Upload 10 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e9c0e5ad3c8a1357f85a9b32c532975bcf4116898f4c1c49c3efbda8d1862b1
+oid sha256:fde7d702f16adb04894390615fe7257ea7e432cc8c618b145e21e24d96027064
 size 377851056

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6121eaa09f943ca12e654e1a00bb5d0dbb6e49a1b9a35853bcc2e7bf3633d103
+oid sha256:440455cfc382222c8f82d640f542694776d5509f7f69dc1794bb575bc34757b6
 size 722217338

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94e82ab90df7341c5bf49bcabc98cefd0b1da5d4cac342d278618b90b206c66e
+oid sha256:cd5bb1d1b8e8892cbeae3b90dba5cfe5683bab00ce5b9e82d9cf9347195d22c8
 size 377899102

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cffe3870e4d630a4512c9362733d36c622403fb58c078a982f82141b7a6268b1
+oid sha256:02d1956f4d55df5967abc0a1cd9710622902be7319dcd00020acbd5d7eca165b
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6eb7c5df2b3c9cf31177c00e1591edde33c48c0171755d909c15a8d97616987
+oid sha256:a30cfca4f11f8b61f9e23428a9d301331ae17ea3f0a4bfb786fddd335aaca9a2
 size 1064

trainer_state.json

@@ -1,156 +1,576 @@
 {
-  "best_metric": 0.16606183350086212,
-  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-1000",
-  "epoch": 14.553014553014552,
+  "best_metric": 0.0727572962641716,
+  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-10500",
+  "epoch": 30.0,
   "eval_steps": 500,
-  "global_step": 3500,
+  "global_step": 14430,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 2.079002079002079,
-      "grad_norm": 1.7982165813446045,
-      "learning_rate": 4.92e-05,
-      "loss": 0.3958,
+      "epoch": 1.0395010395010396,
+      "grad_norm": 2.8997974395751953,
+      "learning_rate": 5e-05,
+      "loss": 0.3549,
       "step": 500
     },
     {
-      "epoch": 2.079002079002079,
-      "eval_cer": 0.06114554303065596,
-      "eval_loss": 0.19113463163375854,
-      "eval_mer": 0.15210595123060308,
-      "eval_runtime": 33.5544,
-      "eval_samples_per_second": 45.806,
-      "eval_steps_per_second": 5.752,
-      "eval_wer": 0.15290554825438546,
-      "eval_wil": 0.26582028780709,
-      "eval_wip": 0.73417971219291,
+      "epoch": 1.0395010395010396,
+      "eval_cer": 0.04969298673090043,
+      "eval_loss": 0.1506405621767044,
+      "eval_mer": 0.1328236435214241,
+      "eval_runtime": 33.659,
+      "eval_samples_per_second": 45.664,
+      "eval_steps_per_second": 5.734,
+      "eval_wer": 0.13357048597165644,
+      "eval_wil": 0.23588589232550194,
+      "eval_wip": 0.764114107674498,
       "step": 500
     },
     {
-      "epoch": 4.158004158004158,
-      "grad_norm": 2.043370246887207,
-      "learning_rate": 4.638805970149254e-05,
-      "loss": 0.3142,
+      "epoch": 2.079002079002079,
+      "grad_norm": 3.4246842861175537,
+      "learning_rate": 4.820890165111271e-05,
+      "loss": 0.3256,
       "step": 1000
     },
     {
-      "epoch": 4.158004158004158,
-      "eval_cer": 0.05659775661023529,
-      "eval_loss": 0.16606183350086212,
-      "eval_mer": 0.14339279627315077,
-      "eval_runtime": 33.4574,
-      "eval_samples_per_second": 45.939,
-      "eval_steps_per_second": 5.769,
-      "eval_wer": 0.1442203302668419,
-      "eval_wil": 0.25109751597729535,
