BM-K commited on
Commit
5df8316
·
verified ·
1 Parent(s): b99fedc

Initial commit

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": 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": true,
9
+ "include_prompt": true
10
+ }
README.md CHANGED
@@ -1,3 +1,1012 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - dense
7
+ - generated_from_trainer
8
+ - dataset_size:1792739
9
+ - loss:CachedGISTEmbedLoss
10
+ widget:
11
+ - source_sentence: 'The New START treaty maintains US nuclear and missile defence.
12
+ The US’ Nuclear armament will be modernized along with New START. “The Obama administration
13
+ has agreed to provide for modernization of the infrastructure essential to maintaining
14
+ our nuclear arsenal. Funding these efforts has become part of the negotiations
15
+ in the ratification process. The administration has put forth a 10-year plan to
16
+ spend $84 billion on the Energy Department''s nuclear weapons complex. Much of
17
+ the credit for getting the administration to add $14 billion to the originally
18
+ proposed $70 billion for modernization goes to Sen. Jon Kyl, the Arizona Republican
19
+ who has been vigilant in this effort. Implementing this modernization program
20
+ in a timely fashion would be important in ensuring that our nuclear arsenal is
21
+ maintained appropriately over the next decade and beyond.” [1] Both US Military
22
+ and civilian leaders insist that the new START treaty will still allow the US
23
+ to deploy effective missile defenses, something which Russia was opposed to, and
24
+ so will not affect US missile defense plans. The main limit on missile defense
25
+ is that the treaty prevents the conversion of existing launchers for this purpose
26
+ this would be more expensive than building new missiles specifically for defense
27
+ purposes. [2] Furthermore, as Joe Biden argues, New START is important to Russian
28
+ cooperation on missile defense: "This [missile defense] system demonstrates America''s
29
+ enduring commitment to Article 5 of the Washington Treaty—that an attack on one
30
+ is an attack on all. NATO missile defense also provides the opportunity for further
31
+ improvements in both NATO-Russian and U.S.-Russian relations. NATO and Russia
32
+ agreed at Lisbon to carry out a joint ballistic missile threat assessment, to
33
+ resume theater missile-defense exercises, and to explore further cooperation on
34
+ territorial missile defense—things that were nearly unimaginable two years ago.
35
+ These agreements underscore the strategic importance the alliance attaches to
36
+ improving its relationship with Russia. But trust and confidence in our relationship
37
+ with Russia would be undermined without Senate approval of the New Start Treaty,
38
+ which reduces strategic nuclear forces to levels not seen since the 1950s, and
39
+ restores important verification mechanisms that ceased when the first Start Treaty
40
+ expired last December." [3] In many ways, in the 21st Century having an abundance
41
+ of nuclear weapons, particularly having too many, is more of a liability than
42
+ an advantage. The United States will be far safer with fewer nuclear weapons in
43
+ the world and a stronger, more stable relationship with Russia under New START,
44
+ and this is desirable. Therefore it is clear that New START maintains the important
45
+ parts of US nuclear capabilities while removing the over-abundance which may become
46
+ a liability due to security and medical concerns, and so New START should be supported.
47
+ [1] Kissinger, Henry A. ; Shultz, George P. ; Baker III, James A’ ; Eagleburger
48
+ , Lawrence S. ; and Powell, Colin L. "The Republican case for ratifying New START".
49
+ Washington Post. 2 December 2010. [2] ibid [3] Biden, Joseph. "The case for ratifying
50
+ New START". Wall Street Journal. 25 November 2010.'
51
+ sentences:
52
+ - 'global law international law politics defence warpeace house supports new New
53
+ START will cause American missile and nuclear capabilities to atrophy, not to
54
+ be maintained. This is because it locks the US in to agreements of defensive reductions
55
+ which are tied into Russian offensive reductions. This could eventually leave
56
+ the US badly under-defended by its missile systems when compared against the offensive
57
+ capabilities of other nuclear states. Moreover, New START leaves in place the
58
+ pre-existing Russian tactical nuclear advantage harming US capabilities by comparison.
59
+ [1] Overall New START hams US missile and nuclear capabilities, and further advantages
60
+ Russia and other nuclear powers, and so should not be supported. As Mitt Romney
61
+ argued in 2010: "Does New START limit America’s options for missile defense? Yes.
62
+ For the first time, we would agree to an interrelationship between strategic offensive
63
+ weapons and missile defense. Moreover, Russia already asserts that the document
64
+ would constitute a binding limit on our missile defense program. But the WikiLeaks
65
+ revelation last weekend that North Korea has supplied Iran with long-range Russian
66
+ missiles confirms that robust missile defense is urgent and indispensable." [2]
67
+ [1] Spring, Baker. "Twelve Flaws of New START That Will Be Difficult to Fix".
68
+ Heritage Foundation, The Foundry. 16 September 2010. [2] Romney, Mitt. "Stop START."
69
+ Boston.com. 3 December 2010.'
70
+ - 'global law international law politics defence warpeace house supports new The
71
+ New START treaty maintains US nuclear and missile defence. The US’ Nuclear armament
72
+ will be modernized along with New START. “The Obama administration has agreed
73
+ to provide for modernization of the infrastructure essential to maintaining our
74
+ nuclear arsenal. Funding these efforts has become part of the negotiations in
75
+ the ratification process. The administration has put forth a 10-year plan to spend
76
+ $84 billion on the Energy Department''s nuclear weapons complex. Much of the credit
77
+ for getting the administration to add $14 billion to the originally proposed $70
78
+ billion for modernization goes to Sen. Jon Kyl, the Arizona Republican who has
79
+ been vigilant in this effort. Implementing this modernization program in a timely
80
+ fashion would be important in ensuring that our nuclear arsenal is maintained
81
+ appropriately over the next decade and beyond.” [1] Both US Military and civilian
82
+ leaders insist that the new START treaty will still allow the US to deploy effective
83
+ missile defenses, something which Russia was opposed to, and so will not affect
84
+ US missile defense plans. The main limit on missile defense is that the treaty
85
+ prevents the conversion of existing launchers for this purpose this would be more
86
+ expensive than building new missiles specifically for defense purposes. [2] Furthermore,
87
+ as Joe Biden argues, New START is important to Russian cooperation on missile
88
+ defense: "This [missile defense] system demonstrates America''s enduring commitment
89
+ to Article 5 of the Washington Treaty—that an attack on one is an attack on all.
90
+ NATO missile defense also provides the opportunity for further improvements in
91
+ both NATO-Russian and U.S.-Russian relations. NATO and Russia agreed at Lisbon
92
+ to carry out a joint ballistic missile threat assessment, to resume theater missile-defense
93
+ exercises, and to explore further cooperation on territorial missile defense—things
94
+ that were nearly unimaginable two years ago. These agreements underscore the strategic
95
+ importance the alliance attaches to improving its relationship with Russia. But
96
+ trust and confidence in our relationship with Russia would be undermined without
97
+ Senate approval of the New Start Treaty, which reduces strategic nuclear forces
98
+ to levels not seen since the 1950s, and restores important verification mechanisms
99
+ that ceased when the first Start Treaty expired last December." [3] In many ways,
100
+ in the 21st Century having an abundance of nuclear weapons, particularly having
101
+ too many, is more of a liability than an advantage. The United States will be
102
+ far safer with fewer nuclear weapons in the world and a stronger, more stable
103
+ relationship with Russia under New START, and this is desirable. Therefore it
104
+ is clear that New START maintains the important parts of US nuclear capabilities
105
+ while removing the over-abundance which may become a liability due to security
106
+ and medical concerns, and so New START should be supported. [1] Kissinger, Henry
107
+ A. ; Shultz, George P. ; Baker III, James A’ ; Eagleburger , Lawrence S. ; and
108
+ Powell, Colin L. "The Republican case for ratifying New START". Washington Post.
109
+ 2 December 2010. [2] ibid [3] Biden, Joseph. "The case for ratifying New START".
110
+ Wall Street Journal. 25 November 2010.'
111
+ - political philosophy house believes civil liberties should be sacrificed If the
112
+ opposition is citing examples from history then there are just as many examples,
113
+ if not more, of western governments resisting the corrupting effects of increased
114
+ power and turning not from good into evil intentions. The fact of the matter is
115
+ that most of today’s western nations have a relatively good track record. It seems
116
+ the opposition is once again forgetting the real enemy – the terrorists. In most
117
+ Western countries we have a fully independent and liberal judiciary, vigorously
118
+ and vigilantly watching for human rights abuses and protecting civil liberties.
119
+ For nearly all Western countries, a slippery slope simply does not exist.
120
+ - source_sentence: Voting is a civic duty Other civic duties also exist “which are
121
+ recognised as necessary in order to live in a better, more cohesive, stable society”
122
+ 1 like paying taxes, attending school, obeying road rules and, in some countries,
123
+ military conscription and jury duty. All of these obligatory activities require
124
+ far more time and effort than voting does, thus compulsory voting can be seen
125
+ as constituting a much smaller intrusion of freedom than any of these other activities.
