ldwang commited on
Commit
9b09f79
1 Parent(s): a32952c
1_Pooling/config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false
7
+ }
README.md CHANGED
@@ -1,3 +1,2839 @@
1
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  license: mit
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ tags:
3
+ - mteb
4
+ - sentence transformers
5
+ model-index:
6
+ - name: bge-small-en
7
+ results:
8
+ - task:
9
+ type: Classification
10
+ dataset:
11
+ type: mteb/amazon_counterfactual
12
+ name: MTEB AmazonCounterfactualClassification (en)
13
+ config: en
14
+ split: test
15
+ revision: e8379541af4e31359cca9fbcf4b00f2671dba205
16
+ metrics:
17
+ - type: accuracy
18
+ value: 74.34328358208955
19
+ - type: ap
20
+ value: 37.59947775195661
21
+ - type: f1
22
+ value: 68.548415491933
23
+ - task:
24
+ type: Classification
25
+ dataset:
26
+ type: mteb/amazon_polarity
27
+ name: MTEB AmazonPolarityClassification
28
+ config: default
29
+ split: test
30
+ revision: e2d317d38cd51312af73b3d32a06d1a08b442046
31
+ metrics:
32
+ - type: accuracy
33
+ value: 93.04527499999999
34
+ - type: ap
35
+ value: 89.60696356772135
36
+ - type: f1
37
+ value: 93.03361469382438
38
+ - task:
39
+ type: Classification
40
+ dataset:
41
+ type: mteb/amazon_reviews_multi
42
+ name: MTEB AmazonReviewsClassification (en)
43
+ config: en
44
+ split: test
45
+ revision: 1399c76144fd37290681b995c656ef9b2e06e26d
46
+ metrics:
47
+ - type: accuracy
48
+ value: 46.08
49
+ - type: f1
50
+ value: 45.66249835363254
51
+ - task:
52
+ type: Retrieval
53
+ dataset:
54
+ type: arguana
55
+ name: MTEB ArguAna
56
+ config: default
57
+ split: test
58
+ revision: None
59
+ metrics:
60
+ - type: map_at_1
61
+ value: 35.205999999999996
62
+ - type: map_at_10
63
+ value: 50.782000000000004
64
+ - type: map_at_100
65
+ value: 51.547
66
+ - type: map_at_1000
67
+ value: 51.554
68
+ - type: map_at_3
69
+ value: 46.515
70
+ - type: map_at_5
71
+ value: 49.296
72
+ - type: mrr_at_1
73
+ value: 35.632999999999996
74
+ - type: mrr_at_10
75
+ value: 50.958999999999996
76
+ - type: mrr_at_100
77
+ value: 51.724000000000004
78
+ - type: mrr_at_1000
79
+ value: 51.731
80
+ - type: mrr_at_3
81
+ value: 46.669
82
+ - type: mrr_at_5
83
+ value: 49.439
84
+ - type: ndcg_at_1
85
+ value: 35.205999999999996
86
+ - type: ndcg_at_10
87
+ value: 58.835
88
+ - type: ndcg_at_100
89
+ value: 62.095
90
+ - type: ndcg_at_1000
91
+ value: 62.255
92
+ - type: ndcg_at_3
93
+ value: 50.255
94
+ - type: ndcg_at_5
95
+ value: 55.296
96
+ - type: precision_at_1
97
+ value: 35.205999999999996
98
+ - type: precision_at_10
99
+ value: 8.421
100
+ - type: precision_at_100
101
+ value: 0.984
102
+ - type: precision_at_1000
103
+ value: 0.1
104
+ - type: precision_at_3
105
+ value: 20.365
106
+ - type: precision_at_5
107
+ value: 14.680000000000001
108
+ - type: recall_at_1
109
+ value: 35.205999999999996
110
+ - type: recall_at_10
111
+ value: 84.211
112
+ - type: recall_at_100
113
+ value: 98.43499999999999
114
+ - type: recall_at_1000
115
+ value: 99.644
116
+ - type: recall_at_3
117
+ value: 61.095
118
+ - type: recall_at_5
119
+ value: 73.4
120
+ - task:
121
+ type: Clustering
122
+ dataset:
123
+ type: mteb/arxiv-clustering-p2p
124
+ name: MTEB ArxivClusteringP2P
125
+ config: default
126
+ split: test
127
+ revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d
128
+ metrics:
129
+ - type: v_measure
130
+ value: 47.52644476278646
131
+ - task:
132
+ type: Clustering
133
+ dataset:
134
+ type: mteb/arxiv-clustering-s2s
135
+ name: MTEB ArxivClusteringS2S
136
+ config: default
137
+ split: test
138
+ revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53
139
+ metrics:
140
+ - type: v_measure
141
+ value: 39.973045724188964
142
+ - task:
143
+ type: Reranking
144
+ dataset:
145
+ type: mteb/askubuntudupquestions-reranking
146
+ name: MTEB AskUbuntuDupQuestions
147
+ config: default
148
+ split: test
149
+ revision: 2000358ca161889fa9c082cb41daa8dcfb161a54
150
+ metrics:
151
+ - type: map
152
+ value: 62.28285314871488
153
+ - type: mrr
154
+ value: 74.52743701358659
155
+ - task:
156
+ type: STS
157
+ dataset:
158
+ type: mteb/biosses-sts
159
+ name: MTEB BIOSSES
160
+ config: default
161
+ split: test
162
+ revision: d3fb88f8f02e40887cd149695127462bbcf29b4a
163
+ metrics:
164
+ - type: cos_sim_pearson
165
+ value: 80.09041909160327
166
+ - type: cos_sim_spearman
167
+ value: 79.96266537706944
168
+ - type: euclidean_pearson
169
+ value: 79.50774978162241
170
+ - type: euclidean_spearman
171
+ value: 79.9144715078551
172
+ - type: manhattan_pearson
173
+ value: 79.2062139879302
174
+ - type: manhattan_spearman
175
+ value: 79.35000081468212
176
+ - task:
177
+ type: Classification
178
+ dataset:
179
+ type: mteb/banking77
180
+ name: MTEB Banking77Classification
181
+ config: default
182
+ split: test
183
+ revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
184
+ metrics:
185
+ - type: accuracy
186
+ value: 85.31493506493506
187
+ - type: f1
188
+ value: 85.2704557977762
189
+ - task:
190
+ type: Clustering
191
+ dataset:
192
+ type: mteb/biorxiv-clustering-p2p
193
+ name: MTEB BiorxivClusteringP2P
194
+ config: default
195
+ split: test
196
+ revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40
197
+ metrics:
198
+ - type: v_measure
199
+ value: 39.6837242810816
200
+ - task:
201
+ type: Clustering
202
+ dataset:
203
+ type: mteb/biorxiv-clustering-s2s
204
+ name: MTEB BiorxivClusteringS2S
205
+ config: default
206
+ split: test
207
+ revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908
208
+ metrics:
209
+ - type: v_measure
210
+ value: 35.38881249555897
211
+ - task:
212
+ type: Retrieval
213
+ dataset:
214
+ type: BeIR/cqadupstack
215
+ name: MTEB CQADupstackAndroidRetrieval
216
+ config: default
217
+ split: test
218
+ revision: None
219
+ metrics:
220
+ - type: map_at_1
221
+ value: 27.884999999999998
222
+ - type: map_at_10
223
+ value: 39.574
224
+ - type: map_at_100
225
+ value: 40.993
226
+ - type: map_at_1000
227
+ value: 41.129
228
+ - type: map_at_3
229
+ value: 36.089
230
+ - type: map_at_5
231
+ value: 38.191
232
+ - type: mrr_at_1
233
+ value: 34.477999999999994
234
+ - type: mrr_at_10
235
+ value: 45.411
236
+ - type: mrr_at_100
237
+ value: 46.089999999999996
238
+ - type: mrr_at_1000
239
+ value: 46.147
240
+ - type: mrr_at_3
241
+ value: 42.346000000000004
242
+ - type: mrr_at_5
243
+ value: 44.292
244
+ - type: ndcg_at_1
245
+ value: 34.477999999999994
246
+ - type: ndcg_at_10
247
+ value: 46.123999999999995
248
+ - type: ndcg_at_100
249
+ value: 51.349999999999994
250
+ - type: ndcg_at_1000
251
+ value: 53.578
252
+ - type: ndcg_at_3
253
+ value: 40.824
254
+ - type: ndcg_at_5
255
+ value: 43.571
256
+ - type: precision_at_1
257
+ value: 34.477999999999994
258
+ - type: precision_at_10
259
+ value: 8.841000000000001
260
+ - type: precision_at_100
261
+ value: 1.4460000000000002
262
+ - type: precision_at_1000
263
+ value: 0.192
264
+ - type: precision_at_3
265
+ value: 19.742
266
+ - type: precision_at_5
267
+ value: 14.421000000000001
268
+ - type: recall_at_1
269
+ value: 27.884999999999998
270
+ - type: recall_at_10
271
+ value: 59.087
272
+ - type: recall_at_100
273
+ value: 80.609
274
+ - type: recall_at_1000
275
+ value: 95.054
276
+ - type: recall_at_3
277
+ value: 44.082
278
+ - type: recall_at_5
279
+ value: 51.593999999999994
280
+ - task:
281
+ type: Retrieval
282
+ dataset:
283
+ type: BeIR/cqadupstack
284
+ name: MTEB CQADupstackEnglishRetrieval
285
+ config: default
286
+ split: test
287
+ revision: None
288
+ metrics:
289
+ - type: map_at_1
290
+ value: 30.639
291
+ - type: map_at_10
292
+ value: 40.047
293
+ - type: map_at_100
294
+ value: 41.302
295
+ - type: map_at_1000
296
+ value: 41.425
297
+ - type: map_at_3
298
+ value: 37.406
299
+ - type: map_at_5
300
+ value: 38.934000000000005
301
+ - type: mrr_at_1
302
+ value: 37.707
303
+ - type: mrr_at_10
304
+ value: 46.082
305
+ - type: mrr_at_100
306
+ value: 46.745
307
+ - type: mrr_at_1000
308
+ value: 46.786
309
+ - type: mrr_at_3
310
+ value: 43.980999999999995
311
+ - type: mrr_at_5
312
+ value: 45.287
313
+ - type: ndcg_at_1
314
+ value: 37.707
315
+ - type: ndcg_at_10
316
+ value: 45.525
317
+ - type: ndcg_at_100
318
+ value: 49.976
319
+ - type: ndcg_at_1000
320
+ value: 51.94499999999999
321
+ - type: ndcg_at_3
322
+ value: 41.704
323
+ - type: ndcg_at_5
324
+ value: 43.596000000000004
325
+ - type: precision_at_1
326
+ value: 37.707
327
+ - type: precision_at_10
328
+ value: 8.465
329
+ - type: precision_at_100
330
+ value: 1.375
331
+ - type: precision_at_1000
332
+ value: 0.183
333
+ - type: precision_at_3
334
+ value: 19.979
335
+ - type: precision_at_5
336
+ value: 14.115
337
+ - type: recall_at_1
338
+ value: 30.639
339
+ - type: recall_at_10
340
+ value: 54.775
341
+ - type: recall_at_100
342
+ value: 73.678
343
+ - type: recall_at_1000
344
+ value: 86.142
345
+ - type: recall_at_3
346
+ value: 43.230000000000004
347
+ - type: recall_at_5
348
+ value: 48.622
349
+ - task:
350
+ type: Retrieval
351
+ dataset:
352
+ type: BeIR/cqadupstack
353
+ name: MTEB CQADupstackGamingRetrieval
354
+ config: default
355
+ split: test
356
+ revision: None
357
+ metrics:
358
+ - type: map_at_1
359
+ value: 38.038
360
+ - type: map_at_10
361
+ value: 49.922
362
+ - type: map_at_100
363
+ value: 51.032
364
+ - type: map_at_1000
365
+ value: 51.085
366
+ - type: map_at_3
367
+ value: 46.664
368
+ - type: map_at_5
369
+ value: 48.588
370
+ - type: mrr_at_1
371
+ value: 43.95
372
+ - type: mrr_at_10
373
+ value: 53.566
374
+ - type: mrr_at_100
375
+ value: 54.318999999999996
376
+ - type: mrr_at_1000
377
+ value: 54.348
378
+ - type: mrr_at_3
379
+ value: 51.066
380
+ - type: mrr_at_5
381
+ value: 52.649
382
+ - type: ndcg_at_1
383
+ value: 43.95
384
+ - type: ndcg_at_10
385
+ value: 55.676
386
+ - type: ndcg_at_100
387
+ value: 60.126000000000005
388
+ - type: ndcg_at_1000
389
+ value: 61.208
390
+ - type: ndcg_at_3
391
+ value: 50.20400000000001
392
+ - type: ndcg_at_5
393
+ value: 53.038
394
+ - type: precision_at_1
395
+ value: 43.95
396
+ - type: precision_at_10
397
+ value: 8.953
398
+ - type: precision_at_100
399
+ value: 1.2109999999999999
400
+ - type: precision_at_1000
401
+ value: 0.135
402
+ - type: precision_at_3
403
+ value: 22.256999999999998
404
+ - type: precision_at_5
405
+ value: 15.524
406
+ - type: recall_at_1
407
+ value: 38.038
408
+ - type: recall_at_10
