JJhooww commited on
Commit
e3dc84a
1 Parent(s): dac10cc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -168
README.md CHANGED
@@ -1,174 +1,20 @@
1
- language:
2
- - pt
3
  license: apache-2.0
4
  datasets:
5
  - nicholasKluge/Pt-Corpus
6
- model-index:
7
- - name: Mistral-7B-v0.2-Base_ptbr
8
- results:
9
- - task:
10
- type: text-generation
11
- name: Text Generation
12
- dataset:
13
- name: ENEM Challenge (No Images)
14
- type: eduagarcia/enem_challenge
15
- split: train
16
- args:
17
- num_few_shot: 3
18
- metrics:
19
- - type: acc
20
- value: 64.94
21
- name: accuracy
22
- source:
23
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
24
- name: Open Portuguese LLM Leaderboard
25
- - task:
26
- type: text-generation
27
- name: Text Generation
28
- dataset:
29
- name: BLUEX (No Images)
30
- type: eduagarcia-temp/BLUEX_without_images
31
- split: train
32
- args:
33
- num_few_shot: 3
34
- metrics:
35
- - type: acc
36
- value: 53.96
37
- name: accuracy
38
- source:
39
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
40
- name: Open Portuguese LLM Leaderboard
41
- - task:
42
- type: text-generation
43
- name: Text Generation
44
- dataset:
45
- name: OAB Exams
46
- type: eduagarcia/oab_exams
47
- split: train
48
- args:
49
- num_few_shot: 3
50
- metrics:
51
- - type: acc
52
- value: 45.42
53
- name: accuracy
54
- source:
55
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
56
- name: Open Portuguese LLM Leaderboard
57
- - task:
58
- type: text-generation
59
- name: Text Generation
60
- dataset:
61
- name: Assin2 RTE
62
- type: assin2
63
- split: test
64
- args:
65
- num_few_shot: 15
66
- metrics:
67
- - type: f1_macro
68
- value: 90.11
69
- name: f1-macro
70
- source:
71
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
72
- name: Open Portuguese LLM Leaderboard
73
- - task:
74
- type: text-generation
75
- name: Text Generation
76
- dataset:
77
- name: Assin2 STS
78
- type: eduagarcia/portuguese_benchmark
79
- split: test
80
- args:
81
- num_few_shot: 15
82
- metrics:
83
- - type: pearson
84
- value: 72.51
85
- name: pearson
86
- source:
87
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
88
- name: Open Portuguese LLM Leaderboard
89
- - task:
90
- type: text-generation
91
- name: Text Generation
92
- dataset:
93
- name: FaQuAD NLI
94
- type: ruanchaves/faquad-nli
95
- split: test
96
- args:
97
- num_few_shot: 15
98
- metrics:
99
- - type: f1_macro
100
- value: 69.04
101
- name: f1-macro
102
- source:
103
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
104
- name: Open Portuguese LLM Leaderboard
105
- - task:
106
- type: text-generation
107
- name: Text Generation
108
- dataset:
109
- name: HateBR Binary
110
- type: ruanchaves/hatebr
111
- split: test
112
- args:
113
- num_few_shot: 25
114
- metrics:
115
- - type: f1_macro
116
- value: 79.62
117
- name: f1-macro
118
- source:
119
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
120
- name: Open Portuguese LLM Leaderboard
121
- - task:
122
- type: text-generation
123
- name: Text Generation
124
- dataset:
125
- name: PT Hate Speech Binary
126
- type: hate_speech_portuguese
127
- split: test
128
- args:
129
- num_few_shot: 25
130
- metrics:
131
- - type: f1_macro
132
- value: 58.52
133
- name: f1-macro
134
- source:
135
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
136
- name: Open Portuguese LLM Leaderboard
137
- - task:
138
- type: text-generation
139
- name: Text Generation
140
- dataset:
141
- name: tweetSentBR
142
- type: eduagarcia/tweetsentbr_fewshot
143
- split: test
144
- args:
145
- num_few_shot: 25
146
- metrics:
147
- - type: f1_macro
148
- value: 62.32
149
- name: f1-macro
150
- source:
151
- url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=JJhooww/Mistral-7B-v0.2-Base_ptbr
152
- name: Open Portuguese LLM Leaderboard
153
  ---
154
 
155
- É um modelo base pré-treinado com cerca de 1b tokens em portugues iniciado com os pesos oficiais do modelo, deve ser utilizado para fine tuning.
156
- | faquad_nli | 68,11 | 47,63 | 20,48 |
157
- | hatebr_offensive_binary | 79,65 | 77,63 | 2,02 |
158
- | oab_exams | 45,42 | 45,24 | 0,18 |
159
- | portuguese_hate_speech_binary| 59,18 | 55,72 | 3,46 |
160
- # Open Portuguese LLM Leaderboard Evaluation Results
161
- Detailed results can be found [here](https://huggingface.co/datasets/eduagarcia-temp/llm_pt_leaderboard_raw_results/tree/main/JJhooww/Mistral-7B-v0.2-Base_ptbr) and on the [🚀 Open Portuguese LLM Leaderboard](https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard)
162
 
163
- | Metric | Value |
164
- |--------------------------|---------|
165
- |Average |**66.27**|
166
- |ENEM Challenge (No Images)| 64.94|
167
- |BLUEX (No Images) | 53.96|
168
- |OAB Exams | 45.42|
169
- |Assin2 RTE | 90.11|
170
- |Assin2 STS | 72.51|
171
- |FaQuAD NLI | 69.04|
172
- |HateBR Binary | 79.62|
173
- |PT Hate Speech Binary | 58.52|
174
- |tweetSentBR | 62.32|
 
1
+ ---
 
2
  license: apache-2.0
3
  datasets:
4
  - nicholasKluge/Pt-Corpus
5
+ language:
6
+ - pt
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  ---
8
 
9
+ É um modelo base pré-treinado com cerca de 1b tokens em portugues iniciado com os pesos oficiais do modelo, o modelo não segue instrução então precisa fazer fine tuning.
 
 
 
 
 
 
10
 
11
+ | | Mistral Base PTBR | Mistral Base | Melhoria |
12
+ |------------------------------|-------------------|--------------|----------|
13
+ | assin2_rte | 90,11 | 87,74 | 2,37 |
14
+ | assin2_sts | 72,51 | 67,05 | 5,46 |
15
+ | bluex | 53,97 | 53,27 | 0,70 |
16
+ | enem | 64,94 | 62,42 | 2,52 |
17
+ | faquad_nli | 69,04 | 47,63 | 21,41 |
18
+ | hatebr_offensive_binary | 79,62 | 77,63 | 1,99 |
19
+ | oab_exams | 45,42 | 45,24 | 0,18 |
20
+ | portuguese_hate_speech_binary| 58,52 | 55,72 | 2,80 |