Barakuga commited on
Commit
ec9144e
·
verified ·
1 Parent(s): a277299

Add new SentenceTransformer model

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 1024,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
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
+ }
README.md ADDED
@@ -0,0 +1,571 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - dense
7
+ - generated_from_trainer
8
+ - dataset_size:19244
9
+ - loss:MultipleNegativesRankingLoss
10
+ base_model: intfloat/multilingual-e5-large
11
+ widget:
12
+ - source_sentence: 'query: @user Alright we would both see eye to eye that contraception
13
+ is ethically better, correct? I would think the most hard‑line #catholics would
14
+ agree with that even though they don''t support it. So sentientce is your line?
15
+ There is proof sentientce at roughly 18 to 25 weeks.'
16
+ sentences:
17
+ - 'passage: title: Aggressive or Moderate Fluid Resuscitation in Acute Pancreatitis
18
+ abstract: Early aggressive hydration is widely recommended for the management
19
+ of acute pancreatitis, but evidence for this practice is limited.'
20
+ - 'passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly
21
+ Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent
22
+ pathogens? Conventional wisdom is that natural selection will remove highly lethal
23
+ pathogens if host death greatly reduces transmission. Vaccines that keep hosts
24
+ alive but still allow transmission could thus allow very virulent strains to circulate
25
+ in a population. Here we show experimentally that immunization of chickens against
26
+ Marek''s disease virus enhances the fitness of more virulent strains, making it
27
+ possible for hyperpathogenic strains to transmit. Immunity elicited by direct
28
+ vaccination or by maternal vaccination prolongs host survival but does not prevent
29
+ infection, viral replication or transmission, thus extending the infectious periods
30
+ of strains otherwise too lethal to persist. Our data show that anti-disease vaccines
31
+ that do not prevent transmission can create conditions that promote the emergence
32
+ of pathogen strains that cause more severe disease in unvaccinated hosts.'
33
+ - 'passage: title: When is the Capacity for Sentience Acquired During Human Fetal
34
+ Development? abstract: The question of when the human fetus develops the capacity
35
+ for sentience is central to many contentious issues. The answer could and should
36
+ influence attitudes toward IVF and embryo experimentation, abortion, and fetal
37
+ and neonatal surgery. For the fetus to be described as sentient, the somatosensory
38
+ pathways from the periphery to the primary somatosensory region of the cerebral
39
+ cortex must be established and functional. Fetal behaviour is described and the
40
+ development of the underlying anatomical substrate and the chemical and electrical
41
+ pathways involved in the detection, transmission, and perception of somatosensory
42
+ stimuli are reviewed.It is concluded that the basic neuronal substrate required
43
+ to transmit somatosensory information develops by mid-gestation (18 to 25 weeks),
44
+ however, the functional capacity of the neural circuitry is limited by the immaturity
45
+ of the system. Thus, 18 to 25 weeks is considered the earliest stage at which
46
+ the lower boundary of sentience could be placed. At this stage of development,
47
+ however, there is little evidence for the central processing of somatosensory
48
+ information. Before 30 weeks gestational age, EEG activity is extremely limited
49
+ and somatosensory evoked potentials are immature, lacking components which correlate
50
+ with information processing within the cerebral cortex. Thus, 30 weeks is considered
51
+ a more plausible stage of fetal development at which the lower boundary for sentience
52
+ could be placed.'
53
+ - source_sentence: 'query: @_Karl_F_ karl ein paar zahlen inzwischen dafür auch schon
54
+ kostspielig ... einige mit dem leben & einige verlor auch seine gesundheit ein...
55
+ magst nur hoffen, daß es nicht alle betroffen, die dazu mehrfach verpflichtet
56
+ & gezwungen wurden.'
57
+ sentences:
58
+ - 'passage: title: Health effects in COPD smokers who switch to electronic cigarettes:
59
+ a retrospective-prospective 3-year follow-up abstract: Health effects of electronic
60
+ cigarette (EC) use in patients with chronic obstructive pulmonary disease (COPD)
61
+ are largely unexplored.We present findings from a long-term prospective assessment
62
+ of respiratory parameters in a cohort of COPD patients who ceased or substantially
63
+ reduced conventional cigarette use with ECs.We prospectively re-evaluated COPD
64
+ exacerbations, spirometric indices, subjective assessments (using the COPD Assessment
65
+ Tool [CAT] scores), physical activity (measured by the 6-minute walk distance
66
+ [6MWD]), and conventional cigarette use in EC users with COPD who were retrospectively
67
+ assessed previously. Baseline measurements prior to switching to EC use were compared
68
+ to follow-up visits at 12, 24, and 36 months. Age- and sex-matched regularly smoking
69
+ COPD patients who were not using ECs were included as reference (control) group.Complete
70
+ data were available from 44 patients. Compared to baseline in the EC-user group,
71
+ there was a marked decline in the use of conventional cigarettes. Although there
72
+ was no change in lung function, significant improvements in COPD exacerbation
73
+ rates, CAT scores, and 6MWD were observed consistently in the EC user group over
74
+ the 3-year period (p<0.01). Similar findings were noted in COPD EC users who also
75
+ smoked conventional cigarettes ("dual users").The present study suggests that
76
+ EC use may ameliorate objective and subjective COPD outcomes and that the benefits
77
+ gained may persist long-term. EC use may reverse some of the harm resulting from
78
+ tobacco smoking in COPD patients.'
