Update README.md

README.md

@@ -13,12 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base](https://huggingface.co/cardiffnlp/twitter-roberta-base) on a dataset of twitter user descriptions.
 It achieves the following results on the evaluation set:
-- eval_loss: 3.2575
-- eval_runtime: 50.3796
-- eval_samples_per_second: 158.933
-- eval_steps_per_second: 2.501
-- epoch: 5.36
-- step: 3798
+- eval_perplexity: 2.33
+- epoch: 15
+- step: 10635
 
 ## Model description
 
@@ -43,7 +40,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 15
 - mixed_precision_training: Native AMP
 
 ### Framework versions