End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- F1: 0.8154
-- Loss: 0.6651
+- Loss: 0.6536
+- F1: 0.8114
 
 ## Model description
 

all_results.json

@@ -2,14 +2,14 @@
     "epoch": 5.0,
     "eval_f1": 0.8113637968918935,
     "eval_loss": 0.6536160707473755,
-    "eval_runtime": 21.9128,
+    "eval_runtime": 21.923,
     "eval_samples": 8605,
-    "eval_samples_per_second": 392.692,
-    "eval_steps_per_second": 6.161,
+    "eval_samples_per_second": 392.509,
+    "eval_steps_per_second": 6.158,
     "total_flos": 1.603929743474688e+17,
-    "train_loss": 0.7819821552716223,
-    "train_runtime": 1776.2036,
+    "train_loss": 0.0,
+    "train_runtime": 0.2765,
     "train_samples": 34420,
-    "train_samples_per_second": 96.892,
-    "train_steps_per_second": 1.514
+    "train_samples_per_second": 622394.733,
+    "train_steps_per_second": 9728.308
 }

eval_results.json

@@ -2,8 +2,8 @@
     "epoch": 5.0,
     "eval_f1": 0.8113637968918935,
     "eval_loss": 0.6536160707473755,
-    "eval_runtime": 21.9128,
+    "eval_runtime": 21.923,
     "eval_samples": 8605,
-    "eval_samples_per_second": 392.692,
-    "eval_steps_per_second": 6.161
+    "eval_samples_per_second": 392.509,
+    "eval_steps_per_second": 6.158
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 5.0,
     "total_flos": 1.603929743474688e+17,
-    "train_loss": 0.7819821552716223,
-    "train_runtime": 1776.2036,
+    "train_loss": 0.0,
+    "train_runtime": 0.2765,
     "train_samples": 34420,
-    "train_samples_per_second": 96.892,
-    "train_steps_per_second": 1.514
+    "train_samples_per_second": 622394.733,
+    "train_steps_per_second": 9728.308
 }

trainer_state.json

@@ -92,10 +92,10 @@
       "epoch": 5.0,
       "step": 2690,
       "total_flos": 1.603929743474688e+17,
-      "train_loss": 0.7819821552716223,
-      "train_runtime": 1776.2036,
-      "train_samples_per_second": 96.892,
-      "train_steps_per_second": 1.514
+      "train_loss": 0.0,
+      "train_runtime": 0.2765,
+      "train_samples_per_second": 622394.733,
+      "train_steps_per_second": 9728.308
     }
   ],
   "logging_steps": 500,