Hugging Face's logo

aari1995
/
German_Semantic_V3b

Sentence Similarity
sentence-transformers
Safetensors
German
bert
feature-extraction
loss:MatryoshkaLoss
custom_code
text-embeddings-inference
Model card Files
xet
Community
3
aari1995 commited on
Commit
1a2c048
·
verified ·
1 Parent(s): 367c957

Add new SentenceTransformer model.

Browse files
Files changed (11) hide show
  1. 1_Pooling/config.json +10 -0
  2. README.md +1217 -0
  3. config.json +34 -0
  4. config_sentence_transformers.json +10 -0
  5. model.safetensors +3 -0
  6. modules.json +14 -0
  7. sentence_bert_config.json +4 -0
  8. special_tokens_map.json +37 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +65 -0
  11. vocab.txt +0 -0
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 1024,
3
+ "pooling_mode_cls_token": false,
4
+ "pooling_mode_mean_tokens": true,
5
+ "pooling_mode_max_tokens": false,
6
+ "pooling_mode_mean_sqrt_len_tokens": false,
7
+ "pooling_mode_weightedmean_tokens": false,
8
+ "pooling_mode_lasttoken": false,
9
+ "include_prompt": true
10
+ }
README.md ADDED
@@ -0,0 +1,1217 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - de
4
+ - en
5
+ - es
6
+ - fr
7
+ - it
8
+ - nl
9
+ - pl
10
+ - pt
11
+ - ru
12
+ - zh
13
+ library_name: sentence-transformers
14
+ tags:
15
+ - sentence-transformers
16
+ - sentence-similarity
17
+ - feature-extraction
18
+ - dataset_size:10K<n<100K
19
+ - loss:MatryoshkaLoss
20
+ - loss:ContrastiveLoss
21
+ base_model: aari1995/gbert-large-nli_mix
22
+ metrics:
23
+ - pearson_cosine
24
+ - spearman_cosine
25
+ - pearson_manhattan
26
+ - spearman_manhattan
27
+ - pearson_euclidean
28
+ - spearman_euclidean
29
+ - pearson_dot
30
+ - spearman_dot
31
+ - pearson_max
32
+ - spearman_max
33
+ widget:
34
+ - source_sentence: Das Tor ist gelb.
35
+ sentences:
36
+ - Das Tor ist blau.
37
+ - Ein Mann mit seinem Hund am Strand.
38
+ - Die Menschen sitzen auf Bänken.
39
+ - source_sentence: Das Tor ist blau.
40
+ sentences:
41
+ - Ein blaues Moped parkt auf dem Bürgersteig.
42
+ - Drei Hunde spielen im weißen Schnee.
43
+ - Bombenanschläge töten 19 Menschen im Irak
44
+ - source_sentence: Ein Mann übt Boxen
45
+ sentences:
46
+ - Ein Fußballspieler versucht ein Tackling.
47
+ - 1 Getötet bei Protest in Bangladesch
48
+ - Das Mädchen sang in ein Mikrofon.
49
+ - source_sentence: Drei Männer tanzen.
50
+ sentences:
51
+ - Ein Mann tanzt.
52
+ - Ein Mann arbeitet an seinem Laptop.
53
+ - Das Mädchen sang in ein Mikrofon.
54
+ - source_sentence: Eine Flagge weht.
55
+ sentences:
56
+ - Die Flagge bewegte sich in der Luft.
57
+ - Zwei Personen beobachten das Wasser.
58
+ - Zwei Frauen sitzen in einem Cafe.
59
+ pipeline_tag: sentence-similarity
60
+ model-index:
61
+ - name: SentenceTransformer based on aari1995/gbert-large-nli_mix
62
+ results:
63
+ - task:
64
+ type: semantic-similarity
65
+ name: Semantic Similarity
66
+ dataset:
67
+ name: sts dev 1024
68
+ type: sts-dev-1024
69
+ metrics:
70
+ - type: pearson_cosine
71
+ value: 0.873823661552029
72
+ name: Pearson Cosine
73
+ - type: spearman_cosine
74
+ value: 0.8803520711782152
75
+ name: Spearman Cosine
76
+ - type: pearson_manhattan
77
+ value: 0.876117767161979
78
+ name: Pearson Manhattan
79
+ - type: spearman_manhattan
80
+ value: 0.8820122762561675
81
+ name: Spearman Manhattan
82
+ - type: pearson_euclidean
83
+ value: 0.8762079650155435
84
+ name: Pearson Euclidean
85
+ - type: spearman_euclidean
86
+ value: 0.8820817487274982
87
+ name: Spearman Euclidean
88
+ - type: pearson_dot
89
+ value: 0.838279478558382
90
+ name: Pearson Dot
91
+ - type: spearman_dot
92
+ value: 0.8381052886607077
93
+ name: Spearman Dot
94
+ - type: pearson_max
95
+ value: 0.8762079650155435
96
+ name: Pearson Max
97
+ - type: spearman_max
98
+ value: 0.8820817487274982
99
+ name: Spearman Max
100
+ - task:
101
+ type: semantic-similarity
102
+ name: Semantic Similarity
103
+ dataset:
104
+ name: sts dev 768
105
+ type: sts-dev-768
106
+ metrics:
107
+ - type: pearson_cosine
108
+ value: 0.8729182431103752
109
+ name: Pearson Cosine
110
+ - type: spearman_cosine
111
+ value: 0.8798510743177114
112
+ name: Spearman Cosine
113
+ - type: pearson_manhattan
114
+ value: 0.8750916595783815
115
+ name: Pearson Manhattan
116
+ - type: spearman_manhattan
117
+ value: 0.8809884317625296
118
+ name: Spearman Manhattan
119
+ - type: pearson_euclidean
120
+ value: 0.8754527585231735
121
+ name: Pearson Euclidean
122
+ - type: spearman_euclidean
123
+ value: 0.8811764170967997
124
+ name: Spearman Euclidean
125
+ - type: pearson_dot
126
+ value: 0.8386088963989539
127
+ name: Pearson Dot
128
+ - type: spearman_dot
129
+ value: 0.8387608674072754
130
+ name: Spearman Dot
131
+ - type: pearson_max
132
+ value: 0.8754527585231735
133
+ name: Pearson Max
134
+ - type: spearman_max
135
+ value: 0.8811764170967997
136
+ name: Spearman Max
137
+ - task:
138
+ type: semantic-similarity
139
+ name: Semantic Similarity
140
+ dataset:
141
+ name: sts dev 512
142
+ type: sts-dev-512
143
+ metrics:
144
+ - type: pearson_cosine
145
+ value: 0.8710783395197956
146
+ name: Pearson Cosine
147
+ - type: spearman_cosine
148
+ value: 0.878639260136433
149
+ name: Spearman Cosine
150
+ - type: pearson_manhattan
151
