Update README.md
Browse files
README.md
CHANGED
@@ -59,9 +59,126 @@ model-index:
|
|
59 |
source:
|
60 |
name: Open Arabic LLM Leaderboard
|
61 |
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
62 |
-
|
63 |
-
|
64 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
|
66 |
|
67 |
---
|
|
|
59 |
source:
|
60 |
name: Open Arabic LLM Leaderboard
|
61 |
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
62 |
+
- task:
|
63 |
+
type: text-generation
|
64 |
+
dataset:
|
65 |
+
name: Arabic_EXAMS
|
66 |
+
type: OALL/Arabic_EXAMS
|
67 |
+
metrics:
|
68 |
+
- name: acc_norm
|
69 |
+
type: loglikelihood_acc_norm
|
70 |
+
value: 51.4
|
71 |
+
source:
|
72 |
+
name: Open Arabic LLM Leaderboard
|
73 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
74 |
+
- task:
|
75 |
+
type: text-generation
|
76 |
+
dataset:
|
77 |
+
name: ARC Easy
|
78 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
79 |
+
metrics:
|
80 |
+
- name: acc_norm
|
81 |
+
type: loglikelihood_acc_norm
|
82 |
+
value: 86
|
83 |
+
source:
|
84 |
+
name: Open Arabic LLM Leaderboard
|
85 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
86 |
+
- task:
|
87 |
+
type: text-generation
|
88 |
+
dataset:
|
89 |
+
name: BOOLQ (Arabic)
|
90 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
91 |
+
metrics:
|
92 |
+
- name: acc_norm
|
93 |
+
type: loglikelihood_acc_norm
|
94 |
+
value: 64.05
|
95 |
+
source:
|
96 |
+
name: Open Arabic LLM Leaderboard
|
97 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
98 |
+
- task:
|
99 |
+
type: text-generation
|
100 |
+
dataset:
|
101 |
+
name: COPA (Arabic)
|
102 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
103 |
+
metrics:
|
104 |
+
- name: acc_norm
|
105 |
+
type: loglikelihood_acc_norm
|
106 |
+
value: 78.89
|
107 |
+
source:
|
108 |
+
name: Open Arabic LLM Leaderboard
|
109 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
110 |
+
- task:
|
111 |
+
type: text-generation
|
112 |
+
dataset:
|
113 |
+
name: HELLASWAG (Arabic)
|
114 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
115 |
+
metrics:
|
116 |
+
- name: acc_norm
|
117 |
+
type: loglikelihood_acc_norm
|
118 |
+
value: 47.64
|
119 |
+
source:
|
120 |
+
name: Open Arabic LLM Leaderboard
|
121 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
122 |
+
- task:
|
123 |
+
type: text-generation
|
124 |
+
dataset:
|
125 |
+
name: OPENBOOK QA (Arabic)
|
126 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
127 |
+
metrics:
|
128 |
+
- name: acc_norm
|
129 |
+
type: loglikelihood_acc_norm
|
130 |
+
value: 72.93
|
131 |
+
source:
|
132 |
+
name: Open Arabic LLM Leaderboard
|
133 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
134 |
+
- task:
|
135 |
+
type: text-generation
|
136 |
+
dataset:
|
137 |
+
name: PIQA (Arabic)
|
138 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
139 |
+
metrics:
|
140 |
+
- name: acc_norm
|
141 |
+
type: loglikelihood_acc_norm
|
142 |
+
value: 71.96
|
143 |
+
source:
|
144 |
+
name: Open Arabic LLM Leaderboard
|
145 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
146 |
+
- task:
|
147 |
+
type: text-generation
|
148 |
+
dataset:
|
149 |
+
name: RACE (Arabic)
|
150 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
151 |
+
metrics:
|
152 |
+
- name: acc_norm
|
153 |
+
type: loglikelihood_acc_norm
|
154 |
+
value: 75.55
|
155 |
+
source:
|
156 |
+
name: Open Arabic LLM Leaderboard
|
157 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
158 |
+
- task:
|
159 |
+
type: text-generation
|
160 |
+
dataset:
|
161 |
+
name: SCIQ (Arabic)
|
162 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
163 |
+
metrics:
|
164 |
+
- name: acc_norm
|
165 |
+
type: loglikelihood_acc_norm
|
166 |
+
value: 91.26
|
167 |
+
source:
|
168 |
+
name: Open Arabic LLM Leaderboard
|
169 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
170 |
+
- task:
|
171 |
+
type: text-generation
|
172 |
+
dataset:
|
173 |
+
name: TOXIGEN (Arabic)
|
174 |
+
type: OALL/AlGhafa-Arabic-LLM-Benchmark-Translated
|
175 |
+
metrics:
|
176 |
+
- name: acc_norm
|
177 |
+
type: loglikelihood_acc_norm
|
178 |
+
value: 67.59
|
179 |
+
source:
|
180 |
+
name: Open Arabic LLM Leaderboard
|
181 |
+
url: https://huggingface.co/spaces/OALL/Open-Arabic-LLM-Leaderboard
|
182 |
|
183 |
|
184 |
---
|