-      "eval_wip": 0.7489024840227047,
+      "epoch": 2.079002079002079,
+      "eval_cer": 0.04905250553459295,
+      "eval_loss": 0.14544856548309326,
+      "eval_mer": 0.13181039406053682,
+      "eval_runtime": 33.3911,
+      "eval_samples_per_second": 46.03,
+      "eval_steps_per_second": 5.78,
+      "eval_wer": 0.13242297320557692,
+      "eval_wil": 0.23395019111086257,
+      "eval_wip": 0.7660498088891374,
       "step": 1000
     },
     {
-      "epoch": 6.237006237006237,
-      "grad_norm": 0.8731489777565002,
-      "learning_rate": 4.2716417910447766e-05,
-      "loss": 0.291,
+      "epoch": 3.1185031185031185,
+      "grad_norm": 1.9387176036834717,
+      "learning_rate": 4.641780330222542e-05,
+      "loss": 0.3255,
       "step": 1500
     },
     {
-      "epoch": 6.237006237006237,
-      "eval_cer": 0.05638977857271606,
-      "eval_loss": 0.20163549482822418,
-      "eval_mer": 0.15495043864646235,
-      "eval_runtime": 33.2557,
-      "eval_samples_per_second": 46.218,
-      "eval_steps_per_second": 5.804,
-      "eval_wer": 0.15541969030341124,
-      "eval_wil": 0.26617091306366814,
-      "eval_wip": 0.7338290869363319,
+      "epoch": 3.1185031185031185,
+      "eval_cer": 0.04656019827070077,
+      "eval_loss": 0.13236825168132782,
+      "eval_mer": 0.12753556939603453,
+      "eval_runtime": 33.3664,
+      "eval_samples_per_second": 46.064,
+      "eval_steps_per_second": 5.784,
+      "eval_wer": 0.12806242469447474,
+      "eval_wil": 0.22709355878713633,
+      "eval_wip": 0.7729064412128637,
       "step": 1500
     },
     {
-      "epoch": 8.316008316008316,
-      "grad_norm": 0.0,
-      "learning_rate": 3.9044776119402986e-05,
-      "loss": 0.6461,
+      "epoch": 4.158004158004158,
+      "grad_norm": 1.5572136640548706,
+      "learning_rate": 4.4626704953338125e-05,
+      "loss": 0.2866,
       "step": 2000
     },
     {
-      "epoch": 8.316008316008316,
-      "eval_cer": 0.0646950348709843,
-      "eval_loss": 0.4501780867576599,
-      "eval_mer": 0.1943082012090795,
-      "eval_runtime": 33.3483,
-      "eval_samples_per_second": 46.089,
-      "eval_steps_per_second": 5.787,
-      "eval_wer": 0.19467459002342724,
-      "eval_wil": 0.3166664651076232,
-      "eval_wip": 0.6833335348923768,
+      "epoch": 4.158004158004158,
+      "eval_cer": 0.04639311621949012,
+      "eval_loss": 0.12307066470384598,
+      "eval_mer": 0.12514292247884748,
+      "eval_runtime": 33.3438,
+      "eval_samples_per_second": 46.096,
+      "eval_steps_per_second": 5.788,
+      "eval_wer": 0.12559527224740374,
+      "eval_wil": 0.22333348147081955,
+      "eval_wip": 0.7766665185291804,
       "step": 2000
     },
     {
-      "epoch": 10.395010395010395,
-      "grad_norm": 0.0,
-      "learning_rate": 3.535074626865672e-05,
-      "loss": 0.7433,
+      "epoch": 5.197505197505198,
+      "grad_norm": 2.1455116271972656,
+      "learning_rate": 4.283201722900216e-05,
+      "loss": 0.2922,
       "step": 2500
     },
     {
-      "epoch": 10.395010395010395,
-      "eval_cer": 0.06426521359344453,
-      "eval_loss": 0.45573943853378296,
-      "eval_mer": 0.1922331204379562,
