End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 19.82,
+    "eval_loss": 0.5335429310798645,
+    "eval_runtime": 100.4145,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 16.402,
+    "eval_steps_per_second": 1.026,
+    "eval_wer": 0.493922990501481,
+    "train_loss": 1.770811665499652,
+    "train_runtime": 22430.6801,
+    "train_samples": 3478,
+    "train_samples_per_second": 3.101,
+    "train_steps_per_second": 0.048
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 19.82,
+    "eval_loss": 0.5335429310798645,
+    "eval_runtime": 100.4145,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 16.402,
+    "eval_steps_per_second": 1.026,
+    "eval_wer": 0.493922990501481
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 19.82,
+    "train_loss": 1.770811665499652,
+    "train_runtime": 22430.6801,
+    "train_samples": 3478,
+    "train_samples_per_second": 3.101,
+    "train_steps_per_second": 0.048
+}

trainer_state.json

@@ -0,0 +1,127 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 19.81651376146789,
+  "global_step": 1080,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.83,
+      "eval_loss": 4.108390808105469,
+      "eval_runtime": 103.6273,
+      "eval_samples_per_second": 15.893,
+      "eval_steps_per_second": 0.994,
+      "eval_wer": 1.0,
+      "step": 100
+    },
+    {
+      "epoch": 3.67,
+      "eval_loss": 3.1519322395324707,
+      "eval_runtime": 103.1635,
+      "eval_samples_per_second": 15.965,
+      "eval_steps_per_second": 0.998,
+      "eval_wer": 1.0,
+      "step": 200
+    },
+    {
+      "epoch": 5.5,
+      "eval_loss": 1.9347835779190063,
+      "eval_runtime": 102.3063,
+      "eval_samples_per_second": 16.099,
+      "eval_steps_per_second": 1.007,
+      "eval_wer": 0.9798794811561639,
+      "step": 300
+    },
+    {
+      "epoch": 7.34,
+      "eval_loss": 0.7185050249099731,
+      "eval_runtime": 102.0354,
+      "eval_samples_per_second": 16.141,
+      "eval_steps_per_second": 1.009,
+      "eval_wer": 0.74895312021244,
+      "step": 400
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 0.0001996,
+      "loss": 3.6165,
+      "step": 500
+    },
+    {
+      "epoch": 9.17,
+      "eval_loss": 0.6041333079338074,
+      "eval_runtime": 101.7041,
+      "eval_samples_per_second": 16.194,
+      "eval_steps_per_second": 1.013,
+      "eval_wer": 0.636809314676744,
+      "step": 500
+    },
+    {
+      "epoch": 11.01,
+      "eval_loss": 0.5609970092773438,
+      "eval_runtime": 102.5678,
+      "eval_samples_per_second": 16.058,
+      "eval_steps_per_second": 1.004,
+      "eval_wer": 0.5770605658257584,
+      "step": 600
+    },
+    {
+      "epoch": 12.84,
+      "eval_loss": 0.5291760563850403,
+      "eval_runtime": 101.4306,
+      "eval_samples_per_second": 16.238,
+      "eval_steps_per_second": 1.015,
+      "eval_wer": 0.5397814319272801,
+      "step": 700
+    },
+    {
+      "epoch": 14.68,
+      "eval_loss": 0.5242128968238831,
+      "eval_runtime": 100.8318,
+      "eval_samples_per_second": 16.334,
+      "eval_steps_per_second": 1.022,
+      "eval_wer": 0.5083239709937698,
+      "step": 800
+    },
+    {
+      "epoch": 16.51,
+      "eval_loss": 0.5442604422569275,
+      "eval_runtime": 101.3912,
+      "eval_samples_per_second": 16.244,
+      "eval_steps_per_second": 1.016,
+      "eval_wer": 0.5037279133898478,
+      "step": 900
+    },
+    {
+      "epoch": 18.35,
+      "learning_rate": 2.7931034482758622e-05,
+      "loss": 0.1894,
+      "step": 1000
+    },
+    {
+      "epoch": 18.35,
+      "eval_loss": 0.5313596725463867,
+      "eval_runtime": 102.9424,
+      "eval_samples_per_second": 15.999,
+      "eval_steps_per_second": 1.001,
+      "eval_wer": 0.49443366356858337,
+      "step": 1000
+    },
+    {
+      "epoch": 19.82,
+      "step": 1080,
+      "total_flos": 8.773864807508535e+18,
+      "train_loss": 1.770811665499652,
+      "train_runtime": 22430.6801,
+      "train_samples_per_second": 3.101,
+      "train_steps_per_second": 0.048
+    }
+  ],
+  "max_steps": 1080,
+  "num_train_epochs": 20,
+  "total_flos": 8.773864807508535e+18,
+  "trial_name": null,
+  "trial_params": null
+}