End of training

all_results.json

@@ -1,8 +1,15 @@
 {
     "epoch": 8.0,
-    "train_loss": 0.4185669529973201,
-    "train_runtime": 25964.0811,
-    "train_samples": 2271,
-    "train_samples_per_second": 0.7,
-    "train_steps_per_second": 0.044
+    "eval_accuracy": 0.5584178313693249,
+    "eval_loss": 3.250995397567749,
+    "eval_runtime": 67.5713,
+    "eval_samples": 123,
+    "eval_samples_per_second": 1.82,
+    "eval_steps_per_second": 0.237,
+    "perplexity": 25.816024339803242,
+    "train_loss": 0.424729940363429,
+    "train_runtime": 26678.4113,
+    "train_samples": 2272,
+    "train_samples_per_second": 0.681,
+    "train_steps_per_second": 0.043
 }

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.5584178313693249,
+    "eval_loss": 3.250995397567749,
+    "eval_runtime": 67.5713,
+    "eval_samples": 123,
+    "eval_samples_per_second": 1.82,
+    "eval_steps_per_second": 0.237,
+    "perplexity": 25.816024339803242
+}

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 8.0,
-    "train_loss": 0.4185669529973201,
-    "train_runtime": 25964.0811,
-    "train_samples": 2271,
-    "train_samples_per_second": 0.7,
-    "train_steps_per_second": 0.044
+    "train_loss": 0.424729940363429,
+    "train_runtime": 26678.4113,
+    "train_samples": 2272,
+    "train_samples_per_second": 0.681,
+    "train_steps_per_second": 0.043
 }

trainer_state.json

@@ -11,36 +11,144 @@
     {
       "epoch": 0.01,
       "learning_rate": 4.995598591549296e-05,
-      "loss": 2.3044,
+      "loss": 2.3251,
       "step": 1
     },
     {
-      "epoch": 3.52,
-      "learning_rate": 2.7992957746478874e-05,
-      "loss": 0.9153,
-      "step": 500
+      "epoch": 1.0,
+      "learning_rate": 4.379401408450705e-05,
+      "loss": 1.8881,
+      "step": 142
     },
     {
-      "epoch": 7.04,
-      "learning_rate": 6.0299295774647895e-06,
-      "loss": 0.0323,
-      "step": 1000
+      "epoch": 1.0,
+      "eval_accuracy": 0.5851831552023385,
+      "eval_loss": 1.8030234575271606,
+      "eval_runtime": 68.3081,
+      "eval_samples_per_second": 1.801,
+      "eval_steps_per_second": 0.234,
+      "step": 142
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.7544014084507044e-05,
+      "loss": 1.0326,
+      "step": 284
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5736850675785703,
+      "eval_loss": 1.9792582988739014,
+      "eval_runtime": 67.3896,
+      "eval_samples_per_second": 1.825,
+      "eval_steps_per_second": 0.237,
+      "step": 284
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.129401408450705e-05,
+      "loss": 0.2979,
+      "step": 426
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5582232177964184,
+      "eval_loss": 2.4750924110412598,
+      "eval_runtime": 67.4719,
+      "eval_samples_per_second": 1.823,
+      "eval_steps_per_second": 0.237,
+      "step": 426
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 2.5044014084507045e-05,
+      "loss": 0.0966,
+      "step": 568
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5602765895758616,
+      "eval_loss": 2.7331745624542236,
+      "eval_runtime": 67.3819,
+      "eval_samples_per_second": 1.825,
+      "eval_steps_per_second": 0.237,
+      "step": 568
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 1.8794014084507046e-05,
+      "loss": 0.0519,
+      "step": 710
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5593313236503151,
+      "eval_loss": 2.9102349281311035,
+      "eval_runtime": 67.3712,
+      "eval_samples_per_second": 1.826,
+      "eval_steps_per_second": 0.237,
+      "step": 710
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 1.2544014084507044e-05,
+      "loss": 0.0211,
+      "step": 852
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5565511297516492,
+      "eval_loss": 3.0963499546051025,
+      "eval_runtime": 67.3971,
+      "eval_samples_per_second": 1.825,
+      "eval_steps_per_second": 0.237,
+      "step": 852
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 6.2940140845070425e-06,
+      "loss": 0.0038,
+      "step": 994
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5585012371862849,
+      "eval_loss": 3.196666955947876,
+      "eval_runtime": 67.4214,
+      "eval_samples_per_second": 1.824,
+      "eval_steps_per_second": 0.237,
+      "step": 994
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.4014084507042253e-08,
+      "loss": 0.0028,
+      "step": 1136
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5584178313693249,
+      "eval_loss": 3.250995397567749,
+      "eval_runtime": 67.4231,
+      "eval_samples_per_second": 1.824,
+      "eval_steps_per_second": 0.237,
+      "step": 1136
     },
     {
       "epoch": 8.0,
       "step": 1136,
-      "total_flos": 2.844093826895708e+18,
-      "train_loss": 0.4185669529973201,
-      "train_runtime": 25964.0811,
-      "train_samples_per_second": 0.7,
-      "train_steps_per_second": 0.044
+      "total_flos": 2.845346179967877e+18,
+      "train_loss": 0.424729940363429,
+      "train_runtime": 26678.4113,
+      "train_samples_per_second": 0.681,
+      "train_steps_per_second": 0.043
     }
   ],
   "logging_steps": 500,
   "max_steps": 1136,
   "num_train_epochs": 8,
   "save_steps": 500,
-  "total_flos": 2.844093826895708e+18,
+  "total_flos": 2.845346179967877e+18,
   "trial_name": null,
   "trial_params": null
 }