79
+ - 'passage: title: A Case Report: Multifocal Necrotizing Encephalitis and Myocarditis
80
+ after BNT162b2 mRNA Vaccination against COVID-19 abstract: The current report
81
+ presents the case of a 76-year-old man with Parkinson''s disease (PD) who died
82
+ three weeks after receiving his third COVID-19 vaccination. The patient was first
83
+ vaccinated in May 2021 with the ChAdOx1 nCov-19 vector vaccine, followed by two
84
+ doses of the BNT162b2 mRNA vaccine in July and December 2021. The family of the
85
+ deceased requested an autopsy due to ambiguous clinical signs before death. PD
86
+ was confirmed by post-mortem examinations. Furthermore, signs of aspiration pneumonia
87
+ and systemic arteriosclerosis were evident. However, histopathological analyses
88
+ of the brain uncovered previously unsuspected findings, including acute vasculitis
89
+ (predominantly lymphocytic) as well as multifocal necrotizing encephalitis of
90
+ unknown etiology with pronounced inflammation including glial and lymphocytic
91
+ reaction. In the heart, signs of chronic cardiomyopathy as well as mild acute
92
+ lympho-histiocytic myocarditis and vasculitis were present. Although there was
93
+ no history of COVID-19 for this patient, immunohistochemistry for SARS-CoV-2 antigens
94
+ (spike and nucleocapsid proteins) was performed. Surprisingly, only spike protein
95
+ but no nucleocapsid protein could be detected within the foci of inflammation
96
+ in both the brain and the heart, particularly in the endothelial cells of small
97
+ blood vessels. Since no nucleocapsid protein could be detected, the presence of
98
+ spike protein must be ascribed to vaccination rather than to viral infection.
99
+ The findings corroborate previous reports of encephalitis and myocarditis caused
100
+ by gene-based COVID-19 vaccines.'
101
+ - 'passage: title: Persistent warm Mediterranean surface waters during the Roman
102
+ period abstract: Abstract Reconstruction of last millennia Sea Surface Temperature
103
+ (SST) evolution is challenging due to the difficulty retrieving good resolution
104
+ marine records and to the several uncertainties in the available proxy tools.
105
+ In this regard, the Roman Period (1 CE to 500 CE) was particularly relevant in
106
+ the socio-cultural development of the Mediterranean region while its climatic
107
+ characteristics remain uncertain. Here we present a new SST reconstruction from
108
+ the Sicily Channel based in Mg/Ca ratios measured on the planktonic foraminifer
109
+ Globigerinoides ruber . This new record is framed in the context of other previously
110
+ published Mediterranean SST records from the Alboran Sea, Minorca Basin and Aegean
111
+ Sea and also compared to a north Hemisphere temperature reconstruction. The most
112
+ solid image that emerges of this trans-Mediterranean comparison is the persistent
113
+ regional occurrence of a distinct warm phase during the Roman Period. This record
114
+ comparison consistently shows the Roman as the warmest period of the last 2 kyr,
115
+ about 2 °C warmer than average values for the late centuries for the Sicily and
116
+ Western Mediterranean regions. After the Roman Period a general cooling trend
117
+ developed in the region with several minor oscillations. We hypothesis the potential
118
+ link between this Roman Climatic Optimum and the expansion and subsequent decline
119
+ of the Roman Empire.'
120
+ - source_sentence: 'query: 4/ Trotzdem versäumten Sie es in mehreren Publikationen
121
+ mit Landt, einen relevanten Interessenkonflikt zu deklarieren. Markantestes Beispiel,
122
+ Ihre Publikation zum #PCR-Protokoll für den Nachweis von #SARSCoV2.'
123
+ sentences:
124
+ - 'passage: title: Cancer risk associated with simian virus 40 contaminated polio
125
+ vaccine. abstract: The presence of SV40 in monkey cell cultures used in the preparation
126
+ of the polio vaccine from 1955 through 1961 is well documented. Investigations
127
+ have consistently demonstrated the oncogenic behavior of SV40 in animal models.
