Ji-Ha commited on
Commit
6895201
1 Parent(s): 04f1d03

Fix some minor errors and wording

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -23,9 +23,9 @@ YAYI 2 是中科闻歌研发的开源大语言模型,包括 Base 和 Chat 版
23
 
24
 
25
 
26
- YAYI 2 is a collection of open-source large language models launched by Wenge Technology. YAYI2-30B is a Transformer-based large language model, and has been pretrained for 2.65 trillion tokens of multilingual data with high quality. The base model is aligned with human values through supervised fine-tuning with millions of instructions and reinforcement learning from human feedback (RLHF).
27
 
28
- We opensource the pre-trained language model in this release, namely **YAYI2-30B**. For more details about the YAYI 2, please refer to our [GitHub](https://github.com/wenge-research/YAYI2) repository. Stay tuned for more technical details in our upcoming technical report! 🔥
29
 
30
 
31
  ## 模型细节/Model Details
@@ -50,7 +50,7 @@ We opensource the pre-trained language model in this release, namely **YAYI2-30B
50
  * python 3.8 and above
51
  * pytorch 2.0.1 and above
52
  * CUDA 11.7 and above are recommended
53
- * To run YAYI2-30B in bf16/fp16, at least 80B GPU memory is required (e.g., 1xA100-80G)
54
 
55
 
56
  ## 快速开始/Quick Start
@@ -79,7 +79,7 @@ We opensource the pre-trained language model in this release, namely **YAYI2-30B
79
 
80
  我们在多个基准数据集上进行了评测,包括 C-Eval、MMLU、 CMMLU、AGIEval、GAOKAO-Bench、GSM8K、MATH、BBH、HumanEval 以及 MBPP。我们考察了模型在语言理解、学科知识、数学推理、逻辑推理以及代码生成方面的表现。YAYI 2 模型在与其规模相近的开源模型中展现出了显著的性能提升。
81
 
82
- We evaluate our model on standard benchmarks, including C-Eval, MMLU, CMMLU, AGIEval, GAOKAO-Bench, GSM8K, MATH, BBH, HumanEval, and MBPP. Our goal is to assess the model's performance in language comprehension, knowledge comprehension, mathematical reasoning, logical reasoning, and code generation. YAYI 2 has demonstrated exceptional performance across models with similar size.
83
 
84
  <table id="myTable">
85
  <!-- Table header -->
@@ -87,7 +87,7 @@ We evaluate our model on standard benchmarks, including C-Eval, MMLU, CMMLU, AGI
87
  <th></th>
88
  <th colspan="5" style="text-align: center;">Knowledge</th>
89
  <th colspan="2" style="text-align: center;">Math</th>
90
- <th colspan="1" style="text-align: center;">Logic reasonning</th>
91
  <th colspan="2" style="text-align: center;">Code</th>
92
  </tr>
93
  <tr>
@@ -238,7 +238,7 @@ We evaluate our model on standard benchmarks, including C-Eval, MMLU, CMMLU, AGI
238
 
239
  我们使用 [OpenCompass Github 仓库](https://github.com/open-compass/opencompass) 提供的源代码进行了评测。对于对比模型,我们列出了他们在 [OpenCompass](https://opencompass.org.cn) 榜单上的评测结果,截止日期为 2023年12月15日。对于其他尚未在 [OpenCompass](https://opencompass.org.cn/leaderboard-llm) 平台参与评测的模型,包括 MPT、Falcon 和 LLaMa 2,我们采用了 [LLaMA 2](https://arxiv.org/abs/2307.09288) 报告的结果。
240
 
241
- We evaluate our model using the source code from the [OpenCompass Github repository](https://github.com/open-compass/opencompass). If available, we report results for comparative models assessed by OpenCompass with the evaluation reference date set to Dec. 15th, 2013. For MPT, Falcon, and Llama, which have not been evaluated by OpenCompass, we use the results reported in the [LLaMA 2](https://arxiv.org/abs/2307.09288) paper.
242
 
