End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.3827837407588959,
+    "eval_runtime": 112.0874,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 14.694,
+    "eval_steps_per_second": 1.838,
+    "eval_wer": 0.34950464712491064,
+    "train_loss": 1.0725414783582774,
+    "train_runtime": 6951.8597,
+    "train_samples": 3478,
+    "train_samples_per_second": 7.504,
+    "train_steps_per_second": 0.235
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 15.0,
+    "eval_loss": 0.3827837407588959,
+    "eval_runtime": 112.0874,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 14.694,
+    "eval_steps_per_second": 1.838,
+    "eval_wer": 0.34950464712491064
+}

runs/Jul26_03-55-58_23c4b8f0470c/events.out.tfevents.1690351073.23c4b8f0470c.3028.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:247ac971f2d10d2ca8ddf224230886b5ff680cb15637859de2c549d1ec6fa5d8
+size 406

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "train_loss": 1.0725414783582774,
+    "train_runtime": 6951.8597,
+    "train_samples": 3478,
+    "train_samples_per_second": 7.504,
+    "train_steps_per_second": 0.235
+}

trainer_state.json

@@ -0,0 +1,187 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 15.0,
+  "global_step": 1635,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.92,
+      "eval_loss": 3.5826058387756348,
+      "eval_runtime": 112.5008,
+      "eval_samples_per_second": 14.64,
+      "eval_steps_per_second": 1.831,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 1.83,
+      "eval_loss": 3.0217819213867188,
+      "eval_runtime": 113.1278,
+      "eval_samples_per_second": 14.559,
+      "eval_steps_per_second": 1.821,
+      "eval_wer": 0.9998978653865795,
+      "step": 200
+    },
+    {
+      "epoch": 2.75,
+      "eval_loss": 0.8984940052032471,
+      "eval_runtime": 111.3066,
+      "eval_samples_per_second": 14.797,
+      "eval_steps_per_second": 1.851,
+      "eval_wer": 0.8035951383924012,
+      "step": 300
+    },
+    {
+      "epoch": 3.67,
+      "eval_loss": 0.5992483496665955,
+      "eval_runtime": 112.2673,
+      "eval_samples_per_second": 14.67,
+      "eval_steps_per_second": 1.835,
+      "eval_wer": 0.619650699622102,
+      "step": 400
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.00029939999999999996,
+      "loss": 3.1629,
+      "step": 500
+    },
+    {
+      "epoch": 4.59,
+      "eval_loss": 0.49684980511665344,
+      "eval_runtime": 112.6664,
+      "eval_samples_per_second": 14.618,
+      "eval_steps_per_second": 1.828,
+      "eval_wer": 0.5339597589623123,
+      "step": 500
+    },
+    {
+      "epoch": 5.5,
+      "eval_loss": 0.4645964503288269,
+      "eval_runtime": 111.3096,
+      "eval_samples_per_second": 14.797,
+      "eval_steps_per_second": 1.851,
+      "eval_wer": 0.5045449902972117,
+      "step": 600
+    },
+    {
+      "epoch": 6.42,
+      "eval_loss": 0.43159034848213196,
+      "eval_runtime": 113.1768,
+      "eval_samples_per_second": 14.552,
+      "eval_steps_per_second": 1.82,
+      "eval_wer": 0.44254927995097537,
+      "step": 700
+    },
+    {
+      "epoch": 7.34,
+      "eval_loss": 0.44998037815093994,
+      "eval_runtime": 111.8678,
+      "eval_samples_per_second": 14.723,
+      "eval_steps_per_second": 1.841,
+      "eval_wer": 0.4734960678173833,
+      "step": 800
+    },
+    {
+      "epoch": 8.26,
+      "eval_loss": 0.41139060258865356,
+      "eval_runtime": 113.568,
+      "eval_samples_per_second": 14.502,
+      "eval_steps_per_second": 1.814,
+      "eval_wer": 0.41231743437851087,
+      "step": 900
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 0.00016810572687224668,
+      "loss": 0.2226,
+      "step": 1000
+    },
+    {
+      "epoch": 9.17,
+      "eval_loss": 0.41622716188430786,
+      "eval_runtime": 112.362,
+      "eval_samples_per_second": 14.658,
+      "eval_steps_per_second": 1.833,
+      "eval_wer": 0.4018997038096211,
+      "step": 1000
+    },
+    {
+      "epoch": 10.09,
+      "eval_loss": 0.3999302089214325,
+      "eval_runtime": 113.8252,
+      "eval_samples_per_second": 14.47,
+      "eval_steps_per_second": 1.81,
+      "eval_wer": 0.38239199264630785,
+      "step": 1100
+    },
+    {
+      "epoch": 11.01,
+      "eval_loss": 0.404796838760376,
+      "eval_runtime": 112.9341,
+      "eval_samples_per_second": 14.584,
+      "eval_steps_per_second": 1.824,
+      "eval_wer": 0.38423041568787664,
+      "step": 1200
+    },
+    {
+      "epoch": 11.93,
+      "eval_loss": 0.3789042532444,
+      "eval_runtime": 112.9251,
+      "eval_samples_per_second": 14.585,
+      "eval_steps_per_second": 1.824,
+      "eval_wer": 0.3602287815340619,
+      "step": 1300
+    },
+    {
+      "epoch": 12.84,
+      "eval_loss": 0.4024040102958679,
+      "eval_runtime": 114.3571,
+      "eval_samples_per_second": 14.402,
+      "eval_steps_per_second": 1.801,
+      "eval_wer": 0.35359003166173014,
+      "step": 1400
+    },
+    {
+      "epoch": 13.76,
+      "learning_rate": 3.594713656387665e-05,
+      "loss": 0.1015,
+      "step": 1500
+    },
+    {
+      "epoch": 13.76,
+      "eval_loss": 0.38988158106803894,
+      "eval_runtime": 113.4392,
+      "eval_samples_per_second": 14.519,
+      "eval_steps_per_second": 1.816,
+      "eval_wer": 0.35747114697170873,
+      "step": 1500
+    },
+    {
+      "epoch": 14.68,
+      "eval_loss": 0.3801647126674652,
+      "eval_runtime": 112.3523,
+      "eval_samples_per_second": 14.659,
+      "eval_steps_per_second": 1.834,
+      "eval_wer": 0.3489939740578082,
+      "step": 1600
+    },
+    {
+      "epoch": 15.0,
+      "step": 1635,
+      "total_flos": 6.468063251673315e+18,
+      "train_loss": 1.0725414783582774,
+      "train_runtime": 6951.8597,
+      "train_samples_per_second": 7.504,
+      "train_steps_per_second": 0.235
+    }
+  ],
+  "max_steps": 1635,
+  "num_train_epochs": 15,
+  "total_flos": 6.468063251673315e+18,
+  "trial_name": null,
+  "trial_params": null
+}