128
+ Early epidemiologic studies were inadequate in demonstrating an increase in cancer
129
+ incidence associated with contaminated vaccine. Recently, investigators have provided
130
+ persuasive evidence that SV40 is present in human ependymomas, choroid plexus
131
+ tumors, bone tumors, and mesotheliomas, however, the etiologic role of the virus
132
+ in tumorigenesis has not been established.Using data from SEER, we analyzed the
133
+ incidence of brain tumors, bone tumors, and mesotheliomas from 1973-1993 and the
134
+ possible relationship of these tumors with the administration of the SV40 contaminated
135
+ vaccine.Our analysis indicates increased rates of ependymomas (37%), osteogenic
136
+ sarcomas (26%), other bone tumors (34%) and mesothelioma (90%) among those in
137
+ the exposed as compared to the unexposed birth cohort.These data suggest that
138
+ there may be an increased incidence of certain cancers among the 98 million persons
139
+ exposed to contaminated polio vaccine in the U.S.; further investigations are
140
+ clearly justified.'
141
+ - 'passage: title: Detection of 2019 novel coronavirus (2019-nCoV) by real-time
142
+ RT-PCR abstract: Background The ongoing outbreak of the recently emerged novel
143
+ coronavirus (2019-nCoV) poses a challenge for public health laboratories as virus
144
+ isolates are unavailable while there is growing evidence that the outbreak is
145
+ more widespread than initially thought, and international spread through travellers
146
+ does already occur. Aim We aimed to develop and deploy robust diagnostic methodology
147
+ for use in public health laboratory settings without having virus material available.
148
+ Methods Here we present a validated diagnostic workflow for 2019-nCoV, its design
149
+ relying on close genetic relatedness of 2019-nCoV with SARS coronavirus, making
150
+ use of synthetic nucleic acid technology. Results The workflow reliably detects
151
+ 2019-nCoV, and further discriminates 2019-nCoV from SARS-CoV. Through coordination
152
+ between academic and public laboratories, we confirmed assay exclusivity based
153
+ on 297 original clinical specimens containing a full spectrum of human respiratory
154
+ viruses. Control material is made available through European Virus Archive – Global
155
+ (EVAg), a European Union infrastructure project. Conclusion The present study
156
+ demonstrates the enormous response capacity achieved through coordination of academic
157
+ and public laboratories in national and European research networks.'
158
+ - 'passage: title: The lifetime cost of driving a car abstract: The car is one of
159
+ the most expensive household consumer goods, yet there is a limited understanding
160
+ of its private (internal) and social (external) cost per vehicle-km, year, or
161
+ lifetime of driving. This paper provides an overview of 23 private and ten social
162
+ cost items, and assesses these for three popular car models in Germany for the
163
+ year 2020. Results confirm that motorists underestimate the full private costs
164
+ of car ownership, while policy makers and planners underestimate social costs.
165
+ For the typical German travel distance of 15,000 car kilometers per year, the
166
+ total lifetime cost of car ownership (50 years) ranges between €599,082 for an
167
+ Opel Corsa to €956,798 for a Mercedes GLC. The share of this cost born by society
168
+ is 41% (€4674 per year) for the Opel Corsa, and 29% (€5273 per year) for the Mercedes
169
+ GLC. Findings suggest that for low-income groups, private car ownership can represent
170
+ a cost equal to housing, consuming a large share of disposable income. This creates
171
+ complexities in perceptions of transport costs, the economic viability of alternative
172
+ transport modes, or the justification of taxes.'
173
+ - source_sentence: 'query: Type 1 Diabetes Incidence and Risk in Children With a Diagnosis
174
+ of COVID-19 As adults we are meant to be shielding the children. We are letting
175
+ them down day by day'
176
+ sentences:
177
+ - 'passage: title: Persistence of neutralizing antibodies a year after SARS‐CoV‐2
178
+ infection in humans abstract: Abstract Most subjects develop antibodies to SARS‐CoV‐2
179
+ following infection. In order to estimate the duration of immunity induced by
180
+ SARS‐CoV‐2 it is important to understand for how long antibodies persist after
181
+ infection in humans. Here, we assessed the persistence of serum antibodies following
182
+ WT SARS‐CoV‐2 infection at 8 and 13 months after diagnosis in 367 individuals.
183
+ The SARS‐CoV‐2 spike IgG (S‐IgG) and nucleoprotein IgG (N‐IgG) concentrations
184
+ and the proportion of subjects with neutralizing antibodies (NAb) were assessed.