409
+ value: 69.15
410
+ - type: recall_at_100
411
+ value: 88.31599999999999
412
+ - type: recall_at_1000
413
+ value: 95.993
414
+ - type: recall_at_3
415
+ value: 54.663
416
+ - type: recall_at_5
417
+ value: 61.373
418
+ - task:
419
+ type: Retrieval
420
+ dataset:
421
+ type: BeIR/cqadupstack
422
+ name: MTEB CQADupstackGisRetrieval
423
+ config: default
424
+ split: test
425
+ revision: None
426
+ metrics:
427
+ - type: map_at_1
428
+ value: 24.872
429
+ - type: map_at_10
430
+ value: 32.912
431
+ - type: map_at_100
432
+ value: 33.972
433
+ - type: map_at_1000
434
+ value: 34.046
435
+ - type: map_at_3
436
+ value: 30.361
437
+ - type: map_at_5
438
+ value: 31.704
439
+ - type: mrr_at_1
440
+ value: 26.779999999999998
441
+ - type: mrr_at_10
442
+ value: 34.812
443
+ - type: mrr_at_100
444
+ value: 35.754999999999995
445
+ - type: mrr_at_1000
446
+ value: 35.809000000000005
447
+ - type: mrr_at_3
448
+ value: 32.335
449
+ - type: mrr_at_5
450
+ value: 33.64
451
+ - type: ndcg_at_1
452
+ value: 26.779999999999998
453
+ - type: ndcg_at_10
454
+ value: 37.623
455
+ - type: ndcg_at_100
456
+ value: 42.924
457
+ - type: ndcg_at_1000
458
+ value: 44.856
459
+ - type: ndcg_at_3
460
+ value: 32.574
461
+ - type: ndcg_at_5
462
+ value: 34.842
463
+ - type: precision_at_1
464
+ value: 26.779999999999998
465
+ - type: precision_at_10
466
+ value: 5.729
467
+ - type: precision_at_100
468
+ value: 0.886
469
+ - type: precision_at_1000
470
+ value: 0.109
471
+ - type: precision_at_3
472
+ value: 13.559
473
+ - type: precision_at_5
474
+ value: 9.469
475
+ - type: recall_at_1
476
+ value: 24.872
477
+ - type: recall_at_10
478
+ value: 50.400999999999996
479
+ - type: recall_at_100
480
+ value: 74.954
481
+ - type: recall_at_1000
482
+ value: 89.56
483
+ - type: recall_at_3
484
+ value: 36.726
485
+ - type: recall_at_5
486
+ value: 42.138999999999996
487
+ - task:
488
+ type: Retrieval
489
+ dataset:
490
+ type: BeIR/cqadupstack
491
+ name: MTEB CQADupstackMathematicaRetrieval
492
+ config: default
493
+ split: test
494
+ revision: None
495
+ metrics:
496
+ - type: map_at_1
497
+ value: 16.803
498
+ - type: map_at_10
499
+ value: 24.348
500
+ - type: map_at_100
501
+ value: 25.56
502
+ - type: map_at_1000
503
+ value: 25.668000000000003
504
+ - type: map_at_3
505
+ value: 21.811
506
+ - type: map_at_5
507
+ value: 23.287
508
+ - type: mrr_at_1
509
+ value: 20.771
510
+ - type: mrr_at_10
511
+ value: 28.961
512
+ - type: mrr_at_100
513
+ value: 29.979
514
+ - type: mrr_at_1000
515
+ value: 30.046
516
+ - type: mrr_at_3
517
+ value: 26.555
518
+ - type: mrr_at_5
519
+ value: 28.060000000000002
520
+ - type: ndcg_at_1
521
+ value: 20.771
522
+ - type: ndcg_at_10
523
+ value: 29.335
524
+ - type: ndcg_at_100
525
+ value: 35.188
526
+ - type: ndcg_at_1000
527
+ value: 37.812
528
+ - type: ndcg_at_3
529
+ value: 24.83
530
+ - type: ndcg_at_5
531
+ value: 27.119
532
+ - type: precision_at_1
533
+ value: 20.771
534
+ - type: precision_at_10
535
+ value: 5.4350000000000005
536
+ - type: precision_at_100
537
+ value: 0.9480000000000001
538
+ - type: precision_at_1000
539
+ value: 0.13
540
+ - type: precision_at_3
541
+ value: 11.982
542
+ - type: precision_at_5
543
+ value: 8.831
544
+ - type: recall_at_1
545
+ value: 16.803
546
+ - type: recall_at_10
547
+ value: 40.039
548
+ - type: recall_at_100
549
+ value: 65.83200000000001
550
+ - type: recall_at_1000
551
+ value: 84.478
552
+ - type: recall_at_3
553
+ value: 27.682000000000002
554
+ - type: recall_at_5
555
+ value: 33.535
556
+ - task:
557
+ type: Retrieval
558
+ dataset:
559
+ type: BeIR/cqadupstack
560
+ name: MTEB CQADupstackPhysicsRetrieval
561
+ config: default
562
+ split: test
563
+ revision: None
564
+ metrics:
565
+ - type: map_at_1
566
+ value: 28.345
567
+ - type: map_at_10
568
+ value: 37.757000000000005
569
+ - type: map_at_100
570
+ value: 39.141
571
+ - type: map_at_1000
572
+ value: 39.262
573
+ - type: map_at_3
574
+ value: 35.183
575
+ - type: map_at_5
576
+ value: 36.592
577
+ - type: mrr_at_1
578
+ value: 34.649
579
+ - type: mrr_at_10
580
+ value: 43.586999999999996
581
+ - type: mrr_at_100
582
+ value: 44.481
583
+ - type: mrr_at_1000
584
+ value: 44.542
585
+ - type: mrr_at_3
586
+ value: 41.29
587
+ - type: mrr_at_5
588
+ value: 42.642
589
+ - type: ndcg_at_1
590
+ value: 34.649
591
+ - type: ndcg_at_10
592
+ value: 43.161
593
+ - type: ndcg_at_100
594
+ value: 48.734
595
+ - type: ndcg_at_1000
596
+ value: 51.046
597
+ - type: ndcg_at_3
598
+ value: 39.118
599
+ - type: ndcg_at_5
600
+ value: 41.022
601
+ - type: precision_at_1
602
+ value: 34.649
603
+ - type: precision_at_10
604
+ value: 7.603
605
+ - type: precision_at_100
606
+ value: 1.209
607
+ - type: precision_at_1000
608
+ value: 0.157
609
+ - type: precision_at_3
610
+ value: 18.319
611
+ - type: precision_at_5
612
+ value: 12.839
613
+ - type: recall_at_1
614
+ value: 28.345
615
+ - type: recall_at_10
616
+ value: 53.367
617
+ - type: recall_at_100
618
+ value: 76.453
619
+ - type: recall_at_1000
620
+ value: 91.82000000000001
621
+ - type: recall_at_3
622
+ value: 41.636
623
+ - type: recall_at_5
624
+ value: 46.760000000000005
625
+ - task:
626
+ type: Retrieval
627
+ dataset:
628
+ type: BeIR/cqadupstack
629
+ name: MTEB CQADupstackProgrammersRetrieval
630
+ config: default
631
+ split: test
632
+ revision: None
633
+ metrics:
634
+ - type: map_at_1
635
+ value: 22.419
636
+ - type: map_at_10
637
+ value: 31.716
638
+ - type: map_at_100
639
+ value: 33.152
640
+ - type: map_at_1000
641
+ value: 33.267
642
+ - type: map_at_3
643
+ value: 28.74
644
+ - type: map_at_5
645
+ value: 30.48
646
+ - type: mrr_at_1
647
+ value: 28.310999999999996
648
+ - type: mrr_at_10
649
+ value: 37.039
650
+ - type: mrr_at_100
651
+ value: 38.09
652
+ - type: mrr_at_1000
653
+ value: 38.145
654
+ - type: mrr_at_3
655
+ value: 34.437
656
+ - type: mrr_at_5
657
+ value: 36.024
658
+ - type: ndcg_at_1
659
+ value: 28.310999999999996
660
+ - type: ndcg_at_10
661
+ value: 37.41
662
+ - type: ndcg_at_100
663
+ value: 43.647999999999996
664
+ - type: ndcg_at_1000
665
+ value: 46.007
666
+ - type: ndcg_at_3
667
+ value: 32.509
668
+ - type: ndcg_at_5
669
+ value: 34.943999999999996
670
+ - type: precision_at_1
671
+ value: 28.310999999999996
672
+ - type: precision_at_10
673
+ value: 6.963
674
+ - type: precision_at_100
675
+ value: 1.1860000000000002
676
+ - type: precision_at_1000
677
+ value: 0.154
678
+ - type: precision_at_3
679
+ value: 15.867999999999999
680
+ - type: precision_at_5
681
+ value: 11.507000000000001
682
+ - type: recall_at_1
683
+ value: 22.419
684
+ - type: recall_at_10
685
+ value: 49.28
686
+ - type: recall_at_100
687
+ value: 75.802
688
+ - type: recall_at_1000
689
+ value: 92.032
690
+ - type: recall_at_3
691
+ value: 35.399
692
+ - type: recall_at_5
693
+ value: 42.027
694
+ - task:
695
+ type: Retrieval
696
+ dataset:
697
+ type: BeIR/cqadupstack
698
+ name: MTEB CQADupstackRetrieval
699
+ config: default
700
+ split: test
701
+ revision: None
702
+ metrics:
703
+ - type: map_at_1
704
+ value: 24.669249999999998
705
+ - type: map_at_10
706
+ value: 33.332583333333325
707
+ - type: map_at_100
708
+ value: 34.557833333333335
709
+ - type: map_at_1000
710
+ value: 34.67141666666666
711
+ - type: map_at_3
712
+ value: 30.663166666666662
713
+ - type: map_at_5
714
+ value: 32.14883333333333
715
+ - type: mrr_at_1
716
+ value: 29.193833333333334
717
+ - type: mrr_at_10
718
+ value: 37.47625
719
+ - type: mrr_at_100
720
+ value: 38.3545
721
+ - type: mrr_at_1000
722
+ value: 38.413166666666676
723
+ - type: mrr_at_3
724
+ value: 35.06741666666667
725
+ - type: mrr_at_5
726
+ value: 36.450666666666656
727
+ - type: ndcg_at_1
728
+ value: 29.193833333333334
729
+ - type: ndcg_at_10
730
+ value: 38.505416666666676
731
+ - type: ndcg_at_100
732
+ value: 43.81125
733
+ - type: ndcg_at_1000
734
+ value: 46.09558333333333
735
+ - type: ndcg_at_3
736
+ value: 33.90916666666667
737
+ - type: ndcg_at_5
738
+ value: 36.07666666666666
739
+ - type: precision_at_1
740
+ value: 29.193833333333334
741
+ - type: precision_at_10
742
+ value: 6.7251666666666665
743
+ - type: precision_at_100
744
+ value: 1.1058333333333332
745
+ - type: precision_at_1000
746
+ value: 0.14833333333333332
747
+ - type: precision_at_3
748
+ value: 15.554166666666665
749
+ - type: precision_at_5
750
+ value: 11.079250000000002
751
+ - type: recall_at_1
752
+ value: 24.669249999999998
753
+ - type: recall_at_10
754
+ value: 49.75583333333332
755
+ - type: recall_at_100
756
+ value: 73.06908333333332
757
+ - type: recall_at_1000
758
+ value: 88.91316666666667
759
+ - type: recall_at_3
760
+ value: 36.913250000000005
761
+ - type: recall_at_5
762
+ value: 42.48641666666666
763
+ - task:
764
+ type: Retrieval
765
+ dataset:
766
+ type: BeIR/cqadupstack
767
+ name: MTEB CQADupstackStatsRetrieval
768
+ config: default
769
+ split: test
770
+ revision: None
771
+ metrics:
772
+ - type: map_at_1
773
+ value: 24.044999999999998
774
+ - type: map_at_10
775
+ value: 30.349999999999998
776
+ - type: map_at_100
777
+ value: 31.273
778
+ - type: map_at_1000
779
+ value: 31.362000000000002
780
+ - type: map_at_3
781
+ value: 28.508
782
+ - type: map_at_5
783
+ value: 29.369
784
+ - type: mrr_at_1
785
+ value: 26.994
786
+ - type: mrr_at_10
787
+ value: 33.12
788
+ - type: mrr_at_100
789
+ value: 33.904
790
+ - type: mrr_at_1000
791
+ value: 33.967000000000006
792
+ - type: mrr_at_3
793
+ value: 31.365
794
+ - type: mrr_at_5
795
+ value: 32.124
796
+ - type: ndcg_at_1
797
+ value: 26.994
798
+ - type: ndcg_at_10
799
+ value: 34.214
800
+ - type: ndcg_at_100
801
+ value: 38.681
802
+ - type: ndcg_at_1000
803
+ value: 40.926
804
+ - type: ndcg_at_3
805
+ value: 30.725
806
+ - type: ndcg_at_5
807
+ value: 31.967000000000002
808
+ - type: precision_at_1
809
+ value: 26.994
810
+ - type: precision_at_10
811
+ value: 5.215
812
+ - type: precision_at_100
813
+ value: 0.807
814
+ - type: precision_at_1000
815
+ value: 0.108
816
+ - type: precision_at_3
817
+ value: 12.986
818
+ - type: precision_at_5
819
+ value: 8.712
820
+ - type: recall_at_1
821
+ value: 24.044999999999998
822
+ - type: recall_at_10
823
+ value: 43.456
824
+ - type: recall_at_100
825
+ value: 63.675000000000004
826
+ - type: recall_at_1000
827
+ value: 80.05499999999999
828
+ - type: recall_at_3
829
+ value: 33.561
830
+ - type: recall_at_5
831
+ value: 36.767
