diff --git a/model_zoo/official/nlp/bert/README.md b/model_zoo/official/nlp/bert/README.md
index ddcfdbc1f44..3073628f35f 100644
--- a/model_zoo/official/nlp/bert/README.md
+++ b/model_zoo/official/nlp/bert/README.md
@@ -608,9 +608,9 @@ The command above will run in the background, you can view training logs in ner_
 If you choose F1 as assessment method, the result will be as follows:
 
 ```text
-Precision 0.920507
-Recall 0.948683
-F1 0.920507
+Precision 0.868245
+Recall 0.865611
+F1 0.866926
 ```
 
 #### evaluation on msra dataset when running on Ascend
diff --git a/model_zoo/official/nlp/bert/README_CN.md b/model_zoo/official/nlp/bert/README_CN.md
index 21f2cd4a7d9..72417e709ee 100644
--- a/model_zoo/official/nlp/bert/README_CN.md
+++ b/model_zoo/official/nlp/bert/README_CN.md
@@ -572,9 +572,9 @@ bash scripts/run_ner.sh
 如您选择F1作为评估方法，可得到如下结果：
 
 ```text
-Precision 0.920507
-Recall 0.948683
-F1 0.920507
+Precision 0.868245
+Recall 0.865611
+F1 0.866926
 ```
 
 #### Ascend处理器上运行后评估msra数据集
diff --git a/model_zoo/official/nlp/bert/pretrain_config.yaml b/model_zoo/official/nlp/bert/pretrain_config.yaml
index 329133d403f..ddb50e782ea 100644
--- a/model_zoo/official/nlp/bert/pretrain_config.yaml
+++ b/model_zoo/official/nlp/bert/pretrain_config.yaml
@@ -127,7 +127,7 @@ large_net_cfg:
     num_hidden_layers: 24
     num_attention_heads: 16
     intermediate_size: 4096
-    hidden_act: "gelu"
+    hidden_act: "fast_gelu"
     hidden_dropout_prob: 0.1
     attention_probs_dropout_prob: 0.1
     max_position_embeddings: 512
@@ -171,4 +171,4 @@ enable_save_ckpt: ["true", "false"]
 enable_lossscale: ["true", "false"]
 do_shuffle: ["true", "false"]
 enable_data_sink: ["true", "false"]
-allreduce_post_accumulation: ["true", "false"]
\ No newline at end of file
+allreduce_post_accumulation: ["true", "false"]
diff --git a/model_zoo/official/recommend/ncf/README.md b/model_zoo/official/recommend/ncf/README.md
index 8c3553d7508..d8437702b8d 100644
--- a/model_zoo/official/recommend/ncf/README.md
+++ b/model_zoo/official/recommend/ncf/README.md
@@ -78,8 +78,8 @@ For FP16 operators, if the input data type is FP32, the backend of MindSpore wil
 
 # [Environment Requirements](#contents)
 
-- Hardware（Ascend/GPU）
-    - Prepare hardware environment with Ascend or GPU processor.
+- Hardware(Ascend）
+    - Prepare hardware environment with Ascend.
 - Framework
     - [MindSpore](https://www.mindspore.cn/install/en)
 - For more information, please check the resources below：
@@ -308,7 +308,7 @@ Inference result is saved in current path, you can find result like this in acc.
 
 ### Inference
 
-If you need to use the trained model to perform inference on multiple hardware platforms, such as GPU, Ascend 910 or Ascend 310, you can refer to this [Link](https://www.mindspore.cn/tutorial/training/en/master/advanced_use/migrate_3rd_scripts.html). Following the steps below, this is a simple example:
+If you need to use the trained model to perform inference on multiple hardware platforms, such as Ascend 910 or Ascend 310, you can refer to this [Link](https://www.mindspore.cn/tutorial/training/en/master/advanced_use/migrate_3rd_scripts.html). Following the steps below, this is a simple example:
 
 <https://www.mindspore.cn/tutorial/inference/en/master/multi_platform_inference.html>