185
+ Moreover, the NAb titers among a smaller subset of participants ( n = 78) against
186
+ a WT virus (B) and variants of concern (VOCs): Alpha (B.1.1.7), Beta (B.1.351),
187
+ and Delta (B.1.617.2) were determined. We found that NAb against the WT virus
188
+ persisted in 89% and S‐IgG in 97% of subjects for at least 13 months after infection.
189
+ Only 36% had N‐IgG by 13 months. The mean S‐IgG concentrations declined from 8
190
+ to 13 months by less than one third; N‐IgG concentrations declined by two‐thirds.
191
+ Subjects with severe infection had markedly higher IgG and NAb levels and are
192
+ expected to remain seropositive for longer. Significantly lower NAb titers against
193
+ the variants compared to the WT virus, especially after a mild disease, suggests
194
+ reduced protection against VOCs.'
195
+ - 'passage: title: Metastasis and Immune Evasion from Extracellular cGAMP Hydrolysis
196
+ abstract: Abstract Cytosolic DNA is characteristic of chromosomally unstable metastatic
197
+ cancer cells, resulting in constitutive activation of the cGAS–STING innate immune
198
+ pathway. How tumors co-opt inflammatory signaling while evading immune surveillance
199
+ remains unknown. Here, we show that the ectonucleotidase ENPP1 promotes metastasis
200
+ by selectively degrading extracellular cGAMP, an immune-stimulatory metabolite
201
+ whose breakdown products include the immune suppressor adenosine. ENPP1 loss suppresses
202
+ metastasis, restores tumor immune infiltration, and potentiates response to immune
203
+ checkpoint blockade in a manner dependent on tumor cGAS and host STING. Conversely,
204
+ overexpression of wild-type ENPP1, but not an enzymatically weakened mutant, promotes
205
+ migration and metastasis, in part through the generation of extracellular adenosine,
206
+ and renders otherwise sensitive tumors completely resistant to immunotherapy.
207
+ In human cancers, ENPP1 expression correlates with reduced immune cell infiltration,
208
+ increased metastasis, and resistance to anti–PD-1/PD-L1 treatment. Thus, cGAMP
209
+ hydrolysis by ENPP1 enables chromosomally unstable tumors to transmute cGAS activation
210
+ into an immune-suppressive pathway. Significance: Chromosomal instability promotes
211
+ metastasis by generating chronic tumor inflammation. ENPP1 facilitates metastasis
212
+ and enables tumor cells to tolerate inflammation by hydrolyzing the immunotransmitter
213
+ cGAMP, preventing its transfer from cancer cells to immune cells. This article
214
+ is highlighted in the In This Issue feature, p. 995'
215
+ - 'passage: title: Type 1 Diabetes Incidence and Risk in Children With a Diagnosis
216
+ of COVID-19 abstract: This study used a population-based individual patient data
217
+ set that included diagnoses of COVID-19 to determine whether there was a temporal
218
+ association between COVID-19 and type 1 diabetes in children.'
219
+ - source_sentence: 'query: @user It’s quite possibly the reverse (Pfizer @ 39%) “Our
220
+ data show that anti-disease vaccines that do not prevent transmission can create
221
+ conditions that promote the emergence of pathogen strains that cause more severe
222
+ disease in unvaccinated hosts” Source:'
223
+ sentences:
224
+ - 'passage: title: Experimental Assessment of Carbon Dioxide Content in Inhaled
225
+ Air With or Without Face Masks in Healthy Children abstract: This randomized clinical
226
+ trial measured inhaled and exhaled carbon dioxide in children with and without
227
+ face masks.'
228
+ - 'passage: title: Access to lifesaving medical resources for African countries:
229
+ COVID-19 testing and response, ethics, and politics abstract: Coronavirus disease
230
+ 2019 (COVID-19) has revealed how strikingly unprepared the world is for a pandemic
231
+ and how easily viruses spread in our interconnected world. A governance crisis
232
+ is unfolding alongside the pandemic as health officials around the world compete
233
+ for access to scarce medical supplies. As governments of African countries, and
234
+ those in low-income and middle-income countries around the world, seek to avoid
235
+ potentially catastrophic epidemics and learn from what has worked in other countries,
236
+ testing and other medical resources are of concern.'
