vin00d commited on
Commit
d09a7b5
·
verified ·
1 Parent(s): 03249d8

Add new SentenceTransformer model

Browse files
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 1024,
3
+ "pooling_mode_cls_token": true,
4
+ "pooling_mode_mean_tokens": false,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
README.md ADDED
@@ -0,0 +1,855 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - generated_from_trainer
7
+ - dataset_size:210
8
+ - loss:MatryoshkaLoss
9
+ - loss:MultipleNegativesRankingLoss
10
+ base_model: Snowflake/snowflake-arctic-embed-l
11
+ widget:
12
+ - source_sentence: 'What does maintenance refer to in the context of providing for
13
+ another person? '
14
+ sentences:
15
+ - '-M-
16
+
17
+ Maintenance: The f urnishing by one person to another the means of living, or
18
+ fo od, clothing,'
19
+ - 'income and expenses to determine if the debtor may proceed under Chapter 7.
20
+
21
+ Chapter 7 trustee
22
+
23
+ A person appointed in a Chapter 7 case to represent the interests of the bankruptcy
24
+ estate
25
+
26
+ and the creditors. The trustee''s responsibilities include reviewing the debtor''s
27
+ petition and
28
+
29
+ schedules, liquidating the property of the estate, and making distributions to
30
+ creditors. The
31
+
32
+ trustee may also bring actions against creditors or the debtor to recover property
33
+ of the
34
+
35
+ bankruptcy estate.
36
+
37
+ Chapter 9'
38
+ - '-19-
39
+
40
+ Trial De Novo: A new trial (See 22NYCRR 28.12).
41
+
42
+ -U-
43
+
44
+ Undertaking: Deposit of a sum of money or filing of a bond in court, to secure
45
+ some actual or
46
+
47
+ potential obligation.
48
+
49
+ -V-
50
+
51
+ Vacate: To set aside or undo a previous action or order.
52
+
53
+ Venire: Technically, a writ summoning persons to court to act as jurors; popularly
54
+ used as meaning
55
+
56
+ the body of names thus summoned.
57
+
58
+ Venue: (a) Geographical place where some legal matter occurs or may be determined.
59
+ (b) The
60
+
61
+ geographical area within which a court has jurisdiction. It relates only to a
62
+ place or territory within
63
+
64
+ which either party may require a case to be tried. A defect in venue may be waived
65
+ by the parties.'
66
+ - source_sentence: 'What does the term "Pro Se" refer to in a legal context? '
67
+ sentences:
68
+ - 'Process: A l egal means, such as a s ummons, used to s ubject a de fendant i
69
+ n a l awsuit to the
70
+
71
+ personal jur isdiction o f the c ourt; broa dly, r efers to all writs iss ued
72
+ i n the c ourse of a le gal
73
+
74
+ proceeding - what is served to obtain jurisdiction.
75
+
76
+ Pro Se (aka Self-Represented): Appearing on one’s own behalf without an attorney.
77
+
78
+ Purge: To atone for or correct an offense, to submit to a court''s mandate (i.e.,
79
+ to purge oneself
80
+
81
+ of contempt of court).
82
+
83
+ -Q-
84
+
85
+ None.
86
+
87
+ -R-
88
+
89
+ Recuse: To disqualify oneself as a judge.
90
+
91
+ Redact: To edit, revise or block out written text.
92
+
93
+ Referee: A person to whom a claim pending in a court is referred by the court
94
+ to take testimony,'
95
+ - '-10-
96
+
97
+ Hearing: A pr eliminary examination where testimony is given and e vidence presented
98
+ for the
99
+
100
+ purpose of determining an issue of fact and reaching a decision on the basis of
101
+ that evidence.
102
+
103
+ Hearsay: Testimony of a witness who relates not what he/she knows personally,
104
+ but what others
105
+
106
+ have told the witness, or what the witness has heard said by others; may be admissible
107
+ or
108
+
109
+ inadmissible in court depending upon rules of evidence.
110
+
111
+ Hung Jury: A jury whose members cannot reconcile their differences of opinion
112
+ and thus cannot
113
+
114
+ reach a verdict.
115
+
116
+ -I-
117
+
118
+ Impaneling: The process by which jurors are selected and sworn to their task.
119
+
120
+ Impleader: An addition of another party to an action by the defendant, a “third
121
+ party” claim.'
