1636602133512497,
+      "eval_runtime": 190.245,
+      "eval_samples_per_second": 25.457,
+      "eval_steps_per_second": 0.799,
+      "eval_wer": 0.1422367300596083,
+      "step": 2800
+    },
+    {
+      "epoch": 30.98,
+      "learning_rate": 0.00016980392156862745,
+      "loss": 0.8607,
+      "step": 2820
+    },
+    {
+      "epoch": 31.21,
+      "learning_rate": 0.00016784313725490196,
+      "loss": 0.8484,
+      "step": 2840
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 0.00016588235294117648,
+      "loss": 0.8335,
+      "step": 2860
+    },
+    {
+      "epoch": 31.64,
+      "learning_rate": 0.00016392156862745098,
+      "loss": 0.8321,
+      "step": 2880
+    },
+    {
+      "epoch": 31.86,
+      "learning_rate": 0.00016196078431372549,
+      "loss": 0.8419,
+      "step": 2900
+    },
+    {
+      "epoch": 31.86,
+      "eval_loss": 0.16137854754924774,
+      "eval_runtime": 190.2053,
+      "eval_samples_per_second": 25.462,
+      "eval_steps_per_second": 0.799,
+      "eval_wer": 0.1397104740278172,
+      "step": 2900
+    },
+    {
+      "epoch": 32.09,
+      "learning_rate": 0.00016,
+      "loss": 0.8634,
+      "step": 2920
+    },
+    {
+      "epoch": 32.31,
+      "learning_rate": 0.00015803921568627451,
+      "loss": 0.8351,
+      "step": 2940
+    },
+    {
+      "epoch": 32.52,
+      "learning_rate": 0.00015607843137254901,
+      "loss": 0.8427,
+      "step": 2960
+    },
+    {
+      "epoch": 32.74,
+      "learning_rate": 0.00015411764705882352,
+      "loss": 0.8324,
+      "step": 2980
+    },
+    {
+      "epoch": 32.96,
+      "learning_rate": 0.00015215686274509804,
+      "loss": 0.8491,
+      "step": 3000
+    },
+    {
+      "epoch": 32.96,
+      "eval_loss": 0.1594749242067337,
+      "eval_runtime": 191.4631,
+      "eval_samples_per_second": 25.295,
+      "eval_steps_per_second": 0.794,
+      "eval_wer": 0.1401362475163213,
+      "step": 3000
     }
   ],
   "max_steps": 4550,
   "num_train_epochs": 50,
-  "total_flos": 4.023333138533005e+19,
+  "total_flos": 4.636386357187625e+19,
   "trial_name": null,
   "trial_params": null
 }