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: 0.4409
-- Accuracy: 0.8388
+- Loss: 0.5190
+- Accuracy: 0.8260
 
 ## 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: 128
-- total_eval_batch_size: 40
+- total_train_batch_size: 32
+- total_eval_batch_size: 10
 - 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 |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.5749        | 1.0   | 574  | 0.4971          | 0.8122   |
-| 0.5427        | 2.0   | 1148 | 0.4358          | 0.8324   |
-| 0.3896        | 3.0   | 1722 | 0.4409          | 0.8388   |
+| 1.0993        | 1.0   | 2298 | 0.5474          | 0.7871   |
+| 0.2222        | 2.0   | 4596 | 0.4744          | 0.8181   |
+| 0.1633        | 3.0   | 6894 | 0.5190          | 0.8260   |
 
 
 ### Framework versions