832
+ - task:
833
+ type: Retrieval
834
+ dataset:
835
+ type: BeIR/cqadupstack
836
+ name: MTEB CQADupstackTexRetrieval
837
+ config: default
838
+ split: test
839
+ revision: None
840
+ metrics:
841
+ - type: map_at_1
842
+ value: 15.672
843
+ - type: map_at_10
844
+ value: 22.641
845
+ - type: map_at_100
846
+ value: 23.75
847
+ - type: map_at_1000
848
+ value: 23.877000000000002
849
+ - type: map_at_3
850
+ value: 20.219
851
+ - type: map_at_5
852
+ value: 21.648
853
+ - type: mrr_at_1
854
+ value: 18.823
855
+ - type: mrr_at_10
856
+ value: 26.101999999999997
857
+ - type: mrr_at_100
858
+ value: 27.038
859
+ - type: mrr_at_1000
860
+ value: 27.118
861
+ - type: mrr_at_3
862
+ value: 23.669
863
+ - type: mrr_at_5
864
+ value: 25.173000000000002
865
+ - type: ndcg_at_1
866
+ value: 18.823
867
+ - type: ndcg_at_10
868
+ value: 27.176000000000002
869
+ - type: ndcg_at_100
870
+ value: 32.42
871
+ - type: ndcg_at_1000
872
+ value: 35.413
873
+ - type: ndcg_at_3
874
+ value: 22.756999999999998
875
+ - type: ndcg_at_5
876
+ value: 25.032
877
+ - type: precision_at_1
878
+ value: 18.823
879
+ - type: precision_at_10
880
+ value: 5.034000000000001
881
+ - type: precision_at_100
882
+ value: 0.895
883
+ - type: precision_at_1000
884
+ value: 0.132
885
+ - type: precision_at_3
886
+ value: 10.771
887
+ - type: precision_at_5
888
+ value: 8.1
889
+ - type: recall_at_1
890
+ value: 15.672
891
+ - type: recall_at_10
892
+ value: 37.296
893
+ - type: recall_at_100
894
+ value: 60.863
895
+ - type: recall_at_1000
896
+ value: 82.234
897
+ - type: recall_at_3
898
+ value: 25.330000000000002
899
+ - type: recall_at_5
900
+ value: 30.964000000000002
901
+ - task:
902
+ type: Retrieval
903
+ dataset:
904
+ type: BeIR/cqadupstack
905
+ name: MTEB CQADupstackUnixRetrieval
906
+ config: default
907
+ split: test
908
+ revision: None
909
+ metrics:
910
+ - type: map_at_1
911
+ value: 24.633
912
+ - type: map_at_10
913
+ value: 32.858
914
+ - type: map_at_100
915
+ value: 34.038000000000004
916
+ - type: map_at_1000
917
+ value: 34.141
918
+ - type: map_at_3
919
+ value: 30.209000000000003
920
+ - type: map_at_5
921
+ value: 31.567
922
+ - type: mrr_at_1
923
+ value: 28.358
924
+ - type: mrr_at_10
925
+ value: 36.433
926
+ - type: mrr_at_100
927
+ value: 37.352000000000004
928
+ - type: mrr_at_1000
929
+ value: 37.41
930
+ - type: mrr_at_3
931
+ value: 34.033
932
+ - type: mrr_at_5
933
+ value: 35.246
934
+ - type: ndcg_at_1
935
+ value: 28.358
936
+ - type: ndcg_at_10
937
+ value: 37.973
938
+ - type: ndcg_at_100
939
+ value: 43.411
940
+ - type: ndcg_at_1000
941
+ value: 45.747
942
+ - type: ndcg_at_3
943
+ value: 32.934999999999995
944
+ - type: ndcg_at_5
945
+ value: 35.013
946
+ - type: precision_at_1
947
+ value: 28.358
948
+ - type: precision_at_10
949
+ value: 6.418
950
+ - type: precision_at_100
951
+ value: 1.02
952
+ - type: precision_at_1000
953
+ value: 0.133
954
+ - type: precision_at_3
955
+ value: 14.677000000000001
956
+ - type: precision_at_5
957
+ value: 10.335999999999999
958
+ - type: recall_at_1
959
+ value: 24.633
960
+ - type: recall_at_10
961
+ value: 50.048
962
+ - type: recall_at_100
963
+ value: 73.821
964
+ - type: recall_at_1000
965
+ value: 90.046
966
+ - type: recall_at_3
967
+ value: 36.284
968
+ - type: recall_at_5
969
+ value: 41.370000000000005
970
+ - task:
971
+ type: Retrieval
972
+ dataset:
973
+ type: BeIR/cqadupstack
974
+ name: MTEB CQADupstackWebmastersRetrieval
975
+ config: default
976
+ split: test
977
+ revision: None
978
+ metrics:
979
+ - type: map_at_1
980
+ value: 23.133
981
+ - type: map_at_10
982
+ value: 31.491999999999997
983
+ - type: map_at_100
984
+ value: 33.062000000000005
985
+ - type: map_at_1000
986
+ value: 33.256
987
+ - type: map_at_3
988
+ value: 28.886
989
+ - type: map_at_5
990
+ value: 30.262
991
+ - type: mrr_at_1
992
+ value: 28.063
993
+ - type: mrr_at_10
994
+ value: 36.144
995
+ - type: mrr_at_100
996
+ value: 37.14
997
+ - type: mrr_at_1000
998
+ value: 37.191
999
+ - type: mrr_at_3
1000
+ value: 33.762
1001
+ - type: mrr_at_5
1002
+ value: 34.997
1003
+ - type: ndcg_at_1
1004
+ value: 28.063
1005
+ - type: ndcg_at_10
1006
+ value: 36.951
1007
+ - type: ndcg_at_100
1008
+ value: 43.287
1009
+ - type: ndcg_at_1000
1010
+ value: 45.777
1011
+ - type: ndcg_at_3
1012
+ value: 32.786
1013
+ - type: ndcg_at_5
1014
+ value: 34.65
1015
+ - type: precision_at_1
1016
+ value: 28.063
1017
+ - type: precision_at_10
1018
+ value: 7.055
1019
+ - type: precision_at_100
1020
+ value: 1.476
1021
+ - type: precision_at_1000
1022
+ value: 0.22899999999999998
1023
+ - type: precision_at_3
1024
+ value: 15.481
1025
+ - type: precision_at_5
1026
+ value: 11.186
1027
+ - type: recall_at_1
1028
+ value: 23.133
1029
+ - type: recall_at_10
1030
+ value: 47.285
1031
+ - type: recall_at_100
1032
+ value: 76.176
1033
+ - type: recall_at_1000
1034
+ value: 92.176
1035
+ - type: recall_at_3
1036
+ value: 35.223
1037
+ - type: recall_at_5
1038
+ value: 40.142
1039
+ - task:
1040
+ type: Retrieval
1041
+ dataset:
1042
+ type: BeIR/cqadupstack
1043
+ name: MTEB CQADupstackWordpressRetrieval
1044
+ config: default
1045
+ split: test
1046
+ revision: None
1047
+ metrics:
1048
+ - type: map_at_1
1049
+ value: 19.547
1050
+ - type: map_at_10
1051
+ value: 26.374
1052
+ - type: map_at_100
1053
+ value: 27.419
1054
+ - type: map_at_1000
1055
+ value: 27.539
1056
+ - type: map_at_3
1057
+ value: 23.882
1058
+ - type: map_at_5
1059
+ value: 25.163999999999998
1060
+ - type: mrr_at_1
1061
+ value: 21.442
1062
+ - type: mrr_at_10
1063
+ value: 28.458
1064
+ - type: mrr_at_100
1065
+ value: 29.360999999999997
1066
+ - type: mrr_at_1000
1067
+ value: 29.448999999999998
1068
+ - type: mrr_at_3
1069
+ value: 25.97
1070
+ - type: mrr_at_5
1071
+ value: 27.273999999999997
1072
+ - type: ndcg_at_1
1073
+ value: 21.442
1074
+ - type: ndcg_at_10
1075
+ value: 30.897000000000002
1076
+ - type: ndcg_at_100
1077
+ value: 35.99
1078
+ - type: ndcg_at_1000
1079
+ value: 38.832
1080
+ - type: ndcg_at_3
1081
+ value: 25.944
1082
+ - type: ndcg_at_5
1083
+ value: 28.126
1084
+ - type: precision_at_1
1085
+ value: 21.442
1086
+ - type: precision_at_10
1087
+ value: 4.9910000000000005
1088
+ - type: precision_at_100
1089
+ value: 0.8109999999999999
1090
+ - type: precision_at_1000
1091
+ value: 0.11800000000000001
1092
+ - type: precision_at_3
1093
+ value: 11.029
1094
+ - type: precision_at_5
1095
+ value: 7.911
1096
+ - type: recall_at_1
1097
+ value: 19.547
1098
+ - type: recall_at_10
1099
+ value: 42.886
1100
+ - type: recall_at_100
1101
+ value: 66.64999999999999
1102
+ - type: recall_at_1000
1103
+ value: 87.368
1104
+ - type: recall_at_3
1105
+ value: 29.143
1106
+ - type: recall_at_5
1107
+ value: 34.544000000000004
1108
+ - task:
1109
+ type: Retrieval
1110
+ dataset:
1111
+ type: climate-fever
1112
+ name: MTEB ClimateFEVER
1113
+ config: default
1114
+ split: test
1115
+ revision: None
1116
+ metrics:
1117
+ - type: map_at_1
1118
+ value: 15.572
1119
+ - type: map_at_10
1120
+ value: 25.312
1121
+ - type: map_at_100
1122
+ value: 27.062
1123
+ - type: map_at_1000
1124
+ value: 27.253
1125
+ - type: map_at_3
1126
+ value: 21.601
1127
+ - type: map_at_5
1128
+ value: 23.473
1129
+ - type: mrr_at_1
1130
+ value: 34.984
1131
+ - type: mrr_at_10
1132
+ value: 46.406
1133
+ - type: mrr_at_100
1134
+ value: 47.179
1135
+ - type: mrr_at_1000
1136
+ value: 47.21
1137
+ - type: mrr_at_3
1138
+ value: 43.485
1139
+ - type: mrr_at_5
1140
+ value: 45.322
1141
+ - type: ndcg_at_1
1142
+ value: 34.984
1143
+ - type: ndcg_at_10
1144
+ value: 34.344
1145
+ - type: ndcg_at_100
1146
+ value: 41.015
1147
+ - type: ndcg_at_1000
1148
+ value: 44.366
1149
+ - type: ndcg_at_3
1150
+ value: 29.119
1151
+ - type: ndcg_at_5
1152
+ value: 30.825999999999997
1153
+ - type: precision_at_1
1154
+ value: 34.984
1155
+ - type: precision_at_10
1156
+ value: 10.358
1157
+ - type: precision_at_100
1158
+ value: 1.762
1159
+ - type: precision_at_1000
1160
+ value: 0.23900000000000002
1161
+ - type: precision_at_3
1162
+ value: 21.368000000000002
1163
+ - type: precision_at_5
1164
+ value: 15.948
1165
+ - type: recall_at_1
1166
+ value: 15.572
1167
+ - type: recall_at_10
1168
+ value: 39.367999999999995
1169
+ - type: recall_at_100
1170
+ value: 62.183
1171
+ - type: recall_at_1000
1172
+ value: 80.92200000000001
1173
+ - type: recall_at_3
1174
+ value: 26.131999999999998
1175
+ - type: recall_at_5
1176
+ value: 31.635999999999996
1177
+ - task:
1178
+ type: Retrieval
1179
+ dataset:
1180
+ type: dbpedia-entity
1181
+ name: MTEB DBPedia
1182
+ config: default
1183
+ split: test
1184
+ revision: None
1185
+ metrics:
1186
+ - type: map_at_1
1187
+ value: 8.848
1188
+ - type: map_at_10
1189
+ value: 19.25
1190
+ - type: map_at_100
1191
+ value: 27.193
1192
+ - type: map_at_1000
1193
+ value: 28.721999999999998
1194
+ - type: map_at_3
1195
+ value: 13.968
1196
+ - type: map_at_5
1197
+ value: 16.283
1198
+ - type: mrr_at_1
1199
+ value: 68.75
1200
+ - type: mrr_at_10
1201
+ value: 76.25
1202
+ - type: mrr_at_100
1203
+ value: 76.534
1204
+ - type: mrr_at_1000
1205
+ value: 76.53999999999999
1206
+ - type: mrr_at_3
1207
+ value: 74.667
1208
+ - type: mrr_at_5
1209
+ value: 75.86699999999999
1210
+ - type: ndcg_at_1
1211
+ value: 56.00000000000001
1212
+ - type: ndcg_at_10
1213
+ value: 41.426
1214
+ - type: ndcg_at_100
1215
+ value: 45.660000000000004
1216
+ - type: ndcg_at_1000
1217
+ value: 53.02
1218
+ - type: ndcg_at_3
1219
+ value: 46.581
1220
+ - type: ndcg_at_5
1221
+ value: 43.836999999999996
1222
+ - type: precision_at_1
1223
+ value: 68.75
1224
+ - type: precision_at_10
1225
+ value: 32.800000000000004
1226
+ - type: precision_at_100
1227
+ value: 10.440000000000001
1228
+ - type: precision_at_1000
1229
+ value: 1.9980000000000002
1230
+ - type: precision_at_3
1231
+ value: 49.667
1232
+ - type: precision_at_5
1233
+ value: 42.25
1234
+ - type: recall_at_1
1235
+ value: 8.848
1236
+ - type: recall_at_10
1237
+ value: 24.467
1238
+ - type: recall_at_100
1239
+ value: 51.344
1240
+ - type: recall_at_1000
1241
+ value: 75.235
1242
+ - type: recall_at_3
1243
+ value: 15.329
1244
+ - type: recall_at_5
1245
+ value: 18.892999999999997
1246
+ - task:
1247
+ type: Classification
1248
+ dataset:
1249
+ type: mteb/emotion
1250
+ name: MTEB EmotionClassification
1251
+ config: default
1252
+ split: test
1253
+ revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37
1254
+ metrics:
1255
+ - type: accuracy
1256
+ value: 48.95
1257
+ - type: f1
1258
+ value: 43.44563593360779
1259
+ - task:
1260
+ type: Retrieval
1261
+ dataset:
1262
+ type: fever
1263
+ name: MTEB FEVER
1264
+ config: default
1265
+ split: test
1266
+ revision: None
1267
+ metrics:
1268
+ - type: map_at_1
1269
+ value: 78.036
1270
+ - type: map_at_10
1271
+ value: 85.639
1272
+ - type: map_at_100
1273
+ value: 85.815
1274
+ - type: map_at_1000
1275
+ value: 85.829
1276
+ - type: map_at_3
1277
+ value: 84.795
1278
+ - type: map_at_5
1279
+ value: 85.336
1280
+ - type: mrr_at_1
1281
+ value: 84.353
1282
+ - type: mrr_at_10
1283
+ value: 90.582
1284
+ - type: mrr_at_100
1285
+ value: 90.617
1286
+ - type: mrr_at_1000
1287
+ value: 90.617
1288
+ - type: mrr_at_3
1289
+ value: 90.132
1290
+ - type: mrr_at_5
1291
+ value: 90.447
1292
+ - type: ndcg_at_1
1293
+ value: 84.353
1294
+ - type: ndcg_at_10
1295
+ value: 89.003
1296
+ - type: ndcg_at_100
1297
+ value: 89.60000000000001
1298
+ - type: ndcg_at_1000
1299
+ value: 89.836
1300
+ - type: ndcg_at_3
1301
+ value: 87.81400000000001
1302
+ - type: ndcg_at_5
1303
+ value: 88.478
1304
+ - type: precision_at_1
1305
+ value: 84.353
1306
+ - type: precision_at_10
1307
+ value: 10.482
1308
+ - type: precision_at_100
1309
+ value: 1.099
1310
+ - type: precision_at_1000
1311
+ value: 0.11399999999999999
1312
+ - type: precision_at_3
1313
+ value: 33.257999999999996
1314
+ - type: precision_at_5
1315
+ value: 20.465
1316
+ - type: recall_at_1
1317
+ value: 78.036
1318
+ - type: recall_at_10
1319
+ value: 94.517
1320
+ - type: recall_at_100
1321
+ value: 96.828
1322
+ - type: recall_at_1000
1323
+ value: 98.261
1324
+ - type: recall_at_3
1325
+ value: 91.12
1326
+ - type: recall_at_5
1327
+ value: 92.946
1328
+ - task:
1329
+ type: Retrieval
1330
+ dataset:
1331
+ type: fiqa
1332
+ name: MTEB FiQA2018
1333
+ config: default
1334
+ split: test
1335
+ revision: None
1336
+ metrics:
1337
+ - type: map_at_1
1338
+ value: 20.191
1339
+ - type: map_at_10
1340
+ value: 32.369
1341
+ - type: map_at_100
1342
+ value: 34.123999999999995
1343
+ - type: map_at_1000
1344
+ value: 34.317
1345
+ - type: map_at_3
1346
+ value: 28.71
1347
+ - type: map_at_5
1348
+ value: 30.607
1349
+ - type: mrr_at_1
1350
+ value: 40.894999999999996
1351
+ - type: mrr_at_10
1352
+ value: 48.842
1353
+ - type: mrr_at_100
1354
+ value: 49.599
1355
+ - type: mrr_at_1000
1356
+ value: 49.647000000000006
1357
+ - type: mrr_at_3
1358
+ value: 46.785
1359
+ - type: mrr_at_5
1360
+ value: 47.672
1361
+ - type: ndcg_at_1
1362
+ value: 40.894999999999996
1363
+ - type: ndcg_at_10
1364
+ value: 39.872
1365
+ - type: ndcg_at_100
1366
+ value: 46.126
1367
+ - type: ndcg_at_1000
1368
+ value: 49.476
1369
+ - type: ndcg_at_3
1370
+ value: 37.153000000000006
1371
+ - type: ndcg_at_5
1372
+ value: 37.433
1373
+ - type: precision_at_1
1374
+ value: 40.894999999999996
1375
+ - type: precision_at_10
1376
+ value: 10.818
1377
+ - type: precision_at_100
1378
+ value: 1.73
1379
+ - type: precision_at_1000
1380
+ value: 0.231
1381
+ - type: precision_at_3
1382
+ value: 25.051000000000002
1383
+ - type: precision_at_5
1384
+ value: 17.531
1385
+ - type: recall_at_1
1386
+ value: 20.191
1387
+ - type: recall_at_10
1388
+ value: 45.768
1389
+ - type: recall_at_100
1390
+ value: 68.82000000000001
1391
+ - type: recall_at_1000
1392
+ value: 89.133
1393
+ - type: recall_at_3
1394
+ value: 33.296
1395
+ - type: recall_at_5
1396
+ value: 38.022
1397
+ - task:
1398
+ type: Retrieval
1399
+ dataset:
1400
+ type: hotpotqa
1401
+ name: MTEB HotpotQA
1402
+ config: default
1403
+ split: test
1404
+ revision: None
1405
+ metrics:
1406
+ - type: map_at_1
1407
+ value: 39.257
1408
+ - type: map_at_10
1409
+ value: 61.467000000000006
1410
+ - type: map_at_100
1411
+ value: 62.364
1412
+ - type: map_at_1000
1413
+ value: 62.424
1414
+ - type: map_at_3
1415
+ value: 58.228
1416
+ - type: map_at_5
1417
+ value: 60.283
1418
+ - type: mrr_at_1
1419
+ value: 78.515
1420
+ - type: mrr_at_10
1421
+ value: 84.191
1422
+ - type: mrr_at_100
1423
+ value: 84.378
1424
+ - type: mrr_at_1000
1425
+ value: 84.385
1426
+ - type: mrr_at_3
1427
+ value: 83.284
1428
+ - type: mrr_at_5
1429
+ value: 83.856
1430
+ - type: ndcg_at_1
1431
+ value: 78.515
1432
+ - type: ndcg_at_10
1433
+ value: 69.78999999999999
1434
+ - type: ndcg_at_100
1435
+ value: 72.886
1436
+ - type: ndcg_at_1000
1437
+ value: 74.015
1438
+ - type: ndcg_at_3
1439
+ value: 65.23
1440
+ - type: ndcg_at_5
1441
+ value: 67.80199999999999
1442
+ - type: precision_at_1
1443
+ value: 78.515
1444
+ - type: precision_at_10
1445
+ value: 14.519000000000002
1446
+ - type: precision_at_100
1447
+ value: 1.694
1448
+ - type: precision_at_1000
1449
+ value: 0.184
1450
+ - type: precision_at_3
1451
+ value: 41.702
1452
+ - type: precision_at_5
1453
+ value: 27.046999999999997
1454
+ - type: recall_at_1
1455
+ value: 39.257
1456
+ - type: recall_at_10
1457
+ value: 72.59299999999999
1458
+ - type: recall_at_100
1459
+ value: 84.679
1460
+ - type: recall_at_1000
1461
+ value: 92.12
1462
+ - type: recall_at_3
1463
+ value: 62.552
1464
+ - type: recall_at_5
1465
+ value: 67.616
1466
+ - task:
1467
+ type: Classification
1468
+ dataset:
1469
+ type: mteb/imdb
1470
+ name: MTEB ImdbClassification
1471
+ config: default
1472
+ split: test
1473
+ revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7
1474
+ metrics:
1475
+ - type: accuracy
1476
+ value: 91.5152
1477
+ - type: ap
1478
+ value: 87.64584669595709
1479
+ - type: f1
1480
+ value: 91.50605576428437
1481
+ - task:
1482
+ type: Retrieval
1483
+ dataset:
1484
+ type: msmarco
1485
+ name: MTEB MSMARCO
1486
+ config: default
1487
+ split: dev
1488
+ revision: None
1489
+ metrics:
1490
+ - type: map_at_1
1491
+ value: 21.926000000000002
1492
+ - type: map_at_10
1493
+ value: 34.049
1494
+ - type: map_at_100
1495
+ value: 35.213
1496
+ - type: map_at_1000
1497
+ value: 35.265
1498
+ - type: map_at_3
1499
+ value: 30.309
1500
+ - type: map_at_5
1501
+ value: 32.407000000000004
1502
+ - type: mrr_at_1
1503
+ value: 22.55
1504
+ - type: mrr_at_10
1505
+ value: 34.657
1506
+ - type: mrr_at_100
1507
+ value: 35.760999999999996
1508
+ - type: mrr_at_1000
1509
+ value: 35.807
1510
+ - type: mrr_at_3
1511
+ value: 30.989
1512
+ - type: mrr_at_5
1513
+ value: 33.039
1514
+ - type: ndcg_at_1
1515
+ value: 22.55
1516
+ - type: ndcg_at_10
1517
+ value: 40.842
1518
+ - type: ndcg_at_100
1519
+ value: 46.436
1520
+ - type: ndcg_at_1000
1521
+ value: 47.721999999999994
1522
+ - type: ndcg_at_3
1523
+ value: 33.209
1524
+ - type: ndcg_at_5
1525
+ value: 36.943
1526
+ - type: precision_at_1
1527
+ value: 22.55
1528
+ - type: precision_at_10
1529
+ value: 6.447
1530
+ - type: precision_at_100
1531
+ value: 0.9249999999999999
1532
+ - type: precision_at_1000
1533
+ value: 0.104
1534
+ - type: precision_at_3
1535
+ value: 14.136000000000001
1536
+ - type: precision_at_5
1537
+ value: 10.381
1538
+ - type: recall_at_1
1539
+ value: 21.926000000000002
1540
+ - type: recall_at_10
1541
+ value: 61.724999999999994
1542
+ - type: recall_at_100
1543
+ value: 87.604
1544
+ - type: recall_at_1000
1545
+ value: 97.421
1546
+ - type: recall_at_3
1547
+ value: 40.944
1548
+ - type: recall_at_5
1549
+ value: 49.915
1550
+ - task:
1551
+ type: Classification
1552
+ dataset:
1553
+ type: mteb/mtop_domain
1554
+ name: MTEB MTOPDomainClassification (en)
1555
+ config: en
1556
+ split: test
1557
+ revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
1558
+ metrics:
1559
+ - type: accuracy
1560
+ value: 93.54765161878704
1561
+ - type: f1
1562
+ value: 93.3298945415573
1563
+ - task:
1564
+ type: Classification
1565
+ dataset:
1566
+ type: mteb/mtop_intent
1567
+ name: MTEB MTOPIntentClassification (en)
1568
+ config: en
1569
+ split: test
1570
+ revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
1571
+ metrics:
1572
+ - type: accuracy
1573
+ value: 75.71591427268582
1574
+ - type: f1
1575
+ value: 59.32113870474471
1576
+ - task:
1577
+ type: Classification
1578
+ dataset:
1579
+ type: mteb/amazon_massive_intent
1580
+ name: MTEB MassiveIntentClassification (en)
1581
+ config: en
1582
+ split: test
1583
+ revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
1584
+ metrics:
1585
+ - type: accuracy
1586
+ value: 75.83053127101547
1587
+ - type: f1
1588
+ value: 73.60757944876475
1589
+ - task:
1590
+ type: Classification
1591
+ dataset:
1592
+ type: mteb/amazon_massive_scenario
1593
+ name: MTEB MassiveScenarioClassification (en)
1594
+ config: en
1595
+ split: test
1596
+ revision: 7d571f92784cd94a019292a1f45445077d0ef634
1597
+ metrics:
1598
+ - type: accuracy
1599
+ value: 78.72562205783457
1600
+ - type: f1
1601
+ value: 78.63761662505502
1602
+ - task:
1603
+ type: Clustering
1604
+ dataset:
1605
+ type: mteb/medrxiv-clustering-p2p
1606
+ name: MTEB MedrxivClusteringP2P
1607
+ config: default
1608
+ split: test
1609
+ revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73
1610
+ metrics:
1611
+ - type: v_measure
1612
+ value: 33.37935633767996
1613
+ - task:
1614
+ type: Clustering
1615
+ dataset:
1616
+ type: mteb/medrxiv-clustering-s2s
1617
+ name: MTEB MedrxivClusteringS2S
1618
+ config: default
1619
+ split: test
1620
+ revision: 35191c8c0dca72d8ff3efcd72aa802307d469663
1621
+ metrics:
1622
+ - type: v_measure
1623
+ value: 31.55270546130387
1624
+ - task:
1625
+ type: Reranking
1626
+ dataset:
1627
+ type: mteb/mind_small
1628
+ name: MTEB MindSmallReranking
1629
+ config: default
1630
+ split: test
1631
+ revision: 3bdac13927fdc888b903db93b2ffdbd90b295a69
1632
+ metrics:
1633
+ - type: map
1634
+ value: 30.462692753143834
1635
+ - type: mrr
1636
+ value: 31.497569753511563
1637
+ - task:
1638
+ type: Retrieval
1639
+ dataset:
1640
+ type: nfcorpus
1641
+ name: MTEB NFCorpus
1642
+ config: default
1643
+ split: test
1644
+ revision: None
1645
+ metrics:
1646
+ - type: map_at_1
1647
+ value: 5.646
1648
+ - type: map_at_10
1649
+ value: 12.498
1650
+ - type: map_at_100
1651
+ value: 15.486
1652
+ - type: map_at_1000
1653
+ value: 16.805999999999997
1654
+ - type: map_at_3
1655
+ value: 9.325
1656
+ - type: map_at_5
1657
+ value: 10.751
1658
+ - type: mrr_at_1
1659
+ value: 43.034
1660
+ - type: mrr_at_10
1661
+ value: 52.662
1662
+ - type: mrr_at_100
1663
+ value: 53.189
1664
+ - type: mrr_at_1000
1665
+ value: 53.25