237
+ - 'passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly
238
+ Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent
239
+ pathogens? Conventional wisdom is that natural selection will remove highly lethal
240
+ pathogens if host death greatly reduces transmission. Vaccines that keep hosts
241
+ alive but still allow transmission could thus allow very virulent strains to circulate
242
+ in a population. Here we show experimentally that immunization of chickens against
243
+ Marek''s disease virus enhances the fitness of more virulent strains, making it
244
+ possible for hyperpathogenic strains to transmit. Immunity elicited by direct
245
+ vaccination or by maternal vaccination prolongs host survival but does not prevent
246
+ infection, viral replication or transmission, thus extending the infectious periods
247
+ of strains otherwise too lethal to persist. Our data show that anti-disease vaccines
248
+ that do not prevent transmission can create conditions that promote the emergence
249
+ of pathogen strains that cause more severe disease in unvaccinated hosts.'
250
+ pipeline_tag: sentence-similarity
251
+ library_name: sentence-transformers
252
+ ---
253
+
254
+ # SentenceTransformer based on intfloat/multilingual-e5-large
255
+
256
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
257
+
258
+ ## Model Details
259
+
260
+ ### Model Description
261
+ - **Model Type:** Sentence Transformer
262
+ - **Base model:** [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) <!-- at revision 3d7cfbdacd47fdda877c5cd8a79fbcc4f2a574f3 -->
263
+ - **Maximum Sequence Length:** 512 tokens
264
+ - **Output Dimensionality:** 1024 dimensions
265
+ - **Similarity Function:** Cosine Similarity
266
+ <!-- - **Training Dataset:** Unknown -->
267
+ <!-- - **Language:** Unknown -->
268
+ <!-- - **License:** Unknown -->
269
+
270
+ ### Model Sources
271
+
272
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
273
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers)
274
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
275
+
276
+ ### Full Model Architecture
277
+
278
+ ```
279
+ SentenceTransformer(
280
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'XLMRobertaModel'})
281
+ (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
282
+ (2): Normalize()
283
+ )
284
+ ```
285
+
286
+ ## Usage
287
+
288
+ ### Direct Usage (Sentence Transformers)
289
+
290
+ First install the Sentence Transformers library:
291
+
292
+ ```bash
293
+ pip install -U sentence-transformers
294
+ ```
295
+
296
+ Then you can load this model and run inference.
297
+ ```python
298
+ from sentence_transformers import SentenceTransformer
299
+
300
+ # Download from the 🤗 Hub
301
+ model = SentenceTransformer("Barakuga/me5-checkthat-task1")
302
+ # Run inference
303
+ sentences = [
304
+ 'query: @user It’s quite possibly the reverse (Pfizer @ 39%) “Our data show that anti-disease vaccines that do not prevent transmission can create conditions that promote the emergence of pathogen strains that cause more severe disease in unvaccinated hosts” Source:',
305
+ "passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent pathogens? Conventional wisdom is that natural selection will remove highly lethal pathogens if host death greatly reduces transmission. Vaccines that keep hosts alive but still allow transmission could thus allow very virulent strains to circulate in a population. Here we show experimentally that immunization of chickens against Marek's disease virus enhances the fitness of more virulent strains, making it possible for hyperpathogenic strains to transmit. Immunity elicited by direct vaccination or by maternal vaccination prolongs host survival but does not prevent infection, viral replication or transmission, thus extending the infectious periods of strains otherwise too lethal to persist. Our data show that anti-disease vaccines that do not prevent transmission can create conditions that promote the emergence of pathogen strains that cause more severe disease in unvaccinated hosts.",
306
+ 'passage: title: Access to lifesaving medical resources for African countries: COVID-19 testing and response, ethics, and politics abstract: Coronavirus disease 2019 (COVID-19) has revealed how strikingly unprepared the world is for a pandemic and how easily viruses spread in our interconnected world. A governance crisis is unfolding alongside the pandemic as health officials around the world compete for access to scarce medical supplies. As governments of African countries, and those in low-income and middle-income countries around the world, seek to avoid potentially catastrophic epidemics and learn from what has worked in other countries, testing and other medical resources are of concern.',
307
+ ]
308
+ embeddings = model.encode(sentences)
309
+ print(embeddings.shape)
310
+ # [3, 1024]
311
+
312
+ # Get the similarity scores for the embeddings
313
+ similarities = model.similarity(embeddings, embeddings)
314
+ print(similarities)
315
+ # tensor([[ 1.0000, 0.7027, -0.0760],
316
+ # [ 0.7027, 1.0000, -0.0821],
317
+ # [-0.0760, -0.0821, 1.0000]])
318
+ ```
319
+
320
+ <!--
321
+ ### Direct Usage (Transformers)
322
+
323
+ <details><summary>Click to see the direct usage in Transformers</summary>
324
+
325
+ </details>
326
+ -->
327
+
328
+ <!--
329
+ ### Downstream Usage (Sentence Transformers)
330
+
331
+ You can finetune this model on your own dataset.
