Update README.md

README.md

@@ -55,18 +55,18 @@ Refer to [ConvLab-3](https://github.com/ConvLab/ConvLab-3) for model description
 
 The following hyperparameters were used during training:
 - learning_rate: 0.001
-- train_batch_size: 128
+- train_batch_size: 64
 - eval_batch_size: 64
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 256
+- total_train_batch_size: 128
 - optimizer: Adafactor
 - lr_scheduler_type: linear
 - num_epochs: 10.0
 
 ### Framework versions
 
-- Transformers 4.18.0
-- Pytorch 1.10.2+cu102
-- Datasets 1.18.3
-- Tokenizers 0.11.0
+- Transformers 4.20.1
+- Pytorch 1.11.0+cu113
+- Datasets 2.3.2
+- Tokenizers 0.12.1