End of training

README.md

@@ -16,6 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
 # run_4
 
 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the wikitext dataset.
+It achieves the following results on the evaluation set:
+- Loss: 7.1830
 
 ## Model description
 
@@ -43,10 +45,16 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 1
+- num_epochs: 5
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 7.8982        | 1.1   | 50   | 7.2421          |
+| 7.2129        | 2.19  | 100  | 7.2225          |
+| 7.1869        | 3.29  | 150  | 7.1976          |
+| 7.1605        | 4.38  | 200  | 7.1830          |
 
 
 ### Framework versions