332
+
333
+ <details><summary>Click to expand</summary>
334
+
335
+ </details>
336
+ -->
337
+
338
+ <!--
339
+ ### Out-of-Scope Use
340
+
341
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
342
+ -->
343
+
344
+ <!--
345
+ ## Bias, Risks and Limitations
346
+
347
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
348
+ -->
349
+
350
+ <!--
351
+ ### Recommendations
352
+
353
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
354
+ -->
355
+
356
+ ## Training Details
357
+
358
+ ### Training Dataset
359
+
360
+ #### Unnamed Dataset
361
+
362
+ * Size: 19,244 training samples
363
+ * Columns: <code>sentence_0</code> and <code>sentence_1</code>
364
+ * Approximate statistics based on the first 1000 samples:
365
+ | | sentence_0 | sentence_1 |
366
+ |:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
367
+ | type | string | string |
368
+ | details | <ul><li>min: 21 tokens</li><li>mean: 59.89 tokens</li><li>max: 134 tokens</li></ul> | <ul><li>min: 30 tokens</li><li>mean: 336.42 tokens</li><li>max: 512 tokens</li></ul> |
369
+ * Samples:
370
+ | sentence_0 | sentence_1 |
371
+ |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
372
+ | <code>query: In what way will Language Modelers such as ChatGPT Impact Jobs and Sectors? by Edward W. Felten, Manav Raj, Robert Seamans :: SSRN</code> | <code>passage: title: How will Language Modelers like ChatGPT Affect Occupations and Industries? abstract: Recent dramatic increases in AI language modeling capabilities has led to many questions about the effect of these technologies on the economy. In this paper we present a methodology to systematically assess the extent to which occupations, industries and geographies are exposed to advances in AI language modeling capabilities. We find that the top occupations exposed to language modeling include telemarketers and a variety of post-secondary teachers such as English language and literature, foreign language and literature, and history teachers. We find the top industries exposed to advances in language modeling are legal services and securities, commodities, and investments. We also find a positive correlation between wages and exposure to AI language modeling.</code> |
373
+ | <code>query: Spannende Studie zu #POTS. Sie verdeutlicht, was man ärztlich häufig wahrnimmt, nämlich dass die geistige Leistungsfähigkeit beim Sitzen und Stehen nachlässt. In diesem Fall waren Konzentration und Ausführungsfunktion gegen Kontrollen vermindert. 1/6</code> | <code>passage: title: Cognitive functioning in postural orthostatic tachycardia syndrome among different body positions: a prospective pilot study (POTSKog study) abstract: Approximately 96% of patients with postural orthostatic tachycardia syndrome (PoTS) report cognitive complaints. We investigated whether cognitive function is impaired during sitting and active standing in 30 patients with PoTS compared with 30 healthy controls (HCs) and whether it will improve with the counter manoeuvre of leg crossing.In this prospective pilot study, patients with PoTS were compared to HCs matched for age, sex, and educational level. Baseline data included norepinephrine plasma levels, autonomic testing and baseline cognitive function in a seated position [the Montreal Cognitive Assessment, the Leistungsprüfsystem (LPS) subtests 1 and 2, and the Test of Attentional Performance (TAP)]. Cognitive functioning was examined in a randomized order in supine, upright and upright legs crossed position. The prima...</code> |
374
+ | <code>query: We now know that Omicron is far from mild. In the unvaccinated it is equally lethal, while being more contagious, as other strains. Most children were and remain not vaccinated. We were aware that in winter 2021. And we know it now.</code> | <code>passage: title: Intrinsic and effective severity of COVID-19 cases infected with the ancestral strain and Omicron BA.2 variant in Hong Kong abstract: ABSTRACT Background Understanding severity of infections with SARS-CoV-2 and its variants is crucial to inform public health measures. Here we used COVID-19 patient data from Hong Kong to characterise the severity profile of COVID-19 and to examine factors associated with fatality of infection. Methods Time-varying and age-specific effective severity measured by case-hospitalization risk and hospitalization risk was estimated with all individual COVID-19 case data collected in Hong Kong from 23 January 2020 through to 26 October 2022 over six epidemic waves, in comparison with estimates of influenza A(H1N1)pdm09 during the 2009 pandemic. The intrinsic severity of Omicron BA.2 was compared with the estimate for the ancestral strain with the data from unvaccinated patients without previous infections. Factors potentially associated with the...</code> |
375
+ * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
376
+ ```json
377
+ {
378
+ "scale": 20.0,
379
+ "similarity_fct": "cos_sim",
380
+ "gather_across_devices": false,
381
+ "directions": [
382
+ "query_to_doc"
383
+ ],
384
+ "partition_mode": "joint",
385
+ "hardness_mode": null,
386
+ "hardness_strength": 0.0
387
+ }
388
+ ```
389
+
390
+ ### Training Hyperparameters
391
+ #### Non-Default Hyperparameters
392
+
393
+ - `per_device_train_batch_size`: 4
394
+ - `per_device_eval_batch_size`: 4
395
+ - `num_train_epochs`: 1
396
+ - `fp16`: True
397
+ - `multi_dataset_batch_sampler`: round_robin
398
+
399
+ #### All Hyperparameters
400
+ <details><summary>Click to expand</summary>
401
+
402
+ - `do_predict`: False
403
+ - `eval_strategy`: no
404
+ - `prediction_loss_only`: True
405
+ - `per_device_train_batch_size`: 4
406
+ - `per_device_eval_batch_size`: 4
407
+ - `gradient_accumulation_steps`: 1
408
+ - `eval_accumulation_steps`: None
409
+ - `torch_empty_cache_steps`: None
410
+ - `learning_rate`: 5e-05
411
+ - `weight_decay`: 0.0
412
+ - `adam_beta1`: 0.9
413
+ - `adam_beta2`: 0.999
414
+ - `adam_epsilon`: 1e-08
415
+ - `max_grad_norm`: 1
416
+ - `num_train_epochs`: 1
417
+ - `max_steps`: -1
418
+ - `lr_scheduler_type`: linear
419
+ - `lr_scheduler_kwargs`: None
420
+ - `warmup_ratio`: None
421
+ - `warmup_steps`: 0
422
+ - `log_level`: passive
423
+ - `log_level_replica`: warning
424
+ - `log_on_each_node`: True
425
+ - `logging_nan_inf_filter`: True
426
+ - `enable_jit_checkpoint`: False
427
+ - `save_on_each_node`: False
428
+ - `save_only_model`: False
429
+ - `restore_callback_states_from_checkpoint`: False
430
+ - `use_cpu`: False
431
+ - `seed`: 42
432
+ - `data_seed`: None
433
+ - `bf16`: False
434
+ - `fp16`: True
435
+ - `bf16_full_eval`: False
436
+ - `fp16_full_eval`: False
437
+ - `tf32`: None
438
+ - `local_rank`: -1
439
+ - `ddp_backend`: None
440
+ - `debug`: []
441
+ - `dataloader_drop_last`: False
442
+ - `dataloader_num_workers`: 0
443
+ - `dataloader_prefetch_factor`: None
444
+ - `disable_tqdm`: False
445
+ - `remove_unused_columns`: True
446
+ - `label_names`: None
447
+ - `load_best_model_at_end`: False
448
+ - `ignore_data_skip`: False
449
+ - `fsdp`: []
450
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
451
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
452
+ - `parallelism_config`: None
453
+ - `deepspeed`: None
454
+ - `label_smoothing_factor`: 0.0
455
+ - `optim`: adamw_torch_fused
456
+ - `optim_args`: None
457
+ - `group_by_length`: False
458
+ - `length_column_name`: length
459
+ - `project`: huggingface
460
+ - `trackio_space_id`: trackio
461
+ - `ddp_find_unused_parameters`: None
462
+ - `ddp_bucket_cap_mb`: None
463
+ - `ddp_broadcast_buffers`: False
464
+ - `dataloader_pin_memory`: True
465
+ - `dataloader_persistent_workers`: False
466
+ - `skip_memory_metrics`: True
467
+ - `push_to_hub`: False
468
+ - `resume_from_checkpoint`: None
469
+ - `hub_model_id`: None
470
+ - `hub_strategy`: every_save
471
+ - `hub_private_repo`: None
472
+ - `hub_always_push`: False
473
+ - `hub_revision`: None
474
+ - `gradient_checkpointing`: False
475
+ - `gradient_checkpointing_kwargs`: None
476
+ - `include_for_metrics`: []
477
+ - `eval_do_concat_batches`: True
478
+ - `auto_find_batch_size`: False
479
+ - `full_determinism`: False
480
+ - `ddp_timeout`: 1800
481
+ - `torch_compile`: False
482
+ - `torch_compile_backend`: None
483
+ - `torch_compile_mode`: None
484
+ - `include_num_input_tokens_seen`: no
485
+ - `neftune_noise_alpha`: None
486
+ - `optim_target_modules`: None
487
+ - `batch_eval_metrics`: False
488
+ - `eval_on_start`: False
489
+ - `use_liger_kernel`: False
490
+ - `liger_kernel_config`: None
491
+ - `eval_use_gather_object`: False
492
+ - `average_tokens_across_devices`: True
493
+ - `use_cache`: False
494
+ - `prompts`: None
495
+ - `batch_sampler`: batch_sampler
496
+ - `multi_dataset_batch_sampler`: round_robin
497
+ - `router_mapping`: {}
498
+ - `learning_rate_mapping`: {}
499
+
500
+ </details>
501
+
502
+ ### Training Logs
503
+ | Epoch | Step | Training Loss |
504
+ |:------:|:----:|:-------------:|
505
+ | 0.1039 | 500 | 0.1880 |
506
+ | 0.2079 | 1000 | 0.1486 |
507
+ | 0.3118 | 1500 | 0.1368 |
508
+ | 0.4157 | 2000 | 0.1392 |
509
+ | 0.5196 | 2500 | 0.1169 |
510
+ | 0.6236 | 3000 | 0.1305 |
511
+ | 0.7275 | 3500 | 0.1070 |
512
+ | 0.8314 | 4000 | 0.1079 |
513
+ | 0.9354 | 4500 | 0.1064 |
514
+
515
+
516
+ ### Framework Versions
517
+ - Python: 3.12.13
518
+ - Sentence Transformers: 5.3.0
519
+ - Transformers: 5.0.0
520
+ - PyTorch: 2.10.0+cu128
521
+ - Accelerate: 1.13.0
522
+ - Datasets: 4.0.0
523
+ - Tokenizers: 0.22.2
524
+
525
+ ## Citation
526
+
527
+ ### BibTeX
528
+
529
+ #### Sentence Transformers
530
+ ```bibtex
531
+ @inproceedings{reimers-2019-sentence-bert,
532
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
533
+ author = "Reimers, Nils and Gurevych, Iryna",
534
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
535
+ month = "11",
536
+ year = "2019",
537
+ publisher = "Association for Computational Linguistics",
538
+ url = "https://arxiv.org/abs/1908.10084",
539
+ }
540
+ ```
541
+
542
+ #### MultipleNegativesRankingLoss
543
+ ```bibtex
544
+ @misc{oord2019representationlearningcontrastivepredictive,
545
+ title={Representation Learning with Contrastive Predictive Coding},
546
+ author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
547
+ year={2019},
548
+ eprint={1807.03748},
549
+ archivePrefix={arXiv},
550
+ primaryClass={cs.LG},
551
+ url={https://arxiv.org/abs/1807.03748},
552
+ }
553
+ ```
554
+
555
+ <!--
556
+ ## Glossary
557
+
558
+ *Clearly define terms in order to be accessible across audiences.*
559
+ -->
560
+
561
+ <!--
562
+ ## Model Card Authors
563
+
564
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
565
+ -->
566
+
567
+ <!--
568
+ ## Model Card Contact
569
+
570
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
571
+ -->
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "XLMRobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "dtype": "float32",
10
+ "eos_token_id": 2,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 4096,
16
+ "is_decoder": false,
17
+ "layer_norm_eps": 1e-05,
18
+ "max_position_embeddings": 514,
19
+ "model_type": "xlm-roberta",
20
+ "num_attention_heads": 16,
21
+ "num_hidden_layers": 24,
22
+ "output_past": true,
23
+ "pad_token_id": 1,
24
+ "position_embedding_type": "absolute",
25
+ "tie_word_embeddings": true,
26
+ "transformers_version": "5.0.0",
27
+ "type_vocab_size": 1,
28
+ "use_cache": true,
29
+ "vocab_size": 250002
30
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "SentenceTransformer",
3
+ "__version__": {
4
+ "sentence_transformers": "5.3.0",
5
+ "transformers": "5.0.0",
6
+ "pytorch": "2.10.0+cu128"
7
+ },
8
+ "prompts": {
9
+ "query": "",
10
+ "document": ""
11
+ },
12
+ "default_prompt_name": null,
13
+ "similarity_fn_name": "cosine"
14
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae083d97adc0b051c7699c537f31b9123c8abfe67848980ca53986303b19ec88
3
+ size 2239607120
modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fe715a86a37cd2b20e5eaeee8b22815bce65de676d1e0cd856114b59dab67fc
3
+ size 16766387
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "clean_up_tokenization_spaces": true,
6
+ "cls_token": "<s>",
7
+ "eos_token": "</s>",
8
+ "is_local": false,
9
+ "mask_token": "<mask>",
10
+ "model_max_length": 512,
11
+ "pad_token": "<pad>",
12
+ "sep_token": "</s>",
13
+ "tokenizer_class": "XLMRobertaTokenizer",
14
+ "unk_token": "<unk>"
15
+ }