Upload 9 files

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:046b4511d9979e59fda0c2409d7b9e4f9f2601e7bb0ebb6bcea6e849db0e7b2e
+oid sha256:55b8ba47c1a14520b9fcd038ff6a20628b01fa4771cb5fb74f72e5ab51d8b1b7
 size 2490594117

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d29291b649665d371303e9cc78b4ce656fb51bbaf143585dfdc11a28235e36ac
+oid sha256:2e0529c7648c521db89e680a68a1d95dc911c30720659cb12acf61bef32a3ad7
 size 1262168365

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98fb9274470e6654c92f2ab92267cb78fced569e3ecb8cb98d8445df5730abdb
-size 14639
+oid sha256:1eb86c4b86a8f8830681aee2bf3f61b818687a4378530eef63dc1a0ab6dba4f3
+size 14575

scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf733a520785e2a26377c7407c8a5281b756e1f9a937911fafbff2630fa6a6d9
+oid sha256:92bad7b4b1284d748ee0f1976606170e2e31c52706571e4ec2ced54d5e289b2f
 size 557

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f98553c888b9ecffc2af30085063b88a8b95684949d549fa8f97627e4abd087
+oid sha256:e926484f096b98b97b945304545e37eda9109685e9310eb8bbddc2a3c7471e8b
 size 627