126
+ The right to vote in a democracy has been fought for throughout modern history
127
+ . In the last century alone the soldiers of numerous wars and the suffragettes
128
+ of many countries fought and died for enfranchisement. It is our duty to respect
129
+ their sacrifice by voting. 1. Liberal Democrat Voice , 2006
130
+ sentences:
131
+ - y political philosophy politics government voting house would make voting Voting
132
+ is a civic duty Other civic duties also exist “which are recognised as necessary
133
+ in order to live in a better, more cohesive, stable society” 1 like paying taxes,
134
+ attending school, obeying road rules and, in some countries, military conscription
135
+ and jury duty. All of these obligatory activities require far more time and effort
136
+ than voting does, thus compulsory voting can be seen as constituting a much smaller
137
+ intrusion of freedom than any of these other activities. The right to vote in
138
+ a democracy has been fought for throughout modern history . In the last century
139
+ alone the soldiers of numerous wars and the suffragettes of many countries fought
140
+ and died for enfranchisement. It is our duty to respect their sacrifice by voting.
141
+ 1. Liberal Democrat Voice , 2006
142
+ - y political philosophy politics government voting house would make voting A democracy
143
+ is based on the principle of respecting basic human rights, such as free choice.
144
+ This principle is directly violated by compulsory voting. With many civil rights
145
+ there is a choice to choose to engage in the activity or not. Voting has carries
146
+ that option, citizens of a democracy have the choice to either vote or not, despite
147
+ being encouraged to vote. It does not matter why a person chooses to vote or not,
148
+ it is the fact of principle that they have the right to choose. Compulsory voting
149
+ goes against such ideas of the freedom of choice, and on that grounds should not
150
+ be compulsory. The proposition speaks of those who died for the right to vote,
151
+ and respecting their sacrifice by voting. Unfortunately the proposition misconstrued
152
+ the point of their sacrifice- to give us the freedom of choice. That right of
153
+ choice must be upheld, as it is the cornerstone of a democratic society. Compulsory
154
+ voting would be infringing upon that.
155
+ - addiction healthcare international africa house believes ghanas ban smoking public
156
+ Even if such a link were true – the campaign against the ban on smoking in public
157
+ places in the UK accept that it’s unlikely that it is the primary cause of closures
158
+ in the UK 1 – the public health benefits would make it worth it. Reductions on
159
+ spending in some areas of the economy is likely to be balanced by increases elsewhere;
160
+ of course there will be losses in some industries – particularly tabacco itself
161
+ but those who stop smoking will have the money to spend elsewhere. Moreover the
162
+ economic effects are likely to be different in Africa; smoking outside in the
163
+ UK, bearing in mind the infamous British weather, is a far less attractive proposition
164
+ than smoking outdoors in many African countries. 1 'Why we want government to
165
+ amend the smoking ban', Save Our Pubs & Clubs,
166
+ - source_sentence: Racial quotas don’t develop new players The quota system could
167
+ lead to moving players from the regional teams who generally have less non-white
168
+ players pilfering them from other unions, rather “Home growing” them [1] . Former
169
+ Springboks coach Peter de Villiers, the first non-white person in that role, has
170
+ described quotas as a “waste of time [2] ”. Depending on the exact phraseology
171
+ of the rules, this could even allow black players from outside South Africa (from,
172
+ for example, England) to be used to fill the quota. [1] McGregor, Liz, ‘New Year,
173
+ new model for SA Rugby? Here’s hoping’, Books Live, 30 December 2013, [2] Peacock,
174
+ James, ‘Peter de Villiers says racial quotas are ‘waste of time’, BBC Sport, 15
175
+ August 2013,
176
+ sentences:
177
+ - ational africa sport team sports house supports racial quotas south african rugby
178
+ Racial quotas don’t develop new players The quota system could lead to moving
179
+ players from the regional teams who generally have less non-white players pilfering
180
+ them from other unions, rather “Home growing” them [1] . Former Springboks coach
181
+ Peter de Villiers, the first non-white person in that role, has described quotas
182
+ as a “waste of time [2] ”. Depending on the exact phraseology of the rules, this
183
+ could even allow black players from outside South Africa (from, for example, England)
184
+ to be used to fill the quota. [1] McGregor, Liz, ‘New Year, new model for SA Rugby?
185
+ Here’s hoping’, Books Live, 30 December 2013, [2] Peacock, James, ‘Peter de Villiers
186
+ says racial quotas are ‘waste of time’, BBC Sport, 15 August 2013,
187
+ - ational africa sport team sports house supports racial quotas south african rugby
188
+ Even if it doesn’t increase the numbers at the grass roots and youth levels, it
189
+ will create more players who can be selected by the provinces for Currie Cup competition.
190
+ This, in turn, could give more non-white players the development and the experience
191
+ they need to make it in to the national team.
192
+ - government house believes governance united states should be split between two
193
+ The reason why a febrile atmosphere has emerged in recent years is because both
194
+ red and blue single-party governments have made unpopular decisions without the
195
+ necessary checks being place upon it. This has made people disenchanted with the
196
+ political system and made them think that it is only looks out for ideological
197
+ elites, causing a backlash in the form of the Tea Party and Occupy movements.
198
+ [1] Divided Government combats this by helping to re-establish consensus between
199
+ the parties over what is best for America, ensuring that policies have the consent
200
+ of a majority of people, thus preventing the overtly ideological backlashes seen
201
+ recently. [1] Miles, Chris, ‘What the Tea Party and Occupy Wall Street Have in
202
+ Common’, policymic,
203
+ - source_sentence: 'Collisions are not as dangerous as they’re feared to be. Some
204
+ hits lead to injury, but the vast majority do not. One commentator challenged
205
+ proponents of a rule change “to name as many as five MLB catchers in the last
206
+ 30 years who have had their careers ended or shortened as a result of a home plate
207
+ collision. Personally, I can’t think of one.” [1] In posing some—though not a
208
+ substantial—risk, home plate collisions are very much like other aspects of the
209
+ sport. Every time a pitcher throws a pitch, the batter could get struck and hurt.
210
+ Every time two outfielders converge on a fly ball, there’s a risk of injury. Baseball,
211
+ as with many other sports, inherently involves the risk of injury. It makes little
212
+ sense to focus on this play, which doesn’t often result in significant injury.
213
+ Moreover, catchers are trained to position their bodies in ways that minimize
214
+ the injury risk from crashes. [2] If catchers do as they’re trained, they’re very
215
+ unlikely to get hurt. [1] Joe Janish, “Buster Posey Aftermath: What Should Be
216
+ Done?,” On Baseball, May 30, 2011, (internal quotation marks omitted). [2] See,
217
+ for example, “Relays, Cutoffs, and Plays at Home,” Baseball-Catcher.com, .'
218
+ sentences:
219
+ - ss economy general international africa house believes africa really rising Bucking
220
+ this trend of increased HDI figures are the states who are currently witnessing,
221
+ or have recently experienced, armed conflict. Africa has observed many well-known
222
+ and lesser known conflicts which have damaged infrastructure and made it significantly
223
+ harder for local populations to access key services such as schools and healthcare.
224
+ Five of seven countries with the poorest nutritional scores are African and have
225
+ recently emerged from armed conflict [1] , they are also rated as some of the
226
+ poorest countries in the world. [1] Smith, ‘Africa is not rising’, 2013
227
+ - 'team sports house believes major league baseball should continue allow collisions
228
+ Collisions are not as dangerous as they’re feared to be. Some hits lead to injury,
229
+ but the vast majority do not. One commentator challenged proponents of a rule
230
+ change “to name as many as five MLB catchers in the last 30 years who have had
231
+ their careers ended or shortened as a result of a home plate collision. Personally,
232
+ I can’t think of one.” [1] In posing some—though not a substantial—risk, home
233
+ plate collisions are very much like other aspects of the sport. Every time a pitcher
234
+ throws a pitch, the batter could get struck and hurt. Every time two outfielders
235
+ converge on a fly ball, there’s a risk of injury. Baseball, as with many other
236
+ sports, inherently involves the risk of injury. It makes little sense to focus
237
+ on this play, which doesn’t often result in significant injury. Moreover, catchers
238
+ are trained to position their bodies in ways that minimize the injury risk from
239
+ crashes. [2] If catchers do as they’re trained, they’re very unlikely to get hurt.
240
+ [1] Joe Janish, “Buster Posey Aftermath: What Should Be Done?,” On Baseball, May
241
+ 30, 2011, (internal quotation marks omitted). [2] See, for example, “Relays, Cutoffs,
242
+ and Plays at Home,” Baseball-Catcher.com, .'
243
+ - 'team sports house believes major league baseball should continue allow collisions
244
+ Collisions increase the risk of injury dramatically. Though it’s true that most
245
+ collisions do not result in significant injury, they result in a higher rate of
246
+ injury than almost any other baseball play. And just because a collision doesn’t
247
+ necessarily result in an injury that derails a player’s entire career does not
248
+ mean that it didn’t take a toll. This is especially true now that we’re learning
249
+ more about concussions, which might be suffered without someone immediately realizing
250
+ it. After a catcher on his team suffered a concussion in a collision, Yankees
251
+ manager Joe Girardi referred to this type of injury as “so unpredictable. That’s
252
+ what’s so scary.... You just don’t know what’s going to happen” with a concussion.