122
+ - '-12-
123
+
124
+ Jurisdiction, Subject Matter: Whether the court has authority over the thing
125
+ or right claimed by
126
+
127
+ one party against another.
128
+
129
+ Jury: A prescribed number of persons selected according to law and sworn to make
130
+ findings of
131
+
132
+ fact.
133
+
134
+ Jury (Advisory): A body of jurors impaneled to hear a case in which the parties
135
+ have no right to
136
+
137
+ a jury trial - the judge remains solely responsible for the findings and may accept
138
+ or reject the
139
+
140
+ jury''s verdict.
141
+
142
+ Jury Instructions: Directions given by the judge to the jury, at the beginning
143
+ and end of trial.
144
+
145
+ -K-
146
+
147
+ None.
148
+
149
+ -L-
150
+
151
+ Laches: The failure to diligently assert a right, which results in a refusal
152
+ to allow the right to be
153
+
154
+ asserted later.
155
+
156
+ Legal Age: Eighteen (18) years of age. See CPLR Section 1206.'
157
+ - source_sentence: What is the purpose of a Chapter 11 bankruptcy filing?
158
+ sentences:
159
+ - 'condemnation, i.e., the legal process by which real estate of a private owner
160
+ is taken for public use
161
+
162
+ without the owner''s consent, but upon the award and payment of just compensation.
163
+
164
+ Enjoin: To require a person, by writ of injunction from a court of equity, to
165
+ perform or to refrain
166
+
167
+ from or cease doing some act.
168
+
169
+ Entry: The formal filing of an order of judgment with the County Clerk.
170
+
171
+ Equitable Action (Equity Matter): An action which may be brought for the purpose
172
+ of restraining'
173
+ - 'A legal claim.
174
+
175
+ Chambers
176
+
177
+ The offices of a judge and his or her staff.
178
+
179
+ Chapter 11
180
+
181
+ A reorganization bankruptcy, usually involving a corporation or partnership. A
182
+ Chapter 11
183
+
184
+ debtor usually proposes a plan of reorganization to keep its business alive and
185
+ pay creditors
186
+
187
+ over time. Individuals or people in business can also seek relief in Chapter 11.
188
+
189
+ Chapter 12
190
+
191
+ The chapter of the Bankruptcy Code providing for adjustment of debts of a "family
192
+ farmer"
193
+
194
+ or "family fisherman," as the terms are defined in the Bankruptcy Code.
195
+
196
+ Chapter 13
197
+
198
+ The chapter of the Bankruptcy Code providing for the adjustment of debts of an
199
+ individual
200
+
201
+ with regular income, often referred to as a "wage-earner" plan. Chapter 13 allows
202
+ a debtor'
203
+ - 'Conviction
204
+
205
+ A judgment of guilt against a criminal defendant.
206
+
207
+ Counsel
208
+
209
+ Legal advice; a term also used to refer to the lawyers in a case.
210
+
211
+ Count
212
+
213
+ An allegation in an indictment or information, charging a defendant with a crime.
214
+ An
215
+
216
+ indictment or information may contain allegations that the defendant committed
217
+ more
218
+
219
+ than one crime. Each allegation is referred to as a count.
220
+
221
+ Court
222
+
223
+ Government entity authorized to resolve legal disputes. Judges sometimes use "court"
224
+ to
225
+
226
+ refer to themselves in the third person, as in "the court has read the briefs."
227
+
228
+ Court reporter
229
+
230
+ A person who makes a word-for-word record of what is said in court, generally
231
+ by using a
232
+
233
+ stenographic machine, shorthand or audio recording, and then produces a transcript
234
+ of the'
235
+ - source_sentence: 'What types of property may a debtor be able to exempt under the
236
+ homestead exemption? '
237
+ sentences:
238
+ - '-2-
239
+
240
+ Affidavit of Service: An affidavit intended to certify or prove that service
241
+ of a writ, notice, or other
242
+
243
+ document has been made.
244
+
245
+ Affirm: An act of declaring something to be true under the penalty of perjury
246
+ by a person who
247
+
248
+ conscientiously declines to take an oath for religious or other pertinent reasons;
249
+ also attorneys are
250
+
251
+ permitted to affirm rather than swear under oath.
252
+
253
+ Affirmation: A solemn and formal declaration under penalties of perjury that
254
+ a statement is true,
255
+
256
+ without an oath.
