Upload 9 files

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fb9fa451e0fdfe61b20ff1a1ff5a502f2c4aed55f4bd2588a3818904242d2b7
+oid sha256:2e21a15fa16cb586a616463ebc418102cb4aefd538b58f7698a10b1432a7f5fd
 size 2490594117

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e87d88f08b88070c69b1f891d632fc2b2539758387567149ecdc8ea1fb5ea28c
+oid sha256:54d05b2a1a51e74b9072bacff21310f9405efe6ff7f507c9d5fbe6bda339b735
 size 1262168365

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ff58ba996d368fe2e3614680ac55c25f343a30115fdd797d09cc9e5021f2ceb
-size 14575
+oid sha256:a66ba13b823b6a1d64e1843e4568cedc137fc46eb9360bf666a96e44d298f3d6
+size 14639

scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5dc9eacfeb00bd0bfeb98934a2309be01be65b288e0d747bbfc423b32679169f
+oid sha256:0fbcebc8f5487b0c117b5dd47f2ea304af3eebf408d297118d9307e1223927e1
 size 557

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c78769c18f0eab807e3f886758f225b7a6742b2333513c189962e3b430c92dd5
+oid sha256:b89c369e4058822d208a19b0783e9b59dfa2d164296df935580354f6ff84ea14
 size 627

