Upload 9 files

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "checkpoint-1200",
+  "_name_or_path": "othrif",
   "activation_dropout": 0.1,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd2f774f6cf1ca2bd300966262722c8cde7aab481aed5e40f8b70664af4d8d6c
+oid sha256:4c79e728d98168d85015d00930e5f1bd407f25c2d89b2d55d9c9bb2f99ce3eee
 size 2490594117

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d1176a167f49d4d997324e388a10a7e83e555a32144196cecdac19812847823
+oid sha256:ccfbee3b3ae99465c34b8091981ba73de06eaab423aefbdb55c7872677393c70
 size 1262195949

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c438dbe3477c483067ae6907d5b0fdcc8f6e66c02f2ef7717fe65f4b5b65eda
-size 14575
+oid sha256:ef516f13eae1058f7d3a5544d2d46b334d6f3f2c0af866334d159bdf2bf78524
+size 14639

scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:573c0e110dfe1e7a3fc11065b73453594c7e0321077f7f0120dfed0a0d70518c
+oid sha256:1cc746e056a378283285a9fcc1e3f23a267ec6b0193f2c4ba34347b78ae0c98f
 size 557

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b5d8f13321c8b4e468ca80ef66eaf9552edb236f9798a14f3c7466d6f8495a4
+oid sha256:6bf093631f76407e20836696a637f5ab6e0d337c65ba9151883e190682646544
 size 627

trainer_state.json

@@ -1,211 +1,886 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 26.395939086294415,
-  "global_step": 5200,
+  "epoch": 37.90553745928339,
+  "global_step": 5800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 2.03,
-      "learning_rate": 0.0002,
-      "loss": 0.2479,
+      "epoch": 0.65,
+      "learning_rate": 0.0001,
+      "loss": 1.0071,
+      "step": 100
+    },
+    {
+      "epoch": 0.65,
+      "eval_loss": 0.3728577196598053,
+      "eval_runtime": 169.6136,
+      "eval_samples_per_second": 19.48,
+      "eval_steps_per_second": 2.435,
+      "eval_wer": 0.5494842925753878,
+      "step": 100
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 9.918652891889694e-05,
+      "loss": 0.6823,
+      "step": 200
+    },
+    {
+      "epoch": 1.31,
+      "eval_loss": 0.33882805705070496,
+      "eval_runtime": 167.327,
+      "eval_samples_per_second": 19.746,
+      "eval_steps_per_second": 2.468,
+      "eval_wer": 0.5344461066057791,
+      "step": 200
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.837305783779387e-05,
+      "loss": 0.6063,
+      "step": 300
+    },
+    {
+      "epoch": 1.96,
+      "eval_loss": 0.3208909332752228,
+      "eval_runtime": 174.661,
+      "eval_samples_per_second": 18.917,
+      "eval_steps_per_second": 2.365,
+      "eval_wer": 0.5098811117234864,
+      "step": 300
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 9.755958675669081e-05,
+      "loss": 0.5326,
       "step": 400
     },
     {
-      "epoch": 2.03,
-      "eval_loss": 0.47397056221961975,
-      "eval_runtime": 121.9231,
-      "eval_samples_per_second": 17.47,
-      "eval_steps_per_second": 2.19,
-      "eval_wer": 0.5506920619898261,
+      "epoch": 2.61,
+      "eval_loss": 0.30963289737701416,
+      "eval_runtime": 177.329,
+      "eval_samples_per_second": 18.632,
+      "eval_steps_per_second": 2.329,
+      "eval_wer": 0.5025588536335721,
       "step": 400
     },
     {
-      "epoch": 4.06,
-      "learning_rate": 0.00018605221657613458,
-      "loss": 0.2306,
+      "epoch": 3.27,
+      "learning_rate": 9.674611567558773e-05,
+      "loss": 0.5074,
+      "step": 500
+    },
+    {
+      "epoch": 3.27,
+      "eval_loss": 0.3118290901184082,
+      "eval_runtime": 173.1863,
+      "eval_samples_per_second": 19.078,
+      "eval_steps_per_second": 2.385,
+      "eval_wer": 0.4959452011652626,
+      "step": 500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 9.593264459448467e-05,
+      "loss": 0.446,
+      "step": 600
+    },
+    {
+      "epoch": 3.92,
+      "eval_loss": 0.30445897579193115,
+      "eval_runtime": 173.9457,
+      "eval_samples_per_second": 18.994,
+      "eval_steps_per_second": 2.374,
+      "eval_wer": 0.4865758601684907,
+      "step": 600
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 9.51191735133816e-05,
+      "loss": 0.4283,
+      "step": 700
+    },
+    {
+      "epoch": 4.57,
+      "eval_loss": 0.30916285514831543,
+      "eval_runtime": 176.2446,
+      "eval_samples_per_second": 18.747,
+      "eval_steps_per_second": 2.343,
+      "eval_wer": 0.48366270372411624,
+      "step": 700
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 9.430570243227855e-05,
+      "loss": 0.41,
       "step": 800
     },
     {
-      "epoch": 4.06,
-      "eval_loss": 0.4647212624549866,
-      "eval_runtime": 123.4969,
-      "eval_samples_per_second": 17.247,
-      "eval_steps_per_second": 2.162,
-      "eval_wer": 0.5497456524310895,
+      "epoch": 5.23,
+      "eval_loss": 0.35956883430480957,
+      "eval_runtime": 169.7235,
+      "eval_samples_per_second": 19.467,
+      "eval_steps_per_second": 2.433,
+      "eval_wer": 0.4932682465947563,
       "step": 800
     },
     {
-      "epoch": 6.09,
-      "learning_rate": 0.00017206938847029964,
-      "loss": 0.201,
+      "epoch": 5.88,
+      "learning_rate": 9.349223135117547e-05,
+      "loss": 0.3802,
+      "step": 900
+    },
+    {
+      "epoch": 5.88,
+      "eval_loss": 0.3234783411026001,
+      "eval_runtime": 171.4522,
+      "eval_samples_per_second": 19.271,
+      "eval_steps_per_second": 2.409,
+      "eval_wer": 0.47752145500354304,
+      "step": 900
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 9.267876027007241e-05,
+      "loss": 0.3852,
+      "step": 1000
+    },
+    {
+      "epoch": 6.53,
+      "eval_loss": 0.32342973351478577,
+      "eval_runtime": 184.4459,
+      "eval_samples_per_second": 17.913,
+      "eval_steps_per_second": 2.239,
+      "eval_wer": 0.47815132666719157,
+      "step": 1000
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 9.186528918896934e-05,
+      "loss": 0.3539,
+      "step": 1100
+    },
+    {
+      "epoch": 7.19,
+      "eval_loss": 0.33684083819389343,
+      "eval_runtime": 170.8359,
+      "eval_samples_per_second": 19.34,
+      "eval_steps_per_second": 2.418,
+      "eval_wer": 0.4796472718683568,
+      "step": 1100
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 9.105181810786628e-05,
+      "loss": 0.3444,
       "step": 1200
     },
     {
-      "epoch": 6.09,
-      "eval_loss": 0.43976637721061707,
-      "eval_runtime": 125.3202,
-      "eval_samples_per_second": 16.996,
-      "eval_steps_per_second": 2.131,
-      "eval_wer": 0.5441854962735123,
+      "epoch": 7.84,
+      "eval_loss": 0.3268304169178009,
+      "eval_runtime": 170.984,
+      "eval_samples_per_second": 19.323,
+      "eval_steps_per_second": 2.415,
+      "eval_wer": 0.4732698212739154,
       "step": 1200
     },
     {
-      "epoch": 8.12,
-      "learning_rate": 0.0001580515156824952,
-      "loss": 0.2676,
+      "epoch": 8.5,
+      "learning_rate": 9.02383470267632e-05,
+      "loss": 0.336,
+      "step": 1300
+    },
+    {
+      "epoch": 8.5,
+      "eval_loss": 0.34285250306129456,
+      "eval_runtime": 171.7981,
+      "eval_samples_per_second": 19.232,
+      "eval_steps_per_second": 2.404,
+      "eval_wer": 0.479883473742225,
+      "step": 1300
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 8.942487594566014e-05,
+      "loss": 0.3041,
+      "step": 1400
+    },
+    {
+      "epoch": 9.15,
+      "eval_loss": 0.35453349351882935,
+      "eval_runtime": 172.0678,
+      "eval_samples_per_second": 19.202,
+      "eval_steps_per_second": 2.4,
+      "eval_wer": 0.46248326903393433,
+      "step": 1400
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 8.861140486455706e-05,
+      "loss": 0.3074,
+      "step": 1500
+    },
+    {
+      "epoch": 9.8,
+      "eval_loss": 0.3339354693889618,
+      "eval_runtime": 172.1803,
+      "eval_samples_per_second": 19.189,
+      "eval_steps_per_second": 2.399,
+      "eval_wer": 0.46319187465553896,
+      "step": 1500
+    },
+    {
+      "epoch": 10.46,
+      "learning_rate": 8.7797933783454e-05,
+      "loss": 0.2948,
       "step": 1600
     },
     {
-      "epoch": 8.12,
-      "eval_loss": 0.46590062975883484,
-      "eval_runtime": 127.6886,
-      "eval_samples_per_second": 16.681,
-      "eval_steps_per_second": 2.091,
-      "eval_wer": 0.5331834851532,
+      "epoch": 10.46,
+      "eval_loss": 0.34325212240219116,
+      "eval_runtime": 171.9876,
+      "eval_samples_per_second": 19.211,
+      "eval_steps_per_second": 2.401,
+      "eval_wer": 0.4646090858987481,
       "step": 1600
     },
     {
-      "epoch": 10.15,
-      "learning_rate": 0.00014403364289469075,
-      "loss": 0.2295,
+      "epoch": 11.11,
+      "learning_rate": 8.698446270235093e-05,
+      "loss": 0.2905,
+      "step": 1700
+    },
+    {
+      "epoch": 11.11,
+      "eval_loss": 0.34282687306404114,
+      "eval_runtime": 172.5719,
+      "eval_samples_per_second": 19.146,
+      "eval_steps_per_second": 2.393,
+      "eval_wer": 0.4641366821510117,
+      "step": 1700
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 8.617099162124787e-05,
+      "loss": 0.296,
+      "step": 1800
+    },
+    {
+      "epoch": 11.76,
+      "eval_loss": 0.35734105110168457,
+      "eval_runtime": 173.1126,
+      "eval_samples_per_second": 19.086,
+      "eval_steps_per_second": 2.386,
+      "eval_wer": 0.4665774348476498,
+      "step": 1800
+    },
+    {
+      "epoch": 12.42,
+      "learning_rate": 8.535752054014479e-05,
+      "loss": 0.2669,
+      "step": 1900
+    },
+    {
+      "epoch": 12.42,
+      "eval_loss": 0.34095147252082825,
+      "eval_runtime": 172.9162,
+      "eval_samples_per_second": 19.108,
+      "eval_steps_per_second": 2.388,
+      "eval_wer": 0.46224706716006614,
+      "step": 1900
+    },
+    {
+      "epoch": 13.07,
+      "learning_rate": 8.454404945904173e-05,
+      "loss": 0.2778,
       "step": 2000
     },
     {
-      "epoch": 10.15,
-      "eval_loss": 0.433741956949234,
-      "eval_runtime": 126.2574,
-      "eval_samples_per_second": 16.87,
-      "eval_steps_per_second": 2.115,
-      "eval_wer": 0.5302259552821483,
+      "epoch": 13.07,
+      "eval_loss": 0.3445983827114105,
+      "eval_runtime": 172.7429,
+      "eval_samples_per_second": 19.127,
+      "eval_steps_per_second": 2.391,
+      "eval_wer": 0.4621683332021101,
       "step": 2000
     },
     {
-      "epoch": 12.18,
-      "learning_rate": 0.00013001577010688628,
-      "loss": 0.2011,
+      "epoch": 13.72,
+      "learning_rate": 8.373057837793867e-05,
+      "loss": 0.2605,
+      "step": 2100
+    },
+    {
+      "epoch": 13.72,
+      "eval_loss": 0.364580363035202,
+      "eval_runtime": 173.3351,
+      "eval_samples_per_second": 19.061,
+      "eval_steps_per_second": 2.383,
+      "eval_wer": 0.4611447917486812,
+      "step": 2100
+    },
+    {
+      "epoch": 14.38,
+      "learning_rate": 8.291710729683561e-05,
+      "loss": 0.2562,
+      "step": 2200
+    },
+    {
+      "epoch": 14.38,
+      "eval_loss": 0.3529307544231415,
+      "eval_runtime": 173.4538,
+      "eval_samples_per_second": 19.048,
+      "eval_steps_per_second": 2.381,
+      "eval_wer": 0.46201086528619795,
+      "step": 2200
+    },
+    {
+      "epoch": 15.03,
+      "learning_rate": 8.210363621573253e-05,
+      "loss": 0.2587,
+      "step": 2300
+    },
+    {
+      "epoch": 15.03,
+      "eval_loss": 0.35722818970680237,
+      "eval_runtime": 173.1723,
+      "eval_samples_per_second": 19.079,
+      "eval_steps_per_second": 2.385,
+      "eval_wer": 0.4694118573340682,
+      "step": 2300
+    },
+    {
+      "epoch": 15.68,
+      "learning_rate": 8.129016513462947e-05,
+      "loss": 0.242,
       "step": 2400
     },
     {
-      "epoch": 12.18,
-      "eval_loss": 0.43763187527656555,
-      "eval_runtime": 127.2021,
-      "eval_samples_per_second": 16.745,
-      "eval_steps_per_second": 2.099,
-      "eval_wer": 0.5218265704483616,
+      "epoch": 15.68,
+      "eval_loss": 0.36534029245376587,
+      "eval_runtime": 173.1065,
+      "eval_samples_per_second": 19.087,
+      "eval_steps_per_second": 2.386,
+      "eval_wer": 0.45894024092591135,
       "step": 2400
     },
     {
-      "epoch": 14.21,
-      "learning_rate": 0.00011599789731908184,
-      "loss": 0.1883,
+      "epoch": 16.34,
+      "learning_rate": 8.047669405352641e-05,
+      "loss": 0.232,
+      "step": 2500
+    },
+    {
+      "epoch": 16.34,
+      "eval_loss": 0.34964719414711,
+      "eval_runtime": 174.2382,
+      "eval_samples_per_second": 18.963,
+      "eval_steps_per_second": 2.37,
+      "eval_wer": 0.4605149200850327,
+      "step": 2500
+    },
+    {
+      "epoch": 16.99,
+      "learning_rate": 7.966322297242333e-05,
+      "loss": 0.2474,
+      "step": 2600
+    },
+    {
+      "epoch": 16.99,
+      "eval_loss": 0.3596344590187073,
+      "eval_runtime": 174.0298,
+      "eval_samples_per_second": 18.985,
+      "eval_steps_per_second": 2.373,
+      "eval_wer": 0.46783717817494685,
+      "step": 2600
+    },
+    {
+      "epoch": 17.64,
+      "learning_rate": 7.884975189132027e-05,
+      "loss": 0.2137,
+      "step": 2700
+    },
+    {
+      "epoch": 17.64,
+      "eval_loss": 0.3547351360321045,
+      "eval_runtime": 174.6108,
+      "eval_samples_per_second": 18.922,
+      "eval_steps_per_second": 2.365,
+      "eval_wer": 0.4609873238327691,
+      "step": 2700
+    },
+    {
+      "epoch": 18.3,
+      "learning_rate": 7.80362808102172e-05,
+      "loss": 0.2261,
       "step": 2800
     },
     {
-      "epoch": 14.21,
-      "eval_loss": 0.4427320659160614,
-      "eval_runtime": 133.553,
-      "eval_samples_per_second": 15.949,
-      "eval_steps_per_second": 1.999,
-      "eval_wer": 0.5275050278007808,
+      "epoch": 18.3,
+      "eval_loss": 0.35713937878608704,
+      "eval_runtime": 173.8691,
+      "eval_samples_per_second": 19.003,
+      "eval_steps_per_second": 2.375,
+      "eval_wer": 0.4579954334304385,
       "step": 2800
     },
     {
-      "epoch": 16.24,
-      "learning_rate": 0.00010198002453127737,
-      "loss": 0.1599,
+      "epoch": 18.95,
+      "learning_rate": 7.723094443992517e-05,
+      "loss": 0.2141,
+      "step": 2900
+    },
+    {
+      "epoch": 18.95,
+      "eval_loss": 0.36411064863204956,
+      "eval_runtime": 174.3463,
+      "eval_samples_per_second": 18.951,
+      "eval_steps_per_second": 2.369,
+      "eval_wer": 0.45563341469175656,
+      "step": 2900
+    },
+    {
+      "epoch": 19.61,
+      "learning_rate": 7.64174733588221e-05,
+      "loss": 0.2201,
+      "step": 3000
+    },
+    {
+      "epoch": 19.61,
+      "eval_loss": 0.34566032886505127,
+      "eval_runtime": 173.9331,
+      "eval_samples_per_second": 18.996,
+      "eval_steps_per_second": 2.374,
+      "eval_wer": 0.45303519407920634,
+      "step": 3000
+    },
+    {
+      "epoch": 20.26,
+      "learning_rate": 7.560400227771903e-05,
+      "loss": 0.2243,
+      "step": 3100
+    },
+    {
+      "epoch": 20.26,
+      "eval_loss": 0.3523178994655609,
+      "eval_runtime": 174.1671,
+      "eval_samples_per_second": 18.97,
+      "eval_steps_per_second": 2.371,
+      "eval_wer": 0.4571293598929218,
+      "step": 3100
+    },
+    {
+      "epoch": 20.91,
+      "learning_rate": 7.479053119661597e-05,
+      "loss": 0.1891,
       "step": 3200
     },
     {
-      "epoch": 16.24,
-      "eval_loss": 0.4528989791870117,
-      "eval_runtime": 130.4751,
-      "eval_samples_per_second": 16.325,
-      "eval_steps_per_second": 2.046,
-      "eval_wer": 0.5167396190701526,
+      "epoch": 20.91,
+      "eval_loss": 0.337533563375473,
+      "eval_runtime": 174.612,
+      "eval_samples_per_second": 18.922,
+      "eval_steps_per_second": 2.365,
+      "eval_wer": 0.4541374694905913,
       "step": 3200
     },
     {
-      "epoch": 18.27,
-      "learning_rate": 8.796215174347293e-05,
-      "loss": 0.1473,
+      "epoch": 21.57,
+      "learning_rate": 7.39770601155129e-05,
+      "loss": 0.2033,
+      "step": 3300
+    },
+    {
+      "epoch": 21.57,
+      "eval_loss": 0.3634466230869293,
+      "eval_runtime": 174.6521,
+      "eval_samples_per_second": 18.918,
+      "eval_steps_per_second": 2.365,
+      "eval_wer": 0.4579166994724825,
+      "step": 3300
+    },
+    {
+      "epoch": 22.22,
+      "learning_rate": 7.316358903440983e-05,
+      "loss": 0.2035,
+      "step": 3400
+    },
+    {
+      "epoch": 22.22,
+      "eval_loss": 0.3793589174747467,
+      "eval_runtime": 174.394,
+      "eval_samples_per_second": 18.946,
+      "eval_steps_per_second": 2.368,
+      "eval_wer": 0.4555546807338005,
+      "step": 3400
+    },
+    {
+      "epoch": 22.87,
+      "learning_rate": 7.235011795330676e-05,
+      "loss": 0.1867,
+      "step": 3500
+    },
+    {
+      "epoch": 22.87,
+      "eval_loss": 0.37910905480384827,
+      "eval_runtime": 174.9971,
+      "eval_samples_per_second": 18.88,
+      "eval_steps_per_second": 2.36,
+      "eval_wer": 0.454924809070152,
+      "step": 3500
+    },
+    {
+      "epoch": 23.53,
+      "learning_rate": 7.15366468722037e-05,
+      "loss": 0.1956,
       "step": 3600
     },
     {
-      "epoch": 18.27,
-      "eval_loss": 0.46878582239151,
-      "eval_runtime": 130.7961,
-      "eval_samples_per_second": 16.285,
-      "eval_steps_per_second": 2.041,
-      "eval_wer": 0.5125990772506802,
+      "epoch": 23.53,
+      "eval_loss": 0.3568515479564667,
+      "eval_runtime": 174.799,
+      "eval_samples_per_second": 18.902,
+      "eval_steps_per_second": 2.363,
+      "eval_wer": 0.45760176364065824,
       "step": 3600
     },
     {
-      "epoch": 20.3,
-      "learning_rate": 7.394427895566848e-05,
-      "loss": 0.1306,
+      "epoch": 24.18,
+      "learning_rate": 7.072317579110062e-05,
+      "loss": 0.1826,
+      "step": 3700
+    },
+    {
+      "epoch": 24.18,
+      "eval_loss": 0.3747410178184509,
+      "eval_runtime": 175.1918,
+      "eval_samples_per_second": 18.859,
+      "eval_steps_per_second": 2.357,
+      "eval_wer": 0.4543736713644595,
+      "step": 3700
+    },
+    {
+      "epoch": 24.83,
+      "learning_rate": 6.99178394208086e-05,
+      "loss": 0.1867,
+      "step": 3800
+    },
+    {
+      "epoch": 24.83,
+      "eval_loss": 0.36731651425361633,
+      "eval_runtime": 175.3726,
+      "eval_samples_per_second": 18.84,
+      "eval_steps_per_second": 2.355,
+      "eval_wer": 0.45366506574285487,
+      "step": 3800
+    },
+    {
+      "epoch": 25.49,
+      "learning_rate": 6.910436833970553e-05,
+      "loss": 0.1902,
+      "step": 3900
+    },
+    {
+      "epoch": 25.49,
+      "eval_loss": 0.3835786283016205,
+      "eval_runtime": 182.8434,
+      "eval_samples_per_second": 18.07,
+      "eval_steps_per_second": 2.259,
+      "eval_wer": 0.4522478544996457,
+      "step": 3900
+    },
+    {
+      "epoch": 26.14,
+      "learning_rate": 6.829089725860246e-05,
+      "loss": 0.1786,
       "step": 4000
     },
     {
-      "epoch": 20.3,
-      "eval_loss": 0.46525096893310547,
-      "eval_runtime": 130.9702,
-      "eval_samples_per_second": 16.263,
-      "eval_steps_per_second": 2.039,
-      "eval_wer": 0.5118892700816279,
+      "epoch": 26.14,
+      "eval_loss": 0.3528241813182831,
+      "eval_runtime": 182.8588,
+      "eval_samples_per_second": 18.069,
+      "eval_steps_per_second": 2.259,
+      "eval_wer": 0.4485473584757106,
       "step": 4000
     },
     {
-      "epoch": 22.34,
-      "learning_rate": 5.9926406167864026e-05,
-      "loss": 0.1222,
+      "epoch": 26.79,
+      "learning_rate": 6.74774261774994e-05,
+      "loss": 0.178,
+      "step": 4100
+    },
+    {
+      "epoch": 26.79,
+      "eval_loss": 0.3756342828273773,
+      "eval_runtime": 183.4843,
+      "eval_samples_per_second": 18.007,
+      "eval_steps_per_second": 2.251,
+      "eval_wer": 0.45303519407920634,
+      "step": 4100
+    },
+    {
+      "epoch": 27.45,
+      "learning_rate": 6.666395509639632e-05,
+      "loss": 0.1783,
+      "step": 4200
+    },
+    {
+      "epoch": 27.45,
+      "eval_loss": 0.38552403450012207,
+      "eval_runtime": 185.3761,
+      "eval_samples_per_second": 17.823,
+      "eval_steps_per_second": 2.228,
+      "eval_wer": 0.4515392488780411,
+      "step": 4200
+    },
+    {
+      "epoch": 28.1,
+      "learning_rate": 6.585048401529326e-05,
+      "loss": 0.1747,
+      "step": 4300
+    },
+    {
+      "epoch": 28.1,
+      "eval_loss": 0.3594723045825958,
+      "eval_runtime": 184.1815,
+      "eval_samples_per_second": 17.939,
+      "eval_steps_per_second": 2.242,
+      "eval_wer": 0.4475238170222817,
+      "step": 4300
+    },
+    {
+      "epoch": 28.76,
+      "learning_rate": 6.503701293419018e-05,
+      "loss": 0.1776,
       "step": 4400
     },
     {
-      "epoch": 22.34,
-      "eval_loss": 0.4552956223487854,
-      "eval_runtime": 132.9038,
-      "eval_samples_per_second": 16.027,
-      "eval_steps_per_second": 2.009,
-      "eval_wer": 0.5014787649355259,
+      "epoch": 28.76,
+      "eval_loss": 0.3899536728858948,
+      "eval_runtime": 183.8028,
+      "eval_samples_per_second": 17.976,
+      "eval_steps_per_second": 2.247,
+      "eval_wer": 0.45303519407920634,
       "step": 4400
     },
     {
-      "epoch": 24.37,
-      "learning_rate": 4.5908533380059577e-05,
-      "loss": 0.1097,
+      "epoch": 29.41,
+      "learning_rate": 6.422354185308712e-05,
+      "loss": 0.1615,
+      "step": 4500
+    },
+    {
+      "epoch": 29.41,
+      "eval_loss": 0.37925612926483154,
+      "eval_runtime": 184.3645,
+      "eval_samples_per_second": 17.921,
+      "eval_steps_per_second": 2.24,
+      "eval_wer": 0.4487048263916227,
+      "step": 4500
+    },
+    {
+      "epoch": 30.07,
+      "learning_rate": 6.341007077198405e-05,
+      "loss": 0.1665,
+      "step": 4600
+    },
+    {
+      "epoch": 30.07,
+      "eval_loss": 0.3769548237323761,
+      "eval_runtime": 185.5661,
+      "eval_samples_per_second": 17.805,
+      "eval_steps_per_second": 2.226,
+      "eval_wer": 0.4504369734666562,
+      "step": 4600
+    },
+    {
+      "epoch": 30.72,
+      "learning_rate": 6.2596599690881e-05,
+      "loss": 0.1562,
+      "step": 4700
+    },
+    {
+      "epoch": 30.72,
+      "eval_loss": 0.38725826144218445,
+      "eval_runtime": 184.2843,
+      "eval_samples_per_second": 17.929,
+      "eval_steps_per_second": 2.241,
+      "eval_wer": 0.45090937721439256,
+      "step": 4700
+    },
+    {
+      "epoch": 31.37,
+      "learning_rate": 6.178312860977793e-05,
+      "loss": 0.1558,
       "step": 4800
     },
     {
-      "epoch": 24.37,
-      "eval_loss": 0.4868864417076111,
-      "eval_runtime": 128.1956,
-      "eval_samples_per_second": 16.615,
-      "eval_steps_per_second": 2.083,
-      "eval_wer": 0.5012421625458418,
+      "epoch": 31.37,
+      "eval_loss": 0.37403690814971924,
+      "eval_runtime": 184.7842,
+      "eval_samples_per_second": 17.88,
+      "eval_steps_per_second": 2.235,
+      "eval_wer": 0.4494134320132273,
       "step": 4800
     },
     {
-      "epoch": 26.4,
-      "learning_rate": 3.189066059225513e-05,
-      "loss": 0.0988,
+      "epoch": 32.03,
+      "learning_rate": 6.0969657528674864e-05,
+      "loss": 0.1574,
+      "step": 4900
+    },
+    {
+      "epoch": 32.03,
+      "eval_loss": 0.38782382011413574,
+      "eval_runtime": 185.4497,
+      "eval_samples_per_second": 17.816,
+      "eval_steps_per_second": 2.227,
+      "eval_wer": 0.44933469805527126,
+      "step": 4900
+    },
+    {
+      "epoch": 32.68,
+      "learning_rate": 6.0156186447571796e-05,
+      "loss": 0.152,
+      "step": 5000
+    },
+    {
+      "epoch": 32.68,
+      "eval_loss": 0.36702463030815125,
+      "eval_runtime": 184.067,
+      "eval_samples_per_second": 17.95,
+      "eval_steps_per_second": 2.244,
+      "eval_wer": 0.44933469805527126,
+      "step": 5000
+    },
+    {
+      "epoch": 33.33,
+      "learning_rate": 5.934271536646873e-05,
+      "loss": 0.1477,
+      "step": 5100
+    },
+    {
+      "epoch": 33.33,
+      "eval_loss": 0.36524683237075806,
+      "eval_runtime": 186.7144,
+      "eval_samples_per_second": 17.695,
+      "eval_steps_per_second": 2.212,
+      "eval_wer": 0.4496496338870955,
+      "step": 5100
+    },
+    {
+      "epoch": 33.98,
+      "learning_rate": 5.852924428536566e-05,
+      "loss": 0.1561,
       "step": 5200
     },
     {
-      "epoch": 26.4,
-      "eval_loss": 0.45241138339042664,
-      "eval_runtime": 127.9868,
-      "eval_samples_per_second": 16.642,
-      "eval_steps_per_second": 2.086,
-      "eval_wer": 0.4968650183366852,
+      "epoch": 33.98,
+      "eval_loss": 0.3987789452075958,
+      "eval_runtime": 177.1487,
+      "eval_samples_per_second": 18.651,
+      "eval_steps_per_second": 2.331,
+      "eval_wer": 0.4535863317848988,
       "step": 5200
+    },
+    {
+      "epoch": 34.64,
+      "learning_rate": 5.771577320426259e-05,
+      "loss": 0.1441,
+      "step": 5300
+    },
+    {
+      "epoch": 34.64,
+      "eval_loss": 0.37290704250335693,
+      "eval_runtime": 178.4122,
+      "eval_samples_per_second": 18.519,
+      "eval_steps_per_second": 2.315,
+      "eval_wer": 0.4471301472325014,
+      "step": 5300
+    },
+    {
+      "epoch": 35.29,
+      "learning_rate": 5.691043683397056e-05,
+      "loss": 0.1462,
+      "step": 5400
+    },
+    {
+      "epoch": 35.29,
+      "eval_loss": 0.3913721740245819,
+      "eval_runtime": 175.0751,
+      "eval_samples_per_second": 18.872,
+      "eval_steps_per_second": 2.359,
+      "eval_wer": 0.4488622943075348,
+      "step": 5400
+    },
+    {
+      "epoch": 35.94,
+      "learning_rate": 5.6096965752867494e-05,
+      "loss": 0.1388,
+      "step": 5500
+    },
+    {
+      "epoch": 35.94,
+      "eval_loss": 0.3886808454990387,
+      "eval_runtime": 175.0289,
+      "eval_samples_per_second": 18.877,
+      "eval_steps_per_second": 2.36,
+      "eval_wer": 0.44807495472797415,
+      "step": 5500
+    },
+    {
+      "epoch": 36.6,
+      "learning_rate": 5.5283494671764426e-05,
+      "loss": 0.1362,
+      "step": 5600
+    },
+    {
+      "epoch": 36.6,
+      "eval_loss": 0.3816515803337097,
+      "eval_runtime": 175.1136,
+      "eval_samples_per_second": 18.868,
+      "eval_steps_per_second": 2.358,
+      "eval_wer": 0.445476734115424,
+      "step": 5600
+    },
+    {
+      "epoch": 37.25,
+      "learning_rate": 5.447002359066136e-05,
+      "loss": 0.1439,
+      "step": 5700
+    },
+    {
+      "epoch": 37.25,
+      "eval_loss": 0.39244014024734497,
+      "eval_runtime": 175.9891,
+      "eval_samples_per_second": 18.774,
+      "eval_steps_per_second": 2.347,
+      "eval_wer": 0.446657743484765,
+      "step": 5700
+    },
+    {
+      "epoch": 37.91,
+      "learning_rate": 5.365655250955829e-05,
+      "loss": 0.1299,
+      "step": 5800
+    },
+    {
+      "epoch": 37.91,
+      "eval_loss": 0.3693729341030121,
+      "eval_runtime": 176.2438,
+      "eval_samples_per_second": 18.747,
+      "eval_steps_per_second": 2.343,
+      "eval_wer": 0.4430359814187859,
+      "step": 5800
     }
   ],
-  "max_steps": 6107,
-  "num_train_epochs": 31,
-  "total_flos": 1.2361810216998281e+19,
+  "max_steps": 12393,
+  "num_train_epochs": 81,
+  "total_flos": 2.5038038866869117e+19,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:402a953d59f3d2c66918c47d238ae1d51eea2ac6120f7939459a514169d7a01a
+oid sha256:429d56e7f33237cdebb585dfadbe12372aa7b7c12ffbf8faf5185cef71f533cb
 size 3003