checkpoint 10000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a61e6e5352efdd5b7f9da781d485d41785a65d97707d8a50f8ed3703500e382
+oid sha256:251cf0da910cf3abee2942ff7eae2edd8dda496b67e50deab3f3aea1ae62d309
 size 1262032932

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d61d798a6ebe2df9f0c91788f23fc918b91a4b2bf0166521c19127eccff65502
+oid sha256:0dee2bf19e6def9c192714575ad6d0c20e33dbb574141996e4fb07e457c5a3fa
 size 2490610241

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70d3644a4cf26f14e27a5301570cd0ec97d359dfc1ab88070196c7de228b37c2
+oid sha256:900edcf97c1b52d6eb2460d65edd6df6a11b8d85295bd757a9f6a84b06841e42
 size 14639

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0fb799b5054cd707634c1757db8b87f65d5adfd51f621867293ee2d7170c2937
+oid sha256:18d9cb2a15a3d9c8a7899c5621a4e08eba4ccc77feac595dd9ae3fd412adcdb1
 size 627

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3642527461051941,
-  "best_model_checkpoint": "./working/checkpoint-9000",
-  "epoch": 107.14285714285714,
+  "best_metric": 0.3630259037017822,
+  "best_model_checkpoint": "./working/checkpoint-10000",
+  "epoch": 119.04761904761905,
   "eval_steps": 1000,
-  "global_step": 9000,
+  "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -727,6 +727,86 @@
       "eval_steps_per_second": 3.997,
       "eval_wer": 0.6248053969901401,
       "step": 9000
+    },
+    {
+      "epoch": 108.33,
+      "grad_norm": 3.5053226947784424,
+      "learning_rate": 3.3975e-06,
+      "loss": 0.9489,
+      "step": 9100
+    },
+    {
+      "epoch": 109.52,
+      "grad_norm": 2.178657054901123,
+      "learning_rate": 3.0225000000000003e-06,
+      "loss": 0.9218,
+      "step": 9200
+    },
+    {
+      "epoch": 110.71,
+      "grad_norm": 2.9659178256988525,
+      "learning_rate": 2.6475e-06,
+      "loss": 0.928,
+      "step": 9300
+    },
+    {
+      "epoch": 111.9,
+      "grad_norm": 3.755510091781616,
+      "learning_rate": 2.2725e-06,
+      "loss": 0.9244,
+      "step": 9400
+    },
+    {
+      "epoch": 113.1,
+      "grad_norm": 2.2474422454833984,
+      "learning_rate": 1.8975e-06,
+      "loss": 0.9269,
+      "step": 9500
+    },
+    {
+      "epoch": 114.29,
+      "grad_norm": 4.289571285247803,
+      "learning_rate": 1.5225000000000002e-06,
+      "loss": 0.9353,
+      "step": 9600
+    },
+    {
+      "epoch": 115.48,
+      "grad_norm": 3.7131989002227783,
+      "learning_rate": 1.1475e-06,
+      "loss": 0.9169,
+      "step": 9700
+    },
+    {
+      "epoch": 116.67,
+      "grad_norm": 2.692605972290039,
+      "learning_rate": 7.725e-07,
+      "loss": 0.9325,
+      "step": 9800
+    },
+    {
+      "epoch": 117.86,
+      "grad_norm": 3.175621509552002,
+      "learning_rate": 3.975e-07,
+      "loss": 0.9163,
+      "step": 9900
+    },
+    {
+      "epoch": 119.05,
+      "grad_norm": 9.366300582885742,
+      "learning_rate": 2.25e-08,
+      "loss": 0.9268,
+      "step": 10000
+    },
+    {
+      "epoch": 119.05,
+      "eval_cer": 0.13159224085673873,
+      "eval_loss": 0.3630259037017822,
+      "eval_runtime": 10.5083,
+      "eval_samples_per_second": 31.975,
+      "eval_steps_per_second": 3.997,
+      "eval_wer": 0.6242864556305138,
+      "step": 10000
     }
   ],
   "logging_steps": 100,
@@ -734,7 +814,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 120,
   "save_steps": 1000,
-  "total_flos": 2.7115316842745717e+19,
+  "total_flos": 3.012648370984383e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null