+ value: 0.8744942112479004
152
+ name: Pearson Manhattan
153
+ - type: spearman_manhattan
154
+ value: 0.880169853184795
155
+ name: Spearman Manhattan
156
+ - type: pearson_euclidean
157
+ value: 0.8750968130873006
158
+ name: Pearson Euclidean
159
+ - type: spearman_euclidean
160
+ value: 0.8805091146806316
161
+ name: Spearman Euclidean
162
+ - type: pearson_dot
163
+ value: 0.8320844036361574
164
+ name: Pearson Dot
165
+ - type: spearman_dot
166
+ value: 0.8320098342545608
167
+ name: Spearman Dot
168
+ - type: pearson_max
169
+ value: 0.8750968130873006
170
+ name: Pearson Max
171
+ - type: spearman_max
172
+ value: 0.8805091146806316
173
+ name: Spearman Max
174
+ - task:
175
+ type: semantic-similarity
176
+ name: Semantic Similarity
177
+ dataset:
178
+ name: sts dev 256
179
+ type: sts-dev-256
180
+ metrics:
181
+ - type: pearson_cosine
182
+ value: 0.8648952635235024
183
+ name: Pearson Cosine
184
+ - type: spearman_cosine
185
+ value: 0.8746516550395731
186
+ name: Spearman Cosine
187
+ - type: pearson_manhattan
188
+ value: 0.8708389858444562
189
+ name: Pearson Manhattan
190
+ - type: spearman_manhattan
191
+ value: 0.876029234462836
192
+ name: Spearman Manhattan
193
+ - type: pearson_euclidean
194
+ value: 0.8719490370119019
195
+ name: Pearson Euclidean
196
+ - type: spearman_euclidean
197
+ value: 0.876707897776359
198
+ name: Spearman Euclidean
199
+ - type: pearson_dot
200
+ value: 0.814982046736955
201
+ name: Pearson Dot
202
+ - type: spearman_dot
203
+ value: 0.8168481427335235
204
+ name: Spearman Dot
205
+ - type: pearson_max
206
+ value: 0.8719490370119019
207
+ name: Pearson Max
208
+ - type: spearman_max
209
+ value: 0.876707897776359
210
+ name: Spearman Max
211
+ - task:
212
+ type: semantic-similarity
213
+ name: Semantic Similarity
214
+ dataset:
215
+ name: sts dev 128
216
+ type: sts-dev-128
217
+ metrics:
218
+ - type: pearson_cosine
219
+ value: 0.8584911759712609
220
+ name: Pearson Cosine
221
+ - type: spearman_cosine
222
+ value: 0.8704026301204416
223
+ name: Spearman Cosine
224
+ - type: pearson_manhattan
225
+ value: 0.8657220587707122
226
+ name: Pearson Manhattan
227
+ - type: spearman_manhattan
228
+ value: 0.869723396167326
229
+ name: Spearman Manhattan
230
+ - type: pearson_euclidean
231
+ value: 0.8680692506297197
232
+ name: Pearson Euclidean
233
+ - type: spearman_euclidean
234
+ value: 0.8718542166801199
235
+ name: Spearman Euclidean
236
+ - type: pearson_dot
237
+ value: 0.8005092818222429
238
+ name: Pearson Dot
239
+ - type: spearman_dot
240
+ value: 0.8021754345558865
241
+ name: Spearman Dot
242
+ - type: pearson_max
243
+ value: 0.8680692506297197
244
+ name: Pearson Max
245
+ - type: spearman_max
246
+ value: 0.8718542166801199
247
+ name: Spearman Max
248
+ - task:
249
+ type: semantic-similarity
250
+ name: Semantic Similarity
251
+ dataset:
252
+ name: sts dev 64
253
+ type: sts-dev-64
254
+ metrics:
255
+ - type: pearson_cosine
256
+ value: 0.8483333803717887
257
+ name: Pearson Cosine
258
+ - type: spearman_cosine
259
+ value: 0.8652221599413363
260
+ name: Spearman Cosine
261
+ - type: pearson_manhattan
262
+ value: 0.8595603525995048
263
+ name: Pearson Manhattan
264
+ - type: spearman_manhattan
265
+ value: 0.863342194337673
266
+ name: Spearman Manhattan
267
+ - type: pearson_euclidean
268
+ value: 0.8635697556624868
269
+ name: Pearson Euclidean
270
+ - type: spearman_euclidean
271
+ value: 0.8668222027396277
272
+ name: Spearman Euclidean
273
+ - type: pearson_dot
274
+ value: 0.7733853267769795
275
+ name: Pearson Dot
276
+ - type: spearman_dot
277
+ value: 0.775678170624028
278
+ name: Spearman Dot
279
+ - type: pearson_max
280
+ value: 0.8635697556624868
281
+ name: Pearson Max
282
+ - type: spearman_max
283
+ value: 0.8668222027396277
284
+ name: Spearman Max
285
+ - task:
286
+ type: semantic-similarity
287
+ name: Semantic Similarity
288
+ dataset:
289
+ name: sts test 1024
290
+ type: sts-test-1024
291
+ metrics:
292
+ - type: pearson_cosine
293
+ value: 0.8538749625112824
294
+ name: Pearson Cosine
295
+ - type: spearman_cosine
296
+ value: 0.8622934726599119
297
+ name: Spearman Cosine
298
+ - type: pearson_manhattan
299
+ value: 0.8554617861095041
300
+ name: Pearson Manhattan
301
+ - type: spearman_manhattan
302
+ value: 0.8632850500504865
303
+ name: Spearman Manhattan
304
+ - type: pearson_euclidean
305
+ value: 0.8554205957277228
306
+ name: Pearson Euclidean
307
+ - type: spearman_euclidean
308
+ value: 0.8630779166725503
309
+ name: Spearman Euclidean
310
+ - type: pearson_dot
311
+ value: 0.8170146846171837
312
+ name: Pearson Dot
313
+ - type: spearman_dot
314
+ value: 0.8149857685956332
315
+ name: Spearman Dot
316
+ - type: pearson_max
317
+ value: 0.8554617861095041
318
+ name: Pearson Max
319
+ - type: spearman_max
320
+ value: 0.8632850500504865
321
+ name: Spearman Max
322
+ - task:
323
+ type: semantic-similarity
324
+ name: Semantic Similarity
325
+ dataset:
326
+ name: sts test 768
327
+ type: sts-test-768
328
+ metrics:
329
+ - type: pearson_cosine
330
+ value: 0.853820621972726
331
+ name: Pearson Cosine
332
+ - type: spearman_cosine
333
+ value: 0.863198271488271
334
+ name: Spearman Cosine
335
+ - type: pearson_manhattan
336
