Training in progress, step 37000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2cb8dabf9a7be0aaf98fdca21d8221dc07d657cdc9101214ce22d983671a048
+oid sha256:82e580b76e08cdfb8448ca6d1803c64c049f1cef861419e8036adf1677312206
 size 223144592

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e6212c976ed04ebbd236c6446302224a7aefbce6d3836bd327b9b5c728fad7f
+oid sha256:51fc0f82c59c2b71d10f095d4baf2da47474fd2fe1bbcbc826cc6c39a81ee2ec
 size 281574266

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3de3a67d18a339a861c6ed23882a029410556f39281d29ad2ad4c3bc7f88fa73
+oid sha256:5f328fe058de8a228ddebf585a5f24f49207d162996c245c7975352b2646f7ee
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3dc7dd0f418853d90ca435bb584a7c3ec5f7940e05d8ede71e5502af2d6925a
+oid sha256:29c8852b8972fe5d4076d17cae40f3bcca1d0af379559b08bc3be9a35e8e8830
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8feb46fcf2857c50ea4a63a9573ecc3d64407b5246f4262683176b26b9e57b06
+oid sha256:c5779eae6320dd873f29bcf99b0bddfe558494d35dfd606a3b158d47caa27415
 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": 21.0896309314587,
+  "epoch": 21.675454012888107,
   "eval_steps": 1000,
-  "global_step": 36000,
+  "global_step": 37000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2852,6 +2852,85 @@
       "eval_steps_per_second": 0.427,
       "eval_wer": 0.18818155025051578,
       "step": 36000
+    },
+    {
+      "epoch": 21.148213239601642,
+      "grad_norm": 6.068084239959717,
+      "learning_rate": 3.972580558104388e-06,
+      "loss": 0.7155,
+      "step": 36100
+    },
+    {
+      "epoch": 21.20679554774458,
+      "grad_norm": 4.509976863861084,
+      "learning_rate": 3.702704161494036e-06,
+      "loss": 0.6615,
+      "step": 36200
+    },
+    {
+      "epoch": 21.26537785588752,
+      "grad_norm": 6.346072673797607,
+      "learning_rate": 3.4328277648836835e-06,
+      "loss": 0.6682,
+      "step": 36300
+    },
+    {
+      "epoch": 21.323960164030463,
+      "grad_norm": 6.095388889312744,
+      "learning_rate": 3.1629513682733306e-06,
+      "loss": 0.6588,
+      "step": 36400
+    },
+    {
+      "epoch": 21.382542472173405,
+      "grad_norm": 5.079260349273682,
+      "learning_rate": 2.8930749716629784e-06,
+      "loss": 0.67,
+      "step": 36500
+    },
+    {
+      "epoch": 21.441124780316343,
+      "grad_norm": 4.323718547821045,
+      "learning_rate": 2.623198575052626e-06,
+      "loss": 0.6606,
+      "step": 36600
+    },
+    {
+      "epoch": 21.499707088459285,
+      "grad_norm": 5.659894943237305,
+      "learning_rate": 2.3533221784422738e-06,
+      "loss": 0.69,
+      "step": 36700
+    },
+    {
+      "epoch": 21.558289396602227,
+      "grad_norm": 7.08272647857666,
+      "learning_rate": 2.083445781831921e-06,
+      "loss": 0.6506,
+      "step": 36800
+    },
+    {
+      "epoch": 21.61687170474517,
+      "grad_norm": 4.9979119300842285,
+      "learning_rate": 1.8135693852215685e-06,
+      "loss": 0.7215,
+      "step": 36900
+    },
+    {
+      "epoch": 21.675454012888107,
+      "grad_norm": 6.872857093811035,
+      "learning_rate": 1.5436929886112161e-06,
+      "loss": 0.7014,
+      "step": 37000
+    },
+    {
+      "epoch": 21.675454012888107,
+      "eval_loss": 0.08349551260471344,
+      "eval_runtime": 147.5946,
+      "eval_samples_per_second": 3.388,
+      "eval_steps_per_second": 0.427,
+      "eval_wer": 0.1861184792219275,
+      "step": 37000
     }
   ],
   "logging_steps": 100,
@@ -2871,7 +2950,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.682615727587328e+19,
+  "total_flos": 4.812697871843328e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null