253
+ [1] When catchers are trained to block the plate, they’re taught how to reduce
254
+ the risk of injury, not how to eliminate the risk of injury. No matter how a catcher
255
+ positions himself, there will still be a risk of injury, and it will still be
256
+ much higher than for any other play in baseball. (Opposition Point #1 elaborates
257
+ more upon the risk of injury.) [1] Mark Feisand, “Yankees manager Joe Girardi
258
+ not counting out catcher Francisco Cervelli for postseason roster,” New York Daily
259
+ News, Sept. 17, 2011, .'
260
+ - source_sentence: 'In the public’s eyes, the government seems to suspect everyone.
261
+ Although the anti-terrorist measures are supposed to be trying to catch certain
262
+ people, it is the whole of the public who have to suffer on a daily basis: an
263
+ abundance of security cameras, security checks, and anti-privacy measures continually
264
+ invade innocent people’s lives and yet it is supposed to be the terrorists who
265
+ are being punished. The issue of justice, and whether it is actually being done,
266
+ has to be fully looked at properly. These measures are not solving the problem
267
+ of terrorism as it does not address the core grievances. Instead other ways such
268
+ as negotiation to address grievances is necessary, as happened in Northern Ireland
269
+ [1] . [1] Bowcott, Owen, ‘Northern Ireland’, The Guardian, 11 May 2007, , accessed
270
+ 9 September 2011'
271
+ sentences:
272
+ - europe global human rights house believes european union should lift its Lifting
273
+ the arms ban will not strengthen China militarily. Not only would sales be restricted
274
+ by the new code of conduct, but China’s Ambassador to the European Union has also
275
+ clearly reiterated "We have stated several times that we do not intend to buy
276
+ European military equipment" as "We cannot afford to buy such expensive weapons".
277
+ [1] Even if China was sold high-tech European equipment, this could even be beneficial
278
+ for the EU as it would make China dependent on the EU for such items and make
279
+ it less likely to pursue its own research and development programmes. [1] Rufino,
280
+ Filipe and Vucheva, Elitsa, ‘EU Arms Embargo is ‘political discrimination, says
281
+ Chinese Ambassador’, 2005.
282
+ - 'political philosophy house believes civil liberties should be sacrificed In the
283
+ public’s eyes, the government seems to suspect everyone. Although the anti-terrorist
284
+ measures are supposed to be trying to catch certain people, it is the whole of
285
+ the public who have to suffer on a daily basis: an abundance of security cameras,
286
+ security checks, and anti-privacy measures continually invade innocent people’s
287
+ lives and yet it is supposed to be the terrorists who are being punished. The
288
+ issue of justice, and whether it is actually being done, has to be fully looked
289
+ at properly. These measures are not solving the problem of terrorism as it does
290
+ not address the core grievances. Instead other ways such as negotiation to address
291
+ grievances is necessary, as happened in Northern Ireland [1] . [1] Bowcott, Owen,
292
+ ‘Northern Ireland’, The Guardian, 11 May 2007, , accessed 9 September 2011'
293
+ - political philosophy house believes civil liberties should be sacrificed This
294
+ is just like any other investigation. Obviously the government has to take a broad
295
+ approach because any loophole could be exploited by the unscrupulous terrorist.
296
+ It is a necessity, albeit one with unfortunate consequences, but a necessity all
297
+ the same. As for negotiations with terrorists, it is the propositions view that
298
+ this option does not exist when dealing with terrorists of a fundamentalist background,
299
+ who are, by definition, not willing to compromise and therefore unable to be negotiated
300
+ with.
301
+ pipeline_tag: sentence-similarity
302
+ library_name: sentence-transformers
303
+ metrics:
304
+ - cosine_accuracy
305
+ model-index:
306
+ - name: SentenceTransformer
307
+ results:
308
+ - task:
309
+ type: triplet
310
+ name: Triplet
311
+ dataset:
312
+ name: dev koen
313
+ type: dev_koen
314
+ metrics:
315
+ - type: cosine_accuracy
316
+ value: 0.6882352828979492
317
+ name: Cosine Accuracy
318
+ ---
319
+
320
+ # SentenceTransformer
321
+
322
+ This is a [sentence-transformers](https://www.SBERT.net) model trained on the json dataset. 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.
323
+
324
+ ## Model Details
325
+
326
+ ### Model Description
327
+ - **Model Type:** Sentence Transformer
328
+ <!-- - **Base model:** [Unknown](https://huggingface.co/unknown) -->
329
+ - **Maximum Sequence Length:** 3072 tokens
330
+ - **Output Dimensionality:** 1024 dimensions
331
+ - **Similarity Function:** Cosine Similarity
332
+ - **Training Dataset:**
333
+ - json
334
+ <!-- - **Language:** Unknown -->
335
+ <!-- - **License:** Unknown -->
336
+
337
+ ### Model Sources
338
+
339
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
340
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
341
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
342
+
343
+ ### Full Model Architecture
344
+
345
+ ```
346
+ SentenceTransformer(
347
+ (0): Transformer({'max_seq_length': 3072, 'do_lower_case': False, 'architecture': 'Qwen3Model'})
348
+ (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': True, 'include_prompt': True})
349
+ (2): Normalize()
350
+ )
351
+ ```
352
+
353
+ ## Usage
354
+
355
+ ### Direct Usage (Sentence Transformers)
356
+
357
+ First install the Sentence Transformers library:
358
+
359
+ ```bash
360
+ pip install -U sentence-transformers
361
+ ```
362
+
363
+ Then you can load this model and run inference.
364
+ ```python
365
+ from sentence_transformers import SentenceTransformer
366
+
367
+ # Download from the 🤗 Hub
368
+ model = SentenceTransformer("sentence_transformers_model_id")
369
+ # Run inference
370
+ queries = [
371
+ "In the public\u2019s eyes, the government seems to suspect everyone. Although the anti-terrorist measures are supposed to be trying to catch certain people, it is the whole of the public who have to suffer on a daily basis: an abundance of security cameras, security checks, and anti-privacy measures continually invade innocent people\u2019s lives and yet it is supposed to be the terrorists who are being punished. The issue of justice, and whether it is actually being done, has to be fully looked at properly. These measures are not solving the problem of terrorism as it does not address the core grievances. Instead other ways such as negotiation to address grievances is necessary, as happened in Northern Ireland [1] . [1] Bowcott, Owen, \u2018Northern Ireland\u2019, The Guardian, 11 May 2007, , accessed 9 September 2011",
372
+ ]
373
+ documents = [
374
+ 'political philosophy house believes civil liberties should be sacrificed This is just like any other investigation. Obviously the government has to take a broad approach because any loophole could be exploited by the unscrupulous terrorist. It is a necessity, albeit one with unfortunate consequences, but a necessity all the same. As for negotiations with terrorists, it is the propositions view that this option does not exist when dealing with terrorists of a fundamentalist background, who are, by definition, not willing to compromise and therefore unable to be negotiated with.',
375
+ 'political philosophy house believes civil liberties should be sacrificed In the public’s eyes, the government seems to suspect everyone. Although the anti-terrorist measures are supposed to be trying to catch certain people, it is the whole of the public who have to suffer on a daily basis: an abundance of security cameras, security checks, and anti-privacy measures continually invade innocent people’s lives and yet it is supposed to be the terrorists who are being punished. The issue of justice, and whether it is actually being done, has to be fully looked at properly. These measures are not solving the problem of terrorism as it does not address the core grievances. Instead other ways such as negotiation to address grievances is necessary, as happened in Northern Ireland [1] . [1] Bowcott, Owen, ‘Northern Ireland’, The Guardian, 11 May 2007, , accessed 9 September 2011',
376
+ 'europe global human rights house believes european union should lift its Lifting the arms ban will not strengthen China militarily. Not only would sales be restricted by the new code of conduct, but China’s Ambassador to the European Union has also clearly reiterated "We have stated several times that we do not intend to buy European military equipment" as "We cannot afford to buy such expensive weapons". [1] Even if China was sold high-tech European equipment, this could even be beneficial for the EU as it would make China dependent on the EU for such items and make it less likely to pursue its own research and development programmes. [1] Rufino, Filipe and Vucheva, Elitsa, ‘EU Arms Embargo is ‘political discrimination, says Chinese Ambassador’, 2005.',
377
+ ]
378
+ query_embeddings = model.encode_query(queries)
379
+ document_embeddings = model.encode_document(documents)
380
+ print(query_embeddings.shape, document_embeddings.shape)
381
+ # [1, 1024] [3, 1024]
382
+
383
+ # Get the similarity scores for the embeddings
384
+ similarities = model.similarity(query_embeddings, document_embeddings)
385
+ print(similarities)
386
+ # tensor([[0.5128, 0.9208, 0.2109]])
387
+ ```
388
+
389
+ <!--
390
+ ### Direct Usage (Transformers)
391
+
392
+ <details><summary>Click to see the direct usage in Transformers</summary>
393
+
394
+ </details>
395
+ -->
396
+
397
+ <!--
398
+ ### Downstream Usage (Sentence Transformers)
399
+
400
+ You can finetune this model on your own dataset.