+ value: 0.8558709278385018
337
+ name: Pearson Manhattan
338
+ - type: spearman_manhattan
339
+ value: 0.8637532036004547
340
+ name: Spearman Manhattan
341
+ - type: pearson_euclidean
342
+ value: 0.8558597695346744
343
+ name: Pearson Euclidean
344
+ - type: spearman_euclidean
345
+ value: 0.8634247094122574
346
+ name: Spearman Euclidean
347
+ - type: pearson_dot
348
+ value: 0.8169163431962185
349
+ name: Pearson Dot
350
+ - type: spearman_dot
351
+ value: 0.8156867907361973
352
+ name: Spearman Dot
353
+ - type: pearson_max
354
+ value: 0.8558709278385018
355
+ name: Pearson Max
356
+ - type: spearman_max
357
+ value: 0.8637532036004547
358
+ name: Spearman Max
359
+ - task:
360
+ type: semantic-similarity
361
+ name: Semantic Similarity
362
+ dataset:
363
+ name: sts test 512
364
+ type: sts-test-512
365
+ metrics:
366
+ - type: pearson_cosine
367
+ value: 0.8502336569709972
368
+ name: Pearson Cosine
369
+ - type: spearman_cosine
370
+ value: 0.8623838162450902
371
+ name: Spearman Cosine
372
+ - type: pearson_manhattan
373
+ value: 0.8547121881183612
374
+ name: Pearson Manhattan
375
+ - type: spearman_manhattan
376
+ value: 0.8628698143219098
377
+ name: Spearman Manhattan
378
+ - type: pearson_euclidean
379
+ value: 0.8546114371189246
380
+ name: Pearson Euclidean
381
+ - type: spearman_euclidean
382
+ value: 0.8625109910600326
383
+ name: Spearman Euclidean
384
+ - type: pearson_dot
385
+ value: 0.8108392647310044
386
+ name: Pearson Dot
387
+ - type: spearman_dot
388
+ value: 0.8103261097232485
389
+ name: Spearman Dot
390
+ - type: pearson_max
391
+ value: 0.8547121881183612
392
+ name: Pearson Max
393
+ - type: spearman_max
394
+ value: 0.8628698143219098
395
+ name: Spearman Max
396
+ - task:
397
+ type: semantic-similarity
398
+ name: Semantic Similarity
399
+ dataset:
400
+ name: sts test 256
401
+ type: sts-test-256
402
+ metrics:
403
+ - type: pearson_cosine
404
+ value: 0.8441242786553879
405
+ name: Pearson Cosine
406
+ - type: spearman_cosine
407
+ value: 0.8582717489671877
408
+ name: Spearman Cosine
409
+ - type: pearson_manhattan
410
+ value: 0.8517415030362573
411
+ name: Pearson Manhattan
412
+ - type: spearman_manhattan
413
+ value: 0.8591688553092182
414
+ name: Spearman Manhattan
415
+ - type: pearson_euclidean
416
+ value: 0.8516965854845419
417
+ name: Pearson Euclidean
418
+ - type: spearman_euclidean
419
+ value: 0.8591770194196562
420
+ name: Spearman Euclidean
421
+ - type: pearson_dot
422
+ value: 0.7901870400809775
423
+ name: Pearson Dot
424
+ - type: spearman_dot
425
+ value: 0.7891397281321177
426
+ name: Spearman Dot
427
+ - type: pearson_max
428
+ value: 0.8517415030362573
429
+ name: Pearson Max
430
+ - type: spearman_max
431
+ value: 0.8591770194196562
432
+ name: Spearman Max
433
+ - task:
434
+ type: semantic-similarity
435
+ name: Semantic Similarity
436
+ dataset:
437
+ name: sts test 128
438
+ type: sts-test-128
439
+ metrics:
440
+ - type: pearson_cosine
441
+ value: 0.8369352495821198
442
+ name: Pearson Cosine
443
+ - type: spearman_cosine
444
+ value: 0.8545806562301762
445
+ name: Spearman Cosine
446
+ - type: pearson_manhattan
447
+ value: 0.8474289413580527
448
+ name: Pearson Manhattan
449
+ - type: spearman_manhattan
450
+ value: 0.8546935424655524
451
+ name: Spearman Manhattan
452
+ - type: pearson_euclidean
453
+ value: 0.8478267316251253
454
+ name: Pearson Euclidean
455
+ - type: spearman_euclidean
456
+ value: 0.8550464936365929
457
+ name: Spearman Euclidean
458
+ - type: pearson_dot
459
+ value: 0.7732663297266509
460
+ name: Pearson Dot
461
+ - type: spearman_dot
462
+ value: 0.7720532782903432
463
+ name: Spearman Dot
464
+ - type: pearson_max
465
+ value: 0.8478267316251253
466
+ name: Pearson Max
467
+ - type: spearman_max
468
+ value: 0.8550464936365929
469
+ name: Spearman Max
470
+ - task:
471
+ type: semantic-similarity
472
+ name: Semantic Similarity
473
+ dataset:
474
+ name: sts test 64
475
+ type: sts-test-64
476
+ metrics:
477
+ - type: pearson_cosine
478
+ value: 0.8282288301025145
479
+ name: Pearson Cosine
480
+ - type: spearman_cosine
481
+ value: 0.8507215646125454
482
+ name: Spearman Cosine
483
+ - type: pearson_manhattan
484
+ value: 0.8404915813802649
485
+ name: Pearson Manhattan
486
+ - type: spearman_manhattan
487
+ value: 0.8482910175231816
488
+ name: Spearman Manhattan
489
+ - type: pearson_euclidean
490
+ value: 0.8425986040609018
491
+ name: Pearson Euclidean
492
+ - type: spearman_euclidean
493
+ value: 0.8498681513437906
494
+ name: Spearman Euclidean
495
+ - type: pearson_dot
496
+ value: 0.7518854418344252
497
+ name: Pearson Dot
498
+ - type: spearman_dot
499
+ value: 0.7518133373839283
500
+ name: Spearman Dot
501
+ - type: pearson_max
502
+ value: 0.8425986040609018
503
+ name: Pearson Max
504
+ - type: spearman_max
505
+ value: 0.8507215646125454
506
+ name: Spearman Max
507
+ ---
508
+
509
+ # SentenceTransformer based on aari1995/gbert-large-nli_mix
510
+
511
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [aari1995/gbert-large-nli_mix](https://huggingface.co/aari1995/gbert-large-nli_mix) on the [PhilipMay/stsb_multi_mt](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt) 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.