243
 
244
 
@@ -254,7 +254,7 @@ The code in this project is open-sourced under the [Apache-2.0](https://github.c
254
 
255
  如果您在工作中使用了我们的模型,请引用我们的论文。
256
 
257
- If you are using the resource for your work, please cite our paper.
258
 
259
  ```
260
  @article{YAYI 2,
 
23
 
24
 
25
 
26
+ YAYI 2 is a collection of open-source large language models launched by Wenge Technology. YAYI2-30B is a Transformer-based large language model, and has been pretrained on 2.65 trillion tokens of high quality multilingual data. The base model has been aligned with human values through supervised fine-tuning with millions of instructions and reinforcement learning from human feedback (RLHF).
27
 
28
+ We open-source the pre-trained language model in this release, namely **YAYI2-30B**. For more details about the YAYI 2, please refer to our [GitHub](https://github.com/wenge-research/YAYI2) repository. Stay tuned for more technical details in our upcoming technical report! 🔥
29
 
30
 
31
  ## 模型细节/Model Details
 
50
  * python 3.8 and above
51
  * pytorch 2.0.1 and above
52
  * CUDA 11.7 and above are recommended
53
+ * To run YAYI2-30B in bf16/fp16, at least 80GB of GPU memory is required (e.g., 1xA100-80G)
54
 
55
 
56
  ## 快速开始/Quick Start
 
79
 
80
  我们在多个基准数据集上进行了评测,包括 C-Eval、MMLU、 CMMLU、AGIEval、GAOKAO-Bench、GSM8K、MATH、BBH、HumanEval 以及 MBPP。我们考察了模型在语言理解、学科知识、数学推理、逻辑推理以及代码生成方面的表现。YAYI 2 模型在与其规模相近的开源模型中展现出了显著的性能提升。
81
 
82
+ We evaluated our model on standard benchmarks, including C-Eval, MMLU, CMMLU, AGIEval, GAOKAO-Bench, GSM8K, MATH, BBH, HumanEval, and MBPP. Our goal is to assess the model's performance in language comprehension, knowledge comprehension, mathematical reasoning, logical reasoning, and code generation. YAYI 2 has demonstrated exceptional performance across models with similar size.
83
 
84
  <table id="myTable">
85
  <!-- Table header -->
 
87
  <th></th>
88
  <th colspan="5" style="text-align: center;">Knowledge</th>
89
  <th colspan="2" style="text-align: center;">Math</th>
90
+ <th colspan="1" style="text-align: center;">Logical reasoning</th>
91
  <th colspan="2" style="text-align: center;">Code</th>
92
  </tr>
93
  <tr>
 
238
 
239
  我们使用 [OpenCompass Github 仓库](https://github.com/open-compass/opencompass) 提供的源代码进行了评测。对于对比模型,我们列出了他们在 [OpenCompass](https://opencompass.org.cn) 榜单上的评测结果,截止日期为 2023年12月15日。对于其他尚未在 [OpenCompass](https://opencompass.org.cn/leaderboard-llm) 平台参与评测的模型,包括 MPT、Falcon 和 LLaMa 2,我们采用了 [LLaMA 2](https://arxiv.org/abs/2307.09288) 报告的结果。
240
 
241
+ We evaluated our model using the source code from the [OpenCompass Github repository](https://github.com/open-compass/opencompass). If available, we report results for comparative models assessed by OpenCompass with the evaluation reference date set to Dec. 15th, 2023. For MPT, Falcon, and Llama, which have not been evaluated by OpenCompass, we used the results reported in the [LLaMA 2](https://arxiv.org/abs/2307.09288) paper.
242
 
243
 
244
 
 
254
 
255
  如果您在工作中使用了我们的模型,请引用我们的论文。
256
 
257
+ If you are using this resource for your work, please cite our paper.
258
 
259
  ```
260
  @article{YAYI 2,