1666
+ - type: mrr_at_3
1667
+ value: 50.929
1668
+ - type: mrr_at_5
1669
+ value: 51.92
1670
+ - type: ndcg_at_1
1671
+ value: 41.796
1672
+ - type: ndcg_at_10
1673
+ value: 33.477000000000004
1674
+ - type: ndcg_at_100
1675
+ value: 29.996000000000002
1676
+ - type: ndcg_at_1000
1677
+ value: 38.864
1678
+ - type: ndcg_at_3
1679
+ value: 38.940000000000005
1680
+ - type: ndcg_at_5
1681
+ value: 36.689
1682
+ - type: precision_at_1
1683
+ value: 43.034
1684
+ - type: precision_at_10
1685
+ value: 24.799
1686
+ - type: precision_at_100
1687
+ value: 7.432999999999999
1688
+ - type: precision_at_1000
1689
+ value: 1.9929999999999999
1690
+ - type: precision_at_3
1691
+ value: 36.842000000000006
1692
+ - type: precision_at_5
1693
+ value: 32.135999999999996
1694
+ - type: recall_at_1
1695
+ value: 5.646
1696
+ - type: recall_at_10
1697
+ value: 15.963
1698
+ - type: recall_at_100
1699
+ value: 29.492
1700
+ - type: recall_at_1000
1701
+ value: 61.711000000000006
1702
+ - type: recall_at_3
1703
+ value: 10.585
1704
+ - type: recall_at_5
1705
+ value: 12.753999999999998
1706
+ - task:
1707
+ type: Retrieval
1708
+ dataset:
1709
+ type: nq
1710
+ name: MTEB NQ
1711
+ config: default
1712
+ split: test
1713
+ revision: None
1714
+ metrics:
1715
+ - type: map_at_1
1716
+ value: 27.602
1717
+ - type: map_at_10
1718
+ value: 41.545
1719
+ - type: map_at_100
1720
+ value: 42.644999999999996
1721
+ - type: map_at_1000
1722
+ value: 42.685
1723
+ - type: map_at_3
1724
+ value: 37.261
1725
+ - type: map_at_5
1726
+ value: 39.706
1727
+ - type: mrr_at_1
1728
+ value: 31.141000000000002
1729
+ - type: mrr_at_10
1730
+ value: 44.139
1731
+ - type: mrr_at_100
1732
+ value: 44.997
1733
+ - type: mrr_at_1000
1734
+ value: 45.025999999999996
1735
+ - type: mrr_at_3
1736
+ value: 40.503
1737
+ - type: mrr_at_5
1738
+ value: 42.64
1739
+ - type: ndcg_at_1
1740
+ value: 31.141000000000002
1741
+ - type: ndcg_at_10
1742
+ value: 48.995
1743
+ - type: ndcg_at_100
1744
+ value: 53.788000000000004
1745
+ - type: ndcg_at_1000
1746
+ value: 54.730000000000004
1747
+ - type: ndcg_at_3
1748
+ value: 40.844
1749
+ - type: ndcg_at_5
1750
+ value: 44.955
1751
+ - type: precision_at_1
1752
+ value: 31.141000000000002
1753
+ - type: precision_at_10
1754
+ value: 8.233
1755
+ - type: precision_at_100
1756
+ value: 1.093
1757
+ - type: precision_at_1000
1758
+ value: 0.11800000000000001
1759
+ - type: precision_at_3
1760
+ value: 18.579
1761
+ - type: precision_at_5
1762
+ value: 13.533999999999999
1763
+ - type: recall_at_1
1764
+ value: 27.602
1765
+ - type: recall_at_10
1766
+ value: 69.216
1767
+ - type: recall_at_100
1768
+ value: 90.252
1769
+ - type: recall_at_1000
1770
+ value: 97.27
1771
+ - type: recall_at_3
1772
+ value: 47.987
1773
+ - type: recall_at_5
1774
+ value: 57.438
1775
+ - task:
1776
+ type: Retrieval
1777
+ dataset:
1778
+ type: quora
1779
+ name: MTEB QuoraRetrieval
1780
+ config: default
1781
+ split: test
1782
+ revision: None
1783
+ metrics:
1784
+ - type: map_at_1
1785
+ value: 70.949
1786
+ - type: map_at_10
1787
+ value: 84.89999999999999
1788
+ - type: map_at_100
1789
+ value: 85.531
1790
+ - type: map_at_1000
1791
+ value: 85.548
1792
+ - type: map_at_3
1793
+ value: 82.027
1794
+ - type: map_at_5
1795
+ value: 83.853
1796
+ - type: mrr_at_1
1797
+ value: 81.69999999999999
1798
+ - type: mrr_at_10
1799
+ value: 87.813
1800
+ - type: mrr_at_100
1801
+ value: 87.917
1802
+ - type: mrr_at_1000
1803
+ value: 87.91799999999999
1804
+ - type: mrr_at_3
1805
+ value: 86.938
1806
+ - type: mrr_at_5
1807
+ value: 87.53999999999999
1808
+ - type: ndcg_at_1
1809
+ value: 81.75
1810
+ - type: ndcg_at_10
1811
+ value: 88.55499999999999
1812
+ - type: ndcg_at_100
1813
+ value: 89.765
1814
+ - type: ndcg_at_1000
1815
+ value: 89.871
1816
+ - type: ndcg_at_3
1817
+ value: 85.905
1818
+ - type: ndcg_at_5
1819
+ value: 87.41
1820
+ - type: precision_at_1
1821
+ value: 81.75
1822
+ - type: precision_at_10
1823
+ value: 13.403
1824
+ - type: precision_at_100
1825
+ value: 1.528
1826
+ - type: precision_at_1000
1827
+ value: 0.157
1828
+ - type: precision_at_3
1829
+ value: 37.597
1830
+ - type: precision_at_5
1831
+ value: 24.69
1832
+ - type: recall_at_1
1833
+ value: 70.949
1834
+ - type: recall_at_10
1835
+ value: 95.423
1836
+ - type: recall_at_100
1837
+ value: 99.509
1838
+ - type: recall_at_1000
1839
+ value: 99.982
1840
+ - type: recall_at_3
1841
+ value: 87.717
1842
+ - type: recall_at_5
1843
+ value: 92.032
1844
+ - task:
1845
+ type: Clustering
1846
+ dataset:
1847
+ type: mteb/reddit-clustering
1848
+ name: MTEB RedditClustering
1849
+ config: default
1850
+ split: test
1851
+ revision: 24640382cdbf8abc73003fb0fa6d111a705499eb
1852
+ metrics:
1853
+ - type: v_measure
1854
+ value: 51.76962893449579
1855
+ - task:
1856
+ type: Clustering
1857
+ dataset:
1858
+ type: mteb/reddit-clustering-p2p
1859
+ name: MTEB RedditClusteringP2P
1860
+ config: default
1861
+ split: test
1862
+ revision: 282350215ef01743dc01b456c7f5241fa8937f16
1863
+ metrics:
1864
+ - type: v_measure
1865
+ value: 62.32897690686379
1866
+ - task:
1867
+ type: Retrieval
1868
+ dataset:
1869
+ type: scidocs
1870
+ name: MTEB SCIDOCS
1871
+ config: default
1872
+ split: test
1873
+ revision: None
1874
+ metrics:
1875
+ - type: map_at_1
1876
+ value: 4.478
1877
+ - type: map_at_10
1878
+ value: 11.994
1879
+ - type: map_at_100
1880
+ value: 13.977
1881
+ - type: map_at_1000
1882
+ value: 14.295
1883
+ - type: map_at_3
1884
+ value: 8.408999999999999
1885
+ - type: map_at_5
1886
+ value: 10.024
1887
+ - type: mrr_at_1
1888
+ value: 22.1
1889
+ - type: mrr_at_10
1890
+ value: 33.526
1891
+ - type: mrr_at_100
1892
+ value: 34.577000000000005
1893
+ - type: mrr_at_1000
1894
+ value: 34.632000000000005
1895
+ - type: mrr_at_3
1896
+ value: 30.217
1897
+ - type: mrr_at_5
1898
+ value: 31.962000000000003
1899
+ - type: ndcg_at_1
1900
+ value: 22.1
1901
+ - type: ndcg_at_10
1902
+ value: 20.191
1903
+ - type: ndcg_at_100
1904
+ value: 27.954
1905
+ - type: ndcg_at_1000
1906
+ value: 33.491
1907
+ - type: ndcg_at_3
1908
+ value: 18.787000000000003
1909
+ - type: ndcg_at_5
1910
+ value: 16.378999999999998
1911
+ - type: precision_at_1
1912
+ value: 22.1
1913
+ - type: precision_at_10
1914
+ value: 10.69
1915
+ - type: precision_at_100
1916
+ value: 2.1919999999999997
1917
+ - type: precision_at_1000
1918
+ value: 0.35200000000000004
1919
+ - type: precision_at_3
1920
+ value: 17.732999999999997
1921
+ - type: precision_at_5
1922
+ value: 14.499999999999998
1923
+ - type: recall_at_1
1924
+ value: 4.478
1925
+ - type: recall_at_10
1926
+ value: 21.657
1927
+ - type: recall_at_100
1928
+ value: 44.54
1929
+ - type: recall_at_1000
1930
+ value: 71.542
1931
+ - type: recall_at_3
1932
+ value: 10.778
1933
+ - type: recall_at_5
1934
+ value: 14.687
1935
+ - task:
1936
+ type: STS
1937
+ dataset:
1938
+ type: mteb/sickr-sts
1939
+ name: MTEB SICK-R
1940
+ config: default
1941
+ split: test
1942
+ revision: a6ea5a8cab320b040a23452cc28066d9beae2cee
1943
+ metrics:
1944
+ - type: cos_sim_pearson
1945
+ value: 82.82325259156718
1946
+ - type: cos_sim_spearman
1947
+ value: 79.2463589100662
1948
+ - type: euclidean_pearson
1949
+ value: 80.48318380496771
1950
+ - type: euclidean_spearman
1951
+ value: 79.34451935199979
1952
+ - type: manhattan_pearson
1953
+ value: 80.39041824178759
1954
+ - type: manhattan_spearman
1955
+ value: 79.23002892700211
1956
+ - task:
1957
+ type: STS
1958
+ dataset:
1959
+ type: mteb/sts12-sts
1960
+ name: MTEB STS12
1961
+ config: default
1962
+ split: test
1963
+ revision: a0d554a64d88156834ff5ae9920b964011b16384
1964
+ metrics:
1965
+ - type: cos_sim_pearson
1966
+ value: 85.74130231431258
1967
+ - type: cos_sim_spearman
1968
+ value: 78.36856568042397
1969
+ - type: euclidean_pearson
1970
+ value: 82.48301631890303
1971
+ - type: euclidean_spearman
1972
+ value: 78.28376980722732
1973
+ - type: manhattan_pearson
1974
+ value: 82.43552075450525
1975
+ - type: manhattan_spearman
1976
+ value: 78.22702443947126
1977
+ - task:
1978
+ type: STS
1979
+ dataset:
1980
+ type: mteb/sts13-sts
1981
+ name: MTEB STS13
1982
+ config: default
1983
+ split: test
1984
+ revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca
1985
+ metrics:
1986
+ - type: cos_sim_pearson
1987
+ value: 79.96138619461459
1988
+ - type: cos_sim_spearman
1989
+ value: 81.85436343502379
1990
+ - type: euclidean_pearson
1991
+ value: 81.82895226665367
1992
+ - type: euclidean_spearman
1993
+ value: 82.22707349602916
1994
+ - type: manhattan_pearson
1995
+ value: 81.66303369445873
1996
+ - type: manhattan_spearman
1997
+ value: 82.05030197179455
1998
+ - task:
1999
+ type: STS
2000
+ dataset:
2001
+ type: mteb/sts14-sts
2002
+ name: MTEB STS14
2003
+ config: default
2004
+ split: test
2005
+ revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375
2006
+ metrics:
2007
+ - type: cos_sim_pearson
2008
+ value: 80.05481244198648
2009
+ - type: cos_sim_spearman
2010
+ value: 80.85052504637808
2011
+ - type: euclidean_pearson
2012
+ value: 80.86728419744497
2013
+ - type: euclidean_spearman
2014
+ value: 81.033786401512
2015
+ - type: manhattan_pearson
2016
+ value: 80.90107531061103
2017
+ - type: manhattan_spearman
2018
+ value: 81.11374116827795
2019
+ - task:
2020
+ type: STS
2021
+ dataset:
2022
+ type: mteb/sts15-sts
2023
+ name: MTEB STS15
2024
+ config: default
2025
+ split: test
2026
+ revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3
2027
+ metrics:
2028
+ - type: cos_sim_pearson
2029
+ value: 84.615220756399
2030
+ - type: cos_sim_spearman
2031
+ value: 86.46858500002092
2032
+ - type: euclidean_pearson
2033
+ value: 86.08307800247586
2034
+ - type: euclidean_spearman
2035
+ value: 86.72691443870013
2036
+ - type: manhattan_pearson
2037
+ value: 85.96155594487269
2038
+ - type: manhattan_spearman
2039
+ value: 86.605909505275
2040
+ - task:
2041
+ type: STS
2042
+ dataset:
2043
+ type: mteb/sts16-sts
2044
+ name: MTEB STS16
2045
+ config: default
2046
+ split: test
2047
+ revision: 4d8694f8f0e0100860b497b999b3dbed754a0513
2048
+ metrics:
2049
+ - type: cos_sim_pearson
2050
+ value: 82.14363913634436
2051
+ - type: cos_sim_spearman
2052
+ value: 84.48430226487102
2053
+ - type: euclidean_pearson
2054
+ value: 83.75303424801902
2055
+ - type: euclidean_spearman
2056
+ value: 84.56762380734538
2057
+ - type: manhattan_pearson
2058
+ value: 83.6135447165928