trainer_state.json

@@ -1,304 +1,64 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 31.64505928853755,
-  "global_step": 5000,
+  "epoch": 4.90727969348659,
+  "global_step": 800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.95,
-      "learning_rate": 9.99968352427369e-06,
-      "loss": 0.0334,
+      "epoch": 0.92,
+      "learning_rate": 9.999693232713664e-06,
+      "loss": 0.0295,
       "step": 150
     },
     {
-      "epoch": 1.9,
-      "learning_rate": 9.998734097094752e-06,
-      "loss": 0.0325,
+      "epoch": 1.84,
+      "learning_rate": 9.998772930854655e-06,
+      "loss": 0.0272,
       "step": 300
     },
     {
-      "epoch": 2.85,
-      "learning_rate": 9.997790999430345e-06,
-      "loss": 0.0318,
-      "step": 450
+      "epoch": 2.45,
+      "eval_loss": 0.23613713681697845,
+      "eval_runtime": 411.8281,
+      "eval_samples_per_second": 11.311,
+      "eval_steps_per_second": 1.416,
+      "eval_wer": 0.15606540535913022,
+      "step": 400
     },
     {
-      "epoch": 3.16,
-      "eval_loss": 0.2169833928346634,
-      "eval_runtime": 414.4566,
-      "eval_samples_per_second": 10.899,
-      "eval_steps_per_second": 1.363,
-      "eval_wer": 0.16305210261835493,
-      "step": 500
+      "epoch": 2.76,
+      "learning_rate": 9.997852628995645e-06,
+      "loss": 0.0272,
+      "step": 450
     },
     {
-      "epoch": 3.8,
-      "learning_rate": 9.99684157225141e-06,
-      "loss": 0.0312,
+      "epoch": 3.68,
+      "learning_rate": 9.996932327136635e-06,
+      "loss": 0.0268,
       "step": 600
     },
     {
-      "epoch": 4.75,
-      "learning_rate": 9.995892145072473e-06,
-      "loss": 0.0312,
+      "epoch": 4.6,
+      "learning_rate": 9.996012025277625e-06,
+      "loss": 0.0261,
       "step": 750
     },
     {
-      "epoch": 5.7,
-      "learning_rate": 9.994942717893538e-06,
-      "loss": 0.0303,
-      "step": 900
-    },
-    {
-      "epoch": 6.33,
-      "eval_loss": 0.21791410446166992,
-      "eval_runtime": 241.3484,
-      "eval_samples_per_second": 18.716,
-      "eval_steps_per_second": 2.341,
-      "eval_wer": 0.16155338093978666,
-      "step": 1000
-    },
-    {
-      "epoch": 6.65,
-      "learning_rate": 9.993993290714603e-06,
-      "loss": 0.0294,
-      "step": 1050
-    },
-    {
-      "epoch": 7.59,
-      "learning_rate": 9.993043863535668e-06,
-      "loss": 0.0297,
-      "step": 1200
-    },
-    {
-      "epoch": 8.54,
-      "learning_rate": 9.992094436356731e-06,
-      "loss": 0.0301,
-      "step": 1350
-    },
-    {
-      "epoch": 9.49,
-      "learning_rate": 9.991145009177796e-06,
-      "loss": 0.0283,
-      "step": 1500
-    },
-    {
-      "epoch": 9.49,
-      "eval_loss": 0.22529518604278564,
-      "eval_runtime": 246.1055,
-      "eval_samples_per_second": 18.354,
-      "eval_steps_per_second": 2.296,
-      "eval_wer": 0.16005465926121837,
-      "step": 1500
-    },
-    {
-      "epoch": 10.44,
-      "learning_rate": 9.990195581998862e-06,
-      "loss": 0.0302,
-      "step": 1650
-    },
-    {
-      "epoch": 11.39,
-      "learning_rate": 9.989246154819927e-06,
-      "loss": 0.0278,
-      "step": 1800
-    },
-    {
-      "epoch": 12.34,
-      "learning_rate": 9.98829672764099e-06,
-      "loss": 0.0282,
-      "step": 1950
-    },
-    {
-      "epoch": 12.66,
-      "eval_loss": 0.22862234711647034,
-      "eval_runtime": 246.4099,
-      "eval_samples_per_second": 18.331,
-      "eval_steps_per_second": 2.293,
-      "eval_wer": 0.16058361985365424,
-      "step": 2000
-    },
-    {
-      "epoch": 13.29,
-      "learning_rate": 9.987353629976582e-06,
-      "loss": 0.0277,
-      "step": 2100
-    },
-    {
-      "epoch": 14.24,
-      "learning_rate": 9.986404202797645e-06,
-      "loss": 0.028,
-      "step": 2250
-    },
-    {
-      "epoch": 15.19,
-      "learning_rate": 9.98545477561871e-06,
-      "loss": 0.0277,
-      "step": 2400
-    },
-    {
-      "epoch": 15.82,
-      "eval_loss": 0.22946597635746002,
-      "eval_runtime": 264.3977,
-      "eval_samples_per_second": 17.084,
-      "eval_steps_per_second": 2.137,
