update model card README.md

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the swag dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5098
-- Accuracy: 0.8273
+- Loss: 0.4382
+- Accuracy: 0.8390
 
 ## Model description
 
@@ -44,8 +44,8 @@ The following hyperparameters were used during training:
 - seed: 42
 - distributed_type: IPU
 - gradient_accumulation_steps: 16
-- total_train_batch_size: 32
-- total_eval_batch_size: 10
+- total_train_batch_size: 128
+- total_eval_batch_size: 40
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 3
@@ -55,9 +55,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 1.0432        | 1.0   | 2298 | 0.5479          | 0.7869   |
-| 0.274         | 2.0   | 4596 | 0.4749          | 0.8155   |
-| 0.0939        | 3.0   | 6894 | 0.5098          | 0.8273   |
+| 0.5707        | 1.0   | 574  | 0.4990          | 0.8097   |
+| 0.5092        | 2.0   | 1148 | 0.4321          | 0.8361   |
+| 0.3597        | 3.0   | 1722 | 0.4382          | 0.8390   |
 
 
 ### Framework versions