2059
+ - type: manhattan_spearman
2060
+ value: 84.39898212616731
2061
+ - task:
2062
+ type: STS
2063
+ dataset:
2064
+ type: mteb/sts17-crosslingual-sts
2065
+ name: MTEB STS17 (en-en)
2066
+ config: en-en
2067
+ split: test
2068
+ revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d
2069
+ metrics:
2070
+ - type: cos_sim_pearson
2071
+ value: 85.09909252554525
2072
+ - type: cos_sim_spearman
2073
+ value: 85.70951402743276
2074
+ - type: euclidean_pearson
2075
+ value: 87.1991936239908
2076
+ - type: euclidean_spearman
2077
+ value: 86.07745840612071
2078
+ - type: manhattan_pearson
2079
+ value: 87.25039137549952
2080
+ - type: manhattan_spearman
2081
+ value: 85.99938746659761
2082
+ - task:
2083
+ type: STS
2084
+ dataset:
2085
+ type: mteb/sts22-crosslingual-sts
2086
+ name: MTEB STS22 (en)
2087
+ config: en
2088
+ split: test
2089
+ revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80
2090
+ metrics:
2091
+ - type: cos_sim_pearson
2092
+ value: 63.529332093413615
2093
+ - type: cos_sim_spearman
2094
+ value: 65.38177340147439
2095
+ - type: euclidean_pearson
2096
+ value: 66.35278011412136
2097
+ - type: euclidean_spearman
2098
+ value: 65.47147267032997
2099
+ - type: manhattan_pearson
2100
+ value: 66.71804682408693
2101
+ - type: manhattan_spearman
2102
+ value: 65.67406521423597
2103
+ - task:
2104
+ type: STS
2105
+ dataset:
2106
+ type: mteb/stsbenchmark-sts
2107
+ name: MTEB STSBenchmark
2108
+ config: default
2109
+ split: test
2110
+ revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831
2111
+ metrics:
2112
+ - type: cos_sim_pearson
2113
+ value: 82.45802942885662
2114
+ - type: cos_sim_spearman
2115
+ value: 84.8853341842566
2116
+ - type: euclidean_pearson
2117
+ value: 84.60915021096707
2118
+ - type: euclidean_spearman
2119
+ value: 85.11181242913666
2120
+ - type: manhattan_pearson
2121
+ value: 84.38600521210364
2122
+ - type: manhattan_spearman
2123
+ value: 84.89045417981723
2124
+ - task:
2125
+ type: Reranking
2126
+ dataset:
2127
+ type: mteb/scidocs-reranking
2128
+ name: MTEB SciDocsRR
2129
+ config: default
2130
+ split: test
2131
+ revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab
2132
+ metrics:
2133
+ - type: map
2134
+ value: 85.92793380635129
2135
+ - type: mrr
2136
+ value: 95.85834191226348
2137
+ - task:
2138
+ type: Retrieval
2139
+ dataset:
2140
+ type: scifact
2141
+ name: MTEB SciFact
2142
+ config: default
2143
+ split: test
2144
+ revision: None
2145
+ metrics:
2146
+ - type: map_at_1
2147
+ value: 55.74400000000001
2148
+ - type: map_at_10
2149
+ value: 65.455
2150
+ - type: map_at_100
2151
+ value: 66.106
2152
+ - type: map_at_1000
2153
+ value: 66.129
2154
+ - type: map_at_3
2155
+ value: 62.719
2156
+ - type: map_at_5
2157
+ value: 64.441
2158
+ - type: mrr_at_1
2159
+ value: 58.667
2160
+ - type: mrr_at_10
2161
+ value: 66.776
2162
+ - type: mrr_at_100
2163
+ value: 67.363
2164
+ - type: mrr_at_1000
2165
+ value: 67.384
2166
+ - type: mrr_at_3
2167
+ value: 64.889
2168
+ - type: mrr_at_5
2169
+ value: 66.122
2170
+ - type: ndcg_at_1
2171
+ value: 58.667
2172
+ - type: ndcg_at_10
2173
+ value: 69.904
2174
+ - type: ndcg_at_100
2175
+ value: 72.807
2176
+ - type: ndcg_at_1000
2177
+ value: 73.423
2178
+ - type: ndcg_at_3
2179
+ value: 65.405
2180
+ - type: ndcg_at_5
2181
+ value: 67.86999999999999
2182
+ - type: precision_at_1
2183
+ value: 58.667
2184
+ - type: precision_at_10
2185
+ value: 9.3
2186
+ - type: precision_at_100
2187
+ value: 1.08
2188
+ - type: precision_at_1000
2189
+ value: 0.11299999999999999
2190
+ - type: precision_at_3
2191
+ value: 25.444
2192
+ - type: precision_at_5
2193
+ value: 17
2194
+ - type: recall_at_1
2195
+ value: 55.74400000000001
2196
+ - type: recall_at_10
2197
+ value: 82.122
2198
+ - type: recall_at_100
2199
+ value: 95.167
2200
+ - type: recall_at_1000
2201
+ value: 100
2202
+ - type: recall_at_3
2203
+ value: 70.14399999999999
2204
+ - type: recall_at_5
2205
+ value: 76.417
2206
+ - task:
2207
+ type: PairClassification
2208
+ dataset:
2209
+ type: mteb/sprintduplicatequestions-pairclassification
2210
+ name: MTEB SprintDuplicateQuestions
2211
+ config: default
2212
+ split: test
2213
+ revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
2214
+ metrics:
2215
+ - type: cos_sim_accuracy
2216
+ value: 99.86534653465347
2217
+ - type: cos_sim_ap
2218
+ value: 96.54142419791388
2219
+ - type: cos_sim_f1
2220
+ value: 93.07535641547861
2221
+ - type: cos_sim_precision
2222
+ value: 94.81327800829875
2223
+ - type: cos_sim_recall
2224
+ value: 91.4
2225
+ - type: dot_accuracy
2226
+ value: 99.86435643564356
2227
+ - type: dot_ap
2228
+ value: 96.53682260449868
2229
+ - type: dot_f1
2230
+ value: 92.98515104966718
2231
+ - type: dot_precision
2232
+ value: 95.27806925498426
2233
+ - type: dot_recall
2234
+ value: 90.8
2235
+ - type: euclidean_accuracy
2236
+ value: 99.86336633663366
2237
+ - type: euclidean_ap
2238
+ value: 96.5228676185697
2239
+ - type: euclidean_f1
2240
+ value: 92.9735234215886
2241
+ - type: euclidean_precision
2242
+ value: 94.70954356846472
2243
+ - type: euclidean_recall
2244
+ value: 91.3
2245
+ - type: manhattan_accuracy
2246
+ value: 99.85841584158416
2247
+ - type: manhattan_ap
2248
+ value: 96.50392760934032
2249
+ - type: manhattan_f1
2250
+ value: 92.84642321160581
2251
+ - type: manhattan_precision
2252
+ value: 92.8928928928929
2253
+ - type: manhattan_recall
2254
+ value: 92.80000000000001
2255
+ - type: max_accuracy
2256
+ value: 99.86534653465347
2257
+ - type: max_ap
2258
+ value: 96.54142419791388
2259
+ - type: max_f1
2260
+ value: 93.07535641547861
2261
+ - task:
2262
+ type: Clustering
2263
+ dataset:
2264
+ type: mteb/stackexchange-clustering
2265
+ name: MTEB StackExchangeClustering
2266
+ config: default
2267
+ split: test
2268
+ revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259
2269
+ metrics:
2270
+ - type: v_measure
2271
+ value: 61.08285408766616
2272
+ - task:
2273
+ type: Clustering
2274
+ dataset:
2275
+ type: mteb/stackexchange-clustering-p2p
2276
+ name: MTEB StackExchangeClusteringP2P
2277
+ config: default
2278
+ split: test
2279
+ revision: 815ca46b2622cec33ccafc3735d572c266efdb44
2280
+ metrics:
2281
+ - type: v_measure
2282
+ value: 35.640675309010604
2283
+ - task:
2284
+ type: Reranking
2285
+ dataset:
2286
+ type: mteb/stackoverflowdupquestions-reranking
2287
+ name: MTEB StackOverflowDupQuestions
2288
+ config: default
2289
+ split: test
2290
+ revision: e185fbe320c72810689fc5848eb6114e1ef5ec69
2291
+ metrics:
2292
+ - type: map
2293
+ value: 53.20333913710715
2294
+ - type: mrr
2295
+ value: 54.088813555725324
2296
+ - task:
2297
+ type: Summarization
2298
+ dataset:
2299
+ type: mteb/summeval
2300
+ name: MTEB SummEval
2301
+ config: default
2302
+ split: test
2303
+ revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c
2304
+ metrics:
2305
+ - type: cos_sim_pearson
2306
+ value: 30.79465221925075
2307
+ - type: cos_sim_spearman
2308
+ value: 30.530816059163634
2309
+ - type: dot_pearson
2310
+ value: 31.364837244718043
2311
+ - type: dot_spearman
2312
+ value: 30.79726823684003
2313
+ - task:
2314
+ type: Retrieval
2315
+ dataset:
2316
+ type: trec-covid
2317
+ name: MTEB TRECCOVID
2318
+ config: default
2319
+ split: test
2320
+ revision: None
2321
+ metrics:
2322
+ - type: map_at_1
2323
+ value: 0.22599999999999998
2324
+ - type: map_at_10
2325
+ value: 1.735
2326
+ - type: map_at_100
2327
+ value: 8.978
2328
+ - type: map_at_1000
2329
+ value: 20.851
2330
+ - type: map_at_3
2331
+ value: 0.613
2332
+ - type: map_at_5
2333
+ value: 0.964
2334
+ - type: mrr_at_1
2335
+ value: 88
2336
+ - type: mrr_at_10
2337
+ value: 92.867
2338
+ - type: mrr_at_100
2339
+ value: 92.867
2340
+ - type: mrr_at_1000
2341
+ value: 92.867
2342
+ - type: mrr_at_3
2343
+ value: 92.667
2344
+ - type: mrr_at_5
2345
+ value: 92.667
2346
+ - type: ndcg_at_1
2347
+ value: 82
2348
+ - type: ndcg_at_10
2349
+ value: 73.164
2350
+ - type: ndcg_at_100
2351
+ value: 51.878
2352
+ - type: ndcg_at_1000
2353
+ value: 44.864
2354
+ - type: ndcg_at_3
2355
+ value: 79.184
2356
+ - type: ndcg_at_5
2357
+ value: 76.39
2358
+ - type: precision_at_1
2359
+ value: 88
2360
+ - type: precision_at_10
2361
+ value: 76.2
2362
+ - type: precision_at_100
2363
+ value: 52.459999999999994
2364
+ - type: precision_at_1000
2365
+ value: 19.692
2366
+ - type: precision_at_3
2367
+ value: 82.667
2368
+ - type: precision_at_5
2369
+ value: 80
2370
+ - type: recall_at_1
2371
+ value: 0.22599999999999998
2372
+ - type: recall_at_10
2373
+ value: 1.942
2374
+ - type: recall_at_100
2375
+ value: 12.342
2376
+ - type: recall_at_1000
2377
+ value: 41.42
2378
+ - type: recall_at_3
2379
+ value: 0.637
2380
+ - type: recall_at_5
2381
+ value: 1.034
2382
+ - task:
2383
+ type: Retrieval
2384
+ dataset:
2385
+ type: webis-touche2020
2386
+ name: MTEB Touche2020
2387
+ config: default
2388
+ split: test
2389
+ revision: None
2390
+ metrics:
2391
+ - type: map_at_1
2392
+ value: 3.567
2393
+ - type: map_at_10
2394
+ value: 13.116
2395
+ - type: map_at_100
2396
+ value: 19.39
2397
+ - type: map_at_1000
2398
+ value: 20.988
2399
+ - type: map_at_3
2400
+ value: 7.109
2401
+ - type: map_at_5
2402
+ value: 9.950000000000001
2403
+ - type: mrr_at_1
2404
+ value: 42.857
2405
+ - type: mrr_at_10
2406
+ value: 57.404999999999994
2407
+ - type: mrr_at_100
2408
+ value: 58.021
2409
+ - type: mrr_at_1000
2410
+ value: 58.021
2411
+ - type: mrr_at_3
2412
+ value: 54.762
2413
+ - type: mrr_at_5
2414
+ value: 56.19
2415
+ - type: ndcg_at_1
2416
+ value: 38.775999999999996
2417
+ - type: ndcg_at_10
2418
+ value: 30.359
2419
+ - type: ndcg_at_100
2420
+ value: 41.284
2421
+ - type: ndcg_at_1000
2422
+ value: 52.30200000000001
2423
+ - type: ndcg_at_3
2424
+ value: 36.744
2425
+ - type: ndcg_at_5
2426
+ value: 34.326
2427
+ - type: precision_at_1
2428
+ value: 42.857
2429
+ - type: precision_at_10
2430
+ value: 26.122
2431
+ - type: precision_at_100
2432
+ value: 8.082
2433
+ - type: precision_at_1000
2434
+ value: 1.559
2435
+ - type: precision_at_3
2436
+ value: 40.136
2437
+ - type: precision_at_5
2438
+ value: 35.510000000000005
2439
+ - type: recall_at_1
2440
+ value: 3.567
2441
+ - type: recall_at_10
2442
+ value: 19.045
2443
+ - type: recall_at_100
2444
+ value: 49.979
2445
+ - type: recall_at_1000
2446
+ value: 84.206
2447
+ - type: recall_at_3
2448
+ value: 8.52
2449
+ - type: recall_at_5