-      "eval_wer": 0.1613329806929384,
-      "step": 2500
-    },
-    {
-      "epoch": 16.14,
-      "learning_rate": 9.984505348439775e-06,
-      "loss": 0.0271,
-      "step": 2550
-    },
-    {
-      "epoch": 17.09,
-      "learning_rate": 9.98355592126084e-06,
-      "loss": 0.0273,
-      "step": 2700
-    },
-    {
-      "epoch": 18.04,
-      "learning_rate": 9.982606494081904e-06,
-      "loss": 0.0268,
-      "step": 2850
-    },
-    {
-      "epoch": 18.99,
-      "learning_rate": 9.981657066902969e-06,
-      "loss": 0.0268,
-      "step": 3000
-    },
-    {
-      "epoch": 18.99,
-      "eval_loss": 0.23895005881786346,
-      "eval_runtime": 263.6411,
-      "eval_samples_per_second": 17.133,
-      "eval_steps_per_second": 2.143,
-      "eval_wer": 0.16243498192717976,
-      "step": 3000
-    },
-    {
-      "epoch": 19.94,
-      "learning_rate": 9.980707639724034e-06,
-      "loss": 0.0256,
-      "step": 3150
-    },
-    {
-      "epoch": 20.89,
-      "learning_rate": 9.979758212545099e-06,
-      "loss": 0.0262,
-      "step": 3300
-    },
-    {
-      "epoch": 21.83,
-      "learning_rate": 9.978808785366162e-06,
-      "loss": 0.0255,
-      "step": 3450
-    },
-    {
-      "epoch": 22.15,
-      "eval_loss": 0.23805132508277893,
-      "eval_runtime": 265.2207,
-      "eval_samples_per_second": 17.031,
-      "eval_steps_per_second": 2.13,
-      "eval_wer": 0.1612007405448294,
-      "step": 3500
-    },
-    {
-      "epoch": 22.78,
-      "learning_rate": 9.977859358187227e-06,
-      "loss": 0.0258,
-      "step": 3600
-    },
-    {
-      "epoch": 23.73,
-      "learning_rate": 9.976909931008292e-06,
-      "loss": 0.0259,
-      "step": 3750
-    },
-    {
-      "epoch": 24.68,
-      "learning_rate": 9.975960503829357e-06,
-      "loss": 0.0259,
-      "step": 3900
-    },
-    {
-      "epoch": 25.32,
-      "eval_loss": 0.24025069177150726,
-      "eval_runtime": 251.4279,
-      "eval_samples_per_second": 17.965,
-      "eval_steps_per_second": 2.247,
-      "eval_wer": 0.1611125804460901,
-      "step": 4000
-    },
-    {
-      "epoch": 25.63,
-      "learning_rate": 9.975011076650422e-06,
-      "loss": 0.0249,
-      "step": 4050
-    },
-    {
-      "epoch": 26.58,
-      "learning_rate": 9.974061649471486e-06,
-      "loss": 0.0257,
-      "step": 4200
-    },
-    {
-      "epoch": 27.53,
-      "learning_rate": 9.97311222229255e-06,
-      "loss": 0.0246,
-      "step": 4350
-    },
-    {
-      "epoch": 28.48,
-      "learning_rate": 9.972162795113616e-06,
-      "loss": 0.0253,
-      "step": 4500
-    },
-    {
-      "epoch": 28.48,
-      "eval_loss": 0.24363693594932556,
-      "eval_runtime": 260.0433,
-      "eval_samples_per_second": 17.37,
-      "eval_steps_per_second": 2.173,
-      "eval_wer": 0.16137706074230804,
-      "step": 4500
-    },
-    {
-      "epoch": 29.43,
-      "learning_rate": 9.97121336793468e-06,
-      "loss": 0.0242,
-      "step": 4650
-    },
-    {
-      "epoch": 30.38,
-      "learning_rate": 9.970263940755744e-06,
-      "loss": 0.0239,
-      "step": 4800
-    },
-    {
-      "epoch": 31.33,
-      "learning_rate": 9.969314513576809e-06,
-      "loss": 0.0235,
-      "step": 4950
-    },
-    {
-      "epoch": 31.65,
-      "eval_loss": 0.24291476607322693,
-      "eval_runtime": 232.435,
-      "eval_samples_per_second": 19.433,
-      "eval_steps_per_second": 2.431,
-      "eval_wer": 0.16115666049545976,
-      "step": 5000
+      "epoch": 4.91,
+      "eval_loss": 0.2315920889377594,
+      "eval_runtime": 257.9698,
+      "eval_samples_per_second": 18.056,
+      "eval_steps_per_second": 2.26,
+      "eval_wer": 0.15478126872699255,
+      "step": 800
     }
   ],
-  "max_steps": 1580000,
+  "max_steps": 1630000,
   "num_train_epochs": 10000,
-  "total_flos": 1.1341132572485426e+20,
+  "total_flos": 1.8155123522141764e+19,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8359e96308fe840448b6443af6dc6fe6a79f6d1d40f9795ba838e7fc0849eec8
+oid sha256:b94603c07847669ce63931a32b056beea3d6cc76af1c8d5dd099819ea13702bd
 size 3323