robinroy03 commited on
Commit
76df3ac
1 Parent(s): 82b5172

embedding model, for some reason HF download always crashes

Browse files
embedding_model/1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 1024,
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
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
embedding_model/README.md ADDED
@@ -0,0 +1,2779 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - allenai/c4
4
+ library_name: transformers
5
+ tags:
6
+ - sentence-transformers
7
+ - gte
8
+ - mteb
9
+ - transformers.js
10
+ - sentence-similarity
11
+ license: apache-2.0
12
+ language:
13
+ - en
14
+ model-index:
15
+ - name: gte-large-en-v1.5
16
+ results:
17
+ - task:
18
+ type: Classification
19
+ dataset:
20
+ type: mteb/amazon_counterfactual
21
+ name: MTEB AmazonCounterfactualClassification (en)
22
+ config: en
23
+ split: test
24
+ revision: e8379541af4e31359cca9fbcf4b00f2671dba205
25
+ metrics:
26
+ - type: accuracy
27
+ value: 73.01492537313432
28
+ - type: ap
29
+ value: 35.05341696659522
30
+ - type: f1
31
+ value: 66.71270310883853
32
+ - task:
33
+ type: Classification
34
+ dataset:
35
+ type: mteb/amazon_polarity
36
+ name: MTEB AmazonPolarityClassification
37
+ config: default
38
+ split: test
39
+ revision: e2d317d38cd51312af73b3d32a06d1a08b442046
40
+ metrics:
41
+ - type: accuracy
42
+ value: 93.97189999999999
43
+ - type: ap
44
+ value: 90.5952493948908
45
+ - type: f1
46
+ value: 93.95848137716877
47
+ - task:
48
+ type: Classification
49
+ dataset:
50
+ type: mteb/amazon_reviews_multi
51
+ name: MTEB AmazonReviewsClassification (en)
52
+ config: en
53
+ split: test
54
+ revision: 1399c76144fd37290681b995c656ef9b2e06e26d
55
+ metrics:
56
+ - type: accuracy
57
+ value: 54.196
58
+ - type: f1
59
+ value: 53.80122334012787
60
+ - task:
61
+ type: Retrieval
62
+ dataset:
63
+ type: mteb/arguana
64
+ name: MTEB ArguAna
65
+ config: default
66
+ split: test
67
+ revision: c22ab2a51041ffd869aaddef7af8d8215647e41a
68
+ metrics:
69
+ - type: map_at_1
70
+ value: 47.297
71
+ - type: map_at_10
72
+ value: 64.303
73
+ - type: map_at_100
74
+ value: 64.541
75
+ - type: map_at_1000
76
+ value: 64.541
77
+ - type: map_at_3
78
+ value: 60.728
79
+ - type: map_at_5
80
+ value: 63.114000000000004
81
+ - type: mrr_at_1
82
+ value: 48.435
83
+ - type: mrr_at_10
84
+ value: 64.657
85
+ - type: mrr_at_100
86
+ value: 64.901
87
+ - type: mrr_at_1000
88
+ value: 64.901
89
+ - type: mrr_at_3
90
+ value: 61.06
91
+ - type: mrr_at_5
92
+ value: 63.514
93
+ - type: ndcg_at_1
94
+ value: 47.297
95
+ - type: ndcg_at_10
96
+ value: 72.107
97
+ - type: ndcg_at_100
98
+ value: 72.963
99
+ - type: ndcg_at_1000
100
+ value: 72.963
101
+ - type: ndcg_at_3
102
+ value: 65.063
103
+ - type: ndcg_at_5
104
+ value: 69.352
105
+ - type: precision_at_1
106
+ value: 47.297
107
+ - type: precision_at_10
108
+ value: 9.623
109
+ - type: precision_at_100
110
+ value: 0.996
111
+ - type: precision_at_1000
112
+ value: 0.1
113
+ - type: precision_at_3
114
+ value: 25.865
115
+ - type: precision_at_5
116
+ value: 17.596
117
+ - type: recall_at_1
118
+ value: 47.297
119
+ - type: recall_at_10
120
+ value: 96.23
121
+ - type: recall_at_100
122
+ value: 99.644
123
+ - type: recall_at_1000
124
+ value: 99.644
125
+ - type: recall_at_3
126
+ value: 77.596
127
+ - type: recall_at_5
128
+ value: 87.98
129
+ - task:
130
+ type: Clustering
131
+ dataset:
132
+ type: mteb/arxiv-clustering-p2p
133
+ name: MTEB ArxivClusteringP2P
134
+ config: default
135
+ split: test
136
+ revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d
137
+ metrics:
138
+ - type: v_measure
139
+ value: 48.467787861077475
140
+ - task:
141
+ type: Clustering
142
+ dataset:
143
+ type: mteb/arxiv-clustering-s2s
144
+ name: MTEB ArxivClusteringS2S
145
+ config: default
146
+ split: test
147
+ revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53
148
+ metrics:
149
+ - type: v_measure
150
+ value: 43.39198391914257
151
+ - task:
152
+ type: Reranking
153
+ dataset:
154
+ type: mteb/askubuntudupquestions-reranking
155
+ name: MTEB AskUbuntuDupQuestions
156
+ config: default
157
+ split: test
158
+ revision: 2000358ca161889fa9c082cb41daa8dcfb161a54
159
+ metrics:
160
+ - type: map
161
+ value: 63.12794820591384
162
+ - type: mrr
163
+ value: 75.9331442641692
164
+ - task:
165
+ type: STS
166
+ dataset:
167
+ type: mteb/biosses-sts
168
+ name: MTEB BIOSSES
169
+ config: default
170
+ split: test
171
+ revision: d3fb88f8f02e40887cd149695127462bbcf29b4a
172
+ metrics:
173
+ - type: cos_sim_pearson
174
+ value: 87.85062993863319
175
+ - type: cos_sim_spearman
176
+ value: 85.39049989733459
177
+ - type: euclidean_pearson
178
+ value: 86.00222680278333
179
+ - type: euclidean_spearman
180
+ value: 85.45556162077396
181
+ - type: manhattan_pearson
182
+ value: 85.88769871785621
183
+ - type: manhattan_spearman
184
+ value: 85.11760211290839
185
+ - task:
186
+ type: Classification
187
+ dataset:
188
+ type: mteb/banking77
189
+ name: MTEB Banking77Classification
190
+ config: default
191
+ split: test
192
+ revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
193
+ metrics:
194
+ - type: accuracy
195
+ value: 87.32792207792208
196
+ - type: f1
197
+ value: 87.29132945999555
198
+ - task:
199
+ type: Clustering
200
+ dataset:
201
+ type: mteb/biorxiv-clustering-p2p
202
+ name: MTEB BiorxivClusteringP2P
203
+ config: default
204
+ split: test
205
+ revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40
206
+ metrics:
207
+ - type: v_measure
208
+ value: 40.5779328301945
209
+ - task:
210
+ type: Clustering
211
+ dataset:
212
+ type: mteb/biorxiv-clustering-s2s
213
+ name: MTEB BiorxivClusteringS2S
214
+ config: default
215
+ split: test
216
+ revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908
217
+ metrics:
218
+ - type: v_measure
219
+ value: 37.94425623865118
220
+ - task:
221
+ type: Retrieval
222
+ dataset:
223
+ type: mteb/cqadupstack-android
224
+ name: MTEB CQADupstackAndroidRetrieval
225
+ config: default
226
+ split: test
227
+ revision: f46a197baaae43b4f621051089b82a364682dfeb
228
+ metrics:
229
+ - type: map_at_1
230
+ value: 32.978
231
+ - type: map_at_10
232
+ value: 44.45
233
+ - type: map_at_100
234
+ value: 46.19
235
+ - type: map_at_1000
236
+ value: 46.303
237
+ - type: map_at_3
238
+ value: 40.849000000000004
239
+ - type: map_at_5
240
+ value: 42.55
241
+ - type: mrr_at_1
242
+ value: 40.629
243
+ - type: mrr_at_10
244
+ value: 50.848000000000006
245
+ - type: mrr_at_100
246
+ value: 51.669
247
+ - type: mrr_at_1000
248
+ value: 51.705
249
+ - type: mrr_at_3
250
+ value: 47.997
251
+ - type: mrr_at_5
252
+ value: 49.506
253
+ - type: ndcg_at_1
254
+ value: 40.629
255
+ - type: ndcg_at_10
256
+ value: 51.102000000000004
257
+ - type: ndcg_at_100
258
+ value: 57.159000000000006
259
+ - type: ndcg_at_1000
260
+ value: 58.669000000000004
261
+ - type: ndcg_at_3
262
+ value: 45.738
263
+ - type: ndcg_at_5
264
+ value: 47.632999999999996
265
+ - type: precision_at_1
266
+ value: 40.629
267
+ - type: precision_at_10
268
+ value: 9.700000000000001
269
+ - type: precision_at_100
270
+ value: 1.5970000000000002
271
+ - type: precision_at_1000
272
+ value: 0.202
273
+ - type: precision_at_3
274
+ value: 21.698
275
+ - type: precision_at_5
276
+ value: 15.393
277
+ - type: recall_at_1
278
+ value: 32.978
279
+ - type: recall_at_10
280
+ value: 63.711
281
+ - type: recall_at_100
282
+ value: 88.39399999999999
283
+ - type: recall_at_1000
284
+ value: 97.513
285
+ - type: recall_at_3
286
+ value: 48.025
287
+ - type: recall_at_5
288
+ value: 53.52
289
+ - task:
290
+ type: Retrieval
291
+ dataset:
292
+ type: mteb/cqadupstack-english
293
+ name: MTEB CQADupstackEnglishRetrieval
294
+ config: default
295
+ split: test
296
+ revision: ad9991cb51e31e31e430383c75ffb2885547b5f0
297
+ metrics:
298
+ - type: map_at_1
299
+ value: 30.767
300
+ - type: map_at_10
301
+ value: 42.195
302
+ - type: map_at_100
303
+ value: 43.541999999999994
304
+ - type: map_at_1000
305
+ value: 43.673
306
+ - type: map_at_3
307
+ value: 38.561
308
+ - type: map_at_5
309
+ value: 40.532000000000004
310
+ - type: mrr_at_1
311
+ value: 38.79
312
+ - type: mrr_at_10
313
+ value: 48.021
314
+ - type: mrr_at_100
315
+ value: 48.735
316
+ - type: mrr_at_1000
317
+ value: 48.776
318
+ - type: mrr_at_3
319
+ value: 45.594
320
+ - type: mrr_at_5
321
+ value: 46.986
322
+ - type: ndcg_at_1
323
+ value: 38.79
324
+ - type: ndcg_at_10
325
+ value: 48.468
326
+ - type: ndcg_at_100
327
+ value: 53.037
328
+ - type: ndcg_at_1000
329
+ value: 55.001999999999995
330
+ - type: ndcg_at_3
331
+ value: 43.409
332
+ - type: ndcg_at_5
333
+ value: 45.654
334
+ - type: precision_at_1
335
+ value: 38.79
336
+ - type: precision_at_10
337
+ value: 9.452
338
+ - type: precision_at_100
339
+ value: 1.518
340
+ - type: precision_at_1000
341
+ value: 0.201
342
+ - type: precision_at_3
343
+ value: 21.21
344
+ - type: precision_at_5
345
+ value: 15.171999999999999
346
+ - type: recall_at_1
347
+ value: 30.767
348
+ - type: recall_at_10
349
+ value: 60.118
350
+ - type: recall_at_100
351
+ value: 79.271
352
+ - type: recall_at_1000
353
+ value: 91.43299999999999
354
+ - type: recall_at_3
355
+ value: 45.36
356
+ - type: recall_at_5
357
+ value: 51.705
358
+ - task:
359
+ type: Retrieval
360
+ dataset:
361
+ type: mteb/cqadupstack-gaming
362
+ name: MTEB CQADupstackGamingRetrieval
363
+ config: default
364
+ split: test
365
+ revision: 4885aa143210c98657558c04aaf3dc47cfb54340
366
+ metrics:
367
+ - type: map_at_1
368
+ value: 40.007
369
+ - type: map_at_10
370
+ value: 53.529
371
+ - type: map_at_100
372
+ value: 54.602
373
+ - type: map_at_1000
374
+ value: 54.647
375
+ - type: map_at_3
376
+ value: 49.951
377
+ - type: map_at_5
378
+ value: 52.066
379
+ - type: mrr_at_1
380
+ value: 45.705
381
+ - type: mrr_at_10
382
+ value: 56.745000000000005
383
+ - type: mrr_at_100
384
