End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 100.0,
+    "eval_loss": 0.9814496636390686,
+    "eval_runtime": 14.3225,
+    "eval_samples": 294,
+    "eval_samples_per_second": 20.527,
+    "eval_steps_per_second": 1.327,
+    "eval_wer": 0.7402486069438491,
+    "train_loss": 1.9762264404296874,
+    "train_runtime": 4294.1559,
+    "train_samples": 624,
+    "train_samples_per_second": 14.531,
+    "train_steps_per_second": 0.466
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 100.0,
+    "eval_loss": 0.9814496636390686,
+    "eval_runtime": 14.3225,
+    "eval_samples": 294,
+    "eval_samples_per_second": 20.527,
+    "eval_steps_per_second": 1.327,
+    "eval_wer": 0.7402486069438491
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 100.0,
+    "train_loss": 1.9762264404296874,
+    "train_runtime": 4294.1559,
+    "train_samples": 624,
+    "train_samples_per_second": 14.531,
+    "train_steps_per_second": 0.466
+}

trainer_state.json

@@ -0,0 +1,82 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 100.0,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 20.0,
+      "eval_loss": 3.14473295211792,
+      "eval_runtime": 14.4908,
+      "eval_samples_per_second": 20.289,
+      "eval_steps_per_second": 1.311,
+      "eval_wer": 1.0,
+      "step": 400
+    },
+    {
+      "epoch": 40.0,
+      "eval_loss": 1.0074288845062256,
+      "eval_runtime": 14.582,
+      "eval_samples_per_second": 20.162,
+      "eval_steps_per_second": 1.303,
+      "eval_wer": 0.8555507929704244,
+      "step": 800
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 0.00018806249999999997,
+      "loss": 3.1278,
+      "step": 1000
+    },
+    {
+      "epoch": 60.0,
+      "eval_loss": 0.9506617784500122,
+      "eval_runtime": 14.9576,
+      "eval_samples_per_second": 19.655,
+      "eval_steps_per_second": 1.27,
+      "eval_wer": 0.7711101585940848,
+      "step": 1200
+    },
+    {
+      "epoch": 80.0,
+      "eval_loss": 0.9729732871055603,
+      "eval_runtime": 14.438,
+      "eval_samples_per_second": 20.363,
+      "eval_steps_per_second": 1.316,
+      "eval_wer": 0.7629661380197171,
+      "step": 1600
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 5.624999999999999e-07,
+      "loss": 0.8247,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "eval_loss": 0.9814496636390686,
+      "eval_runtime": 14.2796,
+      "eval_samples_per_second": 20.589,
+      "eval_steps_per_second": 1.331,
+      "eval_wer": 0.7402486069438491,
+      "step": 2000
+    },
+    {
+      "epoch": 100.0,
+      "step": 2000,
+      "total_flos": 1.1371370438840648e+19,
+      "train_loss": 1.9762264404296874,
+      "train_runtime": 4294.1559,
+      "train_samples_per_second": 14.531,
+      "train_steps_per_second": 0.466
+    }
+  ],
+  "max_steps": 2000,
+  "num_train_epochs": 100,
+  "total_flos": 1.1371370438840648e+19,
+  "trial_name": null,
+  "trial_params": null
+}