401
+
402
+ <details><summary>Click to expand</summary>
403
+
404
+ </details>
405
+ -->
406
+
407
+ <!--
408
+ ### Out-of-Scope Use
409
+
410
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
411
+ -->
412
+
413
+ ## Evaluation
414
+
415
+ ### Metrics
416
+
417
+ #### Triplet
418
+
419
+ * Dataset: `dev_koen`
420
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
421
+
422
+ | Metric | Value |
423
+ |:--------------------|:-----------|
424
+ | **cosine_accuracy** | **0.6882** |
425
+
426
+ <!--
427
+ ## Bias, Risks and Limitations
428
+
429
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
430
+ -->
431
+
432
+ <!--
433
+ ### Recommendations
434
+
435
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
436
+ -->
437
+
438
+ ## Training Details
439
+
440
+ ### Training Dataset
441
+
442
+ #### json
443
+
444
+ * Dataset: json
445
+ * Size: 1,792,739 training samples
446
+ * Columns: <code>anchor</code>, <code>positive</code>, <code>negative_1</code>, <code>negative_2</code>, and <code>negative_3</code>
447
+ * Approximate statistics based on the first 1000 samples:
448
+ | | anchor | positive | negative_1 | negative_2 | negative_3 |
449
+ |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
450
+ | type | string | string | string | string | string |
451
+ | details | <ul><li>min: 2 tokens</li><li>mean: 23.31 tokens</li><li>max: 140 tokens</li></ul> | <ul><li>min: 22 tokens</li><li>mean: 66.84 tokens</li><li>max: 108 tokens</li></ul> | <ul><li>min: 9 tokens</li><li>mean: 63.01 tokens</li><li>max: 110 tokens</li></ul> | <ul><li>min: 14 tokens</li><li>mean: 62.48 tokens</li><li>max: 110 tokens</li></ul> | <ul><li>min: 7 tokens</li><li>mean: 62.98 tokens</li><li>max: 106 tokens</li></ul> |
452
+ * Samples:
453
+ | anchor | positive | negative_1 | negative_2 | negative_3 |
454
+ |:-------------------------------------------------------------------|:-------------------------------------------------------|:--------------------------------------------------------------|:------------------------------------------------------------|:-----------------------------------------------------------|
455
+ | <code>ATP란?</code> | <code>아데노신 삼인산 아데노신 삼인산(, ATP)은 생명체의 주된 에너지원이다.</code> | <code>ATP ATP는 다음 뜻의 약자이다.</code> | <code>해당 실제로 ADP는 ADPMg로, ATP는 ATPMg로 존재한다.</code> | <code>ATE ATE는 다음을 가리킨다.</code> |
456
+ | <code>난촨구와 둥촨구는 어느 나라에 위치해 있습니까?</code> | <code>난촨구(南川区)는 중국 충칭의 구이자 이전의 현이다.</code> | <code>남풍현(南丰县)은 중국 장시성(江西省) 푸저우(福州)에 위치한 군이다.</code> | <code>도교, 광둥 도교(道滘)는 중국 남부 광둥성 동관 시의 관할 하에 있는 도시입니다.</code> | <code>동포구 동포구는 중국 쓰촨성의 구역입니다. 이곳은 메이산시의 관할 하에 있습니다.</code> |
457
+ | <code>가짜대나무(Pseudosasa)와 별꽃(Cerastium)은 모두 자생 식물과 관련이 있습니까?</code> | <code>가짜사사(Pseudosasa)는 풀과에 속하는 동아시아 대나무의 속입니다.</code> | <code>세팔로소루스(Cephalosorus)는 데이지 과에 속하는 꽃이 피는 식물의 속입니다.</code> | <code>가짜기생충속(Pseudoparasitus)은 라엘라피다에 속하는 진드기의 속입니다.</code> | <code>페리타사(Peritassa)는 쐐기풀과(Celastraceae) 식물의 속입니다.</code> |
458
+ * Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
459
+ ```json
460
+ {
461
+ "guide": "SentenceTransformer('dragonkue/snowflake-arctic-embed-l-v2.0-ko')",
462
+ "temperature": 0.01,
463
+ "mini_batch_size": 8,
464
+ "margin_strategy": "absolute",
465
+ "margin": 0.05
466
+ }
467
+ ```
468
+
469
+ ### Evaluation Dataset
470
+
471
+ #### json
472
+
473
+ * Dataset: json
474
+ * Size: 13,430 evaluation samples
475
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
476
+ * Approximate statistics based on the first 1000 samples:
477
+ | | anchor | positive | negative |
478
+ |:--------|:--------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|
479
+ | type | string | string | string |
480
+ | details | <ul><li>min: 54 tokens</li><li>mean: 244.48 tokens</li><li>max: 1090 tokens</li></ul> | <ul><li>min: 37 tokens</li><li>mean: 164.02 tokens</li><li>max: 658 tokens</li></ul> | <ul><li>min: 65 tokens</li><li>mean: 254.71 tokens</li><li>max: 1100 tokens</li></ul> |
481
+ * Samples:
482
+ | anchor | positive | negative |
483
+ |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
484
+ | <code>Being vegetarian helps the environment Becoming a vegetarian is an environmentally friendly thing to do. Modern farming is one of the main sources of pollution in our rivers. Beef farming is one of the main causes of deforestation, and as long as people continue to buy fast food in their billions, there will be a financial incentive to continue cutting down trees to make room for cattle. Because of our desire to eat fish, our rivers and seas are being emptied of fish and many species are facing extinction. Energy resources are used up much more greedily by meat farming than my farming cereals, pulses etc. Eating meat and fish not only causes cruelty to animals, it causes serious harm to the environment and to biodiversity. For example consider Meat production related pollution and deforestation At Toronto’s 1992 Royal Agricultural Winter Fair, Agriculture Canada displayed two contrasting statistics: “it takes four football fields of land (about 1.6 hectares) to feed each Canadian” and ...</code> | <code>animals environment general health health general weight philosophy ethics You don’t have to be vegetarian to be green. Many special environments have been created by livestock farming – for example chalk down land in England and mountain pastures in many countries. Ending livestock farming would see these areas go back to woodland with a loss of many unique plants and animals. Growing crops can also be very bad for the planet, with fertilisers and pesticides polluting rivers, lakes and seas. Most tropical forests are now cut down for timber, or to allow oil palm trees to be grown in plantations, not to create space for meat production. British farmer and former editor Simon Farrell also states: “Many vegans and vegetarians rely on one source from the U.N. calculation that livestock generates 18% of global carbon emissions, but this figure contains basic mistakes. It attributes all deforestation from ranching to cattle, rather than logging or development. It also muddles up one-off emi...</code> | <code>animals environment general health health general weight philosophy ethics Being vegetarian helps the environment Becoming a vegetarian is an environmentally friendly thing to do. Modern farming is one of the main sources of pollution in our rivers. Beef farming is one of the main causes of deforestation, and as long as people continue to buy fast food in their billions, there will be a financial incentive to continue cutting down trees to make room for cattle. Because of our desire to eat fish, our rivers and seas are being emptied of fish and many species are facing extinction. Energy resources are used up much more greedily by meat farming than my farming cereals, pulses etc. Eating meat and fish not only causes cruelty to animals, it causes serious harm to the environment and to biodiversity. For example consider Meat production related pollution and deforestation At Toronto’s 1992 Royal Agricultural Winter Fair, Agriculture Canada displayed two contrasting statistics: “it takes fo...</code> |
485
