Hugging Face's logo

opencsg
/
opencsg-CodeLlama-7b-v0.1

Text Generation
Transformers
Safetensors
code
llama
llama-2
text-generation-inference
Model card Files Community
TomPei commited on
Commit
35f9f88
·
verified ·
1 Parent(s): e94c240

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -3
README.md CHANGED
@@ -42,6 +42,7 @@ This is the repository for the base 7B version finetuned based on [CodeLlama-7b-
42
  | 7B | [opencsg/Opencsg-CodeLlama-7b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-7b-v0.1) |
43
  | 13B | [opencsg/Opencsg-CodeLlama-13b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-13b-v0.1) |
44
  | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.1) |
 
45
 
46
 
47
  ## Model Eval
@@ -65,6 +66,7 @@ To simplify the comparison, we chosed the Pass@1 metric for the Python language,
65
  | opencsg-CodeLlama-13b-v0.1| **51.2%** |
66
  | CodeLlama-34b-hf | 48.2%|
67
  | opencsg-CodeLlama-34b-v0.1| **56.1%** |
 
68
 
69
  **TODO**
70
  - We will provide more benchmark scores on fine-tuned models in the future.
@@ -167,6 +169,7 @@ opencsg-CodeLlama-v0.1是一系列基于CodeLlama的通过全参数微调方法
167
  | 7B | [opencsg/Opencsg-CodeLlama-7b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-7b-v0.1) |
168
  | 13B | [opencsg/Opencsg-CodeLlama-13b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-13b-v0.1) |
169
  | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.1) |
 
170
 
171
 
172
  ## 模型评估
@@ -185,11 +188,12 @@ HumanEval 是评估模型在代码生成方面性能的最常见的基准，尤
185
  | 模型 | HumanEval python pass@1 |
186
  | --- |----------------------------------------------------------------------------- |
187
  | CodeLlama-7b-hf | 30.5%|
188
- | opencsg-CodeLlama-7b-v0.1(4k) | **43.9%** |
189
  | CodeLlama-13b-hf | 36.0%|
190
- | opencsg-CodeLlama-13b-v0.1(4k) | **51.2%** |
191
  | CodeLlama-34b-hf | 48.2%|
192
- | opencsg-CodeLlama-34b-v0.1(4k)| **56.1%** |
 
193
 
194
  **TODO**
195
  - 未来我们将提供更多微调模型的在各基准上的分数。
 
42
  | 7B | [opencsg/Opencsg-CodeLlama-7b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-7b-v0.1) |
43
  | 13B | [opencsg/Opencsg-CodeLlama-13b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-13b-v0.1) |
44
  | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.1) |
45
+ | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.2](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.2) |
46
 
47
 
48
  ## Model Eval
 
66
  | opencsg-CodeLlama-13b-v0.1| **51.2%** |
67
  | CodeLlama-34b-hf | 48.2%|
68
  | opencsg-CodeLlama-34b-v0.1| **56.1%** |
69
+ | opencsg-CodeLlama-34b-v0.1| **64.0%** |
70
 
71
  **TODO**
72
  - We will provide more benchmark scores on fine-tuned models in the future.
 
169
  | 7B | [opencsg/Opencsg-CodeLlama-7b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-7b-v0.1) |
170
  | 13B | [opencsg/Opencsg-CodeLlama-13b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-13b-v0.1) |
171
  | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.1](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.1) |
172
+ | 34B | [opencsg/Opencsg-CodeLlama-34b-v0.2](https://huggingface.co/opencsg/opencsg-CodeLlama-34b-v0.2) |
173
 
174
 
175
  ## 模型评估
 
188
  | 模型 | HumanEval python pass@1 |
189
  | --- |----------------------------------------------------------------------------- |
190
  | CodeLlama-7b-hf | 30.5%|
191
+ | opencsg-CodeLlama-7b-v0.1| **43.9%** |
192
  | CodeLlama-13b-hf | 36.0%|
193
+ | opencsg-CodeLlama-13b-v0.1 | **51.2%** |
194
  | CodeLlama-34b-hf | 48.2%|
195
+ | opencsg-CodeLlama-34b-v0.1| **56.1%** |
196
+ | opencsg-CodeLlama-34b-v0.1| **64.0%** |
197
 
198
  **TODO**
199
  - 未来我们将提供更多微调模型的在各基准上的分数。