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 7.407407407407407,
-  "global_step": 600,
+  "epoch": 30.864197530864196,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -96,11 +96,296 @@
       "eval_steps_per_second": 2.493,
       "eval_wer": 0.2352548138249408,
       "step": 600
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 9.92583436341162e-05,
+      "loss": 0.0381,
+      "step": 700
+    },
+    {
+      "epoch": 8.64,
+      "eval_loss": 0.3883003294467926,
+      "eval_runtime": 164.0468,
+      "eval_samples_per_second": 19.824,
+      "eval_steps_per_second": 2.481,
+      "eval_wer": 0.23683411285913866,
+      "step": 700
+    },
+    {
+      "epoch": 9.88,
+      "learning_rate": 9.913473423980223e-05,
+      "loss": 0.0379,
+      "step": 800
+    },
+    {
+      "epoch": 9.88,
+      "eval_loss": 0.3676028549671173,
+      "eval_runtime": 174.6686,
+      "eval_samples_per_second": 18.618,
+      "eval_steps_per_second": 2.33,
+      "eval_wer": 0.2324606693798214,
+      "step": 800
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 9.901112484548825e-05,
+      "loss": 0.0364,
+      "step": 900
+    },
+    {
+      "epoch": 11.11,
+      "eval_loss": 0.37265580892562866,
+      "eval_runtime": 150.5621,
+      "eval_samples_per_second": 21.599,
+      "eval_steps_per_second": 2.703,
+      "eval_wer": 0.23331106116746644,
+      "step": 900
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 9.88875154511743e-05,
+      "loss": 0.0355,
+      "step": 1000
+    },
+    {
+      "epoch": 12.35,
+      "eval_loss": 0.3740839958190918,
+      "eval_runtime": 180.5633,
+      "eval_samples_per_second": 18.01,
+      "eval_steps_per_second": 2.254,
+      "eval_wer": 0.23282512300309785,
+      "step": 1000
+    },
+    {
+      "epoch": 13.58,
+      "learning_rate": 9.876390605686032e-05,
+      "loss": 0.0365,
+      "step": 1100
+    },
+    {
+      "epoch": 13.58,
+      "eval_loss": 0.37167179584503174,
+      "eval_runtime": 177.8674,
+      "eval_samples_per_second": 18.283,
+      "eval_steps_per_second": 2.288,
+      "eval_wer": 0.23580149425985544,
+      "step": 1100
+    },
+    {
+      "epoch": 14.81,
+      "learning_rate": 9.864029666254637e-05,
+      "loss": 0.0343,
+      "step": 1200
+    },
+    {
+      "epoch": 14.81,
+      "eval_loss": 0.3842860460281372,
+      "eval_runtime": 182.4399,
+      "eval_samples_per_second": 17.825,
+      "eval_steps_per_second": 2.231,
+      "eval_wer": 0.23580149425985544,
+      "step": 1200
+    },
+    {
+      "epoch": 16.05,
+      "learning_rate": 9.851668726823239e-05,
+      "loss": 0.0358,
+      "step": 1300
+    },
+    {
+      "epoch": 16.05,
+      "eval_loss": 0.3594246208667755,
+      "eval_runtime": 180.3747,
+      "eval_samples_per_second": 18.029,
+      "eval_steps_per_second": 2.256,
+      "eval_wer": 0.23616594788313186,
+      "step": 1300
+    },
+    {
+      "epoch": 17.28,
+      "learning_rate": 9.839307787391843e-05,
+      "loss": 0.0343,
+      "step": 1400
+    },
+    {
+      "epoch": 17.28,
+      "eval_loss": 0.4051465690135956,
+      "eval_runtime": 180.0644,
+      "eval_samples_per_second": 18.06,
+      "eval_steps_per_second": 2.26,
+      "eval_wer": 0.2351940715543947,
+      "step": 1400
+    },
+    {
+      "epoch": 18.52,
+      "learning_rate": 9.826946847960445e-05,
+      "loss": 0.0363,
+      "step": 1500
+    },
+    {
+      "epoch": 18.52,
+      "eval_loss": 0.3952256739139557,
+      "eval_runtime": 185.3946,
+      "eval_samples_per_second": 17.541,
+      "eval_steps_per_second": 2.195,
+      "eval_wer": 0.23962825730425805,
+      "step": 1500
+    },
+    {
+      "epoch": 19.75,
+      "learning_rate": 9.814585908529048e-05,
+      "loss": 0.0382,
+      "step": 1600
+    },
+    {
+      "epoch": 19.75,
+      "eval_loss": 0.3582073450088501,
+      "eval_runtime": 179.8373,
+      "eval_samples_per_second": 18.083,
+      "eval_steps_per_second": 2.263,
+      "eval_wer": 0.23756302010569155,
+      "step": 1600
+    },
+    {
+      "epoch": 20.99,
+      "learning_rate": 9.802224969097652e-05,
+      "loss": 0.0337,
+      "step": 1700
+    },
+    {
+      "epoch": 20.99,
+      "eval_loss": 0.3877179026603699,
+      "eval_runtime": 183.3138,
+      "eval_samples_per_second": 17.74,
+      "eval_steps_per_second": 2.22,
+      "eval_wer": 0.2347081333900261,
+      "step": 1700
+    },
+    {
+      "epoch": 22.22,
+      "learning_rate": 9.789864029666255e-05,
+      "loss": 0.0331,
+      "step": 1800
+    },
+    {
+      "epoch": 22.22,
+      "eval_loss": 0.3826364576816559,
+      "eval_runtime": 174.2395,
+      "eval_samples_per_second": 18.664,
+      "eval_steps_per_second": 2.336,
+      "eval_wer": 0.2383526696227905,
+      "step": 1800
+    },
+    {
+      "epoch": 23.46,
+      "learning_rate": 9.777503090234858e-05,
+      "loss": 0.0321,
+      "step": 1900
+    },
+    {
+      "epoch": 23.46,
+      "eval_loss": 0.3872096538543701,
+      "eval_runtime": 181.6396,
+      "eval_samples_per_second": 17.904,
+      "eval_steps_per_second": 2.241,
+      "eval_wer": 0.23847415416388265,
+      "step": 1900
+    },
+    {
+      "epoch": 24.69,
+      "learning_rate": 9.765142150803462e-05,
+      "loss": 0.0342,
+      "step": 2000
+    },
+    {
+      "epoch": 24.69,
+      "eval_loss": 0.4173298180103302,
+      "eval_runtime": 185.9384,
+      "eval_samples_per_second": 17.49,
+      "eval_steps_per_second": 2.189,
+      "eval_wer": 0.24266537083156167,
+      "step": 2000
+    },
+    {
+      "epoch": 25.93,
+      "learning_rate": 9.752781211372065e-05,
+      "loss": 0.0348,
+      "step": 2100
+    },
+    {
+      "epoch": 25.93,
+      "eval_loss": 0.36838769912719727,
+      "eval_runtime": 184.1892,
+      "eval_samples_per_second": 17.656,
+      "eval_steps_per_second": 2.21,
+      "eval_wer": 0.23671262831804654,
+      "step": 2100
+    },
+    {
+      "epoch": 27.16,
+      "learning_rate": 9.740420271940668e-05,
+      "loss": 0.0332,
+      "step": 2200
+    },
+    {
+      "epoch": 27.16,
+      "eval_loss": 0.3941015899181366,
+      "eval_runtime": 184.3624,
+      "eval_samples_per_second": 17.639,
+      "eval_steps_per_second": 2.208,
+      "eval_wer": 0.23233918483872928,
+      "step": 2200
+    },
+    {
+      "epoch": 28.4,
+      "learning_rate": 9.728059332509271e-05,
+      "loss": 0.0339,
+      "step": 2300
+    },
+    {
+      "epoch": 28.4,
+      "eval_loss": 0.3854130506515503,
+      "eval_runtime": 186.7325,
+      "eval_samples_per_second": 17.415,
+      "eval_steps_per_second": 2.18,
+      "eval_wer": 0.24254388629046955,
+      "step": 2300
+    },
+    {
+      "epoch": 29.63,
+      "learning_rate": 9.715698393077875e-05,
+      "loss": 0.0349,
+      "step": 2400
+    },
+    {
+      "epoch": 29.63,
+      "eval_loss": 0.39330288767814636,
+      "eval_runtime": 183.7314,
+      "eval_samples_per_second": 17.7,
+      "eval_steps_per_second": 2.215,
+      "eval_wer": 0.23774524691732976,
+      "step": 2400
+    },
+    {
+      "epoch": 30.86,
+      "learning_rate": 9.703337453646477e-05,
+      "loss": 0.0327,
+      "step": 2500
+    },
+    {
+      "epoch": 30.86,
+      "eval_loss": 0.3882978558540344,
+      "eval_runtime": 186.0182,
+      "eval_samples_per_second": 17.482,
+      "eval_steps_per_second": 2.188,
+      "eval_wer": 0.2347081333900261,
+      "step": 2500
     }
   ],
   "max_steps": 81000,
   "num_train_epochs": 1000,
-  "total_flos": 1.9144703855961313e+19,
+  "total_flos": 7.954928320372541e+19,
   "trial_name": null,
   "trial_params": null
 }