+ | <code>It is immoral to kill animals As evolved human beings it is our moral duty to inflict as little pain as possible for our survival. So if we do not need to inflict pain to animals in order to survive, we should not do it. Farm animals such as chickens, pigs, sheep, and cows are sentient living beings like us - they are our evolutionary cousins and like us they can feel pleasure and pain. The 18th century utilitarian philosopher Jeremy Bentham even believed that animal suffering was just as serious as human suffering and likened the idea of human superiority to racism. It is wrong to farm and kill these animals for food when we do not need to do so. The methods of farming and slaughter of these animals are often barbaric and cruel - even on supposedly 'free range' farms. [1] Ten billion animals were slaughtered for human consumption each year, stated PETA. And unlike the farms long time ago, where animals roamed freely, today, most animals are factory farmed: —crammed into cages where th...</code> | <code>animals environment general health health general weight philosophy ethics There is a great moral difference between humans and animals. Unlike animals, humans are capable of rational thought and can alter the world around them. Other creatures were put on this earth for mankind to use, and that includes eating meat. For all these reasons we say that men and women have rights and that animals don’t. This means that eating meat is in no way like murder. It is natural for human beings to farm, kill, and eat other species. In the wild there is a brutal struggle for existence. The fact that we humans have succeeded in that struggle by exploiting our natural environment means that we have a natural right over lower species. In fact farming animals is much less brutal than the pain and hardship that animals inflict on each other naturally in the wild. Eating meat does not need to mean cruelty to animals. There are a growing number of organic and free-range farms that can provide meat without...</code> | <code>animals environment general health health general weight philosophy ethics It is immoral to kill animals As evolved human beings it is our moral duty to inflict as little pain as possible for our survival. So if we do not need to inflict pain to animals in order to survive, we should not do it. Farm animals such as chickens, pigs, sheep, and cows are sentient living beings like us - they are our evolutionary cousins and like us they can feel pleasure and pain. The 18th century utilitarian philosopher Jeremy Bentham even believed that animal suffering was just as serious as human suffering and likened the idea of human superiority to racism. It is wrong to farm and kill these animals for food when we do not need to do so. The methods of farming and slaughter of these animals are often barbaric and cruel - even on supposedly 'free range' farms. [1] Ten billion animals were slaughtered for human consumption each year, stated PETA. And unlike the farms long time ago, where animals roamed f...</code> |
486
+ | <code>Vegetarianism is healthier There are significant health benefits to 'going veggie'; a vegetarian diet contains high quantities of fibre, vitamins, and minerals, and is low in fat. (A vegan diet is even better since eggs and dairy products are high in cholesterol.) The risk of contracting many forms of cancer is increased by eating meat: in 1996 the American Cancer Society recommended that red meat should be excluded from the diet entirely. Eating meat also increases the risk of heart disease - vegetables contain no cholesterol, which can build up to cause blocked arteries in meat-eaters. An American study found out that: “that men in the highest quintile of red-meat consumption — those who ate about 5 oz. of red meat a day, roughly the equivalent of a small steak had a 31% higher risk of death over a 10-year period than men in the lowest-consumption quintile, who ate less than 1 oz. of red meat per day, or approximately three slices of corned beef.” [1] A vegetarian diet reduces the ri...</code> | <code>animals environment general health health general weight philosophy ethics The key to good health is a balanced diet, not a meat- and fish-free diet. Meat and fish are good sources of protein, iron, and other vitamins and minerals. Most of the health benefits of a vegetarian diet derive from its being high in fibre and low in fat and cholesterol. These can be achieved by avoiding fatty and fried foods, eating only lean grilled meat and fish, and including a large amount of fruit and vegetables in your diet along with meat and fish. In general, raw, unprocessed meat from the muscle is made up of the following: protein 15 - 22 % Fat 3 - 15 % Minerals, carbohydrates 1 - 5 % Water 65 - 75 %, all things that we need in moderation. [1] A meat- and fish-free diet is unbalanced and makes it more likely that you will go short of protein, iron and some minerals such as B12 for which we are primarily dependent on animal foodstuffs. Also, a vegetarian diet, in the West, is a more expensive option ...</code> | <code>animals environment general health health general weight philosophy ethics Vegetarianism is healthier There are significant health benefits to 'going veggie'; a vegetarian diet contains high quantities of fibre, vitamins, and minerals, and is low in fat. (A vegan diet is even better since eggs and dairy products are high in cholesterol.) The risk of contracting many forms of cancer is increased by eating meat: in 1996 the American Cancer Society recommended that red meat should be excluded from the diet entirely. Eating meat also increases the risk of heart disease - vegetables contain no cholesterol, which can build up to cause blocked arteries in meat-eaters. An American study found out that: “that men in the highest quintile of red-meat consumption — those who ate about 5 oz. of red meat a day, roughly the equivalent of a small steak had a 31% higher risk of death over a 10-year period than men in the lowest-consumption quintile, who ate less than 1 oz. of red meat per day, or appro...</code> |
487
+ * Loss: [<code>CachedGISTEmbedLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cachedgistembedloss) with these parameters:
488
+ ```json
489
+ {
490
+ "guide": "SentenceTransformer('dragonkue/snowflake-arctic-embed-l-v2.0-ko')",
491
+ "temperature": 0.01,
492
+ "mini_batch_size": 8,
493
+ "margin_strategy": "absolute",
494
+ "margin": 0.05
495
+ }
496
+ ```
497
+
498
+ ### Training Hyperparameters
499
+ #### Non-Default Hyperparameters
500
+
501
+ - `eval_strategy`: steps
502
+ - `per_device_train_batch_size`: 2048
503
+ - `per_device_eval_batch_size`: 16
504
+ - `learning_rate`: 2e-05
505
+ - `warmup_ratio`: 0.05
506
+ - `bf16`: True
507
+ - `ddp_find_unused_parameters`: True
508
+ - `ddp_timeout`: 7200
509
+ - `prompts`: {'anchor': 'Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery:'}
510
+ - `batch_sampler`: no_duplicates
511
+
512
+ #### All Hyperparameters
513
+ <details><summary>Click to expand</summary>
514
+
515
+ - `overwrite_output_dir`: False
516
+ - `do_predict`: False
517
+ - `eval_strategy`: steps
518
+ - `prediction_loss_only`: True
519
+ - `per_device_train_batch_size`: 2048
520
+ - `per_device_eval_batch_size`: 16
521
+ - `per_gpu_train_batch_size`: None
