leaderboard-pt-pr-bot commited on
Commit
85e19eb
1 Parent(s): 4f92445

Fixing some errors of the leaderboard evaluation results in the ModelCard yaml

Browse files

The name of a few benchmarks are incorrect on the model metadata.

This commit fixes some minor errors of the [last PR](2) on the ModelCard YAML metadata.

Files changed (1) hide show
  1. README.md +27 -1
README.md CHANGED
@@ -68,6 +68,19 @@ model-index:
68
  - type: f1_macro
69
  value: 90.31
70
  name: f1-macro
 
 
 
 
 
 
 
 
 
 
 
 
 
71
  - type: pearson
72
  value: 76.55
73
  name: pearson
@@ -95,7 +108,7 @@ model-index:
95
  name: Text Generation
96
  dataset:
97
  name: HateBR Binary
98
- type: eduagarcia/portuguese_benchmark
99
  split: test
100
  args:
101
  num_few_shot: 25
@@ -103,6 +116,19 @@ model-index:
103
  - type: f1_macro
104
  value: 79.21
105
  name: f1-macro
 
 
 
 
 
 
 
 
 
 
 
 
 
106
  - type: f1_macro
107
  value: 68.87
108
  name: f1-macro
 
68
  - type: f1_macro
69
  value: 90.31
70
  name: f1-macro
71
+ source:
72
+ url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=cnmoro/Mistral-7B-Portuguese
73
+ name: Open Portuguese LLM Leaderboard
74
+ - task:
75
+ type: text-generation
76
+ name: Text Generation
77
+ dataset:
78
+ name: Assin2 STS
79
+ type: eduagarcia/portuguese_benchmark
80
+ split: test
81
+ args:
82
+ num_few_shot: 15
83
+ metrics:
84
  - type: pearson
85
  value: 76.55
86
  name: pearson
 
108
  name: Text Generation
109
  dataset:
110
  name: HateBR Binary
111
+ type: ruanchaves/hatebr
112
  split: test
113
  args:
114
  num_few_shot: 25
 
116
  - type: f1_macro
117
  value: 79.21
118
  name: f1-macro
119
+ source:
120
+ url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=cnmoro/Mistral-7B-Portuguese
121
+ name: Open Portuguese LLM Leaderboard
122
+ - task:
123
+ type: text-generation
124
+ name: Text Generation
125
+ dataset:
126
+ name: PT Hate Speech Binary
127
+ type: hate_speech_portuguese
128
+ split: test
129
+ args:
130
+ num_few_shot: 25
131
+ metrics:
132
  - type: f1_macro
133
  value: 68.87
134
  name: f1-macro