+ value: 57.43899999999999
385
+ - type: mrr_at_1000
386
+ value: 57.462999999999994
387
+ - type: mrr_at_3
388
+ value: 54.25299999999999
389
+ - type: mrr_at_5
390
+ value: 55.842000000000006
391
+ - type: ndcg_at_1
392
+ value: 45.705
393
+ - type: ndcg_at_10
394
+ value: 59.809
395
+ - type: ndcg_at_100
396
+ value: 63.837999999999994
397
+ - type: ndcg_at_1000
398
+ value: 64.729
399
+ - type: ndcg_at_3
400
+ value: 53.994
401
+ - type: ndcg_at_5
402
+ value: 57.028
403
+ - type: precision_at_1
404
+ value: 45.705
405
+ - type: precision_at_10
406
+ value: 9.762
407
+ - type: precision_at_100
408
+ value: 1.275
409
+ - type: precision_at_1000
410
+ value: 0.13899999999999998
411
+ - type: precision_at_3
412
+ value: 24.368000000000002
413
+ - type: precision_at_5
414
+ value: 16.84
415
+ - type: recall_at_1
416
+ value: 40.007
417
+ - type: recall_at_10
418
+ value: 75.017
419
+ - type: recall_at_100
420
+ value: 91.99000000000001
421
+ - type: recall_at_1000
422
+ value: 98.265
423
+ - type: recall_at_3
424
+ value: 59.704
425
+ - type: recall_at_5
426
+ value: 67.109
427
+ - task:
428
+ type: Retrieval
429
+ dataset:
430
+ type: mteb/cqadupstack-gis
431
+ name: MTEB CQADupstackGisRetrieval
432
+ config: default
433
+ split: test
434
+ revision: 5003b3064772da1887988e05400cf3806fe491f2
435
+ metrics:
436
+ - type: map_at_1
437
+ value: 26.639000000000003
438
+ - type: map_at_10
439
+ value: 35.926
440
+ - type: map_at_100
441
+ value: 37.126999999999995
442
+ - type: map_at_1000
443
+ value: 37.202
444
+ - type: map_at_3
445
+ value: 32.989000000000004
446
+ - type: map_at_5
447
+ value: 34.465
448
+ - type: mrr_at_1
449
+ value: 28.475
450
+ - type: mrr_at_10
451
+ value: 37.7
452
+ - type: mrr_at_100
453
+ value: 38.753
454
+ - type: mrr_at_1000
455
+ value: 38.807
456
+ - type: mrr_at_3
457
+ value: 35.066
458
+ - type: mrr_at_5
459
+ value: 36.512
460
+ - type: ndcg_at_1
461
+ value: 28.475
462
+ - type: ndcg_at_10
463
+ value: 41.245
464
+ - type: ndcg_at_100
465
+ value: 46.814
466
+ - type: ndcg_at_1000
467
+ value: 48.571
468
+ - type: ndcg_at_3
469
+ value: 35.528999999999996
470
+ - type: ndcg_at_5
471
+ value: 38.066
472
+ - type: precision_at_1
473
+ value: 28.475
474
+ - type: precision_at_10
475
+ value: 6.497
476
+ - type: precision_at_100
477
+ value: 0.9650000000000001
478
+ - type: precision_at_1000
479
+ value: 0.11499999999999999
480
+ - type: precision_at_3
481
+ value: 15.065999999999999
482
+ - type: precision_at_5
483
+ value: 10.599
484
+ - type: recall_at_1
485
+ value: 26.639000000000003
486
+ - type: recall_at_10
487
+ value: 55.759
488
+ - type: recall_at_100
489
+ value: 80.913
490
+ - type: recall_at_1000
491
+ value: 93.929
492
+ - type: recall_at_3
493
+ value: 40.454
494
+ - type: recall_at_5
495
+ value: 46.439
496
+ - task:
497
+ type: Retrieval
498
+ dataset:
499
+ type: mteb/cqadupstack-mathematica
500
+ name: MTEB CQADupstackMathematicaRetrieval
501
+ config: default
502
+ split: test
503
+ revision: 90fceea13679c63fe563ded68f3b6f06e50061de
504
+ metrics:
505
+ - type: map_at_1
506
+ value: 15.767999999999999
507
+ - type: map_at_10
508
+ value: 24.811
509
+ - type: map_at_100
510
+ value: 26.064999999999998
511
+ - type: map_at_1000
512
+ value: 26.186999999999998
513
+ - type: map_at_3
514
+ value: 21.736
515
+ - type: map_at_5
516
+ value: 23.283
517
+ - type: mrr_at_1
518
+ value: 19.527
519
+ - type: mrr_at_10
520
+ value: 29.179
521
+ - type: mrr_at_100
522
+ value: 30.153999999999996
523
+ - type: mrr_at_1000
524
+ value: 30.215999999999998
525
+ - type: mrr_at_3
526
+ value: 26.223000000000003
527
+ - type: mrr_at_5
528
+ value: 27.733999999999998
529
+ - type: ndcg_at_1
530
+ value: 19.527
531
+ - type: ndcg_at_10
532
+ value: 30.786
533
+ - type: ndcg_at_100
534
+ value: 36.644
535
+ - type: ndcg_at_1000
536
+ value: 39.440999999999995
537
+ - type: ndcg_at_3
538
+ value: 24.958
539
+ - type: ndcg_at_5
540
+ value: 27.392
541
+ - type: precision_at_1
542
+ value: 19.527
543
+ - type: precision_at_10
544
+ value: 5.995
545
+ - type: precision_at_100
546
+ value: 1.03
547
+ - type: precision_at_1000
548
+ value: 0.14100000000000001
549
+ - type: precision_at_3
550
+ value: 12.520999999999999
551
+ - type: precision_at_5
552
+ value: 9.129
553
+ - type: recall_at_1
554
+ value: 15.767999999999999
555
+ - type: recall_at_10
556
+ value: 44.824000000000005
557
+ - type: recall_at_100
558
+ value: 70.186
559
+ - type: recall_at_1000
560
+ value: 89.934
561
+ - type: recall_at_3
562
+ value: 28.607
563
+ - type: recall_at_5
564
+ value: 34.836
565
+ - task:
566
+ type: Retrieval
567
+ dataset:
568
+ type: mteb/cqadupstack-physics
569
+ name: MTEB CQADupstackPhysicsRetrieval
570
+ config: default
571
+ split: test
572
+ revision: 79531abbd1fb92d06c6d6315a0cbbbf5bb247ea4
573
+ metrics:
574
+ - type: map_at_1
575
+ value: 31.952
576
+ - type: map_at_10
577
+ value: 44.438
578
+ - type: map_at_100
579
+ value: 45.778
580
+ - type: map_at_1000
581
+ value: 45.883
582
+ - type: map_at_3
583
+ value: 41.044000000000004
584
+ - type: map_at_5
585
+ value: 42.986000000000004
586
+ - type: mrr_at_1
587
+ value: 39.172000000000004
588
+ - type: mrr_at_10
589
+ value: 49.76
590
+ - type: mrr_at_100
591
+ value: 50.583999999999996
592
+ - type: mrr_at_1000
593
+ value: 50.621
594
+ - type: mrr_at_3
595
+ value: 47.353
596
+ - type: mrr_at_5
597
+ value: 48.739
598
+ - type: ndcg_at_1
599
+ value: 39.172000000000004
600
+ - type: ndcg_at_10
601
+ value: 50.760000000000005
602
+ - type: ndcg_at_100
603
+ value: 56.084
604
+ - type: ndcg_at_1000
605
+ value: 57.865
606
+ - type: ndcg_at_3
607
+ value: 45.663
608
+ - type: ndcg_at_5
609
+ value: 48.178
610
+ - type: precision_at_1
611
+ value: 39.172000000000004
612
+ - type: precision_at_10
613
+ value: 9.22
614
+ - type: precision_at_100
615
+ value: 1.387
616
+ - type: precision_at_1000
617
+ value: 0.17099999999999999
618
+ - type: precision_at_3
619
+ value: 21.976000000000003
620
+ - type: precision_at_5
621
+ value: 15.457
622
+ - type: recall_at_1
623
+ value: 31.952
624
+ - type: recall_at_10
625
+ value: 63.900999999999996
626
+ - type: recall_at_100
627
+ value: 85.676
628
+ - type: recall_at_1000
629
+ value: 97.03699999999999
630
+ - type: recall_at_3
631
+ value: 49.781
632
+ - type: recall_at_5
633
+ value: 56.330000000000005
634
+ - task:
635
+ type: Retrieval
636
+ dataset:
637
+ type: mteb/cqadupstack-programmers
638
+ name: MTEB CQADupstackProgrammersRetrieval
639
+ config: default
640
+ split: test
641
+ revision: 6184bc1440d2dbc7612be22b50686b8826d22b32
642
+ metrics:
643
+ - type: map_at_1
644
+ value: 25.332
645
+ - type: map_at_10
646
+ value: 36.874
647
+ - type: map_at_100
648
+ value: 38.340999999999994
649
+ - type: map_at_1000
650
+ value: 38.452
651
+ - type: map_at_3
652
+ value: 33.068
653
+ - type: map_at_5
654
+ value: 35.324
655
+ - type: mrr_at_1
656
+ value: 30.822
657
+ - type: mrr_at_10
658
+ value: 41.641
659
+ - type: mrr_at_100
660
+ value: 42.519
661
+ - type: mrr_at_1000
662
+ value: 42.573
663
+ - type: mrr_at_3
664
+ value: 38.413000000000004
665
+ - type: mrr_at_5
666
+ value: 40.542
667
+ - type: ndcg_at_1
668
+ value: 30.822
669
+ - type: ndcg_at_10
670
+ value: 43.414
671
+ - type: ndcg_at_100
672
+ value: 49.196
673
+ - type: ndcg_at_1000
674
+ value: 51.237
675
+ - type: ndcg_at_3
676
+ value: 37.230000000000004
677
+ - type: ndcg_at_5
678
+ value: 40.405
679
+ - type: precision_at_1
680
+ value: 30.822
681
+ - type: precision_at_10
682
+ value: 8.379
683
+ - type: precision_at_100
684
+ value: 1.315
685
+ - type: precision_at_1000
686
+ value: 0.168
687
+ - type: precision_at_3
688
+ value: 18.417
689
+ - type: precision_at_5
690
+ value: 13.744
691
+ - type: recall_at_1
692
+ value: 25.332
693
+ - type: recall_at_10
694
+ value: 57.774
695
+ - type: recall_at_100
696
+ value: 82.071
697
+ - type: recall_at_1000
698
+ value: 95.60600000000001
699
+ - type: recall_at_3
700
+ value: 40.722
701
+ - type: recall_at_5
702
+ value: 48.754999999999995
703
+ - task:
704
+ type: Retrieval
705
+ dataset:
706
+ type: mteb/cqadupstack
707
+ name: MTEB CQADupstackRetrieval
708
+ config: default
709
+ split: test
710
+ revision: 4ffe81d471b1924886b33c7567bfb200e9eec5c4
711
+ metrics:
712
+ - type: map_at_1
713
+ value: 25.91033333333334
714
+ - type: map_at_10
715
+ value: 36.23225000000001
716
+ - type: map_at_100
717
+ value: 37.55766666666667
718
+ - type: map_at_1000
719
+ value: 37.672583333333336
720
+ - type: map_at_3
721
+ value: 32.95666666666667
722
+ - type: map_at_5
723
+ value: 34.73375
724
+ - type: mrr_at_1
725
+ value: 30.634
726
+ - type: mrr_at_10
727
+ value: 40.19449999999999
728
+ - type: mrr_at_100
729
+ value: 41.099250000000005
730
+ - type: mrr_at_1000
731
+ value: 41.15091666666667
732
+ - type: mrr_at_3
733
+ value: 37.4615
734
+ - type: mrr_at_5
735
+ value: 39.00216666666667
736
+ - type: ndcg_at_1
737
+ value: 30.634
738
+ - type: ndcg_at_10
739
+ value: 42.162166666666664
740
+ - type: ndcg_at_100
741
+ value: 47.60708333333333
742
+ - type: ndcg_at_1000
743
+ value: 49.68616666666666
744
+ - type: ndcg_at_3
745
+ value: 36.60316666666666
746
+ - type: ndcg_at_5
747
+ value: 39.15616666666668
748
+ - type: precision_at_1
749
+ value: 30.634
750
+ - type: precision_at_10
751
+ value: 7.6193333333333335
752
+ - type: precision_at_100
753
+ value: 1.2198333333333333
754
+ - type: precision_at_1000
755
+ value: 0.15975000000000003
756
+ - type: precision_at_3
757
+ value: 17.087
758
+ - type: precision_at_5
759
+ value: 12.298333333333334
760
+ - type: recall_at_1
761
+ value: 25.91033333333334
762
+ - type: recall_at_10
763
+ value: 55.67300000000001
764
+ - type: recall_at_100
765
+ value: 79.20608333333334
766
+ - type: recall_at_1000
767
+ value: 93.34866666666667
768
+ - type: recall_at_3
769
+ value: 40.34858333333333
770
+ - type: recall_at_5
771
