End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.94,
     "total_flos": 1.1265748270391624e+18,
-    "train_loss": 0.2613685036960401,
-    "train_runtime": 863.1868,
-    "train_samples_per_second": 17.141,
-    "train_steps_per_second": 0.132
+    "train_loss": 0.16734615334293299,
+    "train_runtime": 848.5959,
+    "train_samples_per_second": 17.436,
+    "train_steps_per_second": 0.134
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.94,
     "total_flos": 1.1265748270391624e+18,
-    "train_loss": 0.2613685036960401,
-    "train_runtime": 863.1868,
-    "train_samples_per_second": 17.141,
-    "train_steps_per_second": 0.132
+    "train_loss": 0.16734615334293299,
+    "train_runtime": 848.5959,
+    "train_samples_per_second": 17.436,
+    "train_steps_per_second": 0.134
 }

trainer_state.json

@@ -11,125 +11,125 @@
     {
       "epoch": 0.03,
       "learning_rate": 4.166666666666667e-06,
-      "loss": 0.3991,
+      "loss": 0.1438,
       "step": 1
     },
     {
       "epoch": 0.31,
       "learning_rate": 5e-05,
-      "loss": 0.3808,
+      "loss": 0.146,
       "step": 12
     },
     {
       "epoch": 0.62,
       "learning_rate": 4.411764705882353e-05,
-      "loss": 0.3474,
+      "loss": 0.1822,
       "step": 24
     },
     {
       "epoch": 0.93,
       "learning_rate": 3.8235294117647055e-05,
-      "loss": 0.3093,
+      "loss": 0.1916,
       "step": 36
     },
     {
       "epoch": 0.98,
-      "eval_Macro F1": 0.8820015735129666,
-      "eval_Macro Precision": 0.899385909210207,
-      "eval_Macro Recall": 0.8738252833106378,
-      "eval_Micro F1": 0.8890637293469709,
-      "eval_Micro Precision": 0.8890637293469709,
-      "eval_Micro Recall": 0.8890637293469709,
-      "eval_Weighted F1": 0.8879137250311478,
-      "eval_Weighted Precision": 0.8952120340834557,
-      "eval_Weighted Recall": 0.8890637293469709,
-      "eval_accuracy": 0.8890637293469709,
-      "eval_loss": 0.2252058982849121,
-      "eval_runtime": 64.7982,
-      "eval_samples_per_second": 39.229,
-      "eval_steps_per_second": 1.235,
+      "eval_Macro F1": 0.9408451121862865,
+      "eval_Macro Precision": 0.9411664777049882,
+      "eval_Macro Recall": 0.9427118741264285,
+      "eval_Micro F1": 0.9461054287962234,
+      "eval_Micro Precision": 0.9461054287962234,
+      "eval_Micro Recall": 0.9461054287962234,
+      "eval_Weighted F1": 0.9464960091472664,
+      "eval_Weighted Precision": 0.9487006349374144,
+      "eval_Weighted Recall": 0.9461054287962234,
+      "eval_accuracy": 0.9461054287962234,
+      "eval_loss": 0.13961628079414368,
+      "eval_runtime": 63.2832,
+      "eval_samples_per_second": 40.169,
+      "eval_steps_per_second": 1.264,
       "step": 38
     },
     {
       "epoch": 1.24,
       "learning_rate": 3.235294117647059e-05,
-      "loss": 0.2358,
+      "loss": 0.1565,
       "step": 48
     },
     {
       "epoch": 1.55,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 0.2391,
+      "loss": 0.1692,
       "step": 60
     },
     {
       "epoch": 1.86,
       "learning_rate": 2.058823529411765e-05,
-      "loss": 0.2278,
+      "loss": 0.1597,
       "step": 72
     },
     {
       "epoch": 1.99,
-      "eval_Macro F1": 0.9220395057673447,
-      "eval_Macro Precision": 0.9240847889405631,
-      "eval_Macro Recall": 0.9220981981645411,