257
+
258
+ Affirmed: Upheld, agreed with (e.g.,The Appellate Court affirmed the judgment
259
+ of the City Court);
260
+
261
+ also means a challenge to a court decision or order was rejected.'
262
+ - 'A formal request for the protection of the federal bankruptcy laws. (There is
263
+ an official form
264
+
265
+ for bankruptcy petitions.)
266
+
267
+ Bankruptcy trustee
268
+
269
+ A private individual or corporation appointed in all Chapter 7 and Chapter 13
270
+ cases to
271
+
272
+ represent the interests of the bankruptcy estate and the debtor''s creditors.
273
+
274
+ Bench trial
275
+
276
+ A trial without a jury, in which the judge serves as the fact-finder.
277
+
278
+ Brief
279
+
280
+ A written statement submitted in a trial or appellate proceeding that explains
281
+ one side''s
282
+
283
+ legal and factual arguments.
284
+
285
+ Burden of proof
286
+
287
+ The duty to prove disputed facts. In civil cases, a plaintiff generally has the
288
+ burden of
289
+
290
+ proving his or her case. In criminal cases, the government has the burden of proving
291
+ the
292
+
293
+ defendant''s guilt. (See standard of proof.)'
294
+ - 'residence (homestead exemption), or some or all "tools of the trade" used by
295
+ the debtor to
296
+
297
+ make a living (i.e., auto tools for an auto mechanic or dental tools for a dentist).
298
+ The
299
+
300
+ availability and amount of property the debtor may exempt depends on the state
301
+ the debtor
302
+
303
+ lives in.
304
+
305
+ F
306
+
307
+ Face sheet filing
308
+
309
+ A bankruptcy case filed either without schedules or with incomplete schedules
310
+ listing few
311
+
312
+ creditors and debts. (Face sheet filings are often made for the purpose of delaying
313
+ an'
314
+ - source_sentence: How does a fraudulent transfer relate to a debtor's intent in bankruptcy
315
+ cases?
316
+ sentences:
317
+ - 'Glossary of Legal Terms
318
+
319
+ Find definitions of legal terms to help understand the federal
320
+
321
+ court system.
322
+
323
+ A
324
+
325
+ Acquittal
326
+
327
+ A jury verdict that a criminal defendant is not guilty, or the finding of a judge
328
+ that the
329
+
330
+ evidence is insufficient to support a conviction.
331
+
332
+ Active judge
333
+
334
+ A judge in the full-time service of the court. Compare to senior judge.
335
+
336
+ Administrative Office of the United States Courts (AO)
337
+
338
+ Enter legal term to search for definition
339
+
340
+ Search'
341
+ - 'A serious crime, usually punishable by at least one year in prison.
342
+
343
+ File
344
+
345
+ To place a paper in the official custody of the clerk of court to enter into the
346
+ files or records
347
+
348
+ of a case.
349
+
350
+ Fraudulent transfer
351
+
352
+ A transfer of a debtor''s property made with intent to defraud or for which the
353
+ debtor
354
+
355
+ receives less than the transferred property''s value.
356
+
357
+ Fresh start
358
+
359
+ The characterization of a debtor''s status after bankruptcy, i.e., free of most
360
+ debts. (Giving
361
+
362
+ debtors a fresh start is one purpose of the Bankruptcy Code.)
363
+
364
+ G
365
+
366
+ Grand jury
367
+
368
+ A body of 16-23 citizens who listen to evidence of criminal allegations, which
369
+ is presented by
370
+
371
+ the prosecutors, and determine whether there is probable cause to believe an individual'
372
+ - '-3-
373
+
374
+ Argument: A reason given in proof or rebuttal to persuade a judge or jury.
375
+
376
+ At Issue: Whenever the parties to an action come to a point in the pleadings
377
+ or argument which
378
+
379
+ is affirmed on one side and denied on the other, the points are said to be "at
380
+ issue".
381
+
382
+ Attachment: The taking of property into legal custody by an enforcement officer
383
+ (See specialty
384
+
385
+ section: Recovery of Chattel).
386
+
387
+ Attestation: The act of witnessing an instrument in writing at the request of
388
+ the party making the
389
+
390
+ instrument and signing it as a witness.
391
+
392
+ Attorney of Record: Attorney whose name appears in the court’s records or files
393
+ of a case.