+ value: 46.834083333333325
772
+ - task:
773
+ type: Retrieval
774
+ dataset:
775
+ type: mteb/cqadupstack-stats
776
+ name: MTEB CQADupstackStatsRetrieval
777
+ config: default
778
+ split: test
779
+ revision: 65ac3a16b8e91f9cee4c9828cc7c335575432a2a
780
+ metrics:
781
+ - type: map_at_1
782
+ value: 25.006
783
+ - type: map_at_10
784
+ value: 32.177
785
+ - type: map_at_100
786
+ value: 33.324999999999996
787
+ - type: map_at_1000
788
+ value: 33.419
789
+ - type: map_at_3
790
+ value: 29.952
791
+ - type: map_at_5
792
+ value: 31.095
793
+ - type: mrr_at_1
794
+ value: 28.066999999999997
795
+ - type: mrr_at_10
796
+ value: 34.995
797
+ - type: mrr_at_100
798
+ value: 35.978
799
+ - type: mrr_at_1000
800
+ value: 36.042
801
+ - type: mrr_at_3
802
+ value: 33.103
803
+ - type: mrr_at_5
804
+ value: 34.001
805
+ - type: ndcg_at_1
806
+ value: 28.066999999999997
807
+ - type: ndcg_at_10
808
+ value: 36.481
809
+ - type: ndcg_at_100
810
+ value: 42.022999999999996
811
+ - type: ndcg_at_1000
812
+ value: 44.377
813
+ - type: ndcg_at_3
814
+ value: 32.394
815
+ - type: ndcg_at_5
816
+ value: 34.108
817
+ - type: precision_at_1
818
+ value: 28.066999999999997
819
+ - type: precision_at_10
820
+ value: 5.736
821
+ - type: precision_at_100
822
+ value: 0.9259999999999999
823
+ - type: precision_at_1000
824
+ value: 0.12
825
+ - type: precision_at_3
826
+ value: 13.804
827
+ - type: precision_at_5
828
+ value: 9.508999999999999
829
+ - type: recall_at_1
830
+ value: 25.006
831
+ - type: recall_at_10
832
+ value: 46.972
833
+ - type: recall_at_100
834
+ value: 72.138
835
+ - type: recall_at_1000
836
+ value: 89.479
837
+ - type: recall_at_3
838
+ value: 35.793
839
+ - type: recall_at_5
840
+ value: 39.947
841
+ - task:
842
+ type: Retrieval
843
+ dataset:
844
+ type: mteb/cqadupstack-tex
845
+ name: MTEB CQADupstackTexRetrieval
846
+ config: default
847
+ split: test
848
+ revision: 46989137a86843e03a6195de44b09deda022eec7
849
+ metrics:
850
+ - type: map_at_1
851
+ value: 16.07
852
+ - type: map_at_10
853
+ value: 24.447
854
+ - type: map_at_100
855
+ value: 25.685999999999996
856
+ - type: map_at_1000
857
+ value: 25.813999999999997
858
+ - type: map_at_3
859
+ value: 21.634
860
+ - type: map_at_5
861
+ value: 23.133
862
+ - type: mrr_at_1
863
+ value: 19.580000000000002
864
+ - type: mrr_at_10
865
+ value: 28.127999999999997
866
+ - type: mrr_at_100
867
+ value: 29.119
868
+ - type: mrr_at_1000
869
+ value: 29.192
870
+ - type: mrr_at_3
871
+ value: 25.509999999999998
872
+ - type: mrr_at_5
873
+ value: 26.878
874
+ - type: ndcg_at_1
875
+ value: 19.580000000000002
876
+ - type: ndcg_at_10
877
+ value: 29.804000000000002
878
+ - type: ndcg_at_100
879
+ value: 35.555
880
+ - type: ndcg_at_1000
881
+ value: 38.421
882
+ - type: ndcg_at_3
883
+ value: 24.654999999999998
884
+ - type: ndcg_at_5
885
+ value: 26.881
886
+ - type: precision_at_1
887
+ value: 19.580000000000002
888
+ - type: precision_at_10
889
+ value: 5.736
890
+ - type: precision_at_100
891
+ value: 1.005
892
+ - type: precision_at_1000
893
+ value: 0.145
894
+ - type: precision_at_3
895
+ value: 12.033000000000001
896
+ - type: precision_at_5
897
+ value: 8.871
898
+ - type: recall_at_1
899
+ value: 16.07
900
+ - type: recall_at_10
901
+ value: 42.364000000000004
902
+ - type: recall_at_100
903
+ value: 68.01899999999999
904
+ - type: recall_at_1000
905
+ value: 88.122
906
+ - type: recall_at_3
907
+ value: 27.846
908
+ - type: recall_at_5
909
+ value: 33.638
910
+ - task:
911
+ type: Retrieval
912
+ dataset:
913
+ type: mteb/cqadupstack-unix
914
+ name: MTEB CQADupstackUnixRetrieval
915
+ config: default
916
+ split: test
917
+ revision: 6c6430d3a6d36f8d2a829195bc5dc94d7e063e53
918
+ metrics:
919
+ - type: map_at_1
920
+ value: 26.365
921
+ - type: map_at_10
922
+ value: 36.591
923
+ - type: map_at_100
924
+ value: 37.730000000000004
925
+ - type: map_at_1000
926
+ value: 37.84
927
+ - type: map_at_3
928
+ value: 33.403
929
+ - type: map_at_5
930
+ value: 35.272999999999996
931
+ - type: mrr_at_1
932
+ value: 30.503999999999998
933
+ - type: mrr_at_10
934
+ value: 39.940999999999995
935
+ - type: mrr_at_100
936
+ value: 40.818
937
+ - type: mrr_at_1000
938
+ value: 40.876000000000005
939
+ - type: mrr_at_3
940
+ value: 37.065
941
+ - type: mrr_at_5
942
+ value: 38.814
943
+ - type: ndcg_at_1
944
+ value: 30.503999999999998
945
+ - type: ndcg_at_10
946
+ value: 42.185
947
+ - type: ndcg_at_100
948
+ value: 47.416000000000004
949
+ - type: ndcg_at_1000
950
+ value: 49.705
951
+ - type: ndcg_at_3
952
+ value: 36.568
953
+ - type: ndcg_at_5
954
+ value: 39.416000000000004
955
+ - type: precision_at_1
956
+ value: 30.503999999999998
957
+ - type: precision_at_10
958
+ value: 7.276000000000001
959
+ - type: precision_at_100
960
+ value: 1.118
961
+ - type: precision_at_1000
962
+ value: 0.14300000000000002
963
+ - type: precision_at_3
964
+ value: 16.729
965
+ - type: precision_at_5
966
+ value: 12.107999999999999
967
+ - type: recall_at_1
968
+ value: 26.365
969
+ - type: recall_at_10
970
+ value: 55.616
971
+ - type: recall_at_100
972
+ value: 78.129
973
+ - type: recall_at_1000
974
+ value: 93.95599999999999
975
+ - type: recall_at_3
976
+ value: 40.686
977
+ - type: recall_at_5
978
+ value: 47.668
979
+ - task:
980
+ type: Retrieval
981
+ dataset:
982
+ type: mteb/cqadupstack-webmasters
983
+ name: MTEB CQADupstackWebmastersRetrieval
984
+ config: default
985
+ split: test
986
+ revision: 160c094312a0e1facb97e55eeddb698c0abe3571
987
+ metrics:
988
+ - type: map_at_1
989
+ value: 22.750999999999998
990
+ - type: map_at_10
991
+ value: 33.446
992
+ - type: map_at_100
993
+ value: 35.235
994
+ - type: map_at_1000
995
+ value: 35.478
996
+ - type: map_at_3
997
+ value: 29.358
998
+ - type: map_at_5
999
+ value: 31.525
1000
+ - type: mrr_at_1
1001
+ value: 27.668
1002
+ - type: mrr_at_10
1003
+ value: 37.694
1004
+ - type: mrr_at_100
1005
+ value: 38.732
1006
+ - type: mrr_at_1000
1007
+ value: 38.779
1008
+ - type: mrr_at_3
1009
+ value: 34.223
1010
+ - type: mrr_at_5
1011
+ value: 36.08
1012
+ - type: ndcg_at_1
1013
+ value: 27.668
1014
+ - type: ndcg_at_10
1015
+ value: 40.557
1016
+ - type: ndcg_at_100
1017
+ value: 46.605999999999995
1018
+ - type: ndcg_at_1000
1019
+ value: 48.917
1020
+ - type: ndcg_at_3
1021
+ value: 33.677
1022
+ - type: ndcg_at_5
1023
+ value: 36.85
1024
+ - type: precision_at_1
1025
+ value: 27.668
1026
+ - type: precision_at_10
1027
+ value: 8.3
1028
+ - type: precision_at_100
1029
+ value: 1.6260000000000001
1030
+ - type: precision_at_1000
1031
+ value: 0.253
1032
+ - type: precision_at_3
1033
+ value: 16.008
1034
+ - type: precision_at_5
1035
+ value: 12.292
1036
+ - type: recall_at_1
1037
+ value: 22.750999999999998
1038
+ - type: recall_at_10
1039
+ value: 55.643
1040
+ - type: recall_at_100
1041
+ value: 82.151
1042
+ - type: recall_at_1000
1043
+ value: 95.963
1044
+ - type: recall_at_3
1045
+ value: 36.623
1046
+ - type: recall_at_5
1047
+ value: 44.708
1048
+ - task:
1049
+ type: Retrieval
1050
+ dataset:
1051
+ type: mteb/cqadupstack-wordpress
1052
+ name: MTEB CQADupstackWordpressRetrieval
1053
+ config: default
1054
+ split: test
1055
+ revision: 4ffe81d471b1924886b33c7567bfb200e9eec5c4
1056
+ metrics:
1057
+ - type: map_at_1
1058
+ value: 17.288999999999998
1059
+ - type: map_at_10
1060
+ value: 25.903
1061
+ - type: map_at_100
1062
+ value: 27.071
1063
+ - type: map_at_1000
1064
+ value: 27.173000000000002
1065
+ - type: map_at_3
1066
+ value: 22.935
1067
+ - type: map_at_5
1068
+ value: 24.573
1069
+ - type: mrr_at_1
1070
+ value: 18.669
1071
+ - type: mrr_at_10
1072
+ value: 27.682000000000002
1073
+ - type: mrr_at_100
1074
+ value: 28.691
1075
+ - type: mrr_at_1000
1076
+ value: 28.761
1077
+ - type: mrr_at_3
1078
+ value: 24.738
1079
+ - type: mrr_at_5
1080
+ value: 26.392
1081
+ - type: ndcg_at_1
1082
+ value: 18.669
1083
+ - type: ndcg_at_10
1084
+ value: 31.335
1085
+ - type: ndcg_at_100
1086
+ value: 36.913000000000004
1087
+ - type: ndcg_at_1000
1088
+ value: 39.300000000000004
1089
+ - type: ndcg_at_3
1090
+ value: 25.423000000000002
1091
+ - type: ndcg_at_5
1092
+ value: 28.262999999999998
1093
+ - type: precision_at_1
1094
+ value: 18.669
1095
+ - type: precision_at_10
1096
+ value: 5.379
1097
+ - type: precision_at_100
1098
+ value: 0.876
1099
+ - type: precision_at_1000
1100
+ value: 0.11900000000000001
1101
+ - type: precision_at_3
1102
+ value: 11.214
1103
+ - type: precision_at_5
1104
+ value: 8.466
1105
+ - type: recall_at_1
1106
+ value: 17.288999999999998
1107
+ - type: recall_at_10
1108
+ value: 46.377
1109
+ - type: recall_at_100
1110
+ value: 71.53500000000001
1111
+ - type: recall_at_1000
1112
+ value: 88.947
1113
+ - type: recall_at_3
1114
+ value: 30.581999999999997
1115
+ - type: recall_at_5
1116
+ value: 37.354
1117
+ - task:
1118
+ type: Retrieval
1119
+ dataset:
1120
+ type: mteb/climate-fever
1121
+ name: MTEB ClimateFEVER
1122
+ config: default
1123
+ split: test
1124
+ revision: 47f2ac6acb640fc46020b02a5b59fdda04d39380
1125
+ metrics:
1126
+ - type: map_at_1
1127
+ value: 21.795
1128
+ - type: map_at_10
1129
+ value: 37.614999999999995
1130
+ - type: map_at_100
1131
+ value: 40.037
1132
+ - type: map_at_1000
1133
+ value: 40.184999999999995
1134
+ - type: map_at_3
1135
+ value: 32.221
1136
+ - type: map_at_5
1137
+ value: 35.154999999999994
1138
+ - type: mrr_at_1
1139
+ value: 50.358000000000004
1140
+ - type: mrr_at_10
1141
+ value: 62.129
1142
+ - type: mrr_at_100
1143
+ value: 62.613
1144
+ - type: mrr_at_1000
1145
+ value: 62.62
1146
+ - type: mrr_at_3
1147
+ value: 59.272999999999996
1148
+ - type: mrr_at_5
1149
+ value: 61.138999999999996
1150
+ - type: ndcg_at_1
1151
+ value: 50.358000000000004
1152
+ - type: ndcg_at_10
1153
+ value: 48.362
1154
+ - type: ndcg_at_100
1155
+ value: 55.932
1156
+ - type: ndcg_at_1000
1157
+ value: 58.062999999999995
1158
