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 glue dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4585
-- Matthews Correlation: 0.6333
+- Loss: 0.5386
+- Matthews Correlation: 0.5993
 
 ## 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: 16
-- total_eval_batch_size: 5
+- total_train_batch_size: 64
+- total_eval_batch_size: 20
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 5
@@ -55,16 +55,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.4797        | 1.0   | 534  | 0.4011          | 0.5519               |
-| 0.4283        | 2.0   | 1068 | 0.4207          | 0.6128               |
-| 0.058         | 3.0   | 1602 | 0.5122          | 0.5816               |
-| 0.0678        | 4.0   | 2136 | 0.4585          | 0.6333               |
-| 0.1719        | 5.0   | 2670 | 0.5723          | 0.6170               |
+| 0.3435        | 1.0   | 133  | 0.4041          | 0.5761               |
+| 0.2145        | 2.0   | 266  | 0.4749          | 0.5324               |
+| 0.1297        | 3.0   | 399  | 0.5142          | 0.5564               |
+| 0.0727        | 4.0   | 532  | 0.4985          | 0.5945               |
+| 0.1003        | 5.0   | 665  | 0.5386          | 0.5993               |
 
 
 ### Framework versions
 
-- Transformers 4.20.0
+- Transformers 4.20.1
 - Pytorch 1.10.0+cpu
 - Datasets 2.7.0
 - Tokenizers 0.12.1