522
+ - `per_gpu_eval_batch_size`: None
523
+ - `gradient_accumulation_steps`: 1
524
+ - `eval_accumulation_steps`: None
525
+ - `torch_empty_cache_steps`: None
526
+ - `learning_rate`: 2e-05
527
+ - `weight_decay`: 0.0
528
+ - `adam_beta1`: 0.9
529
+ - `adam_beta2`: 0.999
530
+ - `adam_epsilon`: 1e-08
531
+ - `max_grad_norm`: 1.0
532
+ - `num_train_epochs`: 3
533
+ - `max_steps`: -1
534
+ - `lr_scheduler_type`: linear
535
+ - `lr_scheduler_kwargs`: {}
536
+ - `warmup_ratio`: 0.05
537
+ - `warmup_steps`: 0
538
+ - `log_level`: passive
539
+ - `log_level_replica`: warning
540
+ - `log_on_each_node`: True
541
+ - `logging_nan_inf_filter`: True
542
+ - `save_safetensors`: True
543
+ - `save_on_each_node`: False
544
+ - `save_only_model`: False
545
+ - `restore_callback_states_from_checkpoint`: False
546
+ - `no_cuda`: False
547
+ - `use_cpu`: False
548
+ - `use_mps_device`: False
549
+ - `seed`: 42
550
+ - `data_seed`: None
551
+ - `jit_mode_eval`: False
552
+ - `use_ipex`: False
553
+ - `bf16`: True
554
+ - `fp16`: False
555
+ - `fp16_opt_level`: O1
556
+ - `half_precision_backend`: auto
557
+ - `bf16_full_eval`: False
558
+ - `fp16_full_eval`: False
559
+ - `tf32`: None
560
+ - `local_rank`: 3
561
+ - `ddp_backend`: None
562
+ - `tpu_num_cores`: None
563
+ - `tpu_metrics_debug`: False
564
+ - `debug`: []
565
+ - `dataloader_drop_last`: True
566
+ - `dataloader_num_workers`: 0
567
+ - `dataloader_prefetch_factor`: None
568
+ - `past_index`: -1
569
+ - `disable_tqdm`: False
570
+ - `remove_unused_columns`: True
571
+ - `label_names`: None
572
+ - `load_best_model_at_end`: False
573
+ - `ignore_data_skip`: False
574
+ - `fsdp`: []
575
+ - `fsdp_min_num_params`: 0
576
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
577
+ - `tp_size`: 0
578
+ - `fsdp_transformer_layer_cls_to_wrap`: None
579
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
580
+ - `deepspeed`: None
581
+ - `label_smoothing_factor`: 0.0
582
+ - `optim`: adamw_torch
583
+ - `optim_args`: None
584
+ - `adafactor`: False
585
+ - `group_by_length`: False
586
+ - `length_column_name`: length
587
+ - `ddp_find_unused_parameters`: True
588
+ - `ddp_bucket_cap_mb`: None
589
+ - `ddp_broadcast_buffers`: False
590
+ - `dataloader_pin_memory`: True
591
+ - `dataloader_persistent_workers`: False
592
+ - `skip_memory_metrics`: True
593
+ - `use_legacy_prediction_loop`: False
594
+ - `push_to_hub`: False
595
+ - `resume_from_checkpoint`: None
596
+ - `hub_model_id`: None
597
+ - `hub_strategy`: every_save
598
+ - `hub_private_repo`: None
599
+ - `hub_always_push`: False
600
+ - `gradient_checkpointing`: False
601
+ - `gradient_checkpointing_kwargs`: None
602
+ - `include_inputs_for_metrics`: False
603
+ - `include_for_metrics`: []
604
+ - `eval_do_concat_batches`: True
605
+ - `fp16_backend`: auto
606
+ - `push_to_hub_model_id`: None
607
+ - `push_to_hub_organization`: None
608
+ - `mp_parameters`:
609
+ - `auto_find_batch_size`: False
610
+ - `full_determinism`: False
611
+ - `torchdynamo`: None
612
+ - `ray_scope`: last
613
+ - `ddp_timeout`: 7200
614
+ - `torch_compile`: False
615
+ - `torch_compile_backend`: None
616
+ - `torch_compile_mode`: None
617
+ - `include_tokens_per_second`: False
618
+ - `include_num_input_tokens_seen`: False
619
+ - `neftune_noise_alpha`: None
620
+ - `optim_target_modules`: None
621
+ - `batch_eval_metrics`: False
622
+ - `eval_on_start`: False
623
+ - `use_liger_kernel`: False
624
+ - `eval_use_gather_object`: False
625
+ - `average_tokens_across_devices`: False
626
+ - `prompts`: {'anchor': 'Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery:'}
627
+ - `batch_sampler`: no_duplicates
628
+ - `multi_dataset_batch_sampler`: proportional
629
+ - `router_mapping`: {}
630
+ - `learning_rate_mapping`: {}
631
+
632
+ </details>
633
+
634
+ ### Training Logs
635
+ <details><summary>Click to expand</summary>
636
+
637
+ | Epoch | Step | Training Loss | Validation Loss | dev_koen_cosine_accuracy |
638
+ |:------:|:----:|:-------------:|:---------------:|:------------------------:|
639
+ | -1 | -1 | - | - | 0.6315 |
640
+ | 0.0092 | 1 | 1.2859 | - | - |
641
+ | 0.0183 | 2 | 1.0815 | - | - |
642
+ | 0.0275 | 3 | 1.0462 | - | - |
643
+ | 0.0367 | 4 | 0.9539 | - | - |
644
+ | 0.0459 | 5 | 0.7947 | - | - |
645
+ | 0.0550 | 6 | 0.7215 | - | - |
646
+ | 0.0642 | 7 | 0.6427 | - | - |
647
+ | 0.0734 | 8 | 0.5985 | - | - |
648
+ | 0.0826 | 9 | 0.5854 | - | - |
649
+ | 0.0917 | 10 | 0.5374 | - | - |
650
+ | 0.1009 | 11 | 0.5199 | - | - |
651
+ | 0.1101 | 12 | 0.4608 | - | - |
652
+ | 0.1193 | 13 | 0.4599 | - | - |
653
+ | 0.1284 | 14 | 0.4337 | - | - |
654
+ | 0.1376 | 15 | 0.4132 | - | - |
655
+ | 0.1468 | 16 | 0.4188 | - | - |
656
+ | 0.1560 | 17 | 0.3887 | - | - |
657
+ | 0.1651 | 18 | 0.3626 | - | - |
658
+ | 0.1743 | 19 | 0.3577 | - | - |
659
+ | 0.1835 | 20 | 0.3332 | - | - |
660
+ | 0.1927 | 21 | 0.3139 | - | - |
661
+ | 0.2018 | 22 | 0.2989 | - | - |
662
+ | 0.2110 | 23 | 0.2947 | - | - |
663
+ | 0.2202 | 24 | 0.2724 | - | - |
664
+ | 0.2294 | 25 | 0.2521 | - | - |
665
+ | 0.2385 | 26 | 0.225 | - | - |
666
+ | 0.2477 | 27 | 0.2285 | - | - |
667
+ | 0.2569 | 28 | 0.2326 | - | - |
668
+ | 0.2661 | 29 | 0.2138 | - | - |
669
+ | 0.2752 | 30 | 0.2075 | - | - |
670
+ | 0.2844 | 31 | 0.2148 | - | - |
671
+ | 0.2936 | 32 | 0.2054 | - | - |
672
+ | 0.3028 | 33 | 0.2117 | - | - |
673
+ | 0.3119 | 34 | 0.1802 | - | - |
674
+ | 0.3211 | 35 | 0.206 | - | - |
675
+ | 0.3303 | 36 | 0.1961 | - | - |
676
+ | 0.3394 | 37 | 0.1782 | - | - |
677
+ | 0.3486 | 38 | 0.1746 | - | - |
678
+ | 0.3578 | 39 | 0.1822 | - | - |
679
+ | 0.3670 | 40 | 0.17 | - | - |
680
+ | 0.3761 | 41 | 0.1636 | - | - |
681
+ | 0.3853 | 42 | 0.1617 | - | - |
682
+ | 0.3945 | 43 | 0.1409 | - | - |
683
+ | 0.4037 | 44 | 0.1451 | - | - |
684
+ | 0.4128 | 45 | 0.1359 | - | - |
685
+ | 0.4220 | 46 | 0.1289 | - | - |
686
+ | 0.4312 | 47 | 0.1356 | - | - |
687
+ | 0.4404 | 48 | 0.1326 | - | - |
688
+ | 0.4495 | 49 | 0.1239 | - | - |
689
+ | 0.4587 | 50 | 0.1279 | - | - |
690
+ | 0.4679 | 51 | 0.1105 | - | - |
691
+ | 0.4771 | 52 | 0.124 | - | - |
692
+ | 0.4862 | 53 | 0.1115 | - | - |
693
+ | 0.4954 | 54 | 0.1115 | - | - |
694
+ | 0.5046 | 55 | 0.1111 | 0.2269 | 0.6678 |
695
+ | 0.5138 | 56 | 0.1072 | - | - |
696
+ | 0.5229 | 57 | 0.1052 | - | - |
697
+ | 0.5321 | 58 | 0.1113 | - | - |
698
+ | 0.5413 | 59 | 0.1025 | - | - |
699
+ | 0.5505 | 60 | 0.0998 | - | - |
700
+ | 0.5596 | 61 | 0.1095 | - | - |
701
+ | 0.5688 | 62 | 0.1001 | - | - |
702
+ | 0.5780 | 63 | 0.111 | - | - |
703
+ | 0.5872 | 64 | 0.105 | - | - |
704
+ | 0.5963 | 65 | 0.1303 | - | - |
705
+ | 0.6055 | 66 | 0.1214 | - | - |
706
+ | 0.6147 | 67 | 0.1243 | - | - |
707
+ | 0.6239 | 68 | 0.1234 | - | - |
708
+ | 0.6330 | 69 | 0.1201 | - | - |
709
+ | 0.6422 | 70 | 0.1156 | - | - |
710
+ | 0.6514 | 71 | 0.1268 | - | - |
711
+ | 0.6606 | 72 | 0.1224 | - | - |
712
+ | 0.6697 | 73 | 0.1227 | - | - |
713
+ | 0.6789 | 74 | 0.1153 | - | - |
714
+ | 0.6881 | 75 | 0.1133 | - | - |
715
+ | 0.6972 | 76 | 0.1218 | - | - |
716
+ | 0.7064 | 77 | 0.1369 | - | - |
717
+ | 0.7156 | 78 | 0.128 | - | - |
718
+ | 0.7248 | 79 | 0.1352 | - | - |
719
+ | 0.7339 | 80 | 0.1379 | - | - |
720