+ - type: ndcg_at_3
1159
+ value: 42.111
1160
+ - type: ndcg_at_5
1161
+ value: 44.063
1162
+ - type: precision_at_1
1163
+ value: 50.358000000000004
1164
+ - type: precision_at_10
1165
+ value: 14.677999999999999
1166
+ - type: precision_at_100
1167
+ value: 2.2950000000000004
1168
+ - type: precision_at_1000
1169
+ value: 0.271
1170
+ - type: precision_at_3
1171
+ value: 31.77
1172
+ - type: precision_at_5
1173
+ value: 23.375
1174
+ - type: recall_at_1
1175
+ value: 21.795
1176
+ - type: recall_at_10
1177
+ value: 53.846000000000004
1178
+ - type: recall_at_100
1179
+ value: 78.952
1180
+ - type: recall_at_1000
1181
+ value: 90.41900000000001
1182
+ - type: recall_at_3
1183
+ value: 37.257
1184
+ - type: recall_at_5
1185
+ value: 44.661
1186
+ - task:
1187
+ type: Retrieval
1188
+ dataset:
1189
+ type: mteb/dbpedia
1190
+ name: MTEB DBPedia
1191
+ config: default
1192
+ split: test
1193
+ revision: c0f706b76e590d620bd6618b3ca8efdd34e2d659
1194
+ metrics:
1195
+ - type: map_at_1
1196
+ value: 9.728
1197
+ - type: map_at_10
1198
+ value: 22.691
1199
+ - type: map_at_100
1200
+ value: 31.734
1201
+ - type: map_at_1000
1202
+ value: 33.464
1203
+ - type: map_at_3
1204
+ value: 16.273
1205
+ - type: map_at_5
1206
+ value: 19.016
1207
+ - type: mrr_at_1
1208
+ value: 73.25
1209
+ - type: mrr_at_10
1210
+ value: 80.782
1211
+ - type: mrr_at_100
1212
+ value: 81.01899999999999
1213
+ - type: mrr_at_1000
1214
+ value: 81.021
1215
+ - type: mrr_at_3
1216
+ value: 79.583
1217
+ - type: mrr_at_5
1218
+ value: 80.146
1219
+ - type: ndcg_at_1
1220
+ value: 59.62499999999999
1221
+ - type: ndcg_at_10
1222
+ value: 46.304
1223
+ - type: ndcg_at_100
1224
+ value: 51.23
1225
+ - type: ndcg_at_1000
1226
+ value: 58.048
1227
+ - type: ndcg_at_3
1228
+ value: 51.541000000000004
1229
+ - type: ndcg_at_5
1230
+ value: 48.635
1231
+ - type: precision_at_1
1232
+ value: 73.25
1233
+ - type: precision_at_10
1234
+ value: 36.375
1235
+ - type: precision_at_100
1236
+ value: 11.53
1237
+ - type: precision_at_1000
1238
+ value: 2.23
1239
+ - type: precision_at_3
1240
+ value: 55.583000000000006
1241
+ - type: precision_at_5
1242
+ value: 47.15
1243
+ - type: recall_at_1
1244
+ value: 9.728
1245
+ - type: recall_at_10
1246
+ value: 28.793999999999997
1247
+ - type: recall_at_100
1248
+ value: 57.885
1249
+ - type: recall_at_1000
1250
+ value: 78.759
1251
+ - type: recall_at_3
1252
+ value: 17.79
1253
+ - type: recall_at_5
1254
+ value: 21.733
1255
+ - task:
1256
+ type: Classification
1257
+ dataset:
1258
+ type: mteb/emotion
1259
+ name: MTEB EmotionClassification
1260
+ config: default
1261
+ split: test
1262
+ revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37
1263
+ metrics:
1264
+ - type: accuracy
1265
+ value: 46.775
1266
+ - type: f1
1267
+ value: 41.89794273264891
1268
+ - task:
1269
+ type: Retrieval
1270
+ dataset:
1271
+ type: mteb/fever
1272
+ name: MTEB FEVER
1273
+ config: default
1274
+ split: test
1275
+ revision: bea83ef9e8fb933d90a2f1d5515737465d613e12
1276
+ metrics:
1277
+ - type: map_at_1
1278
+ value: 85.378
1279
+ - type: map_at_10
1280
+ value: 91.51
1281
+ - type: map_at_100
1282
+ value: 91.666
1283
+ - type: map_at_1000
1284
+ value: 91.676
1285
+ - type: map_at_3
1286
+ value: 90.757
1287
+ - type: map_at_5
1288
+ value: 91.277
1289
+ - type: mrr_at_1
1290
+ value: 91.839
1291
+ - type: mrr_at_10
1292
+ value: 95.49
1293
+ - type: mrr_at_100
1294
+ value: 95.493
1295
+ - type: mrr_at_1000
1296
+ value: 95.493
1297
+ - type: mrr_at_3
1298
+ value: 95.345
1299
+ - type: mrr_at_5
1300
+ value: 95.47200000000001
1301
+ - type: ndcg_at_1
1302
+ value: 91.839
1303
+ - type: ndcg_at_10
1304
+ value: 93.806
1305
+ - type: ndcg_at_100
1306
+ value: 94.255
1307
+ - type: ndcg_at_1000
1308
+ value: 94.399
1309
+ - type: ndcg_at_3
1310
+ value: 93.027
1311
+ - type: ndcg_at_5
1312
+ value: 93.51
1313
+ - type: precision_at_1
1314
+ value: 91.839
1315
+ - type: precision_at_10
1316
+ value: 10.93
1317
+ - type: precision_at_100
1318
+ value: 1.1400000000000001
1319
+ - type: precision_at_1000
1320
+ value: 0.117
1321
+ - type: precision_at_3
1322
+ value: 34.873
1323
+ - type: precision_at_5
1324
+ value: 21.44
1325
+ - type: recall_at_1
1326
+ value: 85.378
1327
+ - type: recall_at_10
1328
+ value: 96.814
1329
+ - type: recall_at_100
1330
+ value: 98.386
1331
+ - type: recall_at_1000
1332
+ value: 99.21600000000001
1333
+ - type: recall_at_3
1334
+ value: 94.643
1335
+ - type: recall_at_5
1336
+ value: 95.976
1337
+ - task:
1338
+ type: Retrieval
1339
+ dataset:
1340
+ type: mteb/fiqa
1341
+ name: MTEB FiQA2018
1342
+ config: default
1343
+ split: test
1344
+ revision: 27a168819829fe9bcd655c2df245fb19452e8e06
1345
+ metrics:
1346
+ - type: map_at_1
1347
+ value: 32.190000000000005
1348
+ - type: map_at_10
1349
+ value: 53.605000000000004
1350
+ - type: map_at_100
1351
+ value: 55.550999999999995
1352
+ - type: map_at_1000
1353
+ value: 55.665
1354
+ - type: map_at_3
1355
+ value: 46.62
1356
+ - type: map_at_5
1357
+ value: 50.517999999999994
1358
+ - type: mrr_at_1
1359
+ value: 60.34
1360
+ - type: mrr_at_10
1361
+ value: 70.775
1362
+ - type: mrr_at_100
1363
+ value: 71.238
1364
+ - type: mrr_at_1000
1365
+ value: 71.244
1366
+ - type: mrr_at_3
1367
+ value: 68.72399999999999
1368
+ - type: mrr_at_5
1369
+ value: 69.959
1370
+ - type: ndcg_at_1
1371
+ value: 60.34
1372
+ - type: ndcg_at_10
1373
+ value: 63.226000000000006
1374
+ - type: ndcg_at_100
1375
+ value: 68.60300000000001
1376
+ - type: ndcg_at_1000
1377
+ value: 69.901
1378
+ - type: ndcg_at_3
1379
+ value: 58.048
1380
+ - type: ndcg_at_5
1381
+ value: 59.789
1382
+ - type: precision_at_1
1383
+ value: 60.34
1384
+ - type: precision_at_10
1385
+ value: 17.130000000000003
1386
+ - type: precision_at_100
1387
+ value: 2.29
1388
+ - type: precision_at_1000
1389
+ value: 0.256
1390
+ - type: precision_at_3
1391
+ value: 38.323
1392
+ - type: precision_at_5
1393
+ value: 27.87
1394
+ - type: recall_at_1
1395
+ value: 32.190000000000005
1396
+ - type: recall_at_10
1397
+ value: 73.041
1398
+ - type: recall_at_100
1399
+ value: 91.31
1400
+ - type: recall_at_1000
1401
+ value: 98.104
1402
+ - type: recall_at_3
1403
+ value: 53.70399999999999
1404
+ - type: recall_at_5
1405
+ value: 62.358999999999995
1406
+ - task:
1407
+ type: Retrieval
1408
+ dataset:
1409
+ type: mteb/hotpotqa
1410
+ name: MTEB HotpotQA
1411
+ config: default
1412
+ split: test
1413
+ revision: ab518f4d6fcca38d87c25209f94beba119d02014
1414
+ metrics:
1415
+ - type: map_at_1
1416
+ value: 43.511
1417
+ - type: map_at_10
1418
+ value: 58.15
1419
+ - type: map_at_100
1420
+ value: 58.95399999999999
1421
+ - type: map_at_1000
1422
+ value: 59.018
1423
+ - type: map_at_3
1424
+ value: 55.31700000000001
1425
+ - type: map_at_5
1426
+ value: 57.04900000000001
1427
+ - type: mrr_at_1
1428
+ value: 87.022
1429
+ - type: mrr_at_10
1430
+ value: 91.32000000000001
1431
+ - type: mrr_at_100
1432
+ value: 91.401
1433
+ - type: mrr_at_1000
1434
+ value: 91.403
1435
+ - type: mrr_at_3
1436
+ value: 90.77
1437
+ - type: mrr_at_5
1438
+ value: 91.156
1439
+ - type: ndcg_at_1
1440
+ value: 87.022
1441
+ - type: ndcg_at_10
1442
+ value: 68.183
1443
+ - type: ndcg_at_100
1444
+ value: 70.781
1445
+ - type: ndcg_at_1000
1446
+ value: 72.009
1447
+ - type: ndcg_at_3
1448
+ value: 64.334
1449
+ - type: ndcg_at_5
1450
+ value: 66.449
1451
+ - type: precision_at_1
1452
+ value: 87.022
1453
+ - type: precision_at_10
1454
+ value: 13.406
1455
+ - type: precision_at_100
1456
+ value: 1.542
1457
+ - type: precision_at_1000
1458
+ value: 0.17099999999999999
1459
+ - type: precision_at_3
1460
+ value: 39.023
1461
+ - type: precision_at_5
1462
+ value: 25.080000000000002
1463
+ - type: recall_at_1
1464
+ value: 43.511
1465
+ - type: recall_at_10
1466
+ value: 67.02900000000001
1467
+ - type: recall_at_100
1468
+ value: 77.11
1469
+ - type: recall_at_1000
1470
+ value: 85.294
1471
+ - type: recall_at_3
1472
+ value: 58.535000000000004
1473
+ - type: recall_at_5
1474
+ value: 62.70099999999999
1475
+ - task:
1476
+ type: Classification
1477
+ dataset:
1478
+ type: mteb/imdb
1479
+ name: MTEB ImdbClassification
1480
+ config: default
1481
+ split: test
1482
+ revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7
1483
+ metrics:
1484
+ - type: accuracy
1485
+ value: 92.0996
1486
+ - type: ap
1487
+ value: 87.86206089096373
1488
+ - type: f1
1489
+ value: 92.07554547510763
1490
+ - task:
1491
+ type: Retrieval
1492
+ dataset:
1493
+ type: mteb/msmarco
1494
+ name: MTEB MSMARCO
1495
+ config: default
1496
+ split: dev
1497
+ revision: c5a29a104738b98a9e76336939199e264163d4a0
1498
+ metrics:
1499
+ - type: map_at_1
1500
+ value: 23.179
1501
+ - type: map_at_10
1502
+ value: 35.86
1503
+ - type: map_at_100
1504
+ value: 37.025999999999996
1505
+ - type: map_at_1000
1506
+ value: 37.068
1507
+ - type: map_at_3
1508
+ value: 31.921
1509
+ - type: map_at_5
1510
+ value: 34.172000000000004
1511
+ - type: mrr_at_1
1512
+ value: 23.926
1513
+ - type: mrr_at_10
1514
+ value: 36.525999999999996
1515
+ - type: mrr_at_100
1516
+ value: 37.627
1517
+ - type: mrr_at_1000
1518
+ value: 37.665
1519
+ - type: mrr_at_3
1520
+ value: 32.653
1521
+ - type: mrr_at_5
1522
+ value: 34.897
1523
+ - type: ndcg_at_1
1524
+ value: 23.910999999999998
1525
+ - type: ndcg_at_10
1526
+ value: 42.927
1527
+ - type: ndcg_at_100
1528
+ value: 48.464
1529
+ - type: ndcg_at_1000
1530
+ value: 49.533
1531
+ - type: ndcg_at_3
1532
+ value: 34.910000000000004
1533
+ - type: ndcg_at_5
1534
+ value: 38.937
1535
+ - type: precision_at_1
1536
+ value: 23.910999999999998
1537
+ - type: precision_at_10
1538
+ value: 6.758
1539
+ - type: precision_at_100
1540
+ value: 0.9520000000000001
1541
+ - type: precision_at_1000
1542
+ value: 0.104
1543
+ - type: precision_at_3
1544
+ value: 14.838000000000001
1545
+ - type: precision_at_5
1546
+ value: 10.934000000000001
