End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 125.0,
+    "eval_loss": 0.25108230113983154,
+    "eval_runtime": 55.1407,
+    "eval_samples": 1255,
+    "eval_samples_per_second": 22.76,
+    "eval_steps_per_second": 1.433,
+    "eval_wer": 0.2414934706639691,
+    "train_loss": 1.2395893284997275,
+    "train_runtime": 21849.1477,
+    "train_samples": 2750,
+    "train_samples_per_second": 15.733,
+    "train_steps_per_second": 0.492
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 125.0,
+    "eval_loss": 0.25108230113983154,
+    "eval_runtime": 55.1407,
+    "eval_samples": 1255,
+    "eval_samples_per_second": 22.76,
+    "eval_steps_per_second": 1.433,
+    "eval_wer": 0.2414934706639691
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 125.0,
+    "train_loss": 1.2395893284997275,
+    "train_runtime": 21849.1477,
+    "train_samples": 2750,
+    "train_samples_per_second": 15.733,
+    "train_steps_per_second": 0.492
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 125.0,
+  "global_step": 10750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 5.81,
+      "learning_rate": 1.7395e-05,
+      "loss": 7.7666,
+      "step": 500
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 3.4895e-05,
+      "loss": 3.0049,
+      "step": 1000
+    },
+    {
+      "epoch": 17.44,
+      "learning_rate": 5.2395e-05,
+      "loss": 2.3958,
+      "step": 1500
+    },
+    {
+      "epoch": 17.44,
+      "eval_loss": 0.6807795166969299,
+      "eval_runtime": 55.2294,
+      "eval_samples_per_second": 22.723,
+      "eval_steps_per_second": 1.43,
+      "eval_wer": 0.6521059407761634,
+      "step": 1500
+    },
+    {
+      "epoch": 23.26,
+      "learning_rate": 6.9895e-05,
+      "loss": 1.2466,
+      "step": 2000
+    },
+    {
+      "epoch": 29.07,
+      "learning_rate": 6.6024e-05,
+      "loss": 1.0646,
+      "step": 2500
+    },
+    {
+      "epoch": 34.88,
+      "learning_rate": 6.2024e-05,
+      "loss": 0.9663,
+      "step": 3000
+    },
+    {
+      "epoch": 34.88,
+      "eval_loss": 0.3022635877132416,
+      "eval_runtime": 55.9589,
+      "eval_samples_per_second": 22.427,
+      "eval_steps_per_second": 1.412,
+      "eval_wer": 0.3718043038440316,
+      "step": 3000
+    },
+    {
+      "epoch": 40.7,
+      "learning_rate": 5.802399999999999e-05,
+      "loss": 0.9028,
+      "step": 3500
+    },
+    {
+      "epoch": 46.51,
+      "learning_rate": 5.4023999999999996e-05,
+      "loss": 0.8487,
+      "step": 4000
+    },
+    {
+      "epoch": 52.33,
+      "learning_rate": 5.003199999999999e-05,
+      "loss": 0.7963,
+      "step": 4500
+    },
+    {
+      "epoch": 52.33,
+      "eval_loss": 0.2588391900062561,
+      "eval_runtime": 55.1997,
+      "eval_samples_per_second": 22.736,
+      "eval_steps_per_second": 1.431,
+      "eval_wer": 0.3045797314695604,
+      "step": 4500
+    },
+    {
+      "epoch": 58.14,
+      "learning_rate": 4.6031999999999996e-05,
+      "loss": 0.7531,
+      "step": 5000
+    },
+    {
+      "epoch": 63.95,
+      "learning_rate": 4.2032e-05,
+      "loss": 0.7249,
+      "step": 5500
+    },
+    {
+      "epoch": 69.77,
+      "learning_rate": 3.8032e-05,
+      "loss": 0.6893,
+      "step": 6000
+    },
+    {
+      "epoch": 69.77,
+      "eval_loss": 0.2436135858297348,
+      "eval_runtime": 54.9892,
+      "eval_samples_per_second": 22.823,
+      "eval_steps_per_second": 1.437,
+      "eval_wer": 0.27184108883575503,
+      "step": 6000
+    },
+    {
+      "epoch": 75.58,
+      "learning_rate": 3.403999999999999e-05,
+      "loss": 0.6605,
+      "step": 6500
+    },
+    {
+      "epoch": 81.4,
+      "learning_rate": 3.0039999999999997e-05,
+      "loss": 0.6321,
+      "step": 7000
+    },
+    {
+      "epoch": 87.21,
+      "learning_rate": 2.6039999999999998e-05,
+      "loss": 0.6148,
+      "step": 7500
+    },
+    {
+      "epoch": 87.21,
+      "eval_loss": 0.2521224319934845,
+      "eval_runtime": 55.2601,
+      "eval_samples_per_second": 22.711,
+      "eval_steps_per_second": 1.43,
+      "eval_wer": 0.2572190546257127,
+      "step": 7500
+    },
+    {
+      "epoch": 93.02,
+      "learning_rate": 2.2039999999999996e-05,
+      "loss": 0.5887,
+      "step": 8000
+    },
+    {
+      "epoch": 98.84,
+      "learning_rate": 1.804e-05,
+      "loss": 0.574,
+      "step": 8500
+    },
+    {
+      "epoch": 104.65,
+      "learning_rate": 1.4047999999999998e-05,
+      "loss": 0.5556,
+      "step": 9000
+    },
+    {
+      "epoch": 104.65,
+      "eval_loss": 0.24900510907173157,
+      "eval_runtime": 54.8137,
+      "eval_samples_per_second": 22.896,
+      "eval_steps_per_second": 1.441,
+      "eval_wer": 0.24416038256391392,
+      "step": 9000
+    },
+    {
+      "epoch": 110.47,
+      "learning_rate": 1.0048e-05,
+      "loss": 0.5426,
+      "step": 9500
+    },
+    {
+      "epoch": 116.28,
+      "learning_rate": 6.0479999999999995e-06,
+      "loss": 0.5363,
+      "step": 10000
+    },
+    {
+      "epoch": 122.09,
+      "learning_rate": 2.0479999999999997e-06,
+      "loss": 0.5258,
+      "step": 10500
+    },
+    {
+      "epoch": 122.09,
+      "eval_loss": 0.2514589726924896,
+      "eval_runtime": 54.369,
+      "eval_samples_per_second": 23.083,
+      "eval_steps_per_second": 1.453,
+      "eval_wer": 0.24416038256391392,
+      "step": 10500
+    },
+    {
+      "epoch": 125.0,
+      "step": 10750,
+      "total_flos": 6.008825431704748e+19,
+      "train_loss": 1.2395893284997275,
+      "train_runtime": 21849.1477,
+      "train_samples_per_second": 15.733,
+      "train_steps_per_second": 0.492
+    }
+  ],
+  "max_steps": 10750,
+  "num_train_epochs": 125,
+  "total_flos": 6.008825431704748e+19,
+  "trial_name": null,
+  "trial_params": null
+}