Update README.md

README.md

@@ -134,6 +134,8 @@ This graph displays the training loss for Level 3, where training stabilizes aft
 ![Level 3 Validation Loss](https://huggingface.co/dejanseo/ecommerce-taxonomy-classifier/resolve/main/training/metrics/level-3-val-loss.png)
 The validation loss for Level 3, demonstrating steady improvements as the model converges.
 
+## Level 4
+
 ### Level 4: Training Loss
 ![Level 4 Train Loss](https://huggingface.co/dejanseo/ecommerce-taxonomy-classifier/resolve/main/training/metrics/level-4-train-loss.png)
 The training loss for Level 4 is plotted here, showing the effects of high-dimensional input features at this level.
@@ -141,3 +143,17 @@ The training loss for Level 4 is plotted here, showing the effects of high-dimen
 ### Level 4: Validation Loss
 ![Level 4 Validation Loss](https://huggingface.co/dejanseo/ecommerce-taxonomy-classifier/resolve/main/training/metrics/level-4-val-loss.png)
 Finally, the validation loss for Level 4 is shown, where training seems to stabilize after a longer period.
+
+### Level 4: Validation Loss Per Epoch Table
+| Epoch | Average Training Loss |
+|-------|------------------------|
+| 1     | 5.2803                |
+| 2     | 2.8285                |
+| 3     | 1.5707                |
+| 4     | 0.8696                |
+| 5     | 0.5164                |
+| 6     | 0.3384                |
+| 7     | 0.2408                |
+| 8     | 0.1813                |
+| 9     | 0.1426                |
+