Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Gregor Betz
commited on
Commit
•
749c594
1
Parent(s):
0f51a5c
description
Browse files- src/display/about.py +11 -13
src/display/about.py
CHANGED
@@ -54,14 +54,17 @@ Performance leaderboards like the [🤗 Open LLM Leaderboard](https://huggingfac
|
|
54 |
Unlike these leaderboards, the `/\/` Open CoT Leaderboard assess a model's ability to effectively reason about a `task`:
|
55 |
|
56 |
|
|
|
|
|
|
|
|
|
|
|
57 |
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
d
|
63 |
-
|
64 |
-
|
65 |
|
66 |
|
67 |
## Test dataset selection (`tasks`)
|
@@ -109,12 +112,7 @@ We're populating the Open CoT Leaderboard step by step. The idea is to grow a di
|
|
109 |
|
110 |
"""
|
111 |
|
112 |
-
|
113 |
-
# |:---|:---|
|
114 |
-
# | Can `model` solve `task`? | Can `model` do CoT to improve in `task`? |
|
115 |
-
# | Measures `task` performance. | Measures ability to reason (about `task`). |
|
116 |
-
# | Metric: absolute accuracy. | Metric: relative accuracy gain. |
|
117 |
-
# | Covers broad spectrum of `tasks`. | Focuses on critical thinking `tasks`. |
|
118 |
|
119 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
120 |
CITATION_BUTTON_TEXT = r"""
|
|
|
54 |
Unlike these leaderboards, the `/\/` Open CoT Leaderboard assess a model's ability to effectively reason about a `task`:
|
55 |
|
56 |
|
57 |
+
### 🤗 Open LLM Leaderboard
|
58 |
+
a. Can `model` solve `task`?
|
59 |
+
b. Metric: absolute accuracy.
|
60 |
+
c. Measures `task` performance.
|
61 |
+
d. Covers broad spectrum of `tasks`.
|
62 |
|
63 |
+
### `/\/` Open CoT Leaderboard
|
64 |
+
a. Can `model` do CoT to improve in `task`?
|
65 |
+
b. Metric: relative accuracy gain.
|
66 |
+
c. Measures ability to reason (about `task`).
|
67 |
+
d. Focuses on critical thinking `tasks`.
|
|
|
|
|
68 |
|
69 |
|
70 |
## Test dataset selection (`tasks`)
|
|
|
112 |
|
113 |
"""
|
114 |
|
115 |
+
|
|
|
|
|
|
|
|
|
|
|
116 |
|
117 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
118 |
CITATION_BUTTON_TEXT = r"""
|