1547
+ - type: recall_at_1
1548
+ value: 23.179
1549
+ - type: recall_at_10
1550
+ value: 64.622
1551
+ - type: recall_at_100
1552
+ value: 90.135
1553
+ - type: recall_at_1000
1554
+ value: 98.301
1555
+ - type: recall_at_3
1556
+ value: 42.836999999999996
1557
+ - type: recall_at_5
1558
+ value: 52.512
1559
+ - task:
1560
+ type: Classification
1561
+ dataset:
1562
+ type: mteb/mtop_domain
1563
+ name: MTEB MTOPDomainClassification (en)
1564
+ config: en
1565
+ split: test
1566
+ revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
1567
+ metrics:
1568
+ - type: accuracy
1569
+ value: 96.59598723210215
1570
+ - type: f1
1571
+ value: 96.41913500001952
1572
+ - task:
1573
+ type: Classification
1574
+ dataset:
1575
+ type: mteb/mtop_intent
1576
+ name: MTEB MTOPIntentClassification (en)
1577
+ config: en
1578
+ split: test
1579
+ revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
1580
+ metrics:
1581
+ - type: accuracy
1582
+ value: 82.89557683538533
1583
+ - type: f1
1584
+ value: 63.379319722356264
1585
+ - task:
1586
+ type: Classification
1587
+ dataset:
1588
+ type: mteb/amazon_massive_intent
1589
+ name: MTEB MassiveIntentClassification (en)
1590
+ config: en
1591
+ split: test
1592
+ revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
1593
+ metrics:
1594
+ - type: accuracy
1595
+ value: 78.93745796906524
1596
+ - type: f1
1597
+ value: 75.71616541785902
1598
+ - task:
1599
+ type: Classification
1600
+ dataset:
1601
+ type: mteb/amazon_massive_scenario
1602
+ name: MTEB MassiveScenarioClassification (en)
1603
+ config: en
1604
+ split: test
1605
+ revision: 7d571f92784cd94a019292a1f45445077d0ef634
1606
+ metrics:
1607
+ - type: accuracy
1608
+ value: 81.41223940820443
1609
+ - type: f1
1610
+ value: 81.2877893719078
1611
+ - task:
1612
+ type: Clustering
1613
+ dataset:
1614
+ type: mteb/medrxiv-clustering-p2p
1615
+ name: MTEB MedrxivClusteringP2P
1616
+ config: default
1617
+ split: test
1618
+ revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73
1619
+ metrics:
1620
+ - type: v_measure
1621
+ value: 35.03682528325662
1622
+ - task:
1623
+ type: Clustering
1624
+ dataset:
1625
+ type: mteb/medrxiv-clustering-s2s
1626
+ name: MTEB MedrxivClusteringS2S
1627
+ config: default
1628
+ split: test
1629
+ revision: 35191c8c0dca72d8ff3efcd72aa802307d469663
1630
+ metrics:
1631
+ - type: v_measure
1632
+ value: 32.942529406124
1633
+ - task:
1634
+ type: Reranking
1635
+ dataset:
1636
+ type: mteb/mind_small
1637
+ name: MTEB MindSmallReranking
1638
+ config: default
1639
+ split: test
1640
+ revision: 3bdac13927fdc888b903db93b2ffdbd90b295a69
1641
+ metrics:
1642
+ - type: map
1643
+ value: 31.459949660460317
1644
+ - type: mrr
1645
+ value: 32.70509582031616
1646
+ - task:
1647
+ type: Retrieval
1648
+ dataset:
1649
+ type: mteb/nfcorpus
1650
+ name: MTEB NFCorpus
1651
+ config: default
1652
+ split: test
1653
+ revision: ec0fa4fe99da2ff19ca1214b7966684033a58814
1654
+ metrics:
1655
+ - type: map_at_1
1656
+ value: 6.497
1657
+ - type: map_at_10
1658
+ value: 13.843
1659
+ - type: map_at_100
1660
+ value: 17.713
1661
+ - type: map_at_1000
1662
+ value: 19.241
1663
+ - type: map_at_3
1664
+ value: 10.096
1665
+ - type: map_at_5
1666
+ value: 11.85
1667
+ - type: mrr_at_1
1668
+ value: 48.916
1669
+ - type: mrr_at_10
1670
+ value: 57.764
1671
+ - type: mrr_at_100
1672
+ value: 58.251
1673
+ - type: mrr_at_1000
1674
+ value: 58.282999999999994
1675
+ - type: mrr_at_3
1676
+ value: 55.623999999999995
1677
+ - type: mrr_at_5
1678
+ value: 57.018
1679
+ - type: ndcg_at_1
1680
+ value: 46.594
1681
+ - type: ndcg_at_10
1682
+ value: 36.945
1683
+ - type: ndcg_at_100
1684
+ value: 34.06
1685
+ - type: ndcg_at_1000
1686
+ value: 43.05
1687
+ - type: ndcg_at_3
1688
+ value: 41.738
1689
+ - type: ndcg_at_5
1690
+ value: 39.330999999999996
1691
+ - type: precision_at_1
1692
+ value: 48.916
1693
+ - type: precision_at_10
1694
+ value: 27.43
1695
+ - type: precision_at_100
1696
+ value: 8.616
1697
+ - type: precision_at_1000
1698
+ value: 2.155
1699
+ - type: precision_at_3
1700
+ value: 39.112
1701
+ - type: precision_at_5
1702
+ value: 33.808
1703
+ - type: recall_at_1
1704
+ value: 6.497
1705
+ - type: recall_at_10
1706
+ value: 18.163
1707
+ - type: recall_at_100
1708
+ value: 34.566
1709
+ - type: recall_at_1000
1710
+ value: 67.15
1711
+ - type: recall_at_3
1712
+ value: 11.100999999999999
1713
+ - type: recall_at_5
1714
+ value: 14.205000000000002
1715
+ - task:
1716
+ type: Retrieval
1717
+ dataset:
1718
+ type: mteb/nq
1719
+ name: MTEB NQ
1720
+ config: default
1721
+ split: test
1722
+ revision: b774495ed302d8c44a3a7ea25c90dbce03968f31
1723
+ metrics:
1724
+ - type: map_at_1
1725
+ value: 31.916
1726
+ - type: map_at_10
1727
+ value: 48.123
1728
+ - type: map_at_100
1729
+ value: 49.103
1730
+ - type: map_at_1000
1731
+ value: 49.131
1732
+ - type: map_at_3
1733
+ value: 43.711
1734
+ - type: map_at_5
1735
+ value: 46.323
1736
+ - type: mrr_at_1
1737
+ value: 36.181999999999995
1738
+ - type: mrr_at_10
1739
+ value: 50.617999999999995
1740
+ - type: mrr_at_100
1741
+ value: 51.329
1742
+ - type: mrr_at_1000
1743
+ value: 51.348000000000006
1744
+ - type: mrr_at_3
1745
+ value: 47.010999999999996
1746
+ - type: mrr_at_5
1747
+ value: 49.175000000000004
1748
+ - type: ndcg_at_1
1749
+ value: 36.181999999999995
1750
+ - type: ndcg_at_10
1751
+ value: 56.077999999999996
1752
+ - type: ndcg_at_100
1753
+ value: 60.037
1754
+ - type: ndcg_at_1000
1755
+ value: 60.63499999999999
1756
+ - type: ndcg_at_3
1757
+ value: 47.859
1758
+ - type: ndcg_at_5
1759
+ value: 52.178999999999995
1760
+ - type: precision_at_1
1761
+ value: 36.181999999999995
1762
+ - type: precision_at_10
1763
+ value: 9.284
1764
+ - type: precision_at_100
1765
+ value: 1.149
1766
+ - type: precision_at_1000
1767
+ value: 0.121
1768
+ - type: precision_at_3
1769
+ value: 22.006999999999998
1770
+ - type: precision_at_5
1771
+ value: 15.695
1772
+ - type: recall_at_1
1773
+ value: 31.916
1774
+ - type: recall_at_10
1775
+ value: 77.771
1776
+ - type: recall_at_100
1777
+ value: 94.602
1778
+ - type: recall_at_1000
1779
+ value: 98.967
1780
+ - type: recall_at_3
1781
+ value: 56.528
1782
+ - type: recall_at_5
1783
+ value: 66.527
1784
+ - task:
1785
+ type: Retrieval
1786
+ dataset:
1787
+ type: mteb/quora
1788
+ name: MTEB QuoraRetrieval
1789
+ config: default
1790
+ split: test
1791
+ revision: None
1792
+ metrics:
1793
+ - type: map_at_1
1794
+ value: 71.486
1795
+ - type: map_at_10
1796
+ value: 85.978
1797
+ - type: map_at_100
1798
+ value: 86.587
1799
+ - type: map_at_1000
1800
+ value: 86.598
1801
+ - type: map_at_3
1802
+ value: 83.04899999999999
1803
+ - type: map_at_5
1804
+ value: 84.857
1805
+ - type: mrr_at_1
1806
+ value: 82.32000000000001
1807
+ - type: mrr_at_10
1808
+ value: 88.64
1809
+ - type: mrr_at_100
1810
+ value: 88.702
1811
+ - type: mrr_at_1000
1812
+ value: 88.702
1813
+ - type: mrr_at_3
1814
+ value: 87.735
1815
+ - type: mrr_at_5
1816
+ value: 88.36
1817
+ - type: ndcg_at_1
1818
+ value: 82.34
1819
+ - type: ndcg_at_10
1820
+ value: 89.67
1821
+ - type: ndcg_at_100
1822
+ value: 90.642
1823
+ - type: ndcg_at_1000
1824
+ value: 90.688
1825
+ - type: ndcg_at_3
1826
+ value: 86.932
1827
+ - type: ndcg_at_5
1828
+ value: 88.408
1829
+ - type: precision_at_1
1830
+ value: 82.34
1831
+ - type: precision_at_10
1832
+ value: 13.675999999999998
1833
+ - type: precision_at_100
1834
+ value: 1.544
1835
+ - type: precision_at_1000
1836
+ value: 0.157
1837
+ - type: precision_at_3
1838
+ value: 38.24
1839
+ - type: precision_at_5
1840
+ value: 25.068
1841
+ - type: recall_at_1
1842
+ value: 71.486
1843
+ - type: recall_at_10
1844
+ value: 96.844
1845
+ - type: recall_at_100
1846
+ value: 99.843
1847
+ - type: recall_at_1000
1848
+ value: 99.996
1849
+ - type: recall_at_3
1850
+ value: 88.92099999999999
1851
+ - type: recall_at_5
1852
+ value: 93.215
1853
+ - task:
1854
+ type: Clustering
1855
+ dataset:
1856
+ type: mteb/reddit-clustering
1857
+ name: MTEB RedditClustering
1858
+ config: default
1859
+ split: test
1860
+ revision: 24640382cdbf8abc73003fb0fa6d111a705499eb
1861
+ metrics:
1862
+ - type: v_measure
1863
+ value: 59.75758437908334
1864
+ - task:
1865
+ type: Clustering
1866
+ dataset:
1867
+ type: mteb/reddit-clustering-p2p
1868
+ name: MTEB RedditClusteringP2P
1869
+ config: default
1870
+ split: test
1871
+ revision: 282350215ef01743dc01b456c7f5241fa8937f16
1872
+ metrics:
1873
+ - type: v_measure
1874
+ value: 68.03497914092789
1875
+ - task:
1876
+ type: Retrieval
1877
+ dataset:
1878
+ type: mteb/scidocs
1879
+ name: MTEB SCIDOCS
1880
+ config: default
1881
+ split: test
1882
+ revision: None
1883
+ metrics:
1884
+ - type: map_at_1
1885
+ value: 5.808
1886
+ - type: map_at_10
1887
+ value: 16.059
1888
+ - type: map_at_100
1889
+ value: 19.048000000000002
1890
+ - type: map_at_1000
1891
+ value: 19.43
1892
+ - type: map_at_3
1893
+ value: 10.953
1894
+ - type: map_at_5
1895
+ value: 13.363
1896
+ - type: mrr_at_1
1897
+ value: 28.7
1898
+ - type: mrr_at_10
1899
+ value: 42.436
1900
+ - type: mrr_at_100
1901
+ value: 43.599
1902
+ - type: mrr_at_1000
1903
+ value: 43.62
1904
+ - type: mrr_at_3
1905
+ value: 38.45
1906
+ - type: mrr_at_5
1907
+ value: 40.89
1908
+ - type: ndcg_at_1
1909
+ value: 28.7
1910
+ - type: ndcg_at_10
1911
+ value: 26.346000000000004
1912
+ - type: ndcg_at_100
1913
+ value: 36.758
1914
+ - type: ndcg_at_1000
1915
+ value: 42.113
1916
+ - type: ndcg_at_3
1917
+ value: 24.254
1918
+ - type: ndcg_at_5
1919
+ value: 21.506
1920
+ - type: precision_at_1
1921
+ value: 28.7
1922
+ - type: precision_at_10
1923
+ value: 13.969999999999999
1924
+ - type: precision_at_100
1925
+ value: 2.881
1926
+ - type: precision_at_1000
1927
+ value: 0.414
1928
+ - type: precision_at_3
1929
+ value: 22.933
1930
+ - type: precision_at_5
1931
+ value: 19.220000000000002
1932
+ - type: recall_at_1
1933
+ value: 5.808
1934
+ - type: recall_at_10
1935
+ value: 28.310000000000002
1936
+ - type: recall_at_100
1937
+ value: 58.475
1938
+ - type: recall_at_1000
1939
+ value: 84.072
1940
+ - type: recall_at_3
1941
+ value: 13.957
1942
+ - type: recall_at_5
1943
+ value: 19.515
1944
+ - task:
1945
+ type: STS
1946
+ dataset:
1947
+ type: mteb/sickr-sts
1948
+ name: MTEB SICK-R
1949
+ config: default
1950
+ split: test
1951
+ revision: a6ea5a8cab320b040a23452cc28066d9beae2cee
1952
+ metrics:
1953
+ - type: cos_sim_pearson
1954
+ value: 82.39274129958557
1955
+ - type: cos_sim_spearman
1956
+ value: 79.78021235170053
1957
+ - type: euclidean_pearson
1958
+ value: 79.35335401300166
1959
+ - type: euclidean_spearman
1960
+ value: 79.7271870968275
1961
+ - type: manhattan_pearson
1962
+ value: 79.35256263340601
1963
+ - type: manhattan_spearman
1964
+ value: 79.76036386976321
1965
+ - task:
1966
+ type: STS
1967
+ dataset:
1968
+ type: mteb/sts12-sts
1969
+ name: MTEB STS12
1970
+ config: default
1971
+ split: test
1972
+ revision: a0d554a64d88156834ff5ae9920b964011b16384
1973
+ metrics:
1974
+ - type: cos_sim_pearson
1975
+ value: 83.99130429246708
1976
+ - type: cos_sim_spearman
1977
+ value: 73.88322811171203
1978
+ - type: euclidean_pearson
1979
+ value: 80.7569419170376
1980
+ - type: euclidean_spearman
1981
+ value: 73.82542155409597
1982
+ - type: manhattan_pearson
1983
+ value: 80.79468183847625
1984
+ - type: manhattan_spearman
1985
+ value: 73.87027144047784
1986
+ - task:
1987
+ type: STS
1988
+ dataset:
1989
+ type: mteb/sts13-sts
1990
+ name: MTEB STS13
1991
+ config: default
1992
+ split: test
1993
+ revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca
1994
+ metrics:
1995
+ - type: cos_sim_pearson
1996
+ value: 84.88548789489907
1997
+ - type: cos_sim_spearman
1998
+ value: 85.07535893847255
1999
+ - type: euclidean_pearson
2000
+ value: 84.6637222061494
2001
+ - type: euclidean_spearman
2002
+ value: 85.14200626702456
2003
+ - type: manhattan_pearson
2004
+ value: 84.75327892344734
2005
+ - type: manhattan_spearman
2006
+ value: 85.24406181838596
2007
+ - task:
2008
+ type: STS
2009
+ dataset:
2010
+ type: mteb/sts14-sts
2011
+ name: MTEB STS14
2012
+ config: default
2013
+ split: test
2014
+ revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375
2015
+ metrics:
2016
+ - type: cos_sim_pearson
2017
+ value: 82.88140039325008
2018
+ - type: cos_sim_spearman
2019
+ value: 79.61211268112362
2020
+ - type: euclidean_pearson
2021
+ value: 81.29639728816458
2022
+ - type: euclidean_spearman
2023
+ value: 79.51284578041442
2024
+ - type: manhattan_pearson
2025
+ value: 81.3381797137111
2026
+ - type: manhattan_spearman
2027
+ value: 79.55683684039808
2028
+ - task:
2029
+ type: STS
2030
+ dataset:
2031
+ type: mteb/sts15-sts
2032
+ name: MTEB STS15
2033
+ config: default
2034
+ split: test
2035
+ revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3
2036
+ metrics:
2037
+ - type: cos_sim_pearson
2038
+ value: 85.16716737270485
2039
+ - type: cos_sim_spearman
2040
+ value: 86.14823841857738
2041
+ - type: euclidean_pearson
2042
+ value: 85.36325733440725
2043
+ - type: euclidean_spearman
2044
+ value: 86.04919691402029
2045
+ - type: manhattan_pearson
2046
+ value: 85.3147511385052
2047
+ - type: manhattan_spearman
2048
+ value: 86.00676205857764
2049
+ - task:
2050
+ type: STS
2051
+ dataset:
2052
+ type: mteb/sts16-sts
2053
+ name: MTEB STS16
2054
+ config: default
2055
+ split: test
2056
+ revision: 4d8694f8f0e0100860b497b999b3dbed754a0513
2057
+ metrics:
2058
+ - type: cos_sim_pearson
2059
+ value: 80.34266645861588
2060
+ - type: cos_sim_spearman
2061
+ value: 81.59914035005882
2062
+ - type: euclidean_pearson
2063
+ value: 81.15053076245988
2064
+ - type: euclidean_spearman
2065
+ value: 81.52776915798489
2066
+ - type: manhattan_pearson
2067
+ value: 81.1819647418673
2068
+ - type: manhattan_spearman
2069
+ value: 81.57479527353556
2070
+ - task:
2071
+ type: STS
2072
+ dataset:
2073
+ type: mteb/sts17-crosslingual-sts
2074
+ name: MTEB STS17 (en-en)
2075
+ config: en-en
2076
+ split: test
2077
+ revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d
2078
+ metrics:
2079
+ - type: cos_sim_pearson
2080
+ value: 89.38263326821439
2081
+ - type: cos_sim_spearman
2082
+ value: 89.10946308202642
2083
+ - type: euclidean_pearson
2084
+ value: 88.87831312540068
2085
+ - type: euclidean_spearman
2086
+ value: 89.03615865973664
2087
+ - type: manhattan_pearson
2088
+ value: 88.79835539970384
2089
+ - type: manhattan_spearman
2090
+ value: 88.9766156339753
2091
+ - task:
2092
+ type: STS
2093
+ dataset:
2094
+ type: mteb/sts22-crosslingual-sts
2095
+ name: MTEB STS22 (en)
2096
+ config: en
2097
+ split: test
2098
+ revision: eea2b4fe26a775864c896887d910b76a8098ad3f
2099
+ metrics:
2100
+ - type: cos_sim_pearson
2101
+ value: 70.1574915581685
2102
+ - type: cos_sim_spearman
2103
+ value: 70.59144980004054
2104
+ - type: euclidean_pearson
2105
+ value: 71.43246306918755
2106
+ - type: euclidean_spearman
2107
+ value: 70.5544189562984
2108
+ - type: manhattan_pearson
2109
+ value: 71.4071414609503
2110
+ - type: manhattan_spearman
2111
+ value: 70.31799126163712
2112
+ - task:
2113
+ type: STS
2114
+ dataset:
2115
+ type: mteb/stsbenchmark-sts
2116
+ name: MTEB STSBenchmark
2117
+ config: default
2118
+ split: test
2119
+ revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831
2120
+ metrics:
2121
+ - type: cos_sim_pearson
2122
+ value: 83.36215796635351
2123
+ - type: cos_sim_spearman
2124
+ value: 83.07276756467208
2125
+ - type: euclidean_pearson
2126
+ value: 83.06690453635584
2127
+ - type: euclidean_spearman
2128
+ value: 82.9635366303289
2129
+ - type: manhattan_pearson
2130
+ value: 83.04994049700815
2131
+ - type: manhattan_spearman
2132
+ value: 82.98120125356036
2133
+ - task:
2134
+ type: Reranking
2135
+ dataset:
2136
+ type: mteb/scidocs-reranking
2137
+ name: MTEB SciDocsRR
2138
+ config: default
2139
+ split: test
2140
+ revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab
2141
+ metrics:
2142
+ - type: map
2143
+ value: 86.92530011616722
2144
+ - type: mrr
2145
+ value: 96.21826793395421
2146
+ - task:
2147
+ type: Retrieval
2148
+ dataset:
2149
+ type: mteb/scifact
2150
+ name: MTEB SciFact
2151
+ config: default
2152
+ split: test
2153
+ revision: 0228b52cf27578f30900b9e5271d331663a030d7
2154
+ metrics:
2155
+ - type: map_at_1
2156
+ value: 65.75
2157
+ - type: map_at_10
2158
+ value: 77.701
2159
+ - type: map_at_100
2160
+ value: 78.005
2161
+ - type: map_at_1000
2162
+ value: 78.006
2163
+ - type: map_at_3
2164
+ value: 75.48
2165
+ - type: map_at_5
2166
+ value: 76.927
2167
+ - type: mrr_at_1
2168
+ value: 68.333
2169
+ - type: mrr_at_10
2170
+ value: 78.511
2171
+ - type: mrr_at_100
2172
+ value: 78.704
2173
+ - type: mrr_at_1000
2174
+ value: 78.704
2175
+ - type: mrr_at_3
2176
+ value: 77
2177
+ - type: mrr_at_5
2178
+ value: 78.083
2179
+ - type: ndcg_at_1
2180
+ value: 68.333
2181
+ - type: ndcg_at_10
2182
+ value: 82.42699999999999
2183
+ - type: ndcg_at_100
2184
+ value: 83.486
2185
+ - type: ndcg_at_1000
2186
+ value: 83.511
2187
+ - type: ndcg_at_3
2188
+ value: 78.96300000000001
2189
+ - type: ndcg_at_5
2190
+ value: 81.028
2191
+ - type: precision_at_1
2192
+ value: 68.333
2193
+ - type: precision_at_10
2194
+ value: 10.667
2195
+ - type: precision_at_100
2196
+ value: 1.127
2197
+ - type: precision_at_1000
2198
+ value: 0.11299999999999999
2199
+ - type: precision_at_3
2200
+ value: 31.333
2201
+ - type: precision_at_5
2202
+ value: 20.133000000000003
2203
+ - type: recall_at_1
2204
+ value: 65.75
2205
+ - type: recall_at_10
2206
+ value: 95.578
2207
+ - type: recall_at_100
2208
+ value: 99.833
2209
+ - type: recall_at_1000
2210
+ value: 100
2211
+ - type: recall_at_3
2212
+ value: 86.506
2213
+ - type: recall_at_5
2214
+ value: 91.75
2215
+ - task:
2216
+ type: PairClassification
2217
+ dataset:
2218
+ type: mteb/sprintduplicatequestions-pairclassification
2219
+ name: MTEB SprintDuplicateQuestions
2220
+ config: default
2221
+ split: test
2222
+ revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
2223
+ metrics:
2224
+ - type: cos_sim_accuracy
2225
+ value: 99.75247524752476
2226
+ - type: cos_sim_ap
2227
+ value: 94.16065078045173
2228
+ - type: cos_sim_f1
2229
+ value: 87.22986247544205
2230
+ - type: cos_sim_precision
2231
+ value: 85.71428571428571
2232
+ - type: cos_sim_recall
2233
+ value: 88.8
2234
+ - type: dot_accuracy
2235
+ value: 99.74554455445545
2236
+ - type: dot_ap
2237
+ value: 93.90633887037264
2238
+ - type: dot_f1
2239
+ value: 86.9873417721519
2240
+ - type: dot_precision
2241
+ value: 88.1025641025641
2242
+ - type: dot_recall
2243
+ value: 85.9
2244
+ - type: euclidean_accuracy
2245
+ value: 99.75247524752476
2246
+ - type: euclidean_ap
2247
+ value: 94.17466319018055
2248
+ - type: euclidean_f1
2249
+ value: 87.3405299313052
2250
+ - type: euclidean_precision
2251
+ value: 85.74181117533719
2252
+ - type: euclidean_recall
2253
+ value: 89
2254
+ - type: manhattan_accuracy
2255
+ value: 99.75445544554455
2256
+ - type: manhattan_ap
2257
+ value: 94.27688371923577
2258
+ - type: manhattan_f1
2259
+ value: 87.74002954209749
2260
+ - type: manhattan_precision
2261
+ value: 86.42095053346266
2262
+ - type: manhattan_recall
2263
+ value: 89.1
2264
+ - type: max_accuracy
2265
+ value: 99.75445544554455
2266
+ - type: max_ap
2267
+ value: 94.27688371923577
2268
+ - type: max_f1
2269
+ value: 87.74002954209749
2270
+ - task:
2271
+ type: Clustering
2272
+ dataset:
2273
+ type: mteb/stackexchange-clustering
2274
+ name: MTEB StackExchangeClustering
2275
+ config: default
2276
+ split: test
2277
+ revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259
2278
+ metrics:
2279
+ - type: v_measure
2280
+ value: 71.26500637517056
2281
+ - task:
2282
+ type: Clustering
2283
+ dataset:
2284
+ type: mteb/stackexchange-clustering-p2p
2285
+ name: MTEB StackExchangeClusteringP2P
2286
+ config: default
2287
+ split: test
2288
+ revision: 815ca46b2622cec33ccafc3735d572c266efdb44
2289
+ metrics:
2290
+ - type: v_measure
2291
+ value: 39.17507906280528
2292
+ - task:
2293
+ type: Reranking
2294
+ dataset:
2295
+ type: mteb/stackoverflowdupquestions-reranking
2296
+ name: MTEB StackOverflowDupQuestions
2297
+ config: default
2298
+ split: test
2299
+ revision: e185fbe320c72810689fc5848eb6114e1ef5ec69
2300
+ metrics:
2301
+ - type: map
2302
+ value: 52.4848744828509
2303
+ - type: mrr
2304
+ value: 53.33678168236992
2305
+ - task:
2306
+ type: Summarization
2307
+ dataset:
2308
+ type: mteb/summeval
2309
+ name: MTEB SummEval
2310
+ config: default
2311
+ split: test
2312
+ revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c
2313
+ metrics:
2314
+ - type: cos_sim_pearson
2315
+ value: 30.599864323827887
2316
+ - type: cos_sim_spearman
2317
+ value: 30.91116204665598
2318
+ - type: dot_pearson
2319
+ value: 30.82637894269936
2320
+ - type: dot_spearman
2321
+ value: 30.957573868416066
2322
+ - task:
2323
+ type: Retrieval
2324
+ dataset:
2325
+ type: mteb/trec-covid
2326
+ name: MTEB TRECCOVID
2327
+ config: default
2328
+ split: test
2329
+ revision: None
2330
+ metrics:
2331
+ - type: map_at_1
2332
+ value: 0.23600000000000002
2333
+ - type: map_at_10
2334
+ value: 1.892
2335
+ - type: map_at_100
2336
+ value: 11.586
2337
+ - type: map_at_1000
2338
+ value: 27.761999999999997
2339
+ - type: map_at_3
2340
+ value: 0.653
2341
+ - type: map_at_5
2342
+ value: 1.028
2343
+ - type: mrr_at_1
2344
+ value: 88
2345
+ - type: mrr_at_10
2346
+ value: 94
2347
+ - type: mrr_at_100
2348
+ value: 94
2349
+ - type: mrr_at_1000
2350
+ value: 94
2351
+ - type: mrr_at_3
2352
+ value: 94
2353
+ - type: mrr_at_5
2354
+ value: 94
2355
+ - type: ndcg_at_1
2356
+ value: 82
2357
+ - type: ndcg_at_10
2358
+ value: 77.48899999999999
2359
+ - type: ndcg_at_100
2360
+ value: 60.141
2361
+ - type: ndcg_at_1000
2362
+ value: 54.228
2363
+ - type: ndcg_at_3
2364
+ value: 82.358
2365
+ - type: ndcg_at_5
2366
+ value: 80.449
2367
+ - type: precision_at_1
2368
+ value: 88
2369
+ - type: precision_at_10
2370
+ value: 82.19999999999999
2371
+ - type: precision_at_100
2372
+ value: 61.760000000000005
2373
+ - type: precision_at_1000
2374
+ value: 23.684
2375
+ - type: precision_at_3
2376
+ value: 88
2377
+ - type: precision_at_5
2378
+ value: 85.6
2379
+ - type: recall_at_1
2380
+ value: 0.23600000000000002
2381
+ - type: recall_at_10
2382
+ value: 2.117
2383
+ - type: recall_at_100
2384
+ value: 14.985000000000001
2385
+ - type: recall_at_1000
2386
+ value: 51.107
2387
+ - type: recall_at_3
2388
+ value: 0.688
2389
+ - type: recall_at_5
2390
+ value: 1.1039999999999999
2391
+ - task:
2392
+ type: Retrieval
2393
+ dataset:
2394
+ type: mteb/touche2020
2395
+ name: MTEB Touche2020
2396
+ config: default
2397
+ split: test
2398
+ revision: a34f9a33db75fa0cbb21bb5cfc3dae8dc8bec93f
2399
+ metrics:
2400
+ - type: map_at_1
2401
+ value: 2.3040000000000003
2402
+ - type: map_at_10
2403
+ value: 9.025
2404
+ - type: map_at_100
2405
+ value: 15.312999999999999
2406
+ - type: map_at_1000
2407
+ value: 16.954
2408
+ - type: map_at_3
2409
+ value: 4.981
2410
+ - type: map_at_5
2411
+ value: 6.32
2412
+ - type: mrr_at_1
2413
+ value: 24.490000000000002
2414
+ - type: mrr_at_10
2415
+ value: 39.835
2416
+ - type: mrr_at_100
2417
+ value: 40.8
2418
+ - type: mrr_at_1000
2419
+ value: 40.8
2420
+ - type: mrr_at_3
2421
+ value: 35.034
2422
+ - type: mrr_at_5
2423
+ value: 37.687
2424
+ - type: ndcg_at_1
2425
+ value: 22.448999999999998
2426
+ - type: ndcg_at_10
2427
+ value: 22.545
2428
+ - type: ndcg_at_100
2429
+ value: 35.931999999999995
2430
+ - type: ndcg_at_1000
2431
+ value: 47.665
2432
+ - type: ndcg_at_3
2433
+ value: 23.311
2434
+ - type: ndcg_at_5
2435
+ value: 22.421
2436
+ - type: precision_at_1
2437
+ value: 24.490000000000002
2438
+ - type: precision_at_10
2439
+ value: 20.408
2440
+ - type: precision_at_100
2441
+ value: 7.815999999999999
2442
+ - type: precision_at_1000
2443
+ value: 1.553
2444
+ - type: precision_at_3
2445
+ value: 25.169999999999998
2446
+ - type: precision_at_5
2447
+ value: 23.265
2448
+ - type: recall_at_1
2449
+ value: 2.3040000000000003
2450
+ - type: recall_at_10
2451
+ value: 15.693999999999999
2452
+ - type: recall_at_100
2453
+ value: 48.917
2454
+ - type: recall_at_1000
2455
+ value: 84.964
2456
+ - type: recall_at_3
2457
+ value: 6.026
2458
+ - type: recall_at_5
2459
+ value: 9.066
2460
+ - task:
2461
+ type: Classification
2462
+ dataset:
2463
+ type: mteb/toxic_conversations_50k
2464
+ name: MTEB ToxicConversationsClassification
2465
+ config: default
2466
+ split: test
2467
+ revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c
2468
+ metrics:
2469
+ - type: accuracy
2470
+ value: 82.6074
2471
+ - type: ap
2472
+ value: 23.187467098602013
2473
+ - type: f1
2474
+ value: 65.36829506379657
2475
+ - task:
2476
+ type: Classification
2477
+ dataset:
2478
+ type: mteb/tweet_sentiment_extraction
2479
+ name: MTEB TweetSentimentExtractionClassification
2480
+ config: default
2481
+ split: test
2482
+ revision: d604517c81ca91fe16a244d1248fc021f9ecee7a
2483
+ metrics:
2484
+ - type: accuracy
2485
+ value: 63.16355404640635
2486
+ - type: f1
2487
+ value: 63.534725639863346
2488
+ - task:
2489
+ type: Clustering
2490
+ dataset:
2491
+ type: mteb/twentynewsgroups-clustering
2492
+ name: MTEB TwentyNewsgroupsClustering
2493
+ config: default
2494
+ split: test
2495
+ revision: 6125ec4e24fa026cec8a478383ee943acfbd5449
2496
+ metrics:
2497
+ - type: v_measure
2498
+ value: 50.91004094411276
2499
+ - task:
2500
+ type: PairClassification
2501
+ dataset:
2502
+ type: mteb/twittersemeval2015-pairclassification
2503
+ name: MTEB TwitterSemEval2015
2504
+ config: default
2505
+ split: test
2506
+ revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
2507
+ metrics:
2508
+ - type: cos_sim_accuracy
2509
+ value: 86.55301901412649
2510
+ - type: cos_sim_ap
2511
+ value: 75.25312618556728
2512
+ - type: cos_sim_f1
2513
+ value: 68.76561719140429
2514
+ - type: cos_sim_precision
2515
+ value: 65.3061224489796
2516
+ - type: cos_sim_recall
2517
+ value: 72.61213720316623
2518
+ - type: dot_accuracy
2519
+ value: 86.29671574178936
2520
+ - type: dot_ap
2521
+ value: 75.11910195501207
2522
+ - type: dot_f1
2523
+ value: 68.44048376830045
2524
+ - type: dot_precision
2525
+ value: 66.12546125461255
2526
+ - type: dot_recall
2527
+ value: 70.92348284960423
2528
+ - type: euclidean_accuracy
2529
+ value: 86.5828217202122
2530
+ - type: euclidean_ap
2531
+ value: 75.22986344900924
2532
+ - type: euclidean_f1
2533
+ value: 68.81267797449549
2534
+ - type: euclidean_precision
2535
+ value: 64.8238861674831
2536
+ - type: euclidean_recall
2537
+ value: 73.3245382585752
2538
+ - type: manhattan_accuracy
2539
+ value: 86.61262442629791
2540
+ - type: manhattan_ap
2541
+ value: 75.24401608557328
2542
+ - type: manhattan_f1
2543
+ value: 68.80473982483257
2544
+ - type: manhattan_precision
2545
+ value: 67.21187720181177
2546
+ - type: manhattan_recall
2547
+ value: 70.47493403693932
2548
+ - type: max_accuracy
2549
+ value: 86.61262442629791
2550
+ - type: max_ap
2551
+ value: 75.25312618556728
2552
+ - type: max_f1
2553
+ value: 68.81267797449549
2554
+ - task:
2555
+ type: PairClassification
2556
+ dataset:
2557
+ type: mteb/twitterurlcorpus-pairclassification
2558
+ name: MTEB TwitterURLCorpus
2559
+ config: default
2560
+ split: test
2561
+ revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
2562
+ metrics:
2563
+ - type: cos_sim_accuracy
2564
+ value: 88.10688089416696
2565
+ - type: cos_sim_ap
2566
+ value: 84.17862178779863
2567
+ - type: cos_sim_f1
2568
+ value: 76.17305208781748
2569
+ - type: cos_sim_precision
2570
+ value: 71.31246641590543
2571
+ - type: cos_sim_recall
2572
+ value: 81.74468740375731
2573
+ - type: dot_accuracy
2574
+ value: 88.1844995536927
2575
+ - type: dot_ap
2576
+ value: 84.33816725235876
2577
+ - type: dot_f1
2578
+ value: 76.43554032918746
2579
+ - type: dot_precision
2580
+ value: 74.01557767200346
2581
+ - type: dot_recall
2582
+ value: 79.0190945488143
2583
+ - type: euclidean_accuracy
2584
+ value: 88.07001203089223
2585
+ - type: euclidean_ap
2586
+ value: 84.12267000814985
2587
+ - type: euclidean_f1
2588
+ value: 76.12232600180778
2589
+ - type: euclidean_precision
2590
+ value: 74.50604541433205
2591
+ - type: euclidean_recall
2592
+ value: 77.81028641823221
2593
+ - type: manhattan_accuracy
2594
+ value: 88.06419063142779
2595
+ - type: manhattan_ap
2596
+ value: 84.11648917164187
2597
+ - type: manhattan_f1
2598
+ value: 76.20579953925474
2599
+ - type: manhattan_precision
2600
+ value: 72.56772755762935
2601
+ - type: manhattan_recall
2602
+ value: 80.22790267939637
2603
+ - type: max_accuracy
2604
+ value: 88.1844995536927
2605
+ - type: max_ap
2606
+ value: 84.33816725235876
2607
+ - type: max_f1
2608
+ value: 76.43554032918746
2609
+ ---
2610
+
2611
+ <!-- **English** | [中文](./README_zh.md) -->
2612
+
2613
+ # gte-large-en-v1.5
2614
+
2615
+ We introduce `gte-v1.5` series, upgraded `gte` embeddings that support the context length of up to **8192**, while further enhancing model performance.
2616
+ The models are built upon the `transformer++` encoder [backbone](https://huggingface.co/Alibaba-NLP/new-impl) (BERT + RoPE + GLU).
2617
+
2618
+ The `gte-v1.5` series achieve state-of-the-art scores on the MTEB benchmark within the same model size category and prodvide competitive on the LoCo long-context retrieval tests (refer to [Evaluation](#evaluation)).
2619
+
2620
+ We also present the [`gte-Qwen1.5-7B-instruct`](https://huggingface.co/Alibaba-NLP/gte-Qwen1.5-7B-instruct),
2621
+ a SOTA instruction-tuned multi-lingual embedding model that ranked 2nd in MTEB and 1st in C-MTEB.