394
+
395
+ Award: A decision of an Arbitrator, judge or jury.
396
+
397
+ -B-'
398
+ pipeline_tag: sentence-similarity
399
+ library_name: sentence-transformers
400
+ metrics:
401
+ - cosine_accuracy@1
402
+ - cosine_accuracy@3
403
+ - cosine_accuracy@5
404
+ - cosine_accuracy@10
405
+ - cosine_precision@1
406
+ - cosine_precision@3
407
+ - cosine_precision@5
408
+ - cosine_precision@10
409
+ - cosine_recall@1
410
+ - cosine_recall@3
411
+ - cosine_recall@5
412
+ - cosine_recall@10
413
+ - cosine_ndcg@10
414
+ - cosine_mrr@10
415
+ - cosine_map@100
416
+ model-index:
417
+ - name: SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
418
+ results:
419
+ - task:
420
+ type: information-retrieval
421
+ name: Information Retrieval
422
+ dataset:
423
+ name: Unknown
424
+ type: unknown
425
+ metrics:
426
+ - type: cosine_accuracy@1
427
+ value: 0.9318181818181818
428
+ name: Cosine Accuracy@1
429
+ - type: cosine_accuracy@3
430
+ value: 0.9318181818181818
431
+ name: Cosine Accuracy@3
432
+ - type: cosine_accuracy@5
433
+ value: 0.9545454545454546
434
+ name: Cosine Accuracy@5
435
+ - type: cosine_accuracy@10
436
+ value: 1.0
437
+ name: Cosine Accuracy@10
438
+ - type: cosine_precision@1
439
+ value: 0.9318181818181818
440
+ name: Cosine Precision@1
441
+ - type: cosine_precision@3
442
+ value: 0.3106060606060606
443
+ name: Cosine Precision@3
444
+ - type: cosine_precision@5
445
+ value: 0.1909090909090909
446
+ name: Cosine Precision@5
447
+ - type: cosine_precision@10
448
+ value: 0.09999999999999996
449
+ name: Cosine Precision@10
450
+ - type: cosine_recall@1
451
+ value: 0.9318181818181818
452
+ name: Cosine Recall@1
453
+ - type: cosine_recall@3
454
+ value: 0.9318181818181818
455
+ name: Cosine Recall@3
456
+ - type: cosine_recall@5
457
+ value: 0.9545454545454546
458
+ name: Cosine Recall@5
459
+ - type: cosine_recall@10
460
+ value: 1.0
461
+ name: Cosine Recall@10
462
+ - type: cosine_ndcg@10
463
+ value: 0.9565434941101226
464
+ name: Cosine Ndcg@10
465
+ - type: cosine_mrr@10
466
+ value: 0.9438131313131314
467
+ name: Cosine Mrr@10
468
+ - type: cosine_map@100
469
+ value: 0.9438131313131314
470
+ name: Cosine Map@100
471
+ ---
472
+
473
+ # SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
474
+
475
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l). 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.
476
+
477
+ ## Model Details
478
+
479
+ ### Model Description
480
+ - **Model Type:** Sentence Transformer
481
+ - **Base model:** [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l) <!-- at revision d8fb21ca8d905d2832ee8b96c894d3298964346b -->
482
+ - **Maximum Sequence Length:** 512 tokens
483
+ - **Output Dimensionality:** 1024 dimensions
484
+ - **Similarity Function:** Cosine Similarity
485
+ <!-- - **Training Dataset:** Unknown -->
486
+ <!-- - **Language:** Unknown -->
487
+ <!-- - **License:** Unknown -->
488
+
489
+ ### Model Sources
490
+
491
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
492
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
493
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
494
+
495
+ ### Full Model Architecture
496
+
497
+ ```
498
+ SentenceTransformer(
499
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
500
+ (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, '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': False, 'include_prompt': True})
501
+ (2): Normalize()
502
+ )
503
+ ```
504
+
505
+ ## Usage
506
+
507
+ ### Direct Usage (Sentence Transformers)
508
+
509
+ First install the Sentence Transformers library:
510
+
511
+ ```bash
512
+ pip install -U sentence-transformers
513
+ ```
514
+
515
+ Then you can load this model and run inference.