512
+
513
+ ## Model Details
514
+
515
+ ### Model Description
516
+ - **Model Type:** Sentence Transformer
517
+ - **Base model:** [aari1995/gbert-large-nli_mix](https://huggingface.co/aari1995/gbert-large-nli_mix) <!-- at revision 86b82327d5898d81f9b8caafbf228b803f25abc1 -->
518
+ - **Maximum Sequence Length:** 8192 tokens
519
+ - **Output Dimensionality:** 1024 tokens
520
+ - **Similarity Function:** Cosine Similarity
521
+ - **Training Dataset:**
522
+ - [PhilipMay/stsb_multi_mt](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt)
523
+ - **Languages:** de, en, es, fr, it, nl, pl, pt, ru, zh
524
+ <!-- - **License:** Unknown -->
525
+
526
+ ### Model Sources
527
+
528
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
529
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
530
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
531
+
532
+ ### Full Model Architecture
533
+
534
+ ```
535
+ SentenceTransformer(
536
+ (0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: JinaBertModel
537
+ (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
538
+ )
539
+ ```
540
+
541
+ ## Usage
542
+
543
+ ### Direct Usage (Sentence Transformers)
544
+
545
+ First install the Sentence Transformers library:
546
+
547
+ ```bash
548
+ pip install -U sentence-transformers
549
+ ```
550
+
551
+ Then you can load this model and run inference.
552
+ ```python
553
+ from sentence_transformers import SentenceTransformer
554
+
555
+ # Download from the 🤗 Hub
556
+ model = SentenceTransformer("aari1995/German_Semantic_V3_2_STS_MIX")
557
+ # Run inference
558
+ sentences = [
559
+ 'Eine Flagge weht.',
560
+ 'Die Flagge bewegte sich in der Luft.',
561
+ 'Zwei Personen beobachten das Wasser.',
562
+ ]
563
+ embeddings = model.encode(sentences)
564
+ print(embeddings.shape)
565
+ # [3, 1024]
566
+
567
+ # Get the similarity scores for the embeddings
568
+ similarities = model.similarity(embeddings, embeddings)
569
+ print(similarities.shape)
570
+ # [3, 3]
571
+ ```
572
+
573
+ <!--
574
+ ### Direct Usage (Transformers)
575
+
576
+ <details><summary>Click to see the direct usage in Transformers</summary>
577
+
578
+ </details>
579
+ -->
580
+
581
+ <!--
582
+ ### Downstream Usage (Sentence Transformers)
583
+
584
+ You can finetune this model on your own dataset.
585
+
586
+ <details><summary>Click to expand</summary>
587
+
588
+ </details>
589
+ -->
590
+
591
+ <!--
592
+ ### Out-of-Scope Use
593
+
594
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
595
+ -->
596
+
597
+ ## Evaluation
598
+
599
+ ### Metrics
600
+
601
+ #### Semantic Similarity
602
+ * Dataset: `sts-dev-1024`
603
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
604
+
605
+ | Metric | Value |
606
+ |:--------------------|:-----------|
607
+ | pearson_cosine | 0.8738 |
608
+ | **spearman_cosine** | **0.8804** |
609
+ | pearson_manhattan | 0.8761 |
610
+ | spearman_manhattan | 0.882 |
611
+ | pearson_euclidean | 0.8762 |
612
+ | spearman_euclidean | 0.8821 |
613
+ | pearson_dot | 0.8383 |
614
+ | spearman_dot | 0.8381 |
615
+ | pearson_max | 0.8762 |
616
+ | spearman_max | 0.8821 |
617
+
618
+ #### Semantic Similarity
619
+ * Dataset: `sts-dev-768`
620
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
621
+
622
+ | Metric | Value |
623
+ |:--------------------|:-----------|
624
+ | pearson_cosine | 0.8729 |
625
+ | **spearman_cosine** | **0.8799** |
626
+ | pearson_manhattan | 0.8751 |
627
+ | spearman_manhattan | 0.881 |
628
+ | pearson_euclidean | 0.8755 |
629
+ | spearman_euclidean | 0.8812 |
630
+ | pearson_dot | 0.8386 |
631
+ | spearman_dot | 0.8388 |
632
+ | pearson_max | 0.8755 |
633
+ | spearman_max | 0.8812 |
634
+
635
+ #### Semantic Similarity
636
+ * Dataset: `sts-dev-512`
637
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
638
+
639
+ | Metric | Value |
640
+ |:--------------------|:-----------|
641
+ | pearson_cosine | 0.8711 |
642
+ | **spearman_cosine** | **0.8786** |
643
+ | pearson_manhattan | 0.8745 |
644
+ | spearman_manhattan | 0.8802 |
645
+ | pearson_euclidean | 0.8751 |
646
+ | spearman_euclidean | 0.8805 |
647
+ | pearson_dot | 0.8321 |
648
+ | spearman_dot | 0.832 |
649
+ | pearson_max | 0.8751 |
650
+ | spearman_max | 0.8805 |
651
+
652
+ #### Semantic Similarity
653
+ * Dataset: `sts-dev-256`
654
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
655
+
656
+ | Metric | Value |
657
+ |:--------------------|:-----------|
658
+ | pearson_cosine | 0.8649 |
659
+ | **spearman_cosine** | **0.8747** |
660
+ | pearson_manhattan | 0.8708 |
661
+ | spearman_manhattan | 0.876 |
662
+ | pearson_euclidean | 0.8719 |
663
+ | spearman_euclidean | 0.8767 |
664
+ | pearson_dot | 0.815 |
665
+ | spearman_dot | 0.8168 |
666
+ | pearson_max | 0.8719 |
667
+ | spearman_max | 0.8767 |
668
+
669
+ #### Semantic Similarity
670
+ * Dataset: `sts-dev-128`
671