+ | 0.7431 | 81 | 0.1468 | - | - |
721
+ | 0.7523 | 82 | 0.1489 | - | - |
722
+ | 0.7615 | 83 | 0.1512 | - | - |
723
+ | 0.7706 | 84 | 0.1424 | - | - |
724
+ | 0.7798 | 85 | 0.1507 | - | - |
725
+ | 0.7890 | 86 | 0.157 | - | - |
726
+ | 0.7982 | 87 | 0.1513 | - | - |
727
+ | 0.8073 | 88 | 0.163 | - | - |
728
+ | 0.8165 | 89 | 0.1471 | - | - |
729
+ | 0.8257 | 90 | 0.1636 | - | - |
730
+ | 0.8349 | 91 | 0.1681 | - | - |
731
+ | 0.8440 | 92 | 0.1693 | - | - |
732
+ | 0.8532 | 93 | 0.1835 | - | - |
733
+ | 0.8624 | 94 | 0.1716 | - | - |
734
+ | 0.8716 | 95 | 0.1817 | - | - |
735
+ | 0.8807 | 96 | 0.1999 | - | - |
736
+ | 0.8899 | 97 | 0.21 | - | - |
737
+ | 0.8991 | 98 | 0.2278 | - | - |
738
+ | 0.9083 | 99 | 0.2273 | - | - |
739
+ | 0.9174 | 100 | 0.2469 | - | - |
740
+ | 0.9266 | 101 | 0.2742 | - | - |
741
+ | 0.9358 | 102 | 0.2805 | - | - |
742
+ | 0.9450 | 103 | 0.2932 | - | - |
743
+ | 0.9541 | 104 | 0.2915 | - | - |
744
+ | 0.9633 | 105 | 0.2814 | - | - |
745
+ | 0.9725 | 106 | 0.3003 | - | - |
746
+ | 0.9817 | 107 | 0.3346 | - | - |
747
+ | 0.9908 | 108 | 0.315 | - | - |
748
+ | 1.0 | 109 | 0.9444 | - | - |
749
+ | 1.0092 | 110 | 0.5552 | 0.1111 | 0.7004 |
750
+ | 1.0183 | 111 | 0.4133 | - | - |
751
+ | 1.0275 | 112 | 0.3811 | - | - |
752
+ | 1.0367 | 113 | 0.3525 | - | - |
753
+ | 1.0459 | 114 | 0.2925 | - | - |
754
+ | 1.0550 | 115 | 0.2563 | - | - |
755
+ | 1.0642 | 116 | 0.2417 | - | - |
756
+ | 1.0734 | 117 | 0.2073 | - | - |
757
+ | 1.0826 | 118 | 0.2073 | - | - |
758
+ | 1.0917 | 119 | 0.1958 | - | - |
759
+ | 1.1009 | 120 | 0.1842 | - | - |
760
+ | 1.1101 | 121 | 0.1733 | - | - |
761
+ | 1.1193 | 122 | 0.1799 | - | - |
762
+ | 1.1284 | 123 | 0.15 | - | - |
763
+ | 1.1376 | 124 | 0.1476 | - | - |
764
+ | 1.1468 | 125 | 0.1507 | - | - |
765
+ | 1.1560 | 126 | 0.1448 | - | - |
766
+ | 1.1651 | 127 | 0.1197 | - | - |
767
+ | 1.1743 | 128 | 0.1206 | - | - |
768
+ | 1.1835 | 129 | 0.1223 | - | - |
769
+ | 1.1927 | 130 | 0.102 | - | - |
770
+ | 1.2018 | 131 | 0.0991 | - | - |
771
+ | 1.2110 | 132 | 0.1023 | - | - |
772
+ | 1.2202 | 133 | 0.0914 | - | - |
773
+ | 1.2294 | 134 | 0.085 | - | - |
774
+ | 1.2385 | 135 | 0.0575 | - | - |
775
+ | 1.2477 | 136 | 0.062 | - | - |
776
+ | 1.2569 | 137 | 0.0674 | - | - |
777
+ | 1.2661 | 138 | 0.0617 | - | - |
778
+ | 1.2752 | 139 | 0.0554 | - | - |
779
+ | 1.2844 | 140 | 0.0541 | - | - |
780
+ | 1.2936 | 141 | 0.0605 | - | - |
781
+ | 1.3028 | 142 | 0.0535 | - | - |
782
+ | 1.3119 | 143 | 0.0431 | - | - |
783
+ | 1.3211 | 144 | 0.0527 | - | - |
784
+ | 1.3303 | 145 | 0.0497 | - | - |
785
+ | 1.3394 | 146 | 0.046 | - | - |
786
+ | 1.3486 | 147 | 0.036 | - | - |
787
+ | 1.3578 | 148 | 0.0456 | - | - |
788
+ | 1.3670 | 149 | 0.0441 | - | - |
789
+ | 1.3761 | 150 | 0.0347 | - | - |
790
+ | 1.3853 | 151 | 0.0367 | - | - |
791
+ | 1.3945 | 152 | 0.0287 | - | - |
792
+ | 1.4037 | 153 | 0.0265 | - | - |
793
+ | 1.4128 | 154 | 0.0233 | - | - |
794
+ | 1.4220 | 155 | 0.0222 | - | - |
795
+ | 1.4312 | 156 | 0.0295 | - | - |
796
+ | 1.4404 | 157 | 0.027 | - | - |
797
+ | 1.4495 | 158 | 0.0233 | - | - |
798
+ | 1.4587 | 159 | 0.0243 | - | - |
799
+ | 1.4679 | 160 | 0.019 | - | - |
800
+ | 1.4771 | 161 | 0.0233 | - | - |
801
+ | 1.4862 | 162 | 0.0209 | - | - |
802
+ | 1.4954 | 163 | 0.0211 | - | - |
803
+ | 1.5046 | 164 | 0.0178 | - | - |
804
+ | 1.5138 | 165 | 0.0186 | 0.1471 | 0.6915 |
805
+ | 1.5229 | 166 | 0.0181 | - | - |
806
+ | 1.5321 | 167 | 0.0214 | - | - |
807
+ | 1.5413 | 168 | 0.0179 | - | - |
808
+ | 1.5505 | 169 | 0.0168 | - | - |
809
+ | 1.5596 | 170 | 0.0206 | - | - |
810
+ | 1.5688 | 171 | 0.0175 | - | - |
811
+ | 1.5780 | 172 | 0.0207 | - | - |
812
+ | 1.5872 | 173 | 0.0213 | - | - |
813
+ | 1.5963 | 174 | 0.0291 | - | - |
814
+ | 1.6055 | 175 | 0.0263 | - | - |
815
+ | 1.6147 | 176 | 0.0264 | - | - |
816
+ | 1.6239 | 177 | 0.0265 | - | - |
817
+ | 1.6330 | 178 | 0.0257 | - | - |
818
+ | 1.6422 | 179 | 0.0215 | - | - |
819
+ | 1.6514 | 180 | 0.0275 | - | - |
820
+ | 1.6606 | 181 | 0.0244 | - | - |
821
+ | 1.6697 | 182 | 0.0288 | - | - |
822
+ | 1.6789 | 183 | 0.0234 | - | - |
823
+ | 1.6881 | 184 | 0.0225 | - | - |
824
+ | 1.6972 | 185 | 0.0286 | - | - |
825
+ | 1.7064 | 186 | 0.0333 | - | - |
826
+ | 1.7156 | 187 | 0.0317 | - | - |
827
+ | 1.7248 | 188 | 0.0338 | - | - |
828
+ | 1.7339 | 189 | 0.0344 | - | - |
829
+ | 1.7431 | 190 | 0.0367 | - | - |
830
+ | 1.7523 | 191 | 0.0355 | - | - |
831
+ | 1.7615 | 192 | 0.0434 | - | - |
832
+ | 1.7706 | 193 | 0.0369 | - | - |
833
+ | 1.7798 | 194 | 0.0383 | - | - |
834
+ | 1.7890 | 195 | 0.0416 | - | - |
835
+ | 1.7982 | 196 | 0.0434 | - | - |
836
+ | 1.8073 | 197 | 0.0465 | - | - |
837
+ | 1.8165 | 198 | 0.0387 | - | - |
838
+ | 1.8257 | 199 | 0.0508 | - | - |
839
+ | 1.8349 | 200 | 0.0485 | - | - |
840
+ | 1.8440 | 201 | 0.0487 | - | - |
841
+ | 1.8532 | 202 | 0.056 | - | - |
842
+ | 1.8624 | 203 | 0.0517 | - | - |
843
+ | 1.8716 | 204 | 0.0523 | - | - |
844
+ | 1.8807 | 205 | 0.0711 | - | - |
845
+ | 1.8899 | 206 | 0.0727 | - | - |
846
+ | 1.8991 | 207 | 0.0836 | - | - |
847
+ | 1.9083 | 208 | 0.0782 | - | - |
848
+ | 1.9174 | 209 | 0.0955 | - | - |
849
+ | 1.9266 | 210 | 0.1165 | - | - |
850
+ | 1.9358 | 211 | 0.1154 | - | - |
851
+ | 1.9450 | 212 | 0.1302 | - | - |
852
+ | 1.9541 | 213 | 0.1276 | - | - |
853
+ | 1.9633 | 214 | 0.1305 | - | - |
854
+ | 1.9725 | 215 | 0.1403 | - | - |
855
+ | 1.9817 | 216 | 0.1713 | - | - |
856
+ | 1.9908 | 217 | 0.1908 | - | - |
857
+ | 2.0 | 218 | 0.7015 | - | - |
858
+ | 2.0092 | 219 | 0.4075 | - | - |
859
+ | 2.0183 | 220 | 0.2799 | 0.0935 | 0.7097 |
860
+ | 2.0275 | 221 | 0.2448 | - | - |
861
+ | 2.0367 | 222 | 0.2071 | - | - |
862
+ | 2.0459 | 223 | 0.1706 | - | - |
863
+ | 2.0550 | 224 | 0.1503 | - | - |
864
+ | 2.0642 | 225 | 0.1375 | - | - |
865
+ | 2.0734 | 226 | 0.107 | - | - |
866
+ | 2.0826 | 227 | 0.1033 | - | - |
867
+ | 2.0917 | 228 | 0.1083 | - | - |
868
+ | 2.1009 | 229 | 0.1029 | - | - |
869
+ | 2.1101 | 230 | 0.0772 | - | - |
870
+ | 2.1193 | 231 | 0.0835 | - | - |
871
+ | 2.1284 | 232 | 0.0721 | - | - |
872
+ | 2.1376 | 233 | 0.067 | - | - |
873
+ | 2.1468 | 234 | 0.0682 | - | - |
874
+ | 2.1560 | 235 | 0.0687 | - | - |
875
+ | 2.1651 | 236 | 0.0447 | - | - |
876
+ | 2.1743 | 237 | 0.0476 | - | - |
877
+ | 2.1835 | 238 | 0.0434 | - | - |
878
+ | 2.1927 | 239 | 0.0351 | - | - |
879
+ | 2.2018 | 240 | 0.0371 | - | - |
880
+ | 2.2110 | 241 | 0.0289 | - | - |
881
+ | 2.2202 | 242 | 0.0277 | - | - |
882
+ | 2.2294 | 243 | 0.0261 | - | - |
883
+ | 2.2385 | 244 | 0.0146 | - | - |
884
+ | 2.2477 | 245 | 0.0135 | - | - |
885
+ | 2.2569 | 246 | 0.0188 | - | - |
886
+ | 2.2661 | 247 | 0.0156 | - | - |
887
+ | 2.2752 | 248 | 0.0134 | - | - |
888
+ | 2.2844 | 249 | 0.0134 | - | - |
889
+ | 2.2936 | 250 | 0.0147 | - | - |
890
+ | 2.3028 | 251 | 0.0147 | - | - |
891
+ | 2.3119 | 252 | 0.0114 | - | - |
892
+ | 2.3211 | 253 | 0.0139 | - | - |
893
+ | 2.3303 | 254 | 0.0109 | - | - |
894
+ | 2.3394 | 255 | 0.0113 | - | - |
895
+ | 2.3486 | 256 | 0.01 | - | - |
896
+ | 2.3578 | 257 | 0.0122 | - | - |
897
+ | 2.3670 | 258 | 0.0113 | - | - |
898
+ | 2.3761 | 259 | 0.0092 | - | - |
899
+ | 2.3853 | 260 | 0.0098 | - | - |
900
+ | 2.3945 | 261 | 0.0065 | - | - |
