teddylee777
commited on
Commit
•
d5b0cc8
1
Parent(s):
0e21a65
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
base_model: yanolja/EEVE-Korean-10.8B-v1.0
|
6 |
+
model-index:
|
7 |
+
- name: yanolja/EEVE-Korean-Instruct-10.8B-v1.0
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
|
12 |
+
- Original model is [yanolja/EEVE-Korean-Instruct-10.8B-v1.0](https://huggingface.co/yanolja/EEVE-Korean-Instruct-10.8B-v1.0)
|
13 |
+
- quantized using [llama.cpp](https://github.com/ggerganov/llama.cpp)
|
14 |
+
|
15 |
+
|
16 |
+
## Ollama
|
17 |
+
|
18 |
+
Modelfile
|
19 |
+
|
20 |
+
```
|
21 |
+
FROM EEVE-Korean-Instruct-10.8B-v1.0-Q8_0.gguf
|
22 |
+
|
23 |
+
TEMPLATE """{{- if .System }}
|
24 |
+
<s>{{ .System }}</s>
|
25 |
+
{{- end }}
|
26 |
+
<s>Human:
|
27 |
+
{{ .Prompt }}</s>
|
28 |
+
<s>Assistant:
|
29 |
+
"""
|
30 |
+
|
31 |
+
SYSTEM """A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions."""
|
32 |
+
|
33 |
+
PARAMETER temperature 0
|
34 |
+
PARAMETER num_predict 3000
|
35 |
+
PARAMETER num_ctx 4096
|
36 |
+
PARAMETER stop <s>
|
37 |
+
PARAMETER stop </s>
|
38 |
+
```
|
39 |
+
|
40 |
+
|
41 |
+
|
42 |
+
### Training Data
|
43 |
+
- Korean-translated version of [Open-Orca/SlimOrca-Dedup](https://huggingface.co/datasets/Open-Orca/SlimOrca-Dedup)
|
44 |
+
- Korean-translated version of [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned)
|
45 |
+
- No other dataset was used
|
46 |
+
|
47 |
+
## Citation
|
48 |
+
|
49 |
+
```
|
50 |
+
@misc{kim2024efficient,
|
51 |
+
title={Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models},
|
52 |
+
author={Seungduk Kim and Seungtaek Choi and Myeongho Jeong},
|
53 |
+
year={2024},
|
54 |
+
eprint={2402.14714},
|
55 |
+
archivePrefix={arXiv},
|
56 |
+
primaryClass={cs.CL}
|
57 |
+
}
|
58 |
+
```
|
59 |
+
```
|
60 |
+
@misc{cui2023ultrafeedback,
|
61 |
+
title={UltraFeedback: Boosting Language Models with High-quality Feedback},
|
62 |
+
author={Ganqu Cui and Lifan Yuan and Ning Ding and Guanming Yao and Wei Zhu and Yuan Ni and Guotong Xie and Zhiyuan Liu and Maosong Sun},
|
63 |
+
year={2023},
|
64 |
+
eprint={2310.01377},
|
65 |
+
archivePrefix={arXiv},
|
66 |
+
primaryClass={cs.CL}
|
67 |
+
}
|
68 |
+
```
|
69 |
+
```
|
70 |
+
@misc{SlimOrcaDedup,
|
71 |
+
title = {SlimOrca Dedup: A Deduplicated Subset of SlimOrca},
|
72 |
+
author = {Wing Lian and Guan Wang and Bleys Goodson and Eugene Pentland and Austin Cook and Chanvichet Vong and "Teknium" and Nathan Hoos},
|
73 |
+
year = {2023},
|
74 |
+
publisher = {HuggingFace},
|
75 |
+
url = {https://huggingface.co/datasets/Open-Orca/SlimOrca-Dedup/}
|
76 |
+
}
|
77 |
+
```
|
78 |
+
```
|
79 |
+
@misc{mukherjee2023orca,
|
80 |
+
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
|
81 |
+
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
|
82 |
+
year={2023},
|
83 |
+
eprint={2306.02707},
|
84 |
+
archivePrefix={arXiv},
|
85 |
+
primaryClass={cs.CL}
|
86 |
+
}
|
87 |
+
```
|