update model card README.md

README.md

@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [aubmindlab/araelectra-base-discriminator](https://huggingface.co/aubmindlab/araelectra-base-discriminator) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4828
-- Macro F1: 0.7630
-- Accuracy: 0.7729
-- Precision: 0.7684
-- Recall: 0.7599
+- Loss: 0.5072
+- Macro F1: 0.7679
+- Accuracy: 0.7745
+- Precision: 0.7680
+- Recall: 0.7678
 
 ## Model description
 
@@ -54,9 +54,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Macro F1 | Accuracy | Precision | Recall |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------:|:------:|
-| 0.5269        | 1.0   | 1597 | 0.4987          | 0.7440   | 0.7569   | 0.7533    | 0.7401 |
-| 0.4341        | 2.0   | 3194 | 0.4828          | 0.7630   | 0.7729   | 0.7684    | 0.7599 |
-| 0.3385        | 3.0   | 4791 | 0.5134          | 0.7602   | 0.7648   | 0.7588    | 0.7631 |
+| 0.5316        | 1.0   | 1597 | 0.4992          | 0.7428   | 0.7564   | 0.7535    | 0.7386 |
+| 0.4349        | 2.0   | 3194 | 0.4832          | 0.7523   | 0.7677   | 0.7692    | 0.7470 |
+| 0.3423        | 3.0   | 4791 | 0.5072          | 0.7679   | 0.7745   | 0.7680    | 0.7678 |
+| 0.2826        | 4.0   | 6388 | 0.5890          | 0.7672   | 0.7752   | 0.7693    | 0.7656 |
 
 
 ### Framework versions