2450
+ value: 13.103000000000002
2451
+ - task:
2452
+ type: Classification
2453
+ dataset:
2454
+ type: mteb/toxic_conversations_50k
2455
+ name: MTEB ToxicConversationsClassification
2456
+ config: default
2457
+ split: test
2458
+ revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c
2459
+ metrics:
2460
+ - type: accuracy
2461
+ value: 68.8394
2462
+ - type: ap
2463
+ value: 13.454399712443099
2464
+ - type: f1
2465
+ value: 53.04963076364322
2466
+ - task:
2467
+ type: Classification
2468
+ dataset:
2469
+ type: mteb/tweet_sentiment_extraction
2470
+ name: MTEB TweetSentimentExtractionClassification
2471
+ config: default
2472
+ split: test
2473
+ revision: d604517c81ca91fe16a244d1248fc021f9ecee7a
2474
+ metrics:
2475
+ - type: accuracy
2476
+ value: 60.546123372948514
2477
+ - type: f1
2478
+ value: 60.86952793277713
2479
+ - task:
2480
+ type: Clustering
2481
+ dataset:
2482
+ type: mteb/twentynewsgroups-clustering
2483
+ name: MTEB TwentyNewsgroupsClustering
2484
+ config: default
2485
+ split: test
2486
+ revision: 6125ec4e24fa026cec8a478383ee943acfbd5449
2487
+ metrics:
2488
+ - type: v_measure
2489
+ value: 49.10042955060234
2490
+ - task:
2491
+ type: PairClassification
2492
+ dataset:
2493
+ type: mteb/twittersemeval2015-pairclassification
2494
+ name: MTEB TwitterSemEval2015
2495
+ config: default
2496
+ split: test
2497
+ revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
2498
+ metrics:
2499
+ - type: cos_sim_accuracy
2500
+ value: 85.03308100375514
2501
+ - type: cos_sim_ap
2502
+ value: 71.08284605869684
2503
+ - type: cos_sim_f1
2504
+ value: 65.42539436255494
2505
+ - type: cos_sim_precision
2506
+ value: 64.14807302231237
2507
+ - type: cos_sim_recall
2508
+ value: 66.75461741424802
2509
+ - type: dot_accuracy
2510
+ value: 84.68736961316088
2511
+ - type: dot_ap
2512
+ value: 69.20524036530992
2513
+ - type: dot_f1
2514
+ value: 63.54893953365829
2515
+ - type: dot_precision
2516
+ value: 63.45698500394633
2517
+ - type: dot_recall
2518
+ value: 63.641160949868066
2519
+ - type: euclidean_accuracy
2520
+ value: 85.07480479227513
2521
+ - type: euclidean_ap
2522
+ value: 71.14592761009864
2523
+ - type: euclidean_f1
2524
+ value: 65.43814432989691
2525
+ - type: euclidean_precision
2526
+ value: 63.95465994962216
2527
+ - type: euclidean_recall
2528
+ value: 66.99208443271768
2529
+ - type: manhattan_accuracy
2530
+ value: 85.06288370984085
2531
+ - type: manhattan_ap
2532
+ value: 71.07289742593868
2533
+ - type: manhattan_f1
2534
+ value: 65.37585421412301
2535
+ - type: manhattan_precision
2536
+ value: 62.816147859922175
2537
+ - type: manhattan_recall
2538
+ value: 68.15303430079156
2539
+ - type: max_accuracy
2540
+ value: 85.07480479227513
2541
+ - type: max_ap
2542
+ value: 71.14592761009864
2543
+ - type: max_f1
2544
+ value: 65.43814432989691
2545
+ - task:
2546
+ type: PairClassification
2547
+ dataset:
2548
+ type: mteb/twitterurlcorpus-pairclassification
2549
+ name: MTEB TwitterURLCorpus
2550
+ config: default
2551
+ split: test
2552
+ revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
2553
+ metrics:
2554
+ - type: cos_sim_accuracy
2555
+ value: 87.79058485659952
2556
+ - type: cos_sim_ap
2557
+ value: 83.7183187008759
2558
+ - type: cos_sim_f1
2559
+ value: 75.86921142180798
2560
+ - type: cos_sim_precision
2561
+ value: 73.00683371298405
2562
+ - type: cos_sim_recall
2563
+ value: 78.96519864490298
2564
+ - type: dot_accuracy
2565
+ value: 87.0085768618776
2566
+ - type: dot_ap
2567
+ value: 81.87467488474279
2568
+ - type: dot_f1
2569
+ value: 74.04188363990559
2570
+ - type: dot_precision
2571
+ value: 72.10507114191901
2572
+ - type: dot_recall
2573
+ value: 76.08561749307053
2574
+ - type: euclidean_accuracy
2575
+ value: 87.8332751193387
2576
+ - type: euclidean_ap
2577
+ value: 83.83585648120315
2578
+ - type: euclidean_f1
2579
+ value: 76.02582177042369
2580
+ - type: euclidean_precision
2581
+ value: 73.36388371759989
2582
+ - type: euclidean_recall
2583
+ value: 78.88820449645827
2584
+ - type: manhattan_accuracy
2585
+ value: 87.87208444910156
2586
+ - type: manhattan_ap
2587
+ value: 83.8101950642973
2588
+ - type: manhattan_f1
2589
+ value: 75.90454195535027
2590
+ - type: manhattan_precision
2591
+ value: 72.44419564761039
2592
+ - type: manhattan_recall
2593
+ value: 79.71204188481676
2594
+ - type: max_accuracy
2595
+ value: 87.87208444910156
2596
+ - type: max_ap
2597
+ value: 83.83585648120315
2598
+ - type: max_f1
2599
+ value: 76.02582177042369
2600
  license: mit
2601
+ language:
2602
+ - en
2603
+ pipeline_tag: sentence-similarity
2604
  ---
2605
+
2606
+
2607
+
2608
+ <h1 align="center">FlagEmbedding</h1>
2609
+
2610
+
2611
+ <h4 align="center">
2612
+ <p>
2613
+ <a href=#model-list>Model List</a> |
2614
+ <a href=#usage>Usage</a> |
2615
+ <a href="#evaluation">Evaluation</a> |
2616
+ <a href="#train">Train</a> |
2617
+ <a href="#license">License</a>
2618
+ <p>
2619
+ </h4>
2620
+
2621
+ For more details please refer to our GitHub repo: [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding).
2622
+
2623
+ [English](README.md) | [中文](https://github.com/FlagOpen/FlagEmbedding/blob/master/README_zh.md)
2624
+
2625
+ FlagEmbedding can map any text to a low-dimensional dense vector which can be used for tasks like retrieval, classification, clustering, or semantic search.
2626
+ And it also can be used in vector databases for LLMs.
2627
+
2628
+ ************* 🌟**Updates**🌟 *************
2629
+ - 08/05/2023: Release base-scale and small-scale models, **best performance among the models of the same size 🤗**
2630
+ - 08/02/2023: Release `bge-large-*`(short for BAAI General Embedding) Models, **rank 1st on MTEB and C-MTEB benchmark!** :tada: :tada:
2631
+ - 08/01/2023: We release the [Chinese Massive Text Embedding Benchmark](https://github.com/FlagOpen/FlagEmbedding/blob/master/C_MTEB) (**C-MTEB**), consisting of 31 test dataset.
2632
+
2633
+
2634
+ ## Model List
2635
+
2636
+ `bge` is short for `BAAI general embedding`.
2637
+
2638
+ | Model | Language | Description | query instruction for retrieval |
2639
+ |:-------------------------------|:--------:| :--------:| :--------:|
2640
+ | [BAAI/bge-large-en](https://huggingface.co/BAAI/bge-large-en) | English | :trophy: rank **1st** in [MTEB](https://huggingface.co/spaces/mteb/leaderboard) leaderboard | `Represent this sentence for searching relevant passages: ` |
2641
+ | [BAAI/bge-base-en](https://huggingface.co/BAAI/bge-base-en) | English | rank **2nd** in [MTEB](https://huggingface.co/spaces/mteb/leaderboard) leaderboard | `Represent this sentence for searching relevant passages: ` |
2642
+ | [BAAI/bge-small-en](https://huggingface.co/BAAI/bge-small-en) | English | a small-scale model but with competitive performance | `Represent this sentence for searching relevant passages: ` |
2643
+ | [BAAI/bge-large-zh](https://huggingface.co/BAAI/bge-large-zh) | Chinese | :trophy: rank **1st** in [C-MTEB](https://github.com/FlagOpen/FlagEmbedding/tree/master/C_MTEB) benchmark | `为这个句子生成表示以用于检索相关文章:` |
2644
+ | [BAAI/bge-large-zh-noinstruct](https://huggingface.co/BAAI/bge-large-zh-noinstruct) | Chinese | This model is trained without instruction, and rank **2nd** in [C-MTEB](https://github.com/FlagOpen/FlagEmbedding/tree/master/C_MTEB) benchmark | |
2645
+ | [BAAI/bge-base-zh](https://huggingface.co/BAAI/bge-base-zh) | Chinese | a base-scale model but has similar ability with `bge-large-zh` | `为这个句子生成表示以用于检索相关文章:` |
2646
+ | [BAAI/bge-small-zh](https://huggingface.co/BAAI/bge-small-zh) | Chinese | a small-scale model but with competitive performance | `为这个句子生成表示以用于检索相关文章:` |
2647
+
2648
+
2649
+
2650
+ ## Usage
2651
+
2652
+ * **Using FlagEmbedding**
2653
+ ```
2654
+ pip install -U FlagEmbedding
2655
+ ```
2656
+ See [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding/blob/master/FlagEmbedding/baai_general_embedding/README.md) for more methods to install FlagEmbedding.
2657
+
2658
+ ```python
2659
+ from FlagEmbedding import FlagModel
2660
+ sentences = ["样例数据-1", "样例数据-2"]
2661
+ model = FlagModel('BAAI/bge-large-zh', query_instruction_for_retrieval="为这个句子生成表示以用于检索相关文章:")
2662
+ embeddings = model.encode(sentences)
2663
+ print(embeddings)
2664
+
2665
+ # for retrieval task, please use encode_queries() which will automatically add the instruction to each query
2666
+ # corpus in retrieval task can still use encode() or encode_corpus()
2667
+ queries = ['query_1', 'query_2']
2668
+ passages = ["样例段落-1", "样例段落-2"]
2669
+ q_embeddings = model.encode_queries(queries)
2670
+ p_embeddings = model.encode(passages)
2671
+ scores = q_embeddings @ p_embeddings.T
2672
+ ```
2673
+ The value of argument `query_instruction_for_retrieval` see [Model List](https://github.com/FlagOpen/FlagEmbedding/tree/master#model-list).
2674
+
2675
+ FlagModel will use all available GPUs when encoding, please set `os.environ["CUDA_VISIBLE_DEVICES"]` to choose GPU.
2676
+
2677
+
2678
+ * **Using Sentence-Transformers**
2679
+
2680
+ Using this model also is easy when you have [sentence-transformers](https://www.SBERT.net) installed:
2681
+
2682
+ ```
2683
+ pip install -U sentence-transformers
2684
+ ```
2685
+ ```python
2686
+ from sentence_transformers import SentenceTransformer
2687
+ sentences = ["样例数据-1", "样例数据-2"]
2688
+ model = SentenceTransformer('BAAI/bge-large-zh')
2689
+ embeddings = model.encode(sentences, normalize_embeddings=True)
2690
+ print(embeddings)
2691
+ ```
2692
+ For retrieval task,
2693
+ each query should start with an instruction (instructions see [Model List](https://github.com/FlagOpen/FlagEmbedding/tree/master#model-list)).
2694
+ ```python
2695
+ from sentence_transformers import SentenceTransformer
2696
+ queries = ["手机开不了机怎么办?"]
2697
+ passages = ["样例段落-1", "样例段落-2"]
2698
+ instruction = "为这个句子生成表示以用于检索相关文章:"
2699
+
2700
+ model = SentenceTransformer('BAAI/bge-large-zh')
2701
+ q_embeddings = model.encode([instruction+q for q in queries], normalize_embeddings=True)
2702
+ p_embeddings = model.encode(passages, normalize_embeddings=True)
2703
+ scores = q_embeddings @ p_embeddings.T
2704
+ ```
2705
+
2706
+ * **Using HuggingFace Transformers**
2707
+
2708
+ With transformers package, you can use the model like this: First, you pass your input through the transformer model, then you select the last hidden state of first token (i.e., [CLS]) as the sentence embedding.
