Training in progress, step 6700

config.json

@@ -110,7 +110,7 @@
     1
   ],
   "torch_dtype": "float32",
-  "transformers_version": "4.18.0",
+  "transformers_version": "4.23.1",
   "use_weighted_layer_sum": false,
   "vocab_size": 28,
   "xvector_output_dim": 512

last-checkpoint/config.json

@@ -110,7 +110,7 @@
     1
   ],
   "torch_dtype": "float32",
-  "transformers_version": "4.18.0",
+  "transformers_version": "4.23.1",
   "use_weighted_layer_sum": false,
   "vocab_size": 28,
   "xvector_output_dim": 512

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bbc0864acfd13b716f75d307bf48aada3819f622ff0bd60c9128105ebfcff8e
-size 174303
+oid sha256:f482282dd625d07e4ef01233b70e2e8c1030c46fe2ebeb4b1901cd1670300157
+size 721661957

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:887961831a0f30b4269cdd28f6acde7f7d51f9a06abaabe2de32501df0b1860f
-size 377644503
+oid sha256:65da033b1e874758bedcfe07b3b6d54ea6005c90eead128adc27e7b5d7f21527
+size 377646433

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93198e96e3eab53a5e609a281fb99228f51dc43ea5bfb0348b2d825a7a3ab647
-size 14567
+oid sha256:d4c714b776c16a9ba181a27bc6c2e7d70ca6d2267f5b7991f3fd1c9388171ab7
+size 14639

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61823d7f75e62090d2b8f678678ba83a75ec4c530252db1f2a051fa6494b9b82
-size 559
+oid sha256:bc1873a3085fec4186eecba9b874bba0fd9ad0105505617746123f3e2e0f8c3f
+size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26fc2b87137deb440e0add509cb5c6eaf84f92619ce4a6bab36eea0195aa5885
-size 623
+oid sha256:2310a544508b4524ba2632c6fbf45ea4b35663588ff13c743ebc9b477b456bba
+size 627

last-checkpoint/trainer_state.json

@@ -1,176 +1,32 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 250.0,
-  "global_step": 134000,
+  "epoch": 10.0,
+  "global_step": 6700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 25.0,
-      "learning_rate": 9.474312647289867e-05,
-      "loss": 20.2416,
-      "step": 13400
+      "epoch": 10.0,
+      "learning_rate": 9.474941084053418e-05,
+      "loss": 4.2941,
+      "step": 6700
     },
     {
-      "epoch": 25.0,
-      "eval_cer": 0.9767676767676767,
-      "eval_loss": 3.248976230621338,
-      "eval_runtime": 0.8563,
-      "eval_samples_per_second": 210.199,
-      "eval_steps_per_second": 26.859,
-      "eval_wer": 1.0,
-      "step": 13400
-    },
-    {
-      "epoch": 50.0,
-      "learning_rate": 8.422073841319718e-05,
-      "loss": 2.7136,
-      "step": 26800
-    },
-    {
-      "epoch": 50.0,
-      "eval_cer": 0.9222222222222223,
-      "eval_loss": 3.162766218185425,
-      "eval_runtime": 0.8572,
-      "eval_samples_per_second": 209.98,
-      "eval_steps_per_second": 26.831,
-      "eval_wer": 1.0,
-      "step": 26800
-    },
-    {
-      "epoch": 75.0,
-      "learning_rate": 7.369835035349568e-05,
-      "loss": 2.574,
-      "step": 40200
-    },
-    {
-      "epoch": 75.0,
-      "eval_cer": 0.9141414141414141,
-      "eval_loss": 3.141190528869629,
-      "eval_runtime": 0.8427,
-      "eval_samples_per_second": 213.593,
-      "eval_steps_per_second": 27.292,
-      "eval_wer": 1.0,
-      "step": 40200
-    },
-    {
-      "epoch": 100.0,
-      "learning_rate": 6.317674783974863e-05,
-      "loss": 2.5132,
-      "step": 53600
-    },
-    {
-      "epoch": 100.0,
-      "eval_cer": 0.9101010101010101,
-      "eval_loss": 3.159379720687866,
-      "eval_runtime": 0.8438,
-      "eval_samples_per_second": 213.316,
-      "eval_steps_per_second": 27.257,
-      "eval_wer": 1.0,
-      "step": 53600
-    },
-    {
-      "epoch": 125.0,
-      "learning_rate": 5.2655145326001575e-05,
-      "loss": 2.4764,
-      "step": 67000
-    },
-    {
-      "epoch": 125.0,
-      "eval_cer": 0.9050505050505051,
-      "eval_loss": 3.132108449935913,
-      "eval_runtime": 0.8431,
-      "eval_samples_per_second": 213.503,
-      "eval_steps_per_second": 27.281,
-      "eval_wer": 1.0,
-      "step": 67000
-    },
-    {
-      "epoch": 150.0,
-      "learning_rate": 4.2132757266300084e-05,
-      "loss": 2.4522,
-      "step": 80400
-    },
-    {
-      "epoch": 150.0,
-      "eval_cer": 0.9030303030303031,
-      "eval_loss": 3.149836301803589,
-      "eval_runtime": 0.842,
-      "eval_samples_per_second": 213.776,
-      "eval_steps_per_second": 27.316,
-      "eval_wer": 1.0,
-      "step": 80400
-    },
-    {
-      "epoch": 175.0,
-      "learning_rate": 3.1610369206598586e-05,
-      "loss": 2.4366,
-      "step": 93800
-    },
-    {
-      "epoch": 175.0,
-      "eval_cer": 0.901010101010101,
-      "eval_loss": 3.154324769973755,
-      "eval_runtime": 0.8423,
-      "eval_samples_per_second": 213.693,
-      "eval_steps_per_second": 27.305,
-      "eval_wer": 1.0,
-      "step": 93800
-    },
-    {
-      "epoch": 200.0,
-      "learning_rate": 2.1088766692851532e-05,
-      "loss": 2.426,
-      "step": 107200
-    },
-    {
-      "epoch": 200.0,
-      "eval_cer": 0.901010101010101,
-      "eval_loss": 3.1591243743896484,
-      "eval_runtime": 0.8434,
-      "eval_samples_per_second": 213.431,
-      "eval_steps_per_second": 27.272,
-      "eval_wer": 1.0,
-      "step": 107200
-    },
-    {
-      "epoch": 225.0,
-      "learning_rate": 1.056637863315004e-05,
-      "loss": 2.4199,
-      "step": 120600
-    },
-    {
-      "epoch": 225.0,
-      "eval_cer": 0.898989898989899,
-      "eval_loss": 3.1549627780914307,
-      "eval_runtime": 0.8433,
-      "eval_samples_per_second": 213.458,
-      "eval_steps_per_second": 27.275,
-      "eval_wer": 1.0,
-      "step": 120600
-    },
-    {
-      "epoch": 250.0,
-      "learning_rate": 4.4776119402985075e-08,
-      "loss": 2.4161,
-      "step": 134000
-    },
-    {
-      "epoch": 250.0,
-      "eval_cer": 0.898989898989899,
-      "eval_loss": 3.1552622318267822,
-      "eval_runtime": 0.843,
-      "eval_samples_per_second": 213.521,
-      "eval_steps_per_second": 27.283,
-      "eval_wer": 1.0,
-      "step": 134000
+      "epoch": 10.0,
+      "eval_cer": 0.4717171717171717,
+      "eval_loss": 3.055922508239746,
+      "eval_runtime": 6.8504,
+      "eval_samples_per_second": 26.276,
+      "eval_steps_per_second": 3.357,
+      "eval_wer": 0.8777777777777778,
+      "step": 6700
     }
   ],
-  "max_steps": 134000,
-  "num_train_epochs": 250,
-  "total_flos": 1.2448746848771059e+20,
+  "max_steps": 67000,
+  "num_train_epochs": 100,
+  "total_flos": 4.978806059229573e+18,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1381c52f948e9ae84ef99d734a89cfbb100461d77e8a2a97dc058cd76c194855
-size 3119
+oid sha256:c1b8e65eef88833700c90fcdc65889067bd9eefe856492cda02e2d22b5581adf
+size 3451

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:887961831a0f30b4269cdd28f6acde7f7d51f9a06abaabe2de32501df0b1860f
-size 377644503
+oid sha256:65da033b1e874758bedcfe07b3b6d54ea6005c90eead128adc27e7b5d7f21527
+size 377646433

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1381c52f948e9ae84ef99d734a89cfbb100461d77e8a2a97dc058cd76c194855
-size 3119
+oid sha256:c1b8e65eef88833700c90fcdc65889067bd9eefe856492cda02e2d22b5581adf
+size 3451