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
672
+
673
+ | Metric | Value |
674
+ |:--------------------|:-----------|
675
+ | pearson_cosine | 0.8585 |
676
+ | **spearman_cosine** | **0.8704** |
677
+ | pearson_manhattan | 0.8657 |
678
+ | spearman_manhattan | 0.8697 |
679
+ | pearson_euclidean | 0.8681 |
680
+ | spearman_euclidean | 0.8719 |
681
+ | pearson_dot | 0.8005 |
682
+ | spearman_dot | 0.8022 |
683
+ | pearson_max | 0.8681 |
684
+ | spearman_max | 0.8719 |
685
+
686
+ #### Semantic Similarity
687
+ * Dataset: `sts-dev-64`
688
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
689
+
690
+ | Metric | Value |
691
+ |:--------------------|:-----------|
692
+ | pearson_cosine | 0.8483 |
693
+ | **spearman_cosine** | **0.8652** |
694
+ | pearson_manhattan | 0.8596 |
695
+ | spearman_manhattan | 0.8633 |
696
+ | pearson_euclidean | 0.8636 |
697
+ | spearman_euclidean | 0.8668 |
698
+ | pearson_dot | 0.7734 |
699
+ | spearman_dot | 0.7757 |
700
+ | pearson_max | 0.8636 |
701
+ | spearman_max | 0.8668 |
702
+
703
+ #### Semantic Similarity
704
+ * Dataset: `sts-test-1024`
705
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
706
+
707
+ | Metric | Value |
708
+ |:--------------------|:-----------|
709
+ | pearson_cosine | 0.8539 |
710
+ | **spearman_cosine** | **0.8623** |
711
+ | pearson_manhattan | 0.8555 |
712
+ | spearman_manhattan | 0.8633 |
713
+ | pearson_euclidean | 0.8554 |
714
+ | spearman_euclidean | 0.8631 |
715
+ | pearson_dot | 0.817 |
716
+ | spearman_dot | 0.815 |
717
+ | pearson_max | 0.8555 |
718
+ | spearman_max | 0.8633 |
719
+
720
+ #### Semantic Similarity
721
+ * Dataset: `sts-test-768`
722
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
723
+
724
+ | Metric | Value |
725
+ |:--------------------|:-----------|
726
+ | pearson_cosine | 0.8538 |
727
+ | **spearman_cosine** | **0.8632** |
728
+ | pearson_manhattan | 0.8559 |
729
+ | spearman_manhattan | 0.8638 |
730
+ | pearson_euclidean | 0.8559 |
731
+ | spearman_euclidean | 0.8634 |
732
+ | pearson_dot | 0.8169 |
733
+ | spearman_dot | 0.8157 |
734
+ | pearson_max | 0.8559 |
735
+ | spearman_max | 0.8638 |
736
+
737
+ #### Semantic Similarity
738
+ * Dataset: `sts-test-512`
739
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
740
+
741
+ | Metric | Value |
742
+ |:--------------------|:-----------|
743
+ | pearson_cosine | 0.8502 |
744
+ | **spearman_cosine** | **0.8624** |
745
+ | pearson_manhattan | 0.8547 |
746
+ | spearman_manhattan | 0.8629 |
747
+ | pearson_euclidean | 0.8546 |
748
+ | spearman_euclidean | 0.8625 |
749
+ | pearson_dot | 0.8108 |
750
+ | spearman_dot | 0.8103 |
751
+ | pearson_max | 0.8547 |
752
+ | spearman_max | 0.8629 |
753
+
754
+ #### Semantic Similarity
755
+ * Dataset: `sts-test-256`
756
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
757
+
758
+ | Metric | Value |
759
+ |:--------------------|:-----------|
760
+ | pearson_cosine | 0.8441 |
761
+ | **spearman_cosine** | **0.8583** |
762
+ | pearson_manhattan | 0.8517 |
763
+ | spearman_manhattan | 0.8592 |
764
+ | pearson_euclidean | 0.8517 |
765
+ | spearman_euclidean | 0.8592 |
766
+ | pearson_dot | 0.7902 |
767
+ | spearman_dot | 0.7891 |
768
+ | pearson_max | 0.8517 |
769
+ | spearman_max | 0.8592 |
770
+
771
+ #### Semantic Similarity
772
+ * Dataset: `sts-test-128`
773
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
774
+
775
+ | Metric | Value |
776
+ |:--------------------|:-----------|
777
+ | pearson_cosine | 0.8369 |
778
+ | **spearman_cosine** | **0.8546** |
779
+ | pearson_manhattan | 0.8474 |
780
+ | spearman_manhattan | 0.8547 |
781
+ | pearson_euclidean | 0.8478 |
782
+ | spearman_euclidean | 0.855 |
783
+ | pearson_dot | 0.7733 |
784
+ | spearman_dot | 0.7721 |
785
+ | pearson_max | 0.8478 |
786
+ | spearman_max | 0.855 |
787
+
788
+ #### Semantic Similarity
789
+ * Dataset: `sts-test-64`
790
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
791
+
792
+ | Metric | Value |
793
+ |:--------------------|:-----------|
794
+ | pearson_cosine | 0.8282 |
795
+ | **spearman_cosine** | **0.8507** |
796
+ | pearson_manhattan | 0.8405 |
797
+ | spearman_manhattan | 0.8483 |
798
+ | pearson_euclidean | 0.8426 |
799
+ | spearman_euclidean | 0.8499 |
800
+ | pearson_dot | 0.7519 |
801
+ | spearman_dot | 0.7518 |
802
+ | pearson_max | 0.8426 |
803
+ | spearman_max | 0.8507 |
804
+
805
+ <!--
806
+ ## Bias, Risks and Limitations
807
+
808
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
809
+ -->
810
+
811
+ <!--
812
+ ### Recommendations
813
+
814
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
815
+ -->
816
+
817
+ ## Training Details
818
+
819
+ ### Training Dataset
820
+
821
+ #### PhilipMay/stsb_multi_mt
822
+
823
+ * Dataset: [PhilipMay/stsb_multi_mt](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt) at [3acaa3d](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt/tree/3acaa3dd8c91649e0b8e627ffad891f059e47c8c)
824
+ * Size: 22,996 training samples
825
+ * Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
826
+ * Approximate statistics based on the first 1000 samples:
827
+ | | sentence1 | sentence2 | score |
828
+ |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
829
+ | type | string | string | float |
830
+ | details | <ul><li>min: 6 tokens</li><li>mean: 18.13 tokens</li><li>max: 65 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 18.25 tokens</li><li>max: 90 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.54</li><li>max: 1.0</li></ul> |
831
+ * Samples:
832
+ | sentence1 | sentence2 | score |
833
+ |:-------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:--------------------------------|
834
+ | <code>schütze wegen mordes an schwarzem us-jugendlichen angeklagt</code> | <code>gedanken zu den rassenbeziehungen unter einem schwarzen präsidenten</code> | <code>0.1599999964237213</code> |
835
+ | <code>fußballspieler kicken einen fußball in das tor.</code> | <code>Ein Fußballspieler schießt ein Tor.</code> | <code>0.7599999904632568</code> |
836
+ | <code>obama lockert abschiebungsregeln für junge einwanderer</code> | <code>usa lockert abschiebebestimmungen für jugendliche: napolitano</code> | <code>0.800000011920929</code> |
837
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
838
+ ```json
839
+ {
840
+ "loss": "ContrastiveLoss",
841
+ "matryoshka_dims": [
842
+ 1024,
843
+ 768,
844
+ 512,
845
+ 256,
846
+ 128,
847
+ 64
848
+ ],
849
+ "matryoshka_weights": [
850
+ 1,
851
+ 1,
852
+ 1,
853
+ 1,
854
+ 1,
855
+ 1
856
+ ],
857
+ "n_dims_per_step": -1
858
+ }
859
+ ```
860
+
861
+ ### Evaluation Dataset
862
+
863
+ #### PhilipMay/stsb_multi_mt
864
+
865
+ * Dataset: [PhilipMay/stsb_multi_mt](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt) at [3acaa3d](https://huggingface.co/datasets/PhilipMay/stsb_multi_mt/tree/3acaa3dd8c91649e0b8e627ffad891f059e47c8c)
866
+ * Size: 1,500 evaluation samples
867
+ * Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>score</code>
868
+ * Approximate statistics based on the first 1000 samples:
869
+ | | sentence1 | sentence2 | score |
870
+ |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------|
871
+ | type | string | string | float |
872
+ | details | <ul><li>min: 5 tokens</li><li>mean: 16.54 tokens</li><li>max: 53 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 16.53 tokens</li><li>max: 47 tokens</li></ul> | <ul><li>min: 0.0</li><li>mean: 0.47</li><li>max: 1.0</li></ul> |
873
+ * Samples:
874
+ | sentence1 | sentence2 | score |
875
+ |:-------------------------------------------------------------|:-----------------------------------------------------------|:-------------------------------|
876
+ | <code>Ein Mann mit einem Schutzhelm tanzt.</code> | <code>Ein Mann mit einem Schutzhelm tanzt.</code> | <code>1.0</code> |
877
+ | <code>Ein kleines Kind reitet auf einem Pferd.</code> | <code>Ein Kind reitet auf einem Pferd.</code> | <code>0.949999988079071</code> |
878
+ | <code>Ein Mann verfüttert eine Maus an eine Schlange.</code> | <code>Der Mann füttert die Schlange mit einer Maus.</code> | <code>1.0</code> |
879
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
880
+ ```json
881
+ {
882
+ "loss": "ContrastiveLoss",
883
+ "matryoshka_dims": [
884
+ 1024,
885
+ 768,
886
+ 512,
887
+ 256,
888
+ 128,
889
+ 64
890
+ ],
891
+ "matryoshka_weights": [
892
+ 1,
893
+ 1,
894
+ 1,
895
+ 1,
896
+ 1,
897
+ 1
898
+ ],
899
+ "n_dims_per_step": -1
900
+ }
901
+ ```
902
+
903
+ ### Training Hyperparameters
904
+ #### Non-Default Hyperparameters
905
+
906
+ - `eval_strategy`: steps
907
+ - `learning_rate`: 5e-06
908
+ - `num_train_epochs`: 4
909
+ - `warmup_ratio`: 0.1
910
+
911
+ #### All Hyperparameters
912
+ <details><summary>Click to expand</summary>
913
+
914
+ - `overwrite_output_dir`: False
915
+ - `do_predict`: False
916
+ - `eval_strategy`: steps
917
+ - `prediction_loss_only`: True
918
+ - `per_device_train_batch_size`: 8
919
+ - `per_device_eval_batch_size`: 8
920
+ - `per_gpu_train_batch_size`: None
921
+ - `per_gpu_eval_batch_size`: None
922
+ - `gradient_accumulation_steps`: 1
923
+ - `eval_accumulation_steps`: None
924
+ - `learning_rate`: 5e-06
925
+ - `weight_decay`: 0.0
926
+ - `adam_beta1`: 0.9
927
+ - `adam_beta2`: 0.999
928
+ - `adam_epsilon`: 1e-08
929
+ - `max_grad_norm`: 1.0
930
+ - `num_train_epochs`: 4
931
+ - `max_steps`: -1
932
+ - `lr_scheduler_type`: linear
933
+ - `lr_scheduler_kwargs`: {}
934
+ - `warmup_ratio`: 0.1
935
+ - `warmup_steps`: 0
936
+ - `log_level`: passive
937
+ - `log_level_replica`: warning
938
+ - `log_on_each_node`: True
939
+ - `logging_nan_inf_filter`: True
940
+ - `save_safetensors`: True
941
+ - `save_on_each_node`: False
942
+ - `save_only_model`: False
943
