Update README.md

README.md

@@ -38,7 +38,7 @@ Unlike some models, Phi-2 hasn't been fine-tuned through reinforcement learning
 opencsg-phi-2-v0.1 is a series of models based on phi-2 that have been fine-tuned using full-parameter tuning methods.
 <br>
 
-This is the repository for the base 13B version finetuned based on [CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf). 
+This is the repository for the base 2.7B version finetuned based on [phi-2](https://huggingface.co/microsoft/phi-2). 
 
 | Model Size    | Base Model                                                                    |
 | --- | ----------------------------------------------------------------------------- |
@@ -62,7 +62,7 @@ To simplify the comparison, we chosed the Pass@1 metric for the Python language,
 | Model     | HumanEval python pass@1                                                 |
 | ---  |----------------------------------------------------------------------------- |
 | phi-2 | 48.2% |
-| **opencsg-phi-2-v0.1** |**54.3**|
+| **opencsg-phi-2-v0.1** |**54.3%**|
 
 
 
@@ -143,7 +143,7 @@ opencsg-phi-2-v0.1是是一系列基于phi-2的通过全参数微调方法进行
 
 <br>
 
-这是基于 [CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf) 进行微调的模型版本。
+这是基于 [phi-2](https://huggingface.co/microsoft/phi-2) 进行微调的模型版本。
 
 | 模型大小    | 基座模型                                                                    |
 | --- | ----------------------------------------------------------------------------- |
@@ -166,7 +166,7 @@ HumanEval 是评估模型在代码生成方面性能的最常见的基准，尤
 | 模型     | HumanEval python pass@1                                                 |
 | ---  |----------------------------------------------------------------------------- |
 | phi-2 | 48.2% |
-| **opencsg-phi-2-v0.1** |**54.3**|
+| **opencsg-phi-2-v0.1** |**54.3%**|