901
+ | 2.4037 | 262 | 0.0068 | - | - |
902
+ | 2.4128 | 263 | 0.007 | - | - |
903
+ | 2.4220 | 264 | 0.007 | - | - |
904
+ | 2.4312 | 265 | 0.0084 | - | - |
905
+ | 2.4404 | 266 | 0.009 | - | - |
906
+ | 2.4495 | 267 | 0.0066 | - | - |
907
+ | 2.4587 | 268 | 0.0073 | - | - |
908
+ | 2.4679 | 269 | 0.0065 | - | - |
909
+ | 2.4771 | 270 | 0.0085 | - | - |
910
+ | 2.4862 | 271 | 0.0068 | - | - |
911
+ | 2.4954 | 272 | 0.0072 | - | - |
912
+ | 2.5046 | 273 | 0.0071 | - | - |
913
+ | 2.5138 | 274 | 0.0067 | - | - |
914
+ | 2.5229 | 275 | 0.0066 | 0.1569 | 0.6882 |
915
+ | 2.5321 | 276 | 0.0073 | - | - |
916
+ | 2.5413 | 277 | 0.0071 | - | - |
917
+ | 2.5505 | 278 | 0.0068 | - | - |
918
+ | 2.5596 | 279 | 0.0072 | - | - |
919
+ | 2.5688 | 280 | 0.0066 | - | - |
920
+ | 2.5780 | 281 | 0.0081 | - | - |
921
+ | 2.5872 | 282 | 0.0075 | - | - |
922
+ | 2.5963 | 283 | 0.0086 | - | - |
923
+ | 2.6055 | 284 | 0.009 | - | - |
924
+ | 2.6147 | 285 | 0.0084 | - | - |
925
+ | 2.6239 | 286 | 0.0096 | - | - |
926
+ | 2.6330 | 287 | 0.0082 | - | - |
927
+ | 2.6422 | 288 | 0.0076 | - | - |
928
+ | 2.6514 | 289 | 0.0094 | - | - |
929
+ | 2.6606 | 290 | 0.0094 | - | - |
930
+ | 2.6697 | 291 | 0.0094 | - | - |
931
+ | 2.6789 | 292 | 0.0085 | - | - |
932
+ | 2.6881 | 293 | 0.0102 | - | - |
933
+ | 2.6972 | 294 | 0.0122 | - | - |
934
+ | 2.7064 | 295 | 0.0132 | - | - |
935
+ | 2.7156 | 296 | 0.0131 | - | - |
936
+ | 2.7248 | 297 | 0.0133 | - | - |
937
+ | 2.7339 | 298 | 0.013 | - | - |
938
+ | 2.7431 | 299 | 0.0158 | - | - |
939
+ | 2.7523 | 300 | 0.0144 | - | - |
940
+ | 2.7615 | 301 | 0.0161 | - | - |
941
+ | 2.7706 | 302 | 0.0153 | - | - |
942
+ | 2.7798 | 303 | 0.0167 | - | - |
943
+ | 2.7890 | 304 | 0.0179 | - | - |
944
+ | 2.7982 | 305 | 0.0176 | - | - |
945
+ | 2.8073 | 306 | 0.0178 | - | - |
946
+ | 2.8165 | 307 | 0.0179 | - | - |
947
+ | 2.8257 | 308 | 0.0209 | - | - |
948
+ | 2.8349 | 309 | 0.0194 | - | - |
949
+ | 2.8440 | 310 | 0.022 | - | - |
950
+ | 2.8532 | 311 | 0.0262 | - | - |
951
+ | 2.8624 | 312 | 0.0226 | - | - |
952
+ | 2.8716 | 313 | 0.0226 | - | - |
953
+ | 2.8807 | 314 | 0.0352 | - | - |
954
+ | 2.8899 | 315 | 0.0371 | - | - |
955
+ | 2.8991 | 316 | 0.0427 | - | - |
956
+ | 2.9083 | 317 | 0.0396 | - | - |
957
+ | 2.9174 | 318 | 0.052 | - | - |
958
+ | 2.9266 | 319 | 0.067 | - | - |
959
+ | 2.9358 | 320 | 0.0687 | - | - |
960
+ | 2.9450 | 321 | 0.0866 | - | - |
961
+ | 2.9541 | 322 | 0.0793 | - | - |
962
+ | 2.9633 | 323 | 0.0853 | - | - |
963
+ | 2.9725 | 324 | 0.0918 | - | - |
964
+ | 2.9817 | 325 | 0.1179 | - | - |
965
+ | 2.9908 | 326 | 0.1239 | - | - |
966
+ | 3.0 | 327 | 0.5811 | - | - |
967
+
968
+ </details>
969
+
970
+ ### Framework Versions
971
+ - Python: 3.11.12
972
+ - Sentence Transformers: 5.0.0
973
+ - Transformers: 4.51.3
974
+ - PyTorch: 2.7.0+cu128
975
+ - Accelerate: 1.5.2
976
+ - Datasets: 2.21.0
977
+ - Tokenizers: 0.21.1
978
+
979
+ ## Citation
980
+
981
+ ### BibTeX
982
+
983
+ #### Sentence Transformers
984
+ ```bibtex
985
+ @inproceedings{reimers-2019-sentence-bert,
986
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
987
+ author = "Reimers, Nils and Gurevych, Iryna",
988
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
989
+ month = "11",
990
+ year = "2019",
991
+ publisher = "Association for Computational Linguistics",
992
+ url = "https://arxiv.org/abs/1908.10084",
993
+ }
994
+ ```
995
+
996
+ <!--
997
+ ## Glossary
998
+
999
+ *Clearly define terms in order to be accessible across audiences.*
1000
+ -->
1001
+
1002
+ <!--
1003
+ ## Model Card Authors
1004
+
1005
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
1006
+ -->
1007
+
1008
+ <!--
1009
+ ## Model Card Contact
1010
+
1011
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
1012
+ -->
added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3Model"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 40960,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen3",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 8,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 1000000,
23
+ "sliding_window": null,
24
+ "tie_word_embeddings": true,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.51.3",
27
+ "use_cache": true,
28
+ "use_sliding_window": false,
29
+ "vocab_size": 151936
30
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "SentenceTransformer",
3
+ "__version__": {
4
+ "sentence_transformers": "5.0.0",
5
+ "transformers": "4.51.3",
6
+ "pytorch": "2.7.0+cu128"
7
+ },
8
+ "prompts": {
9
+ "query": "Instruct: Given a web search query, retrieve relevant passages that answer the query\nQuery:",
10
+ "document": ""
11
+ },
12
+ "default_prompt_name": null,
13
+ "similarity_fn_name": "cosine"
14
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:747defaa73c73ea09120679856f38c32a6f3b8db9a7358930e012f6d7a5c3e58
3
+ size 2384233112
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": 8192,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b703c7786128f3188e39ca909a734c4dde9b51819c66ca96dfacabd758c23e42
3
+ size 11423970
tokenizer_config.json ADDED
@@ -0,0 +1,247 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {{- messages[0].content + '\\n\\n' }}\n {%- endif %}\n {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}\n{%- for message in messages[::-1] %}\n {%- set index = (messages|length - 1) - loop.index0 %}\n {%- if ns.multi_step_tool and message.role == \"user\" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}\n {%- set ns.multi_step_tool = false %}\n {%- set ns.last_query_index = index %}\n {%- endif %}\n{%- endfor %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = '' %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {%- set reasoning_content = '' %}\n {%- if message.reasoning_content is string %}\n {%- set reasoning_content = message.reasoning_content %}\n {%- else %}\n {%- if '</think>' in content %}\n {%- set reasoning_content = content.split('</think>')[0].rstrip('\\n').split('<think>')[-1].lstrip('\\n') %}\n {%- set content = content.split('</think>')[-1].lstrip('\\n') %}\n {%- endif %}\n {%- endif %}\n {%- if loop.index0 > ns.last_query_index %}\n {%- if loop.last or (not loop.last and reasoning_content) %}\n {{- '<|im_start|>' + message.role + '\\n<think>\\n' + reasoning_content.strip('\\n') + '\\n</think>\\n\\n' + content.lstrip('\\n') }}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- else %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- endif %}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n {%- if enable_thinking is defined and enable_thinking is false %}\n {{- '<think>\\n\\n</think>\\n\\n' }}\n {%- endif %}\n{%- endif %}",
231
+ "clean_up_tokenization_spaces": false,
232
+ "eos_token": "<|im_end|>",
233
+ "errors": "replace",
234
+ "extra_special_tokens": {},
235
+ "max_length": 2048,
236
+ "model_max_length": 8192,
237
+ "pad_to_multiple_of": null,
238
+ "pad_token": "<|endoftext|>",
239
+ "pad_token_type_id": 0,
240
+ "padding_side": "right",
241
+ "split_special_tokens": false,
242
+ "stride": 0,
243
+ "tokenizer_class": "Qwen2Tokenizer",
244
+ "truncation_side": "right",
245
+ "truncation_strategy": "longest_first",
246
+ "unk_token": null
247
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff