Datasets:

Modalities:
Tabular
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
License:
crazycth commited on
Commit
858b10c
1 Parent(s): c84c91a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -5
README.md CHANGED
@@ -29,18 +29,36 @@ configs:
29
  path: data/train-*
30
  ---
31
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
32
  This math-related data from [Fineweb](https://huggingface.co/spaces/HuggingFaceFW/blogpost-fineweb-v1) was specifically used in [OpenCoder](https://huggingface.co/papers/2411.04905) pre-training.
33
  We employ fastText in three iterative rounds to recall a final dataset of 55B code and math-related data.
34
  You can find code-related data at [OpenCoder-LLM/fineweb-code-corpus](https://huggingface.co/datasets/OpenCoder-LLM/fineweb-code-corpus).
35
 
36
  *This work belongs to [INF](https://www.infly.cn/).*
37
 
38
- ## Citation
 
39
  ```
40
  @inproceedings{Huang2024OpenCoderTO,
41
- title={OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models},
42
- author={Siming Huang and Tianhao Cheng and Jason Klein Liu and Jiaran Hao and Liuyihan Song and Yang Xu and J. Yang and J. H. Liu and Chenchen Zhang and Linzheng Chai and Ruifeng Yuan and Zhaoxiang Zhang and Jie Fu and Qian Liu and Ge Zhang and Zili Wang and Yuan Qi and Yinghui Xu and Wei Chu},
43
- year={2024},
44
- url={https://arxiv.org/pdf/2411.04905}
45
  }
46
  ```
 
29
  path: data/train-*
30
  ---
31
 
32
+ ![image](https://github.com/user-attachments/assets/66e5afec-060d-43c0-937e-dd7b6b1a26ef)
33
+
34
+ # OpenCoder Dataset
35
+ The OpenCoder dataset is composed of the following datasets:
36
+
37
+ * [opc-sft-stage1](https://huggingface.co/datasets/OpenCoder-LLM/opc-sft-stage1): the sft data used for opencoder sft-stage1
38
+ * [opc-sft-stage2](https://huggingface.co/datasets/OpenCoder-LLM/opc-sft-stage2): the sft data used for opencoder sft-stage2
39
+ * [opc-annealing-corpus](https://huggingface.co/datasets/OpenCoder-LLM/opc-annealing-corpus): the synthetic data & algorithmic corpus used for opencoder annealing
40
+ * [opc-fineweb-code-corpus](https://huggingface.co/datasets/OpenCoder-LLM/fineweb-code-corpus): the code-related page recalled from fineweb
41
+ * [opc-fineweb-math-corpus](https://huggingface.co/datasets/OpenCoder-LLM/fineweb-math-corpus): the math-related page recalled from fineweb **<-- you are here**
42
+ * [refineCode-code-corpus-meta](https://huggingface.co/datasets/OpenCoder-LLM/RefineCode-code-corpus-meta): the meta-data of RefineCode
43
+
44
+ Detailed information about the data can be found in our [paper](https://arxiv.org/abs/2411.04905).
45
+
46
+
47
+ ## opc-fineweb-math-corpus summary
48
+
49
  This math-related data from [Fineweb](https://huggingface.co/spaces/HuggingFaceFW/blogpost-fineweb-v1) was specifically used in [OpenCoder](https://huggingface.co/papers/2411.04905) pre-training.
50
  We employ fastText in three iterative rounds to recall a final dataset of 55B code and math-related data.
51
  You can find code-related data at [OpenCoder-LLM/fineweb-code-corpus](https://huggingface.co/datasets/OpenCoder-LLM/fineweb-code-corpus).
52
 
53
  *This work belongs to [INF](https://www.infly.cn/).*
54
 
55
+ ## Citation Information
56
+ Please consider citing our [paper](https://arxiv.org/abs/2411.04905) if you find this dataset useful:
57
  ```
58
  @inproceedings{Huang2024OpenCoderTO,
59
+ title = {OpenCoder: The Open Cookbook for Top-Tier Code Large Language Models},
60
+ author = {Siming Huang and Tianhao Cheng and Jason Klein Liu and Jiaran Hao and Liuyihan Song and Yang Xu and J. Yang and J. H. Liu and Chenchen Zhang and Linzheng Chai and Ruifeng Yuan and Zhaoxiang Zhang and Jie Fu and Qian Liu and Ge Zhang and Zili Wang and Yuan Qi and Yinghui Xu and Wei Chu},
61
+ year = {2024},
62
+ url = {https://arxiv.org/pdf/2411.04905}
63
  }
64
  ```