516
+ ```python
517
+ from sentence_transformers import SentenceTransformer
518
+
519
+ # Download from the 🤗 Hub
520
+ model = SentenceTransformer("vin00d/snowflake-arctic-legal-ft-1")
521
+ # Run inference
522
+ sentences = [
523
+ "How does a fraudulent transfer relate to a debtor's intent in bankruptcy cases?",
524
+ "A serious crime, usually punishable by at least one year in prison.\nFile\nTo place a paper in the official custody of the clerk of court to enter into the files or records\nof a case.\nFraudulent transfer\nA transfer of a debtor's property made with intent to defraud or for which the debtor\nreceives less than the transferred property's value.\nFresh start\nThe characterization of a debtor's status after bankruptcy, i.e., free of most debts. (Giving\ndebtors a fresh start is one purpose of the Bankruptcy Code.)\nG\nGrand jury\nA body of 16-23 citizens who listen to evidence of criminal allegations, which is presented by\nthe prosecutors, and determine whether there is probable cause to believe an individual",
525
+ '-3-\nArgument: A reason given in proof or rebuttal to persuade a judge or jury.\nAt Issue: Whenever the parties to an action come to a point in the pleadings or argument which\nis affirmed on one side and denied on the other, the points are said to be "at issue".\nAttachment: The taking of property into legal custody by an enforcement officer (See specialty\nsection: Recovery of Chattel).\nAttestation: The act of witnessing an instrument in writing at the request of the party making the\ninstrument and signing it as a witness.\nAttorney of Record: Attorney whose name appears in the court’s records or files of a case.\nAward: A decision of an Arbitrator, judge or jury.\n-B-',
526
+ ]
527
+ embeddings = model.encode(sentences)
528
+ print(embeddings.shape)
529
+ # [3, 1024]
530
+
531
+ # Get the similarity scores for the embeddings
532
+ similarities = model.similarity(embeddings, embeddings)
533
+ print(similarities.shape)
534
+ # [3, 3]
535
+ ```
536
+
537
+ <!--
538
+ ### Direct Usage (Transformers)
539
+
540
+ <details><summary>Click to see the direct usage in Transformers</summary>
541
+
542
+ </details>
543
+ -->
544
+
545
+ <!--
546
+ ### Downstream Usage (Sentence Transformers)
547
+
548
+ You can finetune this model on your own dataset.
549
+
550
+ <details><summary>Click to expand</summary>
551
+
552
+ </details>
553
+ -->
554
+
555
+ <!--
556
+ ### Out-of-Scope Use
557
+
558
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
559
+ -->
560
+
561
+ ## Evaluation
562
+
563
+ ### Metrics
564
+
565
+ #### Information Retrieval
566
+
567
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
568
+
569
+ | Metric | Value |
570
+ |:--------------------|:-----------|
571
+ | cosine_accuracy@1 | 0.9318 |
572
+ | cosine_accuracy@3 | 0.9318 |
573
+ | cosine_accuracy@5 | 0.9545 |
574
+ | cosine_accuracy@10 | 1.0 |
575
+ | cosine_precision@1 | 0.9318 |
576
+ | cosine_precision@3 | 0.3106 |
577
+ | cosine_precision@5 | 0.1909 |
578
+ | cosine_precision@10 | 0.1 |
579
+ | cosine_recall@1 | 0.9318 |
580
+ | cosine_recall@3 | 0.9318 |
581
+ | cosine_recall@5 | 0.9545 |
582
+ | cosine_recall@10 | 1.0 |
583
+ | **cosine_ndcg@10** | **0.9565** |
584
+ | cosine_mrr@10 | 0.9438 |
585
+ | cosine_map@100 | 0.9438 |
586
+
587
+ <!--
588
+ ## Bias, Risks and Limitations
589
+
590
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
591
+ -->
592
+
593
+ <!--
594
+ ### Recommendations
595
+
596
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
597
+ -->
598
+
599
+ ## Training Details
600
+
601
+ ### Training Dataset
602
+
603
+ #### Unnamed Dataset
604
+
605
+ * Size: 210 training samples
606
