Merge branch 'main' of https://github.com/carpentries-incubator/fair-…

…explainable-ml

episodes/5b-deep-dive-into-methods.md

@@ -116,8 +116,8 @@ information to make the prediction.
 
 Now, it's time to try implementing these methods yourself! Pick one of the following problems to work on:
 
-- [Train your own linear probe to check if BERT stores the required knowledge for sentiment analysis.](../code/5b-probes.ipynb)
-- [Use GradCAM on a trained model to check if the model is using the right features to make predictions.](../code/5b-gradcam.ipynb)
+- [Train your own linear probe to check if BERT stores the required knowledge for sentiment analysis.](https://carpentries-incubator.github.io/fair-explainable-ml/5c-probes.html)
+- [Use GradCAM on a trained model to check if the model is using the right features to make predictions.](https://carpentries-incubator.github.io/fair-explainable-ml/5d-gradcam.html)
 
 It's time to get your hands dirty now. Good luck, and have fun!
 

episodes/7a-OOD-detection-output-based.md

@@ -66,7 +66,7 @@ Threshold-based methods are one of the simplest and most intuitive approaches fo
 
 ### Distance-based
 - Distance: This method calculates the distance of an instance from the distribution of training data features. If the distance is beyond a certain threshold, the instance is considered OOD.
-
+  
 # Example 1: Softmax scores
 ```python
 n_epochs = 10