Update README.md
Browse files
README.md
CHANGED
|
@@ -122,26 +122,6 @@ For more evaluation details, such as few-shot settings and prompts, please check
|
|
| 122 |
|
| 123 |
</div>
|
| 124 |
|
| 125 |
-
#### Chinese Open Ended Generation Evaluation
|
| 126 |
-
**Alignbench** (https://arxiv.org/abs/2311.18743)
|
| 127 |
-
<div align="center">
|
| 128 |
-
|
| 129 |
-
| **模型** | **开源/闭源** | **总分** | **中文推理** | **中文语言** |
|
| 130 |
-
| :---: | :---: | :---: | :---: | :---: |
|
| 131 |
-
| gpt-4-1106-preview | 闭源 | 8.01 | 7.73 | 8.29 |
|
| 132 |
-
| DeepSeek-V2 Chat (RL) | 开源 | 7.91 | 7.45 | 8.36 |
|
| 133 |
-
| erniebot-4.0-202404 (文心一言) | 闭源 | 7.89 | 7.61 | 8.17 |
|
| 134 |
-
| DeepSeek-V2 Chat (SFT) | 开源 | 7.74 | 7.30 | 8.17 |
|
| 135 |
-
| gpt-4-0613 | 闭源 | 7.53 | 7.47 | 7.59 |
|
| 136 |
-
| erniebot-4.0-202312 (文心一言) | 闭源 | 7.36 | 6.84 | 7.88 |
|
| 137 |
-
| moonshot-v1-32k-202404 (月之暗面) | 闭源 | 7.22 | 6.42 | 8.02 |
|
| 138 |
-
| Qwen1.5-72B-Chat (通义千问) | 开源 | 7.19 | 6.45 | 7.93 |
|
| 139 |
-
| DeepSeek-67B-Chat | 开源 | 6.43 | 5.75 | 7.11 |
|
| 140 |
-
| Yi-34B-Chat (零一万物) | 开源 | 6.12 | 4.86 | 7.38 |
|
| 141 |
-
| gpt-3.5-turbo-0613 | 闭源 | 6.08 | 5.35 | 6.71 |
|
| 142 |
-
| DeepSeek-V2-Lite 16B Chat (SFT) | 开源 | 6.01 | 4.71 | 7.32 |
|
| 143 |
-
|
| 144 |
-
</div>
|
| 145 |
|
| 146 |
## 5. Model Architecture
|
| 147 |
DeepSeek-V2 adopts innovative architectures to guarantee economical training and efficient inference：
|
|
|
|
| 122 |
|
| 123 |
</div>
|
| 124 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 125 |
|
| 126 |
## 5. Model Architecture
|
| 127 |
DeepSeek-V2 adopts innovative architectures to guarantee economical training and efficient inference：
|