-      "eval_Micro F1": 0.9291896144767899,
-      "eval_Micro Precision": 0.9291896144767899,
-      "eval_Micro Recall": 0.9291896144767899,
-      "eval_Weighted F1": 0.929177891947533,
-      "eval_Weighted Precision": 0.9309640796953544,
-      "eval_Weighted Recall": 0.9291896144767899,
-      "eval_accuracy": 0.9291896144767899,
-      "eval_loss": 0.16484124958515167,
-      "eval_runtime": 64.4008,
-      "eval_samples_per_second": 39.472,
-      "eval_steps_per_second": 1.242,
+      "eval_Macro F1": 0.9484522652918805,
+      "eval_Macro Precision": 0.9471910825399321,
+      "eval_Macro Recall": 0.95148334257645,
+      "eval_Micro F1": 0.9520062942564909,
+      "eval_Micro Precision": 0.952006294256491,
+      "eval_Micro Recall": 0.952006294256491,
+      "eval_Weighted F1": 0.9522977862410118,
+      "eval_Weighted Precision": 0.9540997275393984,
+      "eval_Weighted Recall": 0.952006294256491,
+      "eval_accuracy": 0.952006294256491,
+      "eval_loss": 0.12265925854444504,
+      "eval_runtime": 64.9371,
+      "eval_samples_per_second": 39.146,
+      "eval_steps_per_second": 1.232,
       "step": 77
     },
     {
       "epoch": 2.17,
       "learning_rate": 1.4705882352941177e-05,
-      "loss": 0.1989,
+      "loss": 0.1619,
       "step": 84
     },
     {
       "epoch": 2.48,
       "learning_rate": 8.823529411764707e-06,
-      "loss": 0.2251,
+      "loss": 0.1611,
       "step": 96
     },
     {
       "epoch": 2.79,
       "learning_rate": 2.9411764705882355e-06,
-      "loss": 0.2066,
+      "loss": 0.1722,
       "step": 108
     },
     {
       "epoch": 2.94,
-      "eval_Macro F1": 0.9386216440513059,
-      "eval_Macro Precision": 0.9382250515595165,
-      "eval_Macro Recall": 0.9403801182311695,
-      "eval_Micro F1": 0.9425649095200629,
-      "eval_Micro Precision": 0.9425649095200629,
-      "eval_Micro Recall": 0.9425649095200629,
-      "eval_Weighted F1": 0.9426485199157074,
-      "eval_Weighted Precision": 0.9439679024820588,
-      "eval_Weighted Recall": 0.9425649095200629,
-      "eval_accuracy": 0.9425649095200629,
-      "eval_loss": 0.13585534691810608,
-      "eval_runtime": 63.8941,
-      "eval_samples_per_second": 39.785,
-      "eval_steps_per_second": 1.252,
+      "eval_Macro F1": 0.947301058449996,
+      "eval_Macro Precision": 0.9458287584858323,
+      "eval_Macro Recall": 0.9498167521420487,
+      "eval_Micro F1": 0.9512195121951219,
+      "eval_Micro Precision": 0.9512195121951219,
+      "eval_Micro Recall": 0.9512195121951219,
+      "eval_Weighted F1": 0.9515018097434871,
+      "eval_Weighted Precision": 0.9527033488356239,
+      "eval_Weighted Recall": 0.9512195121951219,
+      "eval_accuracy": 0.9512195121951219,
+      "eval_loss": 0.11822786182165146,
+      "eval_runtime": 63.9617,
+      "eval_samples_per_second": 39.743,
+      "eval_steps_per_second": 1.251,
       "step": 114
     },
     {
       "epoch": 2.94,
       "step": 114,
       "total_flos": 1.1265748270391624e+18,
-      "train_loss": 0.2613685036960401,
-      "train_runtime": 863.1868,
-      "train_samples_per_second": 17.141,
-      "train_steps_per_second": 0.132
+      "train_loss": 0.16734615334293299,
+      "train_runtime": 848.5959,
+      "train_samples_per_second": 17.436,
+      "train_steps_per_second": 0.134
     }
   ],
   "logging_steps": 12,