Update README.md

Updated README.md after finetuning on QQPD for 3 epochs.

README.md

@@ -14,16 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # xlnet-base-cased-finetuned-qqp
 
-This model is a fine-tuned version of [xlnet-base-cased](https://huggingface.co/xlnet-base-cased) on the glue dataset.
+This model is a fine-tuned version of [xlnet-base-cased](https://huggingface.co/xlnet-base-cased) on the qqp dataset (part of glue dataset).
 It achieves the following results on the evaluation set:
-- eval_loss: 0.3759
-- eval_accuracy: 0.8292
-- eval_f1: 0.7729
-- eval_runtime: 406.2664
-- eval_samples_per_second: 99.516
-- eval_steps_per_second: 6.22
-- epoch: 0.05
-- step: 1237
+- eval_loss: 0.27
+- eval_accuracy: 0.9087
+- eval_f1: 0.899
+- epoch: 3
 
 ## Model description
 
@@ -45,10 +41,10 @@ The following hyperparameters were used during training:
 - learning_rate: 2e-05
 - train_batch_size: 16
 - eval_batch_size: 16
-- seed: 42
+- weight_decay: 0.01
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 1
+- num_epochs: 3
 
 ### Framework versions