+ * Columns: <code>sentence_0</code> and <code>sentence_1</code>
607
+ * Approximate statistics based on the first 210 samples:
608
+ | | sentence_0 | sentence_1 |
609
+ |:--------|:----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
610
+ | type | string | string |
611
+ | details | <ul><li>min: 9 tokens</li><li>mean: 17.36 tokens</li><li>max: 33 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 122.9 tokens</li><li>max: 192 tokens</li></ul> |
612
+ * Samples:
613
+ | sentence_0 | sentence_1 |
614
+ |:---------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
615
+ | <code>What is the purpose of the glossary of common legal terms provided in the context? </code> | <code>GLOSSARY ‐ COMMON LEGAL TERMS<br>NOTE:  The following definitions are not legal definitions.  Rather, these definitions are<br>intended to give you a general idea of the meanings of common legal words.  For <br>comprehensive Definitions of legal terms, you may wish to consult a legal dictionary<br> “Black’s Law Dictionary” is one such legal dictionary which is usually available at<br> most law libraries.<br>This glossary of common legal terms is also available on‐line at:<br>http://www.nycourts.gov/lawlibraries/glossary.shtml<br> <br>ADDITIONAL ON‐LINE RESOURCES:<br>http://www.nolo.com/glossary.cfm <br>Nolo’s on‐line legal dictionary.<br>http://www.law‐dictionary.org/<br>Free on‐line legal dictionary search engine.<br>http://www.law.cornell.edu/wex</code> |
616
+ | <code>Where can one find a comprehensive legal dictionary for more detailed definitions of legal terms?</code> | <code>GLOSSARY ‐ COMMON LEGAL TERMS<br>NOTE:  The following definitions are not legal definitions.  Rather, these definitions are<br>intended to give you a general idea of the meanings of common legal words.  For <br>comprehensive Definitions of legal terms, you may wish to consult a legal dictionary<br> “Black’s Law Dictionary” is one such legal dictionary which is usually available at<br> most law libraries.<br>This glossary of common legal terms is also available on‐line at:<br>http://www.nycourts.gov/lawlibraries/glossary.shtml<br> <br>ADDITIONAL ON‐LINE RESOURCES:<br>http://www.nolo.com/glossary.cfm <br>Nolo’s on‐line legal dictionary.<br>http://www.law‐dictionary.org/<br>Free on‐line legal dictionary search engine.<br>http://www.law.cornell.edu/wex</code> |
617
+ | <code>What organization maintains the legal dictionary and encyclopedia mentioned in the context? </code> | <code>Legal dictionary and encyclopedia maintained by the<br>Legal Information Institute at Cornell Law School.</code> |
618
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
619
+ ```json
620
+ {
621
+ "loss": "MultipleNegativesRankingLoss",
622
+ "matryoshka_dims": [
623
+ 768,
624
+ 512,
625
+ 256,
626
+ 128,
627
+ 64
628
+ ],
629
+ "matryoshka_weights": [
630
+ 1,
631
+ 1,
632
+ 1,
633
+ 1,
634
+ 1
635
+ ],
636
+ "n_dims_per_step": -1
637
+ }
638
+ ```
639
+
640
+ ### Training Hyperparameters
641
+ #### Non-Default Hyperparameters
642
+
643
+ - `eval_strategy`: steps
644
+ - `per_device_train_batch_size`: 10
645
+ - `per_device_eval_batch_size`: 10
646
+ - `num_train_epochs`: 10
647
+ - `multi_dataset_batch_sampler`: round_robin
648
+
649
+ #### All Hyperparameters
650
+ <details><summary>Click to expand</summary>
651
+
652
+ - `overwrite_output_dir`: False
653
+ - `do_predict`: False
654
+ - `eval_strategy`: steps
655
+ - `prediction_loss_only`: True
656
+ - `per_device_train_batch_size`: 10
657
+ - `per_device_eval_batch_size`: 10
658
+ - `per_gpu_train_batch_size`: None
659
+ - `per_gpu_eval_batch_size`: None
660
+ - `gradient_accumulation_steps`: 1
661
+ - `eval_accumulation_steps`: None
662
