Training in progress, step 17666

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae5d5f2467be1e0f32383e0f24718a2027023a02821399a4ab911c4727fe071e
+oid sha256:dc3dbc93a147703d7016436477b309809daf6cf7dfd5459380d02ef6d6acd536
 size 667849236

runs/May06_08-32-58_ip-10-192-12-5/events.out.tfevents.1746520379.ip-10-192-12-5.1614.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bf97edebecbae8fc943eb5835585041af0dec87a18f100369ec0e37270d0603
-size 9423
+oid sha256:cd6f2c048a39e9b58f1bdb0970c43405dcc4a73b61c2610eb9690cd98318f6e9
+size 10536

trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 16060,
-  "best_metric": 0.7061941356307186,
-  "best_model_checkpoint": "checkpoints/checkpoint-16060",
-  "epoch": 2.4999805424757753,
+  "best_global_step": 17666,
+  "best_metric": 0.706001258948505,
+  "best_model_checkpoint": "checkpoints/checkpoint-17666",
+  "epoch": 2.749970813713663,
   "eval_steps": 1606,
-  "global_step": 16060,
+  "global_step": 17666,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -352,6 +352,39 @@
       "eval_steps_per_second": 1.752,
       "eval_wer": 0.670901504786034,
       "step": 16060
+    },
+    {
+      "epoch": 2.5684710277464298,
+      "grad_norm": 0.032605357468128204,
+      "learning_rate": 4.136843819999117e-05,
+      "loss": 0.038,
+      "step": 16500
+    },
+    {
+      "epoch": 2.6463011246449,
+      "grad_norm": 0.027949590235948563,
+      "learning_rate": 3.977561947162954e-05,
+      "loss": 0.0376,
+      "step": 17000
+    },
+    {
+      "epoch": 2.724131221543371,
+      "grad_norm": 0.026584528386592865,
+      "learning_rate": 3.80846956772673e-05,
+      "loss": 0.0372,
+      "step": 17500
+    },
+    {
+      "epoch": 2.749970813713663,
+      "eval_avg": 0.706001258948505,
+      "eval_cer": 0.6536609451579365,
+      "eval_der": 0.7939834810607371,
+      "eval_loss": 0.030004331842064857,
+      "eval_runtime": 304.4012,
+      "eval_samples_per_second": 445.869,
+      "eval_steps_per_second": 1.744,
+      "eval_wer": 0.6703593506268415,
+      "step": 17666
     }
   ],
   "logging_steps": 500,
@@ -371,7 +404,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.253480973209174e+18,
+  "total_flos": 1.3788271574227354e+18,
   "train_batch_size": 256,
   "trial_name": null,
   "trial_params": null