End of training

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 5.0,
+    "eval_bleu": 0.0065,
+    "eval_gen_len": 251.4632,
+    "eval_loss": 5.67488956451416,
+    "eval_runtime": 1450.6293,
+    "eval_samples": 1999,
+    "eval_samples_per_second": 1.378,
+    "eval_steps_per_second": 0.172,
+    "train_loss": 54.64480890625,
+    "train_runtime": 2612.7385,
+    "train_samples": 10000,
+    "train_samples_per_second": 19.137,
+    "train_steps_per_second": 4.784
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 5.0,
+    "eval_bleu": 0.0065,
+    "eval_gen_len": 251.4632,
+    "eval_loss": 5.67488956451416,
+    "eval_runtime": 1450.6293,
+    "eval_samples": 1999,
+    "eval_samples_per_second": 1.378,
+    "eval_steps_per_second": 0.172
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 54.64480890625,
+    "train_runtime": 2612.7385,
+    "train_samples": 10000,
+    "train_samples_per_second": 19.137,
+    "train_steps_per_second": 4.784
+}

trainer_state.json

@@ -0,0 +1,175 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "global_step": 12500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8e-05,
+      "loss": 244.4444,
+      "step": 500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 90.2438,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 67.008,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.2e-05,
+      "loss": 56.2936,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4e-05,
+      "loss": 50.8351,
+      "step": 2500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.8e-05,
+      "loss": 50.7529,
+      "step": 3000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.6e-05,
+      "loss": 48.0854,
+      "step": 3500
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 46.7528,
+      "step": 4000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 46.4214,
+      "step": 4500
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3e-05,
+      "loss": 42.6723,
+      "step": 5000
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 44.5044,
+      "step": 5500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 44.2767,
+      "step": 6000
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 2.4e-05,
+      "loss": 41.7274,
+      "step": 6500
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 43.7054,
+      "step": 7000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 2e-05,
+      "loss": 41.058,
+      "step": 7500
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 1.8e-05,
+      "loss": 41.684,
+      "step": 8000
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 41.4636,
+      "step": 8500
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 41.6401,
+      "step": 9000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 1.2e-05,
+      "loss": 40.2794,
+      "step": 9500
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 1e-05,
+      "loss": 41.8251,
+      "step": 10000
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 39.2594,
+      "step": 10500
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 6e-06,
+      "loss": 40.5295,
+      "step": 11000
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 39.3806,
+      "step": 11500
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 40.4578,
+      "step": 12000
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.0,
+      "loss": 40.8191,
+      "step": 12500
+    },
+    {
+      "epoch": 5.0,
+      "step": 12500,
+      "total_flos": 2522355797975040.0,
+      "train_loss": 54.64480890625,
+      "train_runtime": 2612.7385,
+      "train_samples_per_second": 19.137,
+      "train_steps_per_second": 4.784
+    }
+  ],
+  "max_steps": 12500,
+  "num_train_epochs": 5,
+  "total_flos": 2522355797975040.0,
+  "trial_name": null,
+  "trial_params": null
+}