-      "eval_runtime": 33.1458,
-      "eval_samples_per_second": 46.371,
-      "eval_steps_per_second": 5.823,
-      "eval_wer": 0.19261756471058797,
-      "eval_wil": 0.3138487894482087,
-      "eval_wip": 0.6861512105517913,
+      "epoch": 5.197505197505198,
+      "eval_cer": 0.04445774912630011,
+      "eval_loss": 0.1228480413556099,
+      "eval_mer": 0.12246998284734134,
+      "eval_runtime": 33.4324,
+      "eval_samples_per_second": 45.973,
+      "eval_steps_per_second": 5.773,
+      "eval_wer": 0.12289861724711687,
+      "eval_wil": 0.2199313920357091,
+      "eval_wip": 0.7800686079642909,
       "step": 2500
     },
     {
-      "epoch": 12.474012474012474,
-      "grad_norm": 0.0,
-      "learning_rate": 3.1671641791044777e-05,
-      "loss": 0.7427,
+      "epoch": 6.237006237006237,
+      "grad_norm": 1.340448260307312,
+      "learning_rate": 4.104091888011487e-05,
+      "loss": 0.2675,
       "step": 3000
     },
     {
-      "epoch": 12.474012474012474,
-      "eval_cer": 0.06422361798594069,
-      "eval_loss": 0.45559772849082947,
-      "eval_mer": 0.1921190693430657,
-      "eval_runtime": 33.7371,
-      "eval_samples_per_second": 45.558,
-      "eval_steps_per_second": 5.721,
-      "eval_wer": 0.19250328552654133,
-      "eval_wil": 0.31369609006868415,
-      "eval_wip": 0.6863039099313158,
+      "epoch": 6.237006237006237,
+      "eval_cer": 0.04433243758789212,
+      "eval_loss": 0.1106928363442421,
+      "eval_mer": 0.12154127601189114,
+      "eval_runtime": 33.3844,
+      "eval_samples_per_second": 46.039,
+      "eval_steps_per_second": 5.781,
+      "eval_wer": 0.12198060703425326,
+      "eval_wil": 0.21841637697011862,
+      "eval_wip": 0.7815836230298814,
       "step": 3000
     },
     {
-      "epoch": 14.553014553014552,
-      "grad_norm": 0.0,
-      "learning_rate": 2.7992537313432837e-05,
-      "loss": 0.7434,
+      "epoch": 7.276507276507276,
+      "grad_norm": 2.0157642364501953,
+      "learning_rate": 3.924982053122757e-05,
+      "loss": 0.2506,
       "step": 3500
     },
     {
-      "epoch": 14.553014553014552,
-      "eval_cer": 0.06425134839094325,
-      "eval_loss": 0.45543229579925537,
-      "eval_mer": 0.1922331204379562,
-      "eval_runtime": 33.1994,
-      "eval_samples_per_second": 46.296,
-      "eval_steps_per_second": 5.813,
-      "eval_wer": 0.19261756471058797,
-      "eval_wil": 0.3138487894482087,
-      "eval_wip": 0.6861512105517913,
+      "epoch": 7.276507276507276,
+      "eval_cer": 0.04295401066540427,
+      "eval_loss": 0.1074899286031723,
+      "eval_mer": 0.1207804096578556,
+      "eval_runtime": 33.3924,
+      "eval_samples_per_second": 46.028,
+      "eval_steps_per_second": 5.78,
+      "eval_wer": 0.12111997245969361,
+      "eval_wil": 0.2170950962978776,
+      "eval_wip": 0.7829049037021224,
       "step": 3500
+    },
+    {
+      "epoch": 8.316008316008316,
+      "grad_norm": 1.8766114711761475,
+      "learning_rate": 3.7458722182340276e-05,
+      "loss": 0.2537,
+      "step": 4000
+    },
+    {
+      "epoch": 8.316008316008316,
+      "eval_cer": 0.04140850169170577,
+      "eval_loss": 0.10360275208950043,
+      "eval_mer": 0.11801029159519726,
+      "eval_runtime": 33.396,
+      "eval_samples_per_second": 46.023,
+      "eval_steps_per_second": 5.779,
+      "eval_wer": 0.11842331745940674,
+      "eval_wil": 0.21248246079158462,
+      "eval_wip": 0.7875175392084154,
+      "step": 4000
+    },
+    {
+      "epoch": 9.355509355509355,
+      "grad_norm": 1.7823630571365356,
+      "learning_rate": 3.566403445800431e-05,
+      "loss": 0.2473,
+      "step": 4500
+    },
+    {
+      "epoch": 9.355509355509355,
+      "eval_cer": 0.04121357263196001,
+      "eval_loss": 0.10312958061695099,
+      "eval_mer": 0.11691058769723302,
+      "eval_runtime": 33.7868,
+      "eval_samples_per_second": 45.491,
+      "eval_steps_per_second": 5.712,
+      "eval_wer": 0.11733318033163119,
+      "eval_wil": 0.21024574000779772,
+      "eval_wip": 0.7897542599922023,
+      "step": 4500
+    },
+    {
+      "epoch": 10.395010395010395,
+      "grad_norm": 3.1602580547332764,
+      "learning_rate": 3.3869346733668345e-05,
+      "loss": 0.2455,
+      "step": 5000
+    },
+    {
+      "epoch": 10.395010395010395,
+      "eval_cer": 0.04086548502527116,
+      "eval_loss": 0.09843221306800842,
+      "eval_mer": 0.11552759508149843,
+      "eval_runtime": 33.4361,
+      "eval_samples_per_second": 45.968,
+      "eval_steps_per_second": 5.772,
+      "eval_wer": 0.11589878937403178,
+      "eval_wil": 0.20822140556647772,
+      "eval_wip": 0.7917785944335223,
+      "step": 5000
+    },
+    {
+      "epoch": 11.434511434511435,
+      "grad_norm": 1.959326148033142,
+      "learning_rate": 3.207465900933237e-05,
+      "loss": 0.2322,
+      "step": 5500
+    },
+    {
+      "epoch": 11.434511434511435,
+      "eval_cer": 0.03993261023934504,
+      "eval_loss": 0.09337513148784637,
+      "eval_mer": 0.11474472585901321,
+      "eval_runtime": 33.5875,
+      "eval_samples_per_second": 45.761,
+      "eval_steps_per_second": 5.746,
+      "eval_wer": 0.1151529060760801,
+      "eval_wil": 0.2070981915905874,
+      "eval_wip": 0.7929018084094126,
+      "step": 5500
+    },
+    {
+      "epoch": 12.474012474012474,
+      "grad_norm": 1.514904499053955,
+      "learning_rate": 3.027997128499641e-05,
+      "loss": 0.251,
+      "step": 6000
+    },
+    {
+      "epoch": 12.474012474012474,
+      "eval_cer": 0.03943136408571309,
+      "eval_loss": 0.09094855934381485,
+      "eval_mer": 0.11248355921541717,
+      "eval_runtime": 33.5995,
+      "eval_samples_per_second": 45.745,
+      "eval_steps_per_second": 5.744,
+      "eval_wer": 0.11285788054392105,
+      "eval_wil": 0.20317729770906068,
+      "eval_wip": 0.7968227022909393,
+      "step": 6000
+    },
+    {
+      "epoch": 13.513513513513514,
+      "grad_norm": 3.5788846015930176,
+      "learning_rate": 2.848528356066045e-05,
+      "loss": 0.2243,
+      "step": 6500
+    },
+    {
+      "epoch": 13.513513513513514,
+      "eval_cer": 0.03856810682112474,
+      "eval_loss": 0.09059835970401764,
+      "eval_mer": 0.11314599668059291,
+      "eval_runtime": 33.4711,
+      "eval_samples_per_second": 45.92,
+      "eval_steps_per_second": 5.766,
+      "eval_wer": 0.11343163692696082,
+      "eval_wil": 0.20495219335271586,
+      "eval_wip": 0.7950478066472841,
+      "step": 6500
+    },
+    {
+      "epoch": 14.553014553014552,
+      "grad_norm": 2.408891201019287,
+      "learning_rate": 2.6690595836324484e-05,
+      "loss": 0.2186,
+      "step": 7000
+    },
+    {
+      "epoch": 14.553014553014552,
+      "eval_cer": 0.0384288717784492,
+      "eval_loss": 0.08281438052654266,
+      "eval_mer": 0.11403860473108425,
+      "eval_runtime": 33.5039,
+      "eval_samples_per_second": 45.875,
+      "eval_steps_per_second": 5.761,
+      "eval_wer": 0.11423489586321647,
+      "eval_wil": 0.20612773409748175,
+      "eval_wip": 0.7938722659025182,
+      "step": 7000
+    },
+    {
+      "epoch": 15.592515592515593,
+      "grad_norm": 1.2376307249069214,
+      "learning_rate": 2.4903086862885857e-05,
+      "loss": 0.2146,
+      "step": 7500
+    },
+    {
+      "epoch": 15.592515592515593,
+      "eval_cer": 0.03816432519736567,
+      "eval_loss": 0.08768957108259201,
+      "eval_mer": 0.11135895676046671,
+      "eval_runtime": 33.5042,
+      "eval_samples_per_second": 45.875,
+      "eval_steps_per_second": 5.76,
+      "eval_wer": 0.11171036777784153,
+      "eval_wil": 0.20133862836387018,
+      "eval_wip": 0.7986613716361298,
+      "step": 7500
+    },
+    {
+      "epoch": 16.632016632016633,
+      "grad_norm": 1.0698959827423096,
+      "learning_rate": 2.3108399138549895e-05,
+      "loss": 0.2076,
+      "step": 8000
+    },
+    {
+      "epoch": 16.632016632016633,
+      "eval_cer": 0.03784408459921193,
+      "eval_loss": 0.08166921883821487,
+      "eval_mer": 0.11109204278931412,
+      "eval_runtime": 33.5384,
+      "eval_samples_per_second": 45.828,
+      "eval_steps_per_second": 5.755,
+      "eval_wer": 0.11142348958632164,
+      "eval_wil": 0.20099477953828748,
+      "eval_wip": 0.7990052204617125,
+      "step": 8000
+    },
+    {
+      "epoch": 17.671517671517673,
+      "grad_norm": 2.2200756072998047,
+      "learning_rate": 2.13173007896626e-05,
+      "loss": 0.2017,
+      "step": 8500
+    },
+    {
+      "epoch": 17.671517671517673,
+      "eval_cer": 0.03691120981328581,
+      "eval_loss": 0.07834411412477493,
+      "eval_mer": 0.11015056964561745,
+      "eval_runtime": 33.5299,
+      "eval_samples_per_second": 45.84,
+      "eval_steps_per_second": 5.756,
+      "eval_wer": 0.11039072809685008,
+      "eval_wil": 0.20021436062866027,
+      "eval_wip": 0.7997856393713397,
+      "step": 8500
+    },
+    {
+      "epoch": 18.71101871101871,
+      "grad_norm": 3.6737000942230225,
+      "learning_rate": 1.9526202440775307e-05,
+      "loss": 0.199,
+      "step": 9000
+    },
+    {
+      "epoch": 18.71101871101871,
+      "eval_cer": 0.037259297419974656,
+      "eval_loss": 0.07848495990037918,
+      "eval_mer": 0.1108757870635375,
+      "eval_runtime": 33.4848,
+      "eval_samples_per_second": 45.901,
+      "eval_steps_per_second": 5.764,
+      "eval_wer": 0.11113661139480177,
+      "eval_wil": 0.20059739778337193,
+      "eval_wip": 0.7994026022166281,
+      "step": 9000
+    },
+    {
+      "epoch": 19.75051975051975,
+      "grad_norm": 2.476435899734497,
+      "learning_rate": 1.773151471643934e-05,
+      "loss": 0.1983,
+      "step": 9500
+    },
+    {
+      "epoch": 19.75051975051975,
+      "eval_cer": 0.037482073488255524,
+      "eval_loss": 0.08044513314962387,
+      "eval_mer": 0.11039295315449293,
+      "eval_runtime": 33.5182,
+      "eval_samples_per_second": 45.856,
+      "eval_steps_per_second": 5.758,
+      "eval_wer": 0.11073498192667393,
+      "eval_wil": 0.19978514503968547,
+      "eval_wip": 0.8002148549603145,
+      "step": 9500
+    },
+    {
+      "epoch": 20.79002079002079,
+      "grad_norm": 1.4843103885650635,
+      "learning_rate": 1.5940416367552046e-05,
+      "loss": 0.1931,
+      "step": 10000
+    },
+    {
+      "epoch": 20.79002079002079,
+      "eval_cer": 0.03720360340290444,
+      "eval_loss": 0.07748907804489136,
+      "eval_mer": 0.11006811287276057,
+      "eval_runtime": 33.4846,
+      "eval_samples_per_second": 45.902,
+      "eval_steps_per_second": 5.764,
+      "eval_wer": 0.1103333524585461,
+      "eval_wil": 0.19979205874359685,
+      "eval_wip": 0.8002079412564032,
+      "step": 10000
+    },
+    {
+      "epoch": 21.82952182952183,
+      "grad_norm": 7.590743064880371,
+      "learning_rate": 1.4156496769562097e-05,
+      "loss": 0.1723,
+      "step": 10500
+    },
+    {
+      "epoch": 21.82952182952183,
+      "eval_cer": 0.03642388716392141,
+      "eval_loss": 0.0727572962641716,
+      "eval_mer": 0.10935531890530173,
+      "eval_runtime": 33.5546,
+      "eval_samples_per_second": 45.806,
+      "eval_steps_per_second": 5.752,
+      "eval_wer": 0.1095874691605944,
+      "eval_wil": 0.1990145874796534,
+      "eval_wip": 0.8009854125203466,
+      "step": 10500
+    },
+    {
+      "epoch": 22.86902286902287,
+      "grad_norm": 1.670258641242981,
+      "learning_rate": 1.2365398420674803e-05,
+      "loss": 0.1996,
+      "step": 11000
+    },
+    {
+      "epoch": 22.86902286902287,
+      "eval_cer": 0.03600618203589479,
+      "eval_loss": 0.07854931801557541,
+      "eval_mer": 0.10902427851580394,
+      "eval_runtime": 33.6003,
+      "eval_samples_per_second": 45.744,
+      "eval_steps_per_second": 5.744,
+      "eval_wer": 0.10924321533077055,
+      "eval_wil": 0.19832511876306813,
+      "eval_wip": 0.8016748812369319,
+      "step": 11000
+    },
+    {
+      "epoch": 23.908523908523907,
+      "grad_norm": 1.4487839937210083,
+      "learning_rate": 1.0577889447236182e-05,
+      "loss": 0.1753,
+      "step": 11500
+    },
+    {
+      "epoch": 23.908523908523907,
+      "eval_cer": 0.03652135169379429,
+      "eval_loss": 0.08054520934820175,
+      "eval_mer": 0.10793342103757936,
+      "eval_runtime": 33.5332,
+      "eval_samples_per_second": 45.835,
+      "eval_steps_per_second": 5.755,
+      "eval_wer": 0.10826782947960296,
+      "eval_wil": 0.19572523190378577,
+      "eval_wip": 0.8042747680962142,
+      "step": 11500
+    },
+    {
+      "epoch": 24.948024948024948,
+      "grad_norm": 1.3358678817749023,
+      "learning_rate": 8.783201722900215e-06,
+      "loss": 0.1781,
+      "step": 12000
+    },
+    {
+      "epoch": 24.948024948024948,
+      "eval_cer": 0.03600618203589479,
+      "eval_loss": 0.07418987154960632,
+      "eval_mer": 0.10750042908633217,
+      "eval_runtime": 33.4349,
+      "eval_samples_per_second": 45.97,
+      "eval_steps_per_second": 5.772,
+      "eval_wer": 0.10780882437317116,
+      "eval_wil": 0.19526624761364586,
+      "eval_wip": 0.8047337523863541,
+      "step": 12000
+    },
+    {
+      "epoch": 25.987525987525988,
+      "grad_norm": 2.8383772373199463,
+      "learning_rate": 6.98851399856425e-06,
+      "loss": 0.1828,
+      "step": 12500
+    },
+    {
+      "epoch": 25.987525987525988,
+      "eval_cer": 0.03568594143774105,
+      "eval_loss": 0.07407771795988083,
+      "eval_mer": 0.10735757228743201,
+      "eval_runtime": 33.4962,
+      "eval_samples_per_second": 45.886,
+      "eval_steps_per_second": 5.762,
+      "eval_wer": 0.10757932181995525,
+      "eval_wil": 0.19555581909035935,
+      "eval_wip": 0.8044441809096406,
+      "step": 12500
+    },
+    {
+      "epoch": 27.027027027027028,
+      "grad_norm": 2.1374740600585938,
+      "learning_rate": 5.197415649676956e-06,
+      "loss": 0.1871,
+      "step": 13000
+    },
+    {
+      "epoch": 27.027027027027028,
+      "eval_cer": 0.03571378844627616,
+      "eval_loss": 0.0738675594329834,
+      "eval_mer": 0.10779711472406686,
+      "eval_runtime": 33.5044,
+      "eval_samples_per_second": 45.875,
+      "eval_steps_per_second": 5.76,
+      "eval_wer": 0.10803832692638705,
+      "eval_wil": 0.1961644777961592,
+      "eval_wip": 0.8038355222038408,
+      "step": 13000
+    },
+    {
+      "epoch": 28.066528066528065,
+      "grad_norm": 1.8452061414718628,
+      "learning_rate": 3.409906676238335e-06,
+      "loss": 0.1686,
+      "step": 13500
+    },
+    {
+      "epoch": 28.066528066528065,
+      "eval_cer": 0.03556062989933306,
+      "eval_loss": 0.0737072303891182,
+      "eval_mer": 0.10779711472406686,
+      "eval_runtime": 33.5159,
+      "eval_samples_per_second": 45.859,
+      "eval_steps_per_second": 5.758,
+      "eval_wer": 0.10803832692638705,
+      "eval_wil": 0.1961644777961592,
+      "eval_wip": 0.8038355222038408,
+      "step": 13500
+    },
+    {
+      "epoch": 29.106029106029105,
+      "grad_norm": 2.0490431785583496,
+      "learning_rate": 1.615218951902369e-06,
+      "loss": 0.1801,
+      "step": 14000
+    },
+    {
+      "epoch": 29.106029106029105,
+      "eval_cer": 0.035518859386530405,
+      "eval_loss": 0.07359343022108078,
+      "eval_mer": 0.10764385914686515,
+      "eval_runtime": 33.4499,
+      "eval_samples_per_second": 45.949,
+      "eval_steps_per_second": 5.77,
+      "eval_wer": 0.10786620001147512,
+      "eval_wil": 0.19607173646103337,
+      "eval_wip": 0.8039282635389666,
+      "step": 14000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 7200,
+  "max_steps": 14430,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 500,
@@ -161,12 +581,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 4.3655496330524303e+18,
+  "total_flos": 9.233840952562297e+18,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd2d2b59a3b86de63e4efe3082cf79448ade9086b5f111247ea4ffa0d38f23ea
+oid sha256:f8aa0ba76070730e78a2e05d51cda9e43beced2b5572af45c731b10724fafd32
 size 5176