Training in progress, step 36000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c40ac5768678ec5e02f1ef67d5a4309a7523549c081d4aa4dcec3a9e62775b9
+oid sha256:f2cb8dabf9a7be0aaf98fdca21d8221dc07d657cdc9101214ce22d983671a048
 size 223144592

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04e97e0d3303ce9166cbde3be58ec202701027a873651cfd877e10f3438e1bef
+oid sha256:7e6212c976ed04ebbd236c6446302224a7aefbce6d3836bd327b9b5c728fad7f
 size 281574266

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9bbeae69b0c3ccad7b2da286244601c7cab8dcaae00fabe926ee3c6e16123c7
+oid sha256:3de3a67d18a339a861c6ed23882a029410556f39281d29ad2ad4c3bc7f88fa73
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f23d0e9ac8ef166c949e157928330a7cb0661782ade2d248d1a928b2ac29805e
+oid sha256:d3dc7dd0f418853d90ca435bb584a7c3ec5f7940e05d8ede71e5502af2d6925a
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d08ed08052d252f71cbe4b116766edbb8b5b9589f8f40bb95cb4854fc9cf848
+oid sha256:8feb46fcf2857c50ea4a63a9573ecc3d64407b5246f4262683176b26b9e57b06
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 28000,
   "best_metric": 0.18110816386678455,
   "best_model_checkpoint": "./distil-whisper/checkpoint-28000",
-  "epoch": 20.50380785002929,
+  "epoch": 21.0896309314587,
   "eval_steps": 1000,
-  "global_step": 35000,
+  "global_step": 36000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2773,6 +2773,85 @@
       "eval_steps_per_second": 0.427,
       "eval_wer": 0.1843501326259947,
       "step": 35000
+    },
+    {
+      "epoch": 20.56239015817223,
+      "grad_norm": 6.862576007843018,
+      "learning_rate": 6.668645760241809e-06,
+      "loss": 0.6942,
+      "step": 35100
+    },
+    {
+      "epoch": 20.620972466315173,
+      "grad_norm": 10.506976127624512,
+      "learning_rate": 6.3987693636314575e-06,
+      "loss": 0.6863,
+      "step": 35200
+    },
+    {
+      "epoch": 20.679554774458115,
+      "grad_norm": 6.648481845855713,
+      "learning_rate": 6.128892967021105e-06,
+      "loss": 0.6987,
+      "step": 35300
+    },
+    {
+      "epoch": 20.738137082601053,
+      "grad_norm": 7.607233047485352,
+      "learning_rate": 5.8590165704107524e-06,
+      "loss": 0.689,
+      "step": 35400
+    },
+    {
+      "epoch": 20.796719390743995,
+      "grad_norm": 6.110599040985107,
+      "learning_rate": 5.589140173800399e-06,
+      "loss": 0.7004,
+      "step": 35500
+    },
+    {
+      "epoch": 20.855301698886937,
+      "grad_norm": 5.632733345031738,
+      "learning_rate": 5.319263777190047e-06,
+      "loss": 0.7299,
+      "step": 35600
+    },
+    {
+      "epoch": 20.91388400702988,
+      "grad_norm": 5.644909381866455,
+      "learning_rate": 5.052086144545798e-06,
+      "loss": 0.6749,
+      "step": 35700
+    },
+    {
+      "epoch": 20.972466315172817,
+      "grad_norm": 5.2816386222839355,
+      "learning_rate": 4.782209747935446e-06,
+      "loss": 0.671,
+      "step": 35800
+    },
+    {
+      "epoch": 21.03104862331576,
+      "grad_norm": 4.735965728759766,
+      "learning_rate": 4.512333351325093e-06,
+      "loss": 0.7137,
+      "step": 35900
+    },
+    {
+      "epoch": 21.0896309314587,
+      "grad_norm": 6.491783618927002,
+      "learning_rate": 4.242456954714741e-06,
+      "loss": 0.6927,
+      "step": 36000
+    },
+    {
+      "epoch": 21.0896309314587,
+      "eval_loss": 0.08344998210668564,
+      "eval_runtime": 147.4098,
+      "eval_samples_per_second": 3.392,
+      "eval_steps_per_second": 0.427,
+      "eval_wer": 0.18818155025051578,
+      "step": 36000
     }
   ],
   "logging_steps": 100,
@@ -2792,7 +2871,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.55254984359936e+19,
+  "total_flos": 4.682615727587328e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null