Training in progress, step 12000

{checkpoint-9000 → checkpoint-12000}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:853a10b701c41859978527142af4efb610899026f2e0fa7952b243e513fbe8cd
+oid sha256:53875c50949cb858b32b38afb434db1546fa61e5e5dcda9462f12c903fe9a8ea
 size 721687499

{checkpoint-9000 → checkpoint-12000}/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81b0f74cc3edaa150bfa1b6a8d50d40730c767b06b515bfe1fe7ed56d12fa4bc
+oid sha256:96d132744211d2c04695c3263e34a2f814f3d08d1bb925aff50c57941723e3b1
 size 377674359

{checkpoint-9000 → checkpoint-12000}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95016d19f94268c2f81c3982cc7610459bc11c4070f1fc44c48b8e16ebb72e44
-size 14709
+oid sha256:24e705e7fbd74f4035ace4fa92507d51e4b05a05c88a5a0718de02484617fc51
+size 14645

{checkpoint-9000 → checkpoint-12000}/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:341b48166e5b847633c7df137f2c55412ed2c1d14b58e81d4d9b894963a415f1
+oid sha256:3e174a76e0e352d264e5f92fd6a6b771293954f5541c746e7e66ccde440cbb3e
 size 1383

{checkpoint-9000 → checkpoint-12000}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e58b2aeb001747b91828136e7425c14c72edfc65e6905aea3a08554cc37fcafd
+oid sha256:055259b9a818f446687e46aed4d609ba8d81f683861514ef91a50adacfd89299
 size 1465

{checkpoint-9000 → checkpoint-12000}/trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.6258692628650904,
-  "global_step": 9000,
+  "epoch": 0.8344923504867872,
+  "global_step": 12000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -141,11 +141,56 @@
       "eval_steps_per_second": 1.891,
       "eval_wer": 0.355694020946573,
       "step": 9000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.561273373512595e-05,
+      "loss": 0.4482,
+      "step": 10000
+    },
+    {
+      "epoch": 0.7,
+      "eval_loss": 0.7687897086143494,
+      "eval_runtime": 1333.2181,
+      "eval_samples_per_second": 14.947,
+      "eval_steps_per_second": 1.868,
+      "eval_wer": 0.343775176169449,
+      "step": 10000
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 2.4840055632823366e-05,
+      "loss": 0.4311,
+      "step": 11000
+    },
+    {
+      "epoch": 0.76,
+      "eval_loss": 0.7420344948768616,
+      "eval_runtime": 1329.3248,
+      "eval_samples_per_second": 14.991,
+      "eval_steps_per_second": 1.874,
+      "eval_wer": 0.3233537972037243,
+      "step": 11000
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 2.4067377530520787e-05,
+      "loss": 0.4272,
+      "step": 12000
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 0.782776415348053,
+      "eval_runtime": 1318.9648,
+      "eval_samples_per_second": 15.109,
+      "eval_steps_per_second": 1.889,
+      "eval_wer": 0.33638421766487525,
+      "step": 12000
     }
   ],
   "max_steps": 43140,
   "num_train_epochs": 3,
-  "total_flos": 5.355539118047232e+18,
+  "total_flos": 7.137862466936832e+18,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6cf498c1fa08223b228f427a79968055619ec91d8255ce93830627bb36991315
+oid sha256:96d132744211d2c04695c3263e34a2f814f3d08d1bb925aff50c57941723e3b1
 size 377674359

runs/Mar10_01-38-16_42fb38fbc845/events.out.tfevents.1773113504.42fb38fbc845.2655.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8dfaedad67200afce2e8e06ba7e7c4adf7e54f74ecf1ab91b12f49039364881
-size 10271
+oid sha256:cd2aab3b538cc43c0d6363401500d463f12630515cca37920b859769dcb3917c
+size 10746