2622
+
2623
+ <!-- Provide a longer summary of what this model is. -->
2624
+
2625
+ - **Developed by:** Institute for Intelligent Computing, Alibaba Group
2626
+ - **Model type:** Text Embeddings
2627
+ - **Paper:** Coming soon.
2628
+
2629
+ <!-- - **Demo [optional]:** [More Information Needed] -->
2630
+
2631
+ ### Model list
2632
+
2633
+ | Models | Language | Model Size | Max Seq. Length | Dimension | MTEB-en | LoCo |
2634
+ |:-----: | :-----: |:-----: |:-----: |:-----: | :-----: | :-----: |
2635
+ |[`gte-Qwen1.5-7B-instruct`](https://huggingface.co/Alibaba-NLP/gte-Qwen1.5-7B-instruct)| Multiple | 7720 | 32768 | 4096 | 67.34 | 87.57 |
2636
+ |[`gte-large-en-v1.5`](https://huggingface.co/Alibaba-NLP/gte-large-en-v1.5) | English | 434 | 8192 | 1024 | 65.39 | 86.71 |
2637
+ |[`gte-base-en-v1.5`](https://huggingface.co/Alibaba-NLP/gte-base-en-v1.5) | English | 137 | 8192 | 768 | 64.11 | 87.44 |
2638
+
2639
+
2640
+ ## How to Get Started with the Model
2641
+
2642
+ Use the code below to get started with the model.
2643
+
2644
+ ```python
2645
+ # Requires transformers>=4.36.0
2646
+
2647
+ import torch.nn.functional as F
2648
+ from transformers import AutoModel, AutoTokenizer
2649
+
2650
+ input_texts = [
2651
+ "what is the capital of China?",
2652
+ "how to implement quick sort in python?",
2653
+ "Beijing",
2654
+ "sorting algorithms"
2655
+ ]
2656
+
2657
+ model_path = 'Alibaba-NLP/gte-large-en-v1.5'
2658
+ tokenizer = AutoTokenizer.from_pretrained(model_path)
2659
+ model = AutoModel.from_pretrained(model_path, trust_remote_code=True)
2660
+
2661
+ # Tokenize the input texts
2662
+ batch_dict = tokenizer(input_texts, max_length=8192, padding=True, truncation=True, return_tensors='pt')
2663
+
2664
+ outputs = model(**batch_dict)
2665
+ embeddings = outputs.last_hidden_state[:, 0]
2666
+
2667
+ # (Optionally) normalize embeddings
2668
+ embeddings = F.normalize(embeddings, p=2, dim=1)
2669
+ scores = (embeddings[:1] @ embeddings[1:].T) * 100
2670
+ print(scores.tolist())
2671
+ ```
2672
+
2673
+ **It is recommended to install xformers and enable unpadding for acceleration, refer to [enable-unpadding-and-xformers](https://huggingface.co/Alibaba-NLP/new-impl#recommendation-enable-unpadding-and-acceleration-with-xformers).**
2674
+
2675
+
2676
+ Use with sentence-transformers:
2677
+
2678
+ ```python
2679
+ # Requires sentence_transformers>=2.7.0
2680
+
2681
+ from sentence_transformers import SentenceTransformer
2682
+ from sentence_transformers.util import cos_sim
2683
+
2684
+ sentences = ['That is a happy person', 'That is a very happy person']
2685
+
2686
+ model = SentenceTransformer('Alibaba-NLP/gte-large-en-v1.5', trust_remote_code=True)
2687
+ embeddings = model.encode(sentences)
2688
+ print(cos_sim(embeddings[0], embeddings[1]))
2689
+ ```
2690
+
2691
+ Use with `transformers.js`:
2692
+
2693
+ ```js
2694
+ // npm i @xenova/transformers
2695
+ import { pipeline, dot } from '@xenova/transformers';
2696
+
2697
+ // Create feature extraction pipeline
2698
+ const extractor = await pipeline('feature-extraction', 'Alibaba-NLP/gte-large-en-v1.5', {
2699
+ quantized: false, // Comment out this line to use the quantized version
2700
+ });
2701
+
2702
+ // Generate sentence embeddings
2703
+ const sentences = [
2704
+ "what is the capital of China?",
2705
+ "how to implement quick sort in python?",
2706
+ "Beijing",
2707
+ "sorting algorithms"
2708
+ ]
2709
+ const output = await extractor(sentences, { normalize: true, pooling: 'cls' });
2710
+
2711
+ // Compute similarity scores
2712
+ const [source_embeddings, ...document_embeddings ] = output.tolist();
2713
+ const similarities = document_embeddings.map(x => 100 * dot(source_embeddings, x));
2714
+ console.log(similarities); // [41.86354093370361, 77.07076371259589, 37.02981979677899]
2715
+ ```
2716
+
2717
+ ## Training Details
2718
+
2719
+ ### Training Data
2720
+
2721
+ - Masked language modeling (MLM): `c4-en`
2722
+ - Weak-supervised contrastive (WSC) pre-training: [GTE](https://arxiv.org/pdf/2308.03281.pdf) pre-training data
2723
+ - Supervised contrastive fine-tuning: GTE(https://arxiv.org/pdf/2308.03281.pdf) fine-tuning data
2724
+
2725
+ ### Training Procedure
2726
+
2727
+ To enable the backbone model to support a context length of 8192, we adopted a multi-stage training strategy.
2728
+ The model first undergoes preliminary MLM pre-training on shorter lengths.
2729
+ And then, we resample the data, reducing the proportion of short texts, and continue the MLM pre-training.
2730
+
2731
+ The entire training process is as follows:
2732
+ - MLM-512: lr 2e-4, mlm_probability 0.3, batch_size 4096, num_steps 300000, rope_base 10000
2733
+ - MLM-2048: lr 5e-5, mlm_probability 0.3, batch_size 4096, num_steps 30000, rope_base 10000
2734
+ - MLM-8192: lr 5e-5, mlm_probability 0.3, batch_size 1024, num_steps 30000, rope_base 160000
2735
+ - WSC: max_len 512, lr 5e-5, batch_size 28672, num_steps 100000
2736
+ - Fine-tuning: TODO
2737
+
2738
+
2739
+ ## Evaluation
2740
+
2741
+
2742
+ ### MTEB
2743
+
2744
+ The results of other models are retrieved from [MTEB leaderboard](https://huggingface.co/spaces/mteb/leaderboard).
2745
+
2746
+ The gte evaluation setting: `mteb==1.2.0, fp16 auto mix precision, max_length=8192`, and set ntk scaling factor to 2 (equivalent to rope_base * 2).
2747
+
2748
+ | Model Name | Param Size (M) | Dimension | Sequence Length | Average (56) | Class. (12) | Clust. (11) | Pair Class. (3) | Reran. (4) | Retr. (15) | STS (10) | Summ. (1) |
2749
+ |:----:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
2750
+ | [**gte-large-en-v1.5**](https://huggingface.co/Alibaba-NLP/gte-large-en-v1.5) | 409 | 1024 | 8192 | **65.39** | 77.75 | 47.95 | 84.63 | 58.50 | 57.91 | 81.43 | 30.91 |
2751
+ | [mxbai-embed-large-v1](https://huggingface.co/mixedbread-ai/mxbai-embed-large-v1) | 335 | 1024 | 512 | 64.68 | 75.64 | 46.71 | 87.2 | 60.11 | 54.39 | 85 | 32.71 |
2752
+ | [multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) | 560 | 1024 | 514 | 64.41 | 77.56 | 47.1 | 86.19 | 58.58 | 52.47 | 84.78 | 30.39 |
2753
+ | [bge-large-en-v1.5](https://huggingface.co/BAAI/bge-large-en-v1.5)| 335 | 1024 | 512 | 64.23 | 75.97 | 46.08 | 87.12 | 60.03 | 54.29 | 83.11 | 31.61 |
2754
+ | [**gte-base-en-v1.5**](https://huggingface.co/Alibaba-NLP/gte-base-en-v1.5) | 137 | 768 | 8192 | **64.11** | 77.17 | 46.82 | 85.33 | 57.66 | 54.09 | 81.97 | 31.17 |
2755
+ | [bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5)| 109 | 768 | 512 | 63.55 | 75.53 | 45.77 | 86.55 | 58.86 | 53.25 | 82.4 | 31.07 |
2756
+
2757
+
2758
+ ### LoCo
2759
+
2760
+ | Model Name | Dimension | Sequence Length | Average (5) | QsmsumRetrieval | SummScreenRetrieval | QasperAbastractRetrieval | QasperTitleRetrieval | GovReportRetrieval |
2761
+ |:----:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|:---:|
2762
+ | [gte-qwen1.5-7b](https://huggingface.co/Alibaba-NLP/gte-qwen1.5-7b) | 4096 | 32768 | 87.57 | 49.37 | 93.10 | 99.67 | 97.54 | 98.21 |
2763
+ | [gte-large-v1.5](https://huggingface.co/Alibaba-NLP/gte-large-v1.5) |1024 | 8192 | 86.71 | 44.55 | 92.61 | 99.82 | 97.81 | 98.74 |
2764
+ | [gte-base-v1.5](https://huggingface.co/Alibaba-NLP/gte-base-v1.5) | 768 | 8192 | 87.44 | 49.91 | 91.78 | 99.82 | 97.13 | 98.58 |
2765
+
2766
+
2767
+
2768
+ ## Citation
2769
+
2770
+ If you find our paper or models helpful, please consider citing them as follows:
2771
+
2772
+ ```
2773
+ @article{li2023towards,
2774
+ title={Towards general text embeddings with multi-stage contrastive learning},
2775
+ author={Li, Zehan and Zhang, Xin and Zhang, Yanzhao and Long, Dingkun and Xie, Pengjun and Zhang, Meishan},
2776
+ journal={arXiv preprint arXiv:2308.03281},
2777
+ year={2023}
2778
+ }
2779
+ ```
embedding_model/config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Alibaba-NLP/gte-large-en-v1.5",
3
+ "architectures": [
4
+ "NewModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "auto_map": {
8
+ "AutoConfig": "Alibaba-NLP/new-impl--configuration.NewConfig",
9
+ "AutoModel": "Alibaba-NLP/new-impl--modeling.NewModel",
10
+ "AutoModelForMaskedLM": "Alibaba-NLP/new-impl--modeling.NewForMaskedLM",
11
+ "AutoModelForMultipleChoice": "Alibaba-NLP/new-impl--modeling.NewForMultipleChoice",
12
+ "AutoModelForQuestionAnswering": "Alibaba-NLP/new-impl--modeling.NewForQuestionAnswering",
13
+ "AutoModelForSequenceClassification": "Alibaba-NLP/new-impl--modeling.NewForSequenceClassification",
14
+ "AutoModelForTokenClassification": "Alibaba-NLP/new-impl--modeling.NewForTokenClassification"
15
+ },
16
+ "classifier_dropout": null,
17
+ "hidden_act": "gelu",
18
+ "hidden_dropout_prob": 0.1,
19
+ "hidden_size": 1024,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 4096,
22
+ "layer_norm_eps": 1e-12,
23
+ "layer_norm_type": "layer_norm",
24
+ "logn_attention_clip1": false,
25
+ "logn_attention_scale": false,
26
+ "max_position_embeddings": 8192,
27
+ "model_type": "new",
28
+ "num_attention_heads": 16,
29
+ "num_hidden_layers": 24,
30
+ "pack_qkv": true,
31
+ "pad_token_id": 0,
32
+ "position_embedding_type": "rope",
33
+ "rope_scaling": {
34
+ "factor": 2.0,
35
+ "type": "ntk"
36
+ },
37
+ "rope_theta": 160000,
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.41.2",
40
+ "type_vocab_size": 2,
41
+ "unpad_inputs": false,
42
+ "use_memory_efficient_attention": false,
43
+ "vocab_size": 30528
44
+ }
embedding_model/config_sentence_transformers.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.7.0",
4
+ "transformers": "4.41.2",
5
+ "pytorch": "2.3.0+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null
9
+ }
embedding_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe6e4200b833d5332b7c61859d7f4ff204211b1583d732353efe1b7594176cf2
3
+ size 1736585680
embedding_model/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
+ ]
embedding_model/sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 8192,
3
+ "do_lower_case": false
4
+ }
embedding_model/special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
embedding_model/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
embedding_model/tokenizer_config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "mask_token": "[MASK]",
48
+ "max_length": 8000,
49
+ "model_max_length": 32768,
50
+ "pad_to_multiple_of": null,
51
+ "pad_token": "[PAD]",
52
+ "pad_token_type_id": 0,
53
+ "padding_side": "right",
54
+ "sep_token": "[SEP]",
55
+ "stride": 0,
56
+ "strip_accents": null,
57
+ "tokenize_chinese_chars": true,
58
+ "tokenizer_class": "BertTokenizer",
59
+ "truncation_side": "right",
60
+ "truncation_strategy": "longest_first",
61
+ "unk_token": "[UNK]"
62
+ }
embedding_model/vocab.txt ADDED
The diff for this file is too large to render. See raw diff