Training in progress, step 9400, checkpoint

checkpoint-9400/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c08b169dbc301670a5a7437983e2bdb309c761725c4ad00a4e504f279b35b470
+oid sha256:74fc93f66189b6167184fea73b2e23a9c55c2a1934f0bad20d21ddf0ea8fa2b2
 size 1261975580

checkpoint-9400/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24db415858cc06e5a8a6e3adb939b78fcf984184af2c12f89746aca356d5075d
+oid sha256:64c942ac8f092000faab3d1a2b8aca0f1e57ba65e6951fad7df8f8921102fc6e
 size 2490495926

checkpoint-9400/trainer_state.json

@@ -6358,146 +6358,146 @@
     },
     {
       "epoch": 15.0,
-      "learning_rate": 1.920199501246883e-05,
-      "loss": 9.0274,
+      "learning_rate": 7.629139072847681e-05,
+      "loss": 6.7067,
       "step": 9210
     },
     {
       "epoch": 15.02,
-      "learning_rate": 1.8890274314214465e-05,
-      "loss": 4.9159,
+      "learning_rate": 7.60430463576159e-05,
+      "loss": 3.5775,
       "step": 9220
     },
     {
       "epoch": 15.03,
-      "learning_rate": 1.8578553615960096e-05,
-      "loss": 3.9926,
+      "learning_rate": 7.579470198675495e-05,
+      "loss": 3.1504,
       "step": 9230
     },
     {
       "epoch": 15.05,
-      "learning_rate": 1.8266832917705734e-05,
-      "loss": 3.6284,
+      "learning_rate": 7.554635761589404e-05,
+      "loss": 3.0605,
       "step": 9240
     },
     {
       "epoch": 15.06,
-      "learning_rate": 1.795511221945137e-05,
-      "loss": 3.4183,
+      "learning_rate": 7.52980132450331e-05,
+      "loss": 3.0235,
       "step": 9250
     },
     {
       "epoch": 15.08,
-      "learning_rate": 1.7643391521197004e-05,
-      "loss": 3.1265,
+      "learning_rate": 7.504966887417217e-05,
+      "loss": 2.9939,
       "step": 9260
     },
     {
       "epoch": 15.1,
-      "learning_rate": 1.7331670822942642e-05,
-      "loss": 2.7977,
+      "learning_rate": 7.480132450331126e-05,
+      "loss": 2.9837,
       "step": 9270
     },
     {
       "epoch": 15.11,
-      "learning_rate": 1.7019950124688277e-05,
-      "loss": 2.0169,
+      "learning_rate": 7.455298013245033e-05,
+      "loss": 2.8137,
       "step": 9280
     },
     {
       "epoch": 15.13,
-      "learning_rate": 1.6708229426433912e-05,
-      "loss": 1.3587,
+      "learning_rate": 7.43046357615894e-05,
+      "loss": 1.9888,
       "step": 9290
     },
     {
       "epoch": 15.15,
-      "learning_rate": 1.639650872817955e-05,
-      "loss": 0.9968,
+      "learning_rate": 7.405629139072846e-05,
+      "loss": 1.0386,
       "step": 9300
     },
     {
       "epoch": 15.15,
-      "eval_loss": 0.7115535140037537,
-      "eval_runtime": 1033.0002,
-      "eval_samples_per_second": 11.238,
-      "eval_steps_per_second": 0.703,
-      "eval_wer": 0.9390480317474799,
+      "eval_loss": 0.43749570846557617,
+      "eval_runtime": 1037.1435,
+      "eval_samples_per_second": 11.193,
+      "eval_steps_per_second": 0.7,
+      "eval_wer": 0.679286593988049,
       "step": 9300
     },
     {
       "epoch": 15.16,
-      "learning_rate": 1.6084788029925186e-05,
-      "loss": 0.7074,
+      "learning_rate": 7.380794701986755e-05,
+      "loss": 0.4016,
       "step": 9310
     },
     {
       "epoch": 15.18,
-      "learning_rate": 1.577306733167082e-05,
-      "loss": 0.5215,
+      "learning_rate": 7.355960264900662e-05,
+      "loss": 0.23,
       "step": 9320
     },
     {
       "epoch": 15.19,
-      "learning_rate": 1.5461346633416455e-05,
-      "loss": 0.446,
+      "learning_rate": 7.331125827814569e-05,
+      "loss": 0.1834,
       "step": 9330
     },
     {
       "epoch": 15.21,
-      "learning_rate": 1.5149625935162094e-05,
-      "loss": 0.3456,
+      "learning_rate": 7.306291390728476e-05,
+      "loss": 0.149,
       "step": 9340
     },
     {
       "epoch": 15.23,
-      "learning_rate": 1.4837905236907729e-05,
-      "loss": 0.2667,
+      "learning_rate": 7.281456953642384e-05,
+      "loss": 0.1275,
       "step": 9350
     },
     {
       "epoch": 15.24,
-      "learning_rate": 1.4526184538653365e-05,
-      "loss": 0.2608,
+      "learning_rate": 7.256622516556291e-05,
+      "loss": 0.1483,
       "step": 9360
     },
     {
       "epoch": 15.26,
-      "learning_rate": 1.4214463840399e-05,
-      "loss": 0.1953,
+      "learning_rate": 7.231788079470198e-05,
+      "loss": 0.1038,
       "step": 9370
     },
     {
       "epoch": 15.28,
-      "learning_rate": 1.3902743142144637e-05,
-      "loss": 0.204,
+      "learning_rate": 7.206953642384106e-05,
+      "loss": 0.1185,
       "step": 9380
     },
     {
       "epoch": 15.29,
-      "learning_rate": 1.3591022443890273e-05,
-      "loss": 0.1784,
+      "learning_rate": 7.182119205298013e-05,
+      "loss": 0.1099,
       "step": 9390
     },
     {
       "epoch": 15.31,
-      "learning_rate": 1.3279301745635908e-05,
-      "loss": 0.1715,
+      "learning_rate": 7.15728476821192e-05,
+      "loss": 0.116,
       "step": 9400
     },
     {
       "epoch": 15.31,
-      "eval_loss": 0.16349774599075317,
-      "eval_runtime": 1030.7944,
-      "eval_samples_per_second": 11.262,
-      "eval_steps_per_second": 0.704,
-      "eval_wer": 0.3316266021986042,
+      "eval_loss": 0.12657363712787628,
+      "eval_runtime": 1045.2349,
+      "eval_samples_per_second": 11.107,
+      "eval_steps_per_second": 0.695,
+      "eval_wer": 0.30423527801851935,
       "step": 9400
     }
   ],
   "logging_steps": 10,
-  "max_steps": 9824,
-  "num_train_epochs": 16,
+  "max_steps": 12280,
+  "num_train_epochs": 20,
   "save_steps": 200,
   "total_flos": 9.855811400623335e+19,
   "trial_name": null,