Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
AkimfromParis
commited on
Update src/about.py
Browse files- src/about.py +10 -15
src/about.py
CHANGED
@@ -68,7 +68,6 @@ class Tasks(Enum):
|
|
68 |
NUM_FEWSHOT = 0 # Change with your few shot
|
69 |
# ---------------------------------------------------
|
70 |
|
71 |
-
|
72 |
# Your leaderboard name
|
73 |
TITLE = """<h1 align="center" id="space-title">🇯🇵 Open Japanese LLM Leaderboard 🌸<br>「オープンソース大規模言語モデルのリーダーボード」</h1>"""
|
74 |
|
@@ -77,6 +76,16 @@ INTRODUCTION_TEXT = """
|
|
77 |
The __Open Japanese LLM Leaderboard__ by __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__ evaluates the performance of Japanese Large Language Models (LLMs) with more than 16 tasks from classical to modern NLP tasks. The __Open Japanese LLM Leaderboard__ was built by open-source contributors of __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__, a cross-organizational project for the research and development of Japanese LLMs supported by the _National Institute of Informatics_ in Tokyo, Japan.
|
78 |
|
79 |
On the __"LLM Benchmark"__ page, the question mark **"?"** refers to the parameters that are unknown in the model card on Hugging Face. For more information about datasets, please consult the __"About"__ page or refer to the website of __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__. And on the __"Submit here!"__ page, you can evaluate the performance of your model, and be part of the leaderboard.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
"""
|
81 |
|
82 |
# Which evaluations are you running? how can people reproduce what you have?
|
@@ -193,20 +202,6 @@ If your model is displayed in the `FAILED` category, its execution stopped.
|
|
193 |
Make sure you have followed the above steps first.
|
194 |
If everything is done, check you can launch the EleutherAIHarness on your model locally, using the above command without modifications (you can add `--limit` to limit the number of examples per task).
|
195 |
"""
|
196 |
-
|
197 |
-
BOTTOM_LOGO = """
|
198 |
-
<div style="display: flex; flex-direction: row; justify-content">
|
199 |
-
<a href="https://llm-jp.nii.ac.jp/en/">
|
200 |
-
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/LLM-jp-Logo-Oct-2024.png" alt="LLM-jp" style="max-height: 100px">
|
201 |
-
</a>
|
202 |
-
<a href="https://mdx.jp/">
|
203 |
-
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/MDX-Logo-Oct-2024.jpg" alt="MDX" style="max-height: 100px">
|
204 |
-
</a>
|
205 |
-
<a href="https://huggingface.co/">
|
206 |
-
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/HuggingFace-Logo-Oct-2024.png" alt="HuggingFace" style="max-height: 100px">
|
207 |
-
</a>
|
208 |
-
"""
|
209 |
-
|
210 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
211 |
CITATION_BUTTON_TEXT = r"""@misc{OJLL,
|
212 |
author = {Miyao, Yusuke and Ishida, Shigeki and Okamoto, Takumi and Han, Namgi and Mousterou, Akim and Fourrier, Clémentine and Hayashi, Toshihiro and Tachibana, Yuichiro},
|
|
|
68 |
NUM_FEWSHOT = 0 # Change with your few shot
|
69 |
# ---------------------------------------------------
|
70 |
|
|
|
71 |
# Your leaderboard name
|
72 |
TITLE = """<h1 align="center" id="space-title">🇯🇵 Open Japanese LLM Leaderboard 🌸<br>「オープンソース大規模言語モデルのリーダーボード」</h1>"""
|
73 |
|
|
|
76 |
The __Open Japanese LLM Leaderboard__ by __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__ evaluates the performance of Japanese Large Language Models (LLMs) with more than 16 tasks from classical to modern NLP tasks. The __Open Japanese LLM Leaderboard__ was built by open-source contributors of __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__, a cross-organizational project for the research and development of Japanese LLMs supported by the _National Institute of Informatics_ in Tokyo, Japan.
|
77 |
|
78 |
On the __"LLM Benchmark"__ page, the question mark **"?"** refers to the parameters that are unknown in the model card on Hugging Face. For more information about datasets, please consult the __"About"__ page or refer to the website of __[LLM-Jp](https://llm-jp.nii.ac.jp/en/)__. And on the __"Submit here!"__ page, you can evaluate the performance of your model, and be part of the leaderboard.
|
79 |
+
<div style="display: flex; flex-direction: row; justify-content: space-around; align-items: center" dir="ltr">
|
80 |
+
<a href="https://llm-jp.nii.ac.jp/en/">
|
81 |
+
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/LLM-jp-Logo-Oct-2024.png" alt="LLM-jp" style="max-height: 100px">
|
82 |
+
</a>
|
83 |
+
<a href="https://mdx.jp/">
|
84 |
+
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/MDX-Logo-Oct-2024.jpg" alt="MDX" style="max-height: 100px">
|
85 |
+
</a>
|
86 |
+
<a href="https://huggingface.co/">
|
87 |
+
<img src="https://raw.githubusercontent.com/AkimfromParis/akimfromparis/refs/heads/main/images/HuggingFace-Logo-Oct-2024.png" alt="HuggingFace" style="max-height: 100px">
|
88 |
+
</a>
|
89 |
"""
|
90 |
|
91 |
# Which evaluations are you running? how can people reproduce what you have?
|
|
|
202 |
Make sure you have followed the above steps first.
|
203 |
If everything is done, check you can launch the EleutherAIHarness on your model locally, using the above command without modifications (you can add `--limit` to limit the number of examples per task).
|
204 |
"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
205 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
206 |
CITATION_BUTTON_TEXT = r"""@misc{OJLL,
|
207 |
author = {Miyao, Yusuke and Ishida, Shigeki and Okamoto, Takumi and Han, Namgi and Mousterou, Akim and Fourrier, Clémentine and Hayashi, Toshihiro and Tachibana, Yuichiro},
|