2709
+
2710
+ ```python
2711
+ from transformers import AutoTokenizer, AutoModel
2712
+ import torch
2713
+ # Sentences we want sentence embeddings for
2714
+ sentences = ["样例数据-1", "样例数据-2"]
2715
+
2716
+ # Load model from HuggingFace Hub
2717
+ tokenizer = AutoTokenizer.from_pretrained('BAAI/bge-large-zh')
2718
+ model = AutoModel.from_pretrained('BAAI/bge-large-zh')
2719
+
2720
+ # Tokenize sentences
2721
+ encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
2722
+ # for retrieval task, add an instruction to query
2723
+ # encoded_input = tokenizer([instruction + q for q in queries], padding=True, truncation=True, return_tensors='pt')
2724
+
2725
+ # Compute token embeddings
2726
+ with torch.no_grad():
2727
+ model_output = model(**encoded_input)
2728
+ # Perform pooling. In this case, cls pooling.
2729
+ sentence_embeddings = model_output[0][:, 0]
2730
+ # normalize embeddings
2731
+ sentence_embeddings = torch.nn.functional.normalize(sentence_embeddings, p=2, dim=1)
2732
+ print("Sentence embeddings:", sentence_embeddings)
2733
+ ```
2734
+
2735
+
2736
+ ## Evaluation
2737
+ `baai-general-embedding` models achieve **state-of-the-art performance on both MTEB and C-MTEB leaderboard!**
2738
+ More details and evaluation tools see our [scripts](https://github.com/FlagOpen/FlagEmbedding/blob/master/C_MTEB/README.md).
2739
+
2740
+ - **MTEB**:
2741
+
2742
+ | Model Name | Dimension | Sequence Length | Average (56) | Retrieval (15) |Clustering (11) | Pair Classification (3) | Reranking (4) | STS (10) | Summarization (1) | Classification (12) |
2743
+ |:----:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
2744
+ | [**bge-large-en**](https://huggingface.co/BAAI/bge-large-en) | 1024 | 512 | **63.98** | **53.9** | **46.98** | 85.8 | **59.48** | 81.56 | 32.06 | **76.21** |
2745
+ | [**bge-base-en**](https://huggingface.co/BAAI/bge-base-en) | 768 | 512 | 63.36 | 53.0 | 46.32 | 85.86 | 58.7 | 81.84 | 29.27 | 75.27 |
2746
+ | [gte-large](https://huggingface.co/thenlper/gte-large) | 1024 | 512 | 63.13 | 52.22 | 46.84 | 85.00 | 59.13 | 83.35 | 31.66 | 73.33 |
2747
+ | [gte-base](https://huggingface.co/thenlper/gte-base) | 768 | 512 | 62.39 | 51.14 | 46.2 | 84.57 | 58.61 | 82.3 | 31.17 | 73.01 |
2748
+ | [e5-large-v2](https://huggingface.co/intfloat/e5-large-v2) | 1024| 512 | 62.25 | 50.56 | 44.49 | 86.03 | 56.61 | 82.05 | 30.19 | 75.24 |
2749
+ | [**bge-small-en**](https://huggingface.co/BAAI/bge-small-en) | 384 | 512 | 62.11 | 51.82 | 44.31 | 83.78 | 57.97 | 80.72 | 30.53 | 74.37 |
2750
+ | [instructor-xl](https://huggingface.co/hkunlp/instructor-xl) | 768 | 512 | 61.79 | 49.26 | 44.74 | 86.62 | 57.29 | 83.06 | 32.32 | 61.79 |
2751
+ | [e5-base-v2](https://huggingface.co/intfloat/e5-base-v2) | 768 | 512 | 61.5 | 50.29 | 43.80 | 85.73 | 55.91 | 81.05 | 30.28 | 73.84 |
2752
+ | [gte-small](https://huggingface.co/thenlper/gte-small) | 384 | 512 | 61.36 | 49.46 | 44.89 | 83.54 | 57.7 | 82.07 | 30.42 | 72.31 |
2753
+ | [text-embedding-ada-002](https://platform.openai.com/docs/guides/embeddings) | 1536 | 8192 | 60.99 | 49.25 | 45.9 | 84.89 | 56.32 | 80.97 | 30.8 | 70.93 |
2754
+ | [e5-small-v2](https://huggingface.co/intfloat/e5-base-v2) | 384 | 512 | 59.93 | 49.04 | 39.92 | 84.67 | 54.32 | 80.39 | 31.16 | 72.94 |
2755
+ | [sentence-t5-xxl](https://huggingface.co/sentence-transformers/sentence-t5-xxl) | 768 | 512 | 59.51 | 42.24 | 43.72 | 85.06 | 56.42 | 82.63 | 30.08 | 73.42 |
2756
+ | [all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) | 768 | 514 | 57.78 | 43.81 | 43.69 | 83.04 | 59.36 | 80.28 | 27.49 | 65.07 |
2757
+ | [sgpt-bloom-7b1-msmarco](https://huggingface.co/bigscience/sgpt-bloom-7b1-msmarco) | 4096 | 2048 | 57.59 | 48.22 | 38.93 | 81.9 | 55.65 | 77.74 | 33.6 | 66.19 |
2758
+ | [all-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L12-v2) | 384 | 512 | 56.53 | 42.69 | 41.81 | 82.41 | 58.44 | 79.8 | 27.9 | 63.21 |
2759
+ | [all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) | 384 | 512 | 56.26 | 41.95 | 42.35 | 82.37 | 58.04 | 78.9 | 30.81 | 63.05 |
2760
+ | [contriever-base-msmarco](https://huggingface.co/nthakur/contriever-base-msmarco) | 768 | 512 | 56.00 | 41.88 | 41.1 | 82.54 | 53.14 | 76.51 | 30.36 | 66.68 |
2761
+ | [sentence-t5-base](https://huggingface.co/sentence-transformers/sentence-t5-base) | 768 | 512 | 55.27 | 33.63 | 40.21 | 85.18 | 53.09 | 81.14 | 31.39 | 69.81 |
2762
+
2763
+
2764
+
2765
+ - **C-MTEB**:
2766
+ We create a benchmark C-MTEB for Chinese text embedding which consists of 31 datasets from 6 tasks.
2767
+ Please refer to [C_MTEB](https://github.com/FlagOpen/FlagEmbedding/blob/master/C_MTEB/README.md) for a detailed introduction.
2768
+
2769
+ | Model | Embedding dimension | Avg | Retrieval | STS | PairClassification | Classification | Reranking | Clustering |
2770
+ |:-------------------------------|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|
2771
+ | [**bge-large-zh**](https://huggingface.co/BAAI/bge-large-zh) | 1024 | **64.20** | **71.53** | **53.23** | **78.94** | 72.26 | **65.11** | 48.39 |
2772
+ | [**bge-large-zh-noinstruct**](https://huggingface.co/BAAI/bge-large-zh-noinstruct) | 1024 | 63.53 | 70.55 | 50.98 | 76.77 | **72.49** | 64.91 | **50.01** |
2773
+ | [**BAAI/bge-base-zh**](https://huggingface.co/BAAI/bge-base-zh) | 768 | 62.96 | 69.53 | 52.05 | 77.5 | 70.98 | 64.91 | 47.63 |
2774
+ | [**BAAI/bge-small-zh**](https://huggingface.co/BAAI/bge-small-zh) | 512 | 58.27 | 63.07 | 46.87 | 70.35 | 67.78 | 61.48 | 45.09 |
2775
+ | [m3e-base](https://huggingface.co/moka-ai/m3e-base) | 768 | 57.10 |56.91 | 48.15 | 63.99 | 70.28 | 59.34 | 47.68 |
2776
+ | [m3e-large](https://huggingface.co/moka-ai/m3e-large) | 1024 | 57.05 |54.75 | 48.64 | 64.3 | 71.22 | 59.66 | 48.88 |
2777
+ | [text-embedding-ada-002(OpenAI)](https://platform.openai.com/docs/guides/embeddings/what-are-embeddings) | 1536 | 53.02 | 52.0 | 40.61 | 69.56 | 67.38 | 54.28 | 45.68 |
2778
+ | [luotuo](https://huggingface.co/silk-road/luotuo-bert-medium) | 1024 | 49.37 | 44.4 | 39.41 | 66.62 | 65.29 | 49.25 | 44.39 |
2779
+ | [text2vec](https://huggingface.co/shibing624/text2vec-base-chinese) | 768 | 47.63 | 38.79 | 41.71 | 67.41 | 65.18 | 49.45 | 37.66 |
2780
+ | [text2vec-large](https://huggingface.co/GanymedeNil/text2vec-large-chinese) | 1024 | 47.36 | 41.94 | 41.98 | 70.86 | 63.42 | 49.16 | 30.02 |
2781
+
2782
+
2783
+
2784
+ ## Train
2785
+ This section will introduce the way we used to train the general embedding.
2786
+ The training scripts are in [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding/blob/master/FlagEmbedding/baai_general_embedding/README.md),
2787
+ and we provide some examples to do [pre-train](https://github.com/FlagOpen/FlagEmbedding/blob/master/examples/pretrain/README.md) and [fine-tune](https://github.com/FlagOpen/FlagEmbedding/blob/master/examples/finetune/README.md).
2788
+
2789
+
2790
+ **1. RetroMAE Pre-train**
2791
+ We pre-train the model following the method [retromae](https://github.com/staoxiao/RetroMAE),
2792
+ which shows promising improvement in retrieval task ([paper](https://aclanthology.org/2022.emnlp-main.35.pdf)).
2793
+ The pre-training was conducted on 24 A100(40G) GPUs with a batch size of 720.
2794
+ In retromae, the mask ratio of encoder and decoder are 0.3, and 0.5 respectively.
2795
+ We used the AdamW optimizer and the learning rate is 2e-5.
2796
+
2797
+ **Pre-training data**:
2798
+ - English:
2799
+ - [Pile](https://pile.eleuther.ai/)
2800
+ - [wikipedia](https://huggingface.co/datasets/wikipedia)
2801
+ - [msmarco](https://huggingface.co/datasets/Tevatron/msmarco-passage-corpus)
2802
+ - Chinese:
2803
+ - Subset of [wudao](https://github.com/BAAI-WuDao/Data)
2804
+ - [baidu-baike](https://baike.baidu.com/)
2805
+
2806
+
2807
+ **2. Finetune**
2808
+ We fine-tune the model using a contrastive objective.
2809
+ The format of input data is a triple`(query, positive, negative)`.
2810
+ Besides the negative in the triple, we also adopt in-batch negatives strategy.
2811
+ We employ the cross-device negatives sharing method to share negatives among different GPUs,
2812
+ which can dramatically **increase the number of negatives**.
2813
+
2814
+ We trained our model on 48 A100(40G) GPUs with a large batch size of 32,768 (so there are **65,535** negatives for each query in a batch).
2815
+ We used the AdamW optimizer and the learning rate is 1e-5.
2816
+ The temperature for contrastive loss is 0.01.
2817
+
2818
+ For the version with `*-instrcution`, we add instruction to the query for retrieval task in the training.
2819
+ For english, the instruction is `Represent this sentence for searching relevant passages: `;
2820
+ For chinese, the instruction is `为这个句子生成表示以用于检索相关文章:`.
2821
+ In the evaluation, the instruction should be added for sentence to passages retrieval task, not be added for other tasks.
2822
+
2823
+
2824
+ The finetune script is accessible in this repository: [FlagEmbedding](https://github.com/FlagOpen/FlagEmbedding/blob/master/FlagEmbedding/baai_general_embedding/README.md).
2825
+ You can easily finetune your model with it.
2826
+
2827
+ **Training data**:
2828
+
2829
+ - For English, we collect 230M text pairs from [wikipedia](https://huggingface.co/datasets/wikipedia), [cc-net](https://github.com/facebookresearch/cc_net), and so on.
2830
+
2831
+ - For chinese, we collect 120M text pairs from [wudao](https://github.com/BAAI-WuDao/Data), [simclue](https://github.com/CLUEbenchmark/SimCLUE) and so on.
2832
+
2833
+ **The data collection is to be released in the future.**
2834
+
2835
+ We will continually update the embedding models and training codes,
2836
+ hoping to promote the development of the embedding model community.
2837
+
2838
+ ## License
2839
+ FlagEmbedding is licensed under [MIT License](https://github.com/FlagOpen/FlagEmbedding/blob/master/LICENSE). The released models can be used for commercial purposes free of charge.
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/root/.cache/torch/sentence_transformers/BAAI_bge-small-en/",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "id2label": {
12
+ "0": "LABEL_0"
13
+ },
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 1536,
16
+ "label2id": {
17
+ "LABEL_0": 0
18
+ },
19
+ "layer_norm_eps": 1e-12,
20
+ "max_position_embeddings": 512,
21
+ "model_type": "bert",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 0,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.30.0",
28
+ "type_vocab_size": 2,
29
+ "use_cache": true,
30
+ "vocab_size": 30522
31
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.2.2",
4
+ "transformers": "4.28.1",
5
+ "pytorch": "1.13.0+cu117"
6
+ }
7
+ }
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ }
14
+ ]
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84868a67d847f7f37d1df745b73a4c20b5bc0a797c87a23b265f8a0131728b88
3
+ size 133508397
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": true
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "clean_up_tokenization_spaces": true,
3
+ "cls_token": "[CLS]",
4
+ "do_basic_tokenize": true,
5
+ "do_lower_case": true,
6
+ "mask_token": "[MASK]",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "never_split": null,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff