cp 28000

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2097409e22b1fab03277ae466406bfb2f48efaa2987c211dc571d6aeb8354e54
+oid sha256:b0cbfe228f29e909f61ba70ddc910e6e4fa3b3e4fd586609bf658a29a5c3b944
 size 2490339591

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11e8cefb76774bf50b86c1bc9486e5ec95567c8f849a411556992285820275c2
+oid sha256:d94dfee73ef732a05a4ecf0c7d027d89711722bce6f408c40b54fd07a5866fa9
 size 1262065048

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e58b5205a60751338b8db83fc32e97ce49b5dbb4327655a27ca0de316c1b3f76
+oid sha256:01f8d3a80f7b9dd9e7c50e7b2de049b01c32890f1166f11413fd758673a399cb
 size 623

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 21.9735503560529,
-  "global_step": 21600,
+  "epoch": 28.484231943031535,
+  "global_step": 28000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -762,11 +762,235 @@
       "eval_samples_per_second": 7.866,
       "eval_wer": 0.3369164619164619,
       "step": 21600
+    },
+    {
+      "epoch": 22.38,
+      "learning_rate": 7.750948602966539e-05,
+      "loss": 0.0466,
+      "step": 22000
+    },
+    {
+      "epoch": 22.38,
+      "eval_loss": 0.5171410441398621,
+      "eval_runtime": 254.3593,
+      "eval_samples_per_second": 7.356,
+      "eval_wer": 0.34121621621621623,
+      "step": 22000
+    },
+    {
+      "epoch": 22.79,
+      "learning_rate": 7.33701276302173e-05,
+      "loss": 0.0456,
+      "step": 22400
+    },
+    {
+      "epoch": 22.79,
+      "eval_loss": 0.5072354674339294,
+      "eval_runtime": 240.0638,
+      "eval_samples_per_second": 7.794,
+      "eval_wer": 0.33507371007371006,
+      "step": 22400
+    },
+    {
+      "epoch": 23.19,
+      "learning_rate": 6.923076923076922e-05,
+      "loss": 0.0459,
+      "step": 22800
+    },
+    {
+      "epoch": 23.19,
+      "eval_loss": 0.521458625793457,
+      "eval_runtime": 457.6179,
+      "eval_samples_per_second": 4.089,
+      "eval_wer": 0.34170761670761673,
+      "step": 22800
+    },
+    {
+      "epoch": 23.6,
+      "learning_rate": 6.509141083132114e-05,
+      "loss": 0.0404,
+      "step": 23200
+    },
+    {
+      "epoch": 23.6,
+      "eval_loss": 0.5035232901573181,
+      "eval_runtime": 991.4067,
+      "eval_samples_per_second": 1.887,
+      "eval_wer": 0.33746928746928745,
+      "step": 23200
+    },
+    {
+      "epoch": 24.01,
+      "learning_rate": 6.095205243187306e-05,
+      "loss": 0.0426,
+      "step": 23600
+    },
+    {
+      "epoch": 24.01,
+      "eval_loss": 0.5209127068519592,
+      "eval_runtime": 809.3899,
+      "eval_samples_per_second": 2.312,
+      "eval_wer": 0.3356879606879607,
+      "step": 23600
+    },
+    {
+      "epoch": 24.42,
+      "learning_rate": 5.6812694032424966e-05,
+      "loss": 0.0407,
+      "step": 24000
+    },
+    {
+      "epoch": 24.42,
+      "eval_loss": 0.521318256855011,
+      "eval_runtime": 685.3958,
+      "eval_samples_per_second": 2.73,
+      "eval_wer": 0.3367936117936118,
+      "step": 24000
+    },
+    {
+      "epoch": 24.82,
+      "learning_rate": 5.2673335632976885e-05,
+      "loss": 0.0384,
+      "step": 24400
+    },
+    {
+      "epoch": 24.82,
+      "eval_loss": 0.533201277256012,
+      "eval_runtime": 995.2009,
+      "eval_samples_per_second": 1.88,
+      "eval_wer": 0.3352579852579853,
+      "step": 24400
+    },
+    {
+      "epoch": 25.23,
+      "learning_rate": 4.8533977233528796e-05,
+      "loss": 0.0381,
+      "step": 24800
+    },
+    {
+      "epoch": 25.23,
+      "eval_loss": 0.5235264897346497,
+      "eval_runtime": 297.7118,
+      "eval_samples_per_second": 6.285,
+      "eval_wer": 0.33175675675675675,
+      "step": 24800
+    },
+    {
+      "epoch": 25.64,
+      "learning_rate": 4.4394618834080715e-05,
+      "loss": 0.0392,
+      "step": 25200
+    },
+    {
+      "epoch": 25.64,
+      "eval_loss": 0.5155506134033203,
+      "eval_runtime": 236.6244,
+      "eval_samples_per_second": 7.907,
+      "eval_wer": 0.3305896805896806,
+      "step": 25200
+    },
+    {
+      "epoch": 26.04,
+      "learning_rate": 4.025526043463263e-05,
+      "loss": 0.0374,
+      "step": 25600
+    },
+    {
+      "epoch": 26.04,
+      "eval_loss": 0.52613765001297,
+      "eval_runtime": 238.4752,
+      "eval_samples_per_second": 7.846,
+      "eval_wer": 0.3332309582309582,
+      "step": 25600
+    },
+    {
+      "epoch": 26.45,
+      "learning_rate": 3.611590203518454e-05,
+      "loss": 0.0352,
+      "step": 26000
+    },
+    {
+      "epoch": 26.45,
+      "eval_loss": 0.5194851756095886,
+      "eval_runtime": 237.5911,
+      "eval_samples_per_second": 7.875,
+      "eval_wer": 0.33132678132678134,
+      "step": 26000
+    },
+    {
+      "epoch": 26.86,
+      "learning_rate": 3.197654363573646e-05,
+      "loss": 0.0357,
+      "step": 26400
+    },
+    {
+      "epoch": 26.86,
+      "eval_loss": 0.5337009429931641,
+      "eval_runtime": 236.9359,
+      "eval_samples_per_second": 7.897,
+      "eval_wer": 0.33218673218673217,
+      "step": 26400
+    },
+    {
+      "epoch": 27.26,
+      "learning_rate": 2.7837185236288373e-05,
+      "loss": 0.0342,
+      "step": 26800
+    },
+    {
+      "epoch": 27.26,
+      "eval_loss": 0.5221392512321472,
+      "eval_runtime": 237.1237,
+      "eval_samples_per_second": 7.89,
+      "eval_wer": 0.3324324324324324,
+      "step": 26800
+    },
+    {
+      "epoch": 27.67,
+      "learning_rate": 2.3697826836840288e-05,
+      "loss": 0.0346,
+      "step": 27200
+    },
+    {
+      "epoch": 27.67,
+      "eval_loss": 0.5248429775238037,
+      "eval_runtime": 238.7028,
+      "eval_samples_per_second": 7.838,
+      "eval_wer": 0.32843980343980345,
+      "step": 27200
+    },
+    {
+      "epoch": 28.08,
+      "learning_rate": 1.95584684373922e-05,
+      "loss": 0.0343,
+      "step": 27600
+    },
+    {
+      "epoch": 28.08,
+      "eval_loss": 0.5455237030982971,
+      "eval_runtime": 238.0797,
+      "eval_samples_per_second": 7.859,
+      "eval_wer": 0.3278869778869779,
+      "step": 27600
+    },
+    {
+      "epoch": 28.48,
+      "learning_rate": 1.541911003794412e-05,
+      "loss": 0.034,
+      "step": 28000
+    },
+    {
+      "epoch": 28.48,
+      "eval_loss": 0.5291683077812195,
+      "eval_runtime": 300.3419,
+      "eval_samples_per_second": 6.23,
+      "eval_wer": 0.3269041769041769,
+      "step": 28000
     }
   ],
   "max_steps": 29490,
   "num_train_epochs": 30,
-  "total_flos": 1.1986301431319814e+20,
+  "total_flos": 1.5535560057099262e+20,
   "trial_name": null,
   "trial_params": null
 }