+ - `torch_empty_cache_steps`: None
663
+ - `learning_rate`: 5e-05
664
+ - `weight_decay`: 0.0
665
+ - `adam_beta1`: 0.9
666
+ - `adam_beta2`: 0.999
667
+ - `adam_epsilon`: 1e-08
668
+ - `max_grad_norm`: 1
669
+ - `num_train_epochs`: 10
670
+ - `max_steps`: -1
671
+ - `lr_scheduler_type`: linear
672
+ - `lr_scheduler_kwargs`: {}
673
+ - `warmup_ratio`: 0.0
674
+ - `warmup_steps`: 0
675
+ - `log_level`: passive
676
+ - `log_level_replica`: warning
677
+ - `log_on_each_node`: True
678
+ - `logging_nan_inf_filter`: True
679
+ - `save_safetensors`: True
680
+ - `save_on_each_node`: False
681
+ - `save_only_model`: False
682
+ - `restore_callback_states_from_checkpoint`: False
683
+ - `no_cuda`: False
684
+ - `use_cpu`: False
685
+ - `use_mps_device`: False
686
+ - `seed`: 42
687
+ - `data_seed`: None
688
+ - `jit_mode_eval`: False
689
+ - `use_ipex`: False
690
+ - `bf16`: False
691
+ - `fp16`: False
692
+ - `fp16_opt_level`: O1
693
+ - `half_precision_backend`: auto
694
+ - `bf16_full_eval`: False
695
+ - `fp16_full_eval`: False
696
+ - `tf32`: None
697
+ - `local_rank`: 0
698
+ - `ddp_backend`: None
699
+ - `tpu_num_cores`: None
700
+ - `tpu_metrics_debug`: False
701
+ - `debug`: []
702
+ - `dataloader_drop_last`: False
703
+ - `dataloader_num_workers`: 0
704
+ - `dataloader_prefetch_factor`: None
705
+ - `past_index`: -1
706
+ - `disable_tqdm`: False
707
+ - `remove_unused_columns`: True
708
+ - `label_names`: None
709
+ - `load_best_model_at_end`: False
710
+ - `ignore_data_skip`: False
711
+ - `fsdp`: []
712
+ - `fsdp_min_num_params`: 0
713
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
714
+ - `fsdp_transformer_layer_cls_to_wrap`: None
715
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
716
+ - `deepspeed`: None
717
+ - `label_smoothing_factor`: 0.0
718
+ - `optim`: adamw_torch
719
+ - `optim_args`: None
720
+ - `adafactor`: False
721
+ - `group_by_length`: False
722
+ - `length_column_name`: length
723
+ - `ddp_find_unused_parameters`: None
724
+ - `ddp_bucket_cap_mb`: None
725
+ - `ddp_broadcast_buffers`: False
726
+ - `dataloader_pin_memory`: True
727
+ - `dataloader_persistent_workers`: False
728
+ - `skip_memory_metrics`: True
729
+ - `use_legacy_prediction_loop`: False
730
+ - `push_to_hub`: False
731
+ - `resume_from_checkpoint`: None
732
+ - `hub_model_id`: None
733
+ - `hub_strategy`: every_save
734
+ - `hub_private_repo`: None
735
+ - `hub_always_push`: False
736
+ - `gradient_checkpointing`: False
737
+ - `gradient_checkpointing_kwargs`: None
738
+ - `include_inputs_for_metrics`: False
739
+ - `include_for_metrics`: []
740
+ - `eval_do_concat_batches`: True
741
+ - `fp16_backend`: auto
742
+ - `push_to_hub_model_id`: None
743
+ - `push_to_hub_organization`: None
744
+ - `mp_parameters`:
745
+ - `auto_find_batch_size`: False
746
+ - `full_determinism`: False
747
+ - `torchdynamo`: None
748
+ - `ray_scope`: last
749
+ - `ddp_timeout`: 1800
750
+ - `torch_compile`: False
751
+ - `torch_compile_backend`: None
752
+ - `torch_compile_mode`: None
753
+ - `dispatch_batches`: None
754
+ - `split_batches`: None
755
+ - `include_tokens_per_second`: False
756
+ - `include_num_input_tokens_seen`: False
757
+ - `neftune_noise_alpha`: None
758
+ - `optim_target_modules`: None
759
+ - `batch_eval_metrics`: False
760
+ - `eval_on_start`: False
761
+ - `use_liger_kernel`: False
762
+ - `eval_use_gather_object`: False
763
+ - `average_tokens_across_devices`: False
764
+ - `prompts`: None
765
+ - `batch_sampler`: batch_sampler
766
+ - `multi_dataset_batch_sampler`: round_robin
767
+
768
+ </details>
769
+
770
+ ### Training Logs
771
+ | Epoch | Step | cosine_ndcg@10 |
772
+ |:------:|:----:|:--------------:|
773
+ | 1.0 | 21 | 0.9240 |
774
+ | 2.0 | 42 | 0.9628 |
775
+ | 2.3810 | 50 | 0.9628 |
776
+ | 3.0 | 63 | 0.9502 |
777
+ | 4.0 | 84 | 0.9569 |
778
+ | 4.7619 | 100 | 0.9563 |
779
+ | 5.0 | 105 | 0.9556 |
780
+ | 6.0 | 126 | 0.9569 |
781
+ | 7.0 | 147 | 0.9555 |
782
+ | 7.1429 | 150 | 0.9555 |
783
+ | 8.0 | 168 | 0.9565 |
784
+ | 9.0 | 189 | 0.9565 |
785
+ | 9.5238 | 200 | 0.9565 |
786
+ | 10.0 | 210 | 0.9565 |
787
+
788
+
789
+ ### Framework Versions
790
+ - Python: 3.11.11
791
+ - Sentence Transformers: 3.4.1
792
+ - Transformers: 4.48.3
793
+ - PyTorch: 2.5.1+cu124
794
+ - Accelerate: 1.3.0
795
+ - Datasets: 3.3.2
796
+ - Tokenizers: 0.21.0
797
+
798
+ ## Citation
799
+
800
+ ### BibTeX
801
+
802
+ #### Sentence Transformers
803
+ ```bibtex
804
+ @inproceedings{reimers-2019-sentence-bert,
805
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
806
+ author = "Reimers, Nils and Gurevych, Iryna",
807
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
808
+ month = "11",
809
+ year = "2019",
810
+ publisher = "Association for Computational Linguistics",
811
+ url = "https://arxiv.org/abs/1908.10084",
812
+ }
813
+ ```
814
+
815
+ #### MatryoshkaLoss
816
+ ```bibtex
817
+ @misc{kusupati2024matryoshka,
818
+ title={Matryoshka Representation Learning},
819
+ author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
820
+ year={2024},
821
+ eprint={2205.13147},
822
+ archivePrefix={arXiv},
823
+ primaryClass={cs.LG}
824
+ }
825
+ ```
826
+
827
+ #### MultipleNegativesRankingLoss
828
+ ```bibtex
829
+ @misc{henderson2017efficient,
830
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
831
+ author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
832
+ year={2017},
833
+ eprint={1705.00652},
834
+ archivePrefix={arXiv},
835
+ primaryClass={cs.CL}
836
+ }
837
+ ```
838
+
839
+ <!--
840
+ ## Glossary
841
+
842
+ *Clearly define terms in order to be accessible across audiences.*
843
+ -->
844
+
845
+ <!--
846
+ ## Model Card Authors
847
+
848
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
849
+ -->
850
+
851
+ <!--
852
+ ## Model Card Contact
853
+
854
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
855
+ -->
config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Snowflake/snowflake-arctic-embed-l",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 1024,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 4096,
13
+ "layer_norm_eps": 1e-12,
14
+ "max_position_embeddings": 512,
15
+ "model_type": "bert",
16
+ "num_attention_heads": 16,
17
+ "num_hidden_layers": 24,
18
+ "pad_token_id": 0,
19
+ "position_embedding_type": "absolute",
20
+ "torch_dtype": "float32",
21
+ "transformers_version": "4.48.3",
22
+ "type_vocab_size": 2,
23
+ "use_cache": true,
24
+ "vocab_size": 30522
25
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.4.1",
4
+ "transformers": "4.48.3",
5
+ "pytorch": "2.5.1+cu124"
6
+ },
7
+ "prompts": {
8
+ "query": "Represent this sentence for searching relevant passages: "
9
+ },
10
+ "default_prompt_name": null,
11
+ "similarity_fn_name": "cosine"
12
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1555b48c1bf42a197ec1d33b0f0f5da39f4ad202b3a5a5372cabf87f6d845c18
3
+ size 1336413848
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
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "pad_to_multiple_of": null,
52
+ "pad_token": "[PAD]",
53
+ "pad_token_type_id": 0,
54
+ "padding_side": "right",
55
+ "sep_token": "[SEP]",
56
+ "stride": 0,
57
+ "strip_accents": null,
58
+ "tokenize_chinese_chars": true,
59
+ "tokenizer_class": "BertTokenizer",
60
+ "truncation_side": "right",
61
+ "truncation_strategy": "longest_first",
62
+ "unk_token": "[UNK]"
63
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff