Training in progress, step 2000

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfae1ce0483ee7d709f85d979ef48ca34acff837cb53b6b0b0b5738dbe80615f
+oid sha256:681a60ed57add5fb21363a44a2f5591041a0fe5d518d5c483aa0636c868c46a7
 size 1934161157

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52d2c8e89b76968dd8da7a9713eae29748d3ac37683f9a4695606aeea8480f2f
+oid sha256:f9dd712d07b073f0caf4eec287bd0224534565171a49c28cbf3ca0b4b9b7cb98
 size 967102729

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4654d9957152370ac9eb45d2aa8d04af5e9438ba0e7571164ecc0dbb97b880c8
+oid sha256:1ff2483baa43e96dd429264d4ff11728434987ac928368cb400847576898fcc7
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90dd2338805bd3a69fa663686537f5e5065d09a0615f56ffc3e4278e4e48b9dd
+oid sha256:9536b63cd02db26062d122a42fc284c452283fd0cb8cc2a36d3aee25ee14708d
 size 627

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 123.61493123772102,
-  "best_model_checkpoint": "./whisper-vietnamese-3/checkpoint-1000",
-  "epoch": 7.194244604316546,
-  "global_step": 1000,
+  "best_metric": 111.96070726915521,
+  "best_model_checkpoint": "./whisper-vietnamese-3/checkpoint-2000",
+  "epoch": 14.388489208633093,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -256,11 +256,261 @@
       "eval_steps_per_second": 0.125,
       "eval_wer": 123.61493123772102,
       "step": 1000
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 8.505714285714287e-06,
+      "loss": 0.0215,
+      "step": 1025
+    },
+    {
+      "epoch": 7.55,
+      "learning_rate": 8.434285714285716e-06,
+      "loss": 0.0214,
+      "step": 1050
+    },
+    {
+      "epoch": 7.73,
+      "learning_rate": 8.362857142857143e-06,
+      "loss": 0.0222,
+      "step": 1075
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 8.291428571428572e-06,
+      "loss": 0.0234,
+      "step": 1100
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 8.220000000000001e-06,
+      "loss": 0.016,
+      "step": 1125
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 8.148571428571428e-06,
+      "loss": 0.0135,
+      "step": 1150
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 8.077142857142857e-06,
+      "loss": 0.0144,
+      "step": 1175
+    },
+    {
+      "epoch": 8.63,
+      "learning_rate": 8.005714285714286e-06,
+      "loss": 0.0139,
+      "step": 1200
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 7.934285714285715e-06,
+      "loss": 0.0126,
+      "step": 1225
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 7.862857142857143e-06,
+      "loss": 0.0146,
+      "step": 1250
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 7.791428571428572e-06,
+      "loss": 0.0089,
+      "step": 1275
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 7.72e-06,
+      "loss": 0.0081,
+      "step": 1300
+    },
+    {
+      "epoch": 9.53,
+      "learning_rate": 7.64857142857143e-06,
+      "loss": 0.0094,
+      "step": 1325
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 7.577142857142857e-06,
+      "loss": 0.0084,
+      "step": 1350
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 7.505714285714286e-06,
+      "loss": 0.0075,
+      "step": 1375
+    },
+    {
+      "epoch": 10.07,
+      "learning_rate": 7.434285714285715e-06,
+      "loss": 0.0084,
+      "step": 1400
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 7.362857142857144e-06,
+      "loss": 0.0063,
+      "step": 1425
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 7.291428571428571e-06,
+      "loss": 0.0057,
+      "step": 1450
+    },
+    {
+      "epoch": 10.61,
+      "learning_rate": 7.22e-06,
+      "loss": 0.0055,
+      "step": 1475
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 7.148571428571429e-06,
+      "loss": 0.0062,
+      "step": 1500
+    },
+    {
+      "epoch": 10.97,
+      "learning_rate": 7.077142857142858e-06,
+      "loss": 0.0068,
+      "step": 1525
+    },
+    {
+      "epoch": 11.15,
+      "learning_rate": 7.0057142857142865e-06,
+      "loss": 0.0041,
+      "step": 1550
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 6.934285714285715e-06,
+      "loss": 0.0045,
+      "step": 1575
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 6.862857142857144e-06,
+      "loss": 0.0047,
+      "step": 1600
+    },
+    {
+      "epoch": 11.69,
+      "learning_rate": 6.791428571428572e-06,
+      "loss": 0.0048,
+      "step": 1625
+    },
+    {
+      "epoch": 11.87,
+      "learning_rate": 6.720000000000001e-06,
+      "loss": 0.0037,
+      "step": 1650
+    },
+    {
+      "epoch": 12.05,
+      "learning_rate": 6.648571428571429e-06,
+      "loss": 0.0034,
+      "step": 1675
+    },
+    {
+      "epoch": 12.23,
+      "learning_rate": 6.577142857142857e-06,
+      "loss": 0.0031,
+      "step": 1700
+    },
+    {
+      "epoch": 12.41,
+      "learning_rate": 6.505714285714286e-06,
+      "loss": 0.003,
+      "step": 1725
+    },
+    {
+      "epoch": 12.59,
+      "learning_rate": 6.434285714285715e-06,
+      "loss": 0.0044,
+      "step": 1750
+    },
+    {
+      "epoch": 12.77,
+      "learning_rate": 6.3628571428571426e-06,
+      "loss": 0.0033,
+      "step": 1775
+    },
+    {
+      "epoch": 12.95,
+      "learning_rate": 6.2914285714285716e-06,
+      "loss": 0.0032,
+      "step": 1800
+    },
+    {
+      "epoch": 13.13,
+      "learning_rate": 6.220000000000001e-06,
+      "loss": 0.0032,
+      "step": 1825
+    },
+    {
+      "epoch": 13.31,
+      "learning_rate": 6.14857142857143e-06,
+      "loss": 0.0027,
+      "step": 1850
+    },
+    {
+      "epoch": 13.49,
+      "learning_rate": 6.077142857142858e-06,
+      "loss": 0.0029,
+      "step": 1875
+    },
+    {
+      "epoch": 13.67,
+      "learning_rate": 6.005714285714286e-06,
+      "loss": 0.0028,
+      "step": 1900
+    },
+    {
+      "epoch": 13.85,
+      "learning_rate": 5.934285714285715e-06,
+      "loss": 0.0023,
+      "step": 1925
+    },
+    {
+      "epoch": 14.03,
+      "learning_rate": 5.862857142857143e-06,
+      "loss": 0.0019,
+      "step": 1950
+    },
+    {
+      "epoch": 14.21,
+      "learning_rate": 5.791428571428572e-06,
+      "loss": 0.0029,
+      "step": 1975
+    },
+    {
+      "epoch": 14.39,
+      "learning_rate": 5.72e-06,
+      "loss": 0.0022,
+      "step": 2000
+    },
+    {
+      "epoch": 14.39,
+      "eval_cer": 104.70151047347889,
+      "eval_loss": 0.5671390295028687,
+      "eval_runtime": 247.6486,
+      "eval_samples_per_second": 0.997,
+      "eval_steps_per_second": 0.125,
+      "eval_wer": 111.96070726915521,
+      "step": 2000
     }
   ],
   "max_steps": 4000,
   "num_train_epochs": 29,
-  "total_flos": 4.60322572050432e+18,
+  "total_flos": 9.20645144100864e+18,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52d2c8e89b76968dd8da7a9713eae29748d3ac37683f9a4695606aeea8480f2f
+oid sha256:f9dd712d07b073f0caf4eec287bd0224534565171a49c28cbf3ca0b4b9b7cb98
 size 967102729

runs/Aug17_16-46-37_8582233f681d/events.out.tfevents.1692290809.8582233f681d.216.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:009f2e6ffda944e495480f1aae52f4224acf27ead27752075824589e45dbf9bb
-size 11531
+oid sha256:181579ae0c49e470dcaaaddd157e87cc76dad37db6713ce767a401b7c6770256
+size 18176