+ - `restore_callback_states_from_checkpoint`: False
944
+ - `no_cuda`: False
945
+ - `use_cpu`: False
946
+ - `use_mps_device`: False
947
+ - `seed`: 42
948
+ - `data_seed`: None
949
+ - `jit_mode_eval`: False
950
+ - `use_ipex`: False
951
+ - `bf16`: False
952
+ - `fp16`: False
953
+ - `fp16_opt_level`: O1
954
+ - `half_precision_backend`: auto
955
+ - `bf16_full_eval`: False
956
+ - `fp16_full_eval`: False
957
+ - `tf32`: None
958
+ - `local_rank`: 0
959
+ - `ddp_backend`: None
960
+ - `tpu_num_cores`: None
961
+ - `tpu_metrics_debug`: False
962
+ - `debug`: []
963
+ - `dataloader_drop_last`: False
964
+ - `dataloader_num_workers`: 0
965
+ - `dataloader_prefetch_factor`: None
966
+ - `past_index`: -1
967
+ - `disable_tqdm`: False
968
+ - `remove_unused_columns`: True
969
+ - `label_names`: None
970
+ - `load_best_model_at_end`: False
971
+ - `ignore_data_skip`: False
972
+ - `fsdp`: []
973
+ - `fsdp_min_num_params`: 0
974
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
975
+ - `fsdp_transformer_layer_cls_to_wrap`: None
976
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
977
+ - `deepspeed`: None
978
+ - `label_smoothing_factor`: 0.0
979
+ - `optim`: adamw_torch
980
+ - `optim_args`: None
981
+ - `adafactor`: False
982
+ - `group_by_length`: False
983
+ - `length_column_name`: length
984
+ - `ddp_find_unused_parameters`: None
985
+ - `ddp_bucket_cap_mb`: None
986
+ - `ddp_broadcast_buffers`: False
987
+ - `dataloader_pin_memory`: True
988
+ - `dataloader_persistent_workers`: False
989
+ - `skip_memory_metrics`: True
990
+ - `use_legacy_prediction_loop`: False
991
+ - `push_to_hub`: False
992
+ - `resume_from_checkpoint`: None
993
+ - `hub_model_id`: None
994
+ - `hub_strategy`: every_save
995
+ - `hub_private_repo`: False
996
+ - `hub_always_push`: False
997
+ - `gradient_checkpointing`: False
998
+ - `gradient_checkpointing_kwargs`: None
999
+ - `include_inputs_for_metrics`: False
1000
+ - `eval_do_concat_batches`: True
1001
+ - `fp16_backend`: auto
1002
+ - `push_to_hub_model_id`: None
1003
+ - `push_to_hub_organization`: None
1004
+ - `mp_parameters`:
1005
+ - `auto_find_batch_size`: False
1006
+ - `full_determinism`: False
1007
+ - `torchdynamo`: None
1008
+ - `ray_scope`: last
1009
+ - `ddp_timeout`: 1800
1010
+ - `torch_compile`: False
1011
+ - `torch_compile_backend`: None
1012
+ - `torch_compile_mode`: None
1013
+ - `dispatch_batches`: None
1014
+ - `split_batches`: None
1015
+ - `include_tokens_per_second`: False
1016
+ - `include_num_input_tokens_seen`: False
1017
+ - `neftune_noise_alpha`: None
1018
+ - `optim_target_modules`: None
1019
+ - `batch_eval_metrics`: False
1020
+ - `eval_on_start`: False
1021
+ - `batch_sampler`: batch_sampler
1022
+ - `multi_dataset_batch_sampler`: proportional
1023
+
1024
+ </details>
1025
+
1026
+ ### Training Logs
1027
+ <details><summary>Click to expand</summary>
1028
+
1029
+ | Epoch | Step | Training Loss | loss | sts-dev-1024_spearman_cosine | sts-dev-128_spearman_cosine | sts-dev-256_spearman_cosine | sts-dev-512_spearman_cosine | sts-dev-64_spearman_cosine | sts-dev-768_spearman_cosine | sts-test-1024_spearman_cosine | sts-test-128_spearman_cosine | sts-test-256_spearman_cosine | sts-test-512_spearman_cosine | sts-test-64_spearman_cosine | sts-test-768_spearman_cosine |
1030
+ |:------:|:-----:|:-------------:|:------:|:----------------------------:|:---------------------------:|:---------------------------:|:---------------------------:|:--------------------------:|:---------------------------:|:-----------------------------:|:----------------------------:|:----------------------------:|:----------------------------:|:---------------------------:|:----------------------------:|
1031
+ | 0.0348 | 100 | 0.2334 | 0.2530 | 0.8329 | 0.8219 | 0.8274 | 0.8292 | 0.8148 | 0.8317 | - | - | - | - | - | - |
1032
+ | 0.0696 | 200 | 0.1959 | 0.1921 | 0.8285 | 0.8183 | 0.8234 | 0.8250 | 0.8121 | 0.8275 | - | - | - | - | - | - |
1033
+ | 0.1043 | 300 | 0.1468 | 0.1592 | 0.8346 | 0.8267 | 0.8305 | 0.8319 | 0.8227 | 0.8334 | - | - | - | - | - | - |
1034
+ | 0.1391 | 400 | 0.1346 | 0.1511 | 0.8513 | 0.8451 | 0.8486 | 0.8497 | 0.8418 | 0.8505 | - | - | - | - | - | - |
1035
+ | 0.1739 | 500 | 0.1333 | 0.1480 | 0.8590 | 0.8526 | 0.8563 | 0.8576 | 0.8502 | 0.8583 | - | - | - | - | - | - |
1036
+ | 0.2087 | 600 | 0.1328 | 0.1478 | 0.8626 | 0.8557 | 0.8595 | 0.8612 | 0.8530 | 0.8620 | - | - | - | - | - | - |
1037
+ | 0.2435 | 700 | 0.1345 | 0.1451 | 0.8631 | 0.8563 | 0.8599 | 0.8618 | 0.8548 | 0.8626 | - | - | - | - | - | - |
1038
+ | 0.2783 | 800 | 0.1282 | 0.1423 | 0.8705 | 0.8625 | 0.8671 | 0.8692 | 0.8601 | 0.8698 | - | - | - | - | - | - |
1039
+ | 0.3130 | 900 | 0.1317 | 0.1416 | 0.8724 | 0.8639 | 0.8690 | 0.8714 | 0.8619 | 0.8716 | - | - | - | - | - | - |
1040
+ | 0.3478 | 1000 | 0.1295 | 0.1422 | 0.8641 | 0.8577 | 0.8617 | 0.8637 | 0.8556 | 0.8639 | - | - | - | - | - | - |
1041
+ | 0.3826 | 1100 | 0.1267 | 0.1427 | 0.8675 | 0.8603 | 0.8644 | 0.8666 | 0.8581 | 0.8671 | - | - | - | - | - | - |
1042
+ | 0.4174 | 1200 | 0.127 | 0.1417 | 0.8674 | 0.8589 | 0.8635 | 0.8664 | 0.8570 | 0.8671 | - | - | - | - | - | - |
1043
+ | 0.4522 | 1300 | 0.1292 | 0.1419 | 0.8756 | 0.8663 | 0.8711 | 0.8739 | 0.8641 | 0.8748 | - | - | - | - | - | - |
1044
+ | 0.4870 | 1400 | 0.1281 | 0.1411 | 0.8726 | 0.8646 | 0.8686 | 0.8713 | 0.8616 | 0.8721 | - | - | - | - | - | - |
1045
+ | 0.5217 | 1500 | 0.1292 | 0.1407 | 0.8738 | 0.8654 | 0.8698 | 0.8727 | 0.8617 | 0.8739 | - | - | - | - | - | - |
1046
+ | 0.5565 | 1600 | 0.1251 | 0.1419 | 0.8732 | 0.8643 | 0.8686 | 0.8720 | 0.8605 | 0.8731 | - | - | - | - | - | - |
1047
+ | 0.5913 | 1700 | 0.1288 | 0.1412 | 0.8782 | 0.8682 | 0.8731 | 0.8769 | 0.8652 | 0.8779 | - | - | - | - | - | - |
1048
+ | 0.6261 | 1800 | 0.1306 | 0.1405 | 0.8755 | 0.8664 | 0.8710 | 0.8744 | 0.8632 | 0.8751 | - | - | - | - | - | - |
1049
+ | 0.6609 | 1900 | 0.1289 | 0.1410 | 0.8739 | 0.8647 | 0.8691 | 0.8727 | 0.8614 | 0.8736 | - | - | - | - | - | - |
1050
+ | 0.6957 | 2000 | 0.1287 | 0.1403 | 0.8773 | 0.8669 | 0.8719 | 0.8758 | 0.8637 | 0.8769 | - | - | - | - | - | - |
1051
+ | 0.7304 | 2100 | 0.126 | 0.1402 | 0.8773 | 0.8675 | 0.8722 | 0.8758 | 0.8635 | 0.8772 | - | - | - | - | - | - |
1052
+ | 0.7652 | 2200 | 0.1274 | 0.1401 | 0.8799 | 0.8693 | 0.8743 | 0.8784 | 0.8652 | 0.8797 | - | - | - | - | - | - |
1053
+ | 0.8 | 2300 | 0.1234 | 0.1399 | 0.8777 | 0.8686 | 0.8729 | 0.8767 | 0.8650 | 0.8778 | - | - | - | - | - | - |
1054
+ | 0.8348 | 2400 | 0.128 | 0.1401 | 0.8769 | 0.8660 | 0.8712 | 0.8759 | 0.8621 | 0.8768 | - | - | - | - | - | - |
1055
+ | 0.8696 | 2500 | 0.1269 | 0.1403 | 0.8756 | 0.8648 | 0.8698 | 0.8742 | 0.8605 | 0.8750 | - | - | - | - | - | - |
1056
+ | 0.9043 | 2600 | 0.1243 | 0.1401 | 0.8762 | 0.8665 | 0.8711 | 0.8751 | 0.8622 | 0.8760 | - | - | - | - | - | - |
1057
+ | 0.9391 | 2700 | 0.1277 | 0.1406 | 0.8742 | 0.8649 | 0.8693 | 0.8725 | 0.8613 | 0.8738 | - | - | - | - | - | - |
1058
+ | 0.9739 | 2800 | 0.1287 | 0.1394 | 0.8789 | 0.8689 | 0.8738 | 0.8773 | 0.8648 | 0.8785 | - | - | - | - | - | - |
1059
+ | 1.0087 | 2900 | 0.1274 | 0.1397 | 0.8784 | 0.8682 | 0.8731 | 0.8769 | 0.8632 | 0.8782 | - | - | - | - | - | - |
1060
+ | 1.0435 | 3000 | 0.129 | 0.1401 | 0.8800 | 0.8693 | 0.8743 | 0.8782 | 0.8653 | 0.8795 | - | - | - | - | - | - |
1061
+ | 1.0783 | 3100 | 0.121 | 0.1408 | 0.8785 | 0.8682 | 0.8731 | 0.8769 | 0.8638 | 0.8782 | - | - | - | - | - | - |
1062
+ | 1.1130 | 3200 | 0.1249 | 0.1399 | 0.8773 | 0.8668 | 0.8722 | 0.8759 | 0.8625 | 0.8771 | - | - | - | - | - | - |
1063
+ | 1.1478 | 3300 | 0.1252 | 0.1404 | 0.8740 | 0.8643 | 0.8688 | 0.8724 | 0.8593 | 0.8737 | - | - | - | - | - | - |
1064
+ | 1.1826 | 3400 | 0.126 | 0.1398 | 0.8761 | 0.8657 | 0.8707 | 0.8745 | 0.8610 | 0.8758 | - | - | - | - | - | - |
1065
+ | 1.2174 | 3500 | 0.1279 | 0.1400 | 0.8760 | 0.8661 | 0.8708 | 0.8745 | 0.8617 | 0.8759 | - | - | - | - | - | - |
1066
+ | 1.2522 | 3600 | 0.1264 | 0.1399 | 0.8786 | 0.8684 | 0.8734 | 0.8768 | 0.8633 | 0.8783 | - | - | - | - | - | - |
1067
+ | 1.2870 | 3700 | 0.126 | 0.1395 | 0.8789 | 0.8690 | 0.8734 | 0.8773 | 0.8643 | 0.8786 | - | - | - | - | - | - |
1068
+ | 1.3217 | 3800 | 0.1234 | 0.1399 | 0.8777 | 0.8669 | 0.8723 | 0.8760 | 0.8625 | 0.8775 | - | - | - | - | - | - |
1069
+ | 1.3565 | 3900 | 0.1269 | 0.1397 | 0.8777 | 0.8671 | 0.8725 | 0.8760 | 0.8630 | 0.8773 | - | - | - | - | - | - |
1070
+ | 1.3913 | 4000 | 0.1223 | 0.1393 | 0.8806 | 0.8694 | 0.8751 | 0.8789 | 0.8654 | 0.8802 | - | - | - | - | - | - |
1071
+ | 1.4261 | 4100 | 0.1227 | 0.1399 | 0.8775 | 0.8671 | 0.8728 | 0.8764 | 0.8622 | 0.8774 | - | - | - | - | - | - |
1072
+ | 1.4609 | 4200 | 0.1263 | 0.1402 | 0.8771 | 0.8669 | 0.8724 | 0.8756 | 0.8619 | 0.8769 | - | - | - | - | - | - |
1073
+ | 1.4957 | 4300 | 0.1263 | 0.1400 | 0.8781 | 0.8674 | 0.8730 | 0.8766 | 0.8627 | 0.8778 | - | - | - | - | - | - |
1074
+ | 1.5304 | 4400 | 0.1302 | 0.1396 | 0.8778 | 0.8675 | 0.8728 | 0.8761 | 0.8628 | 0.8775 | - | - | - | - | - | - |
1075
+ | 1.5652 | 4500 | 0.1274 | 0.1393 | 0.8789 | 0.8685 | 0.8736 | 0.8770 | 0.8637 | 0.8784 | - | - | - | - | - | - |
1076
+ | 1.6 | 4600 | 0.1273 | 0.1394 | 0.8794 | 0.8683 | 0.8737 | 0.8773 | 0.8637 | 0.8789 | - | - | - | - | - | - |
1077
+ | 1.6348 | 4700 | 0.1297 | 0.1391 | 0.8822 | 0.8712 | 0.8764 | 0.8800 | 0.8666 | 0.8817 | - | - | - | - | - | - |
1078
+ | 1.6696 | 4800 | 0.1249 | 0.1392 | 0.8804 | 0.8694 | 0.8748 | 0.8785 | 0.8643 | 0.8802 | - | - | - | - | - | - |
1079
+ | 1.7043 | 4900 | 0.1286 | 0.1390 | 0.8803 | 0.8693 | 0.8746 | 0.8784 | 0.8643 | 0.8800 | - | - | - | - | - | - |
1080
+ | 1.7391 | 5000 | 0.1271 | 0.1392 | 0.8799 | 0.8697 | 0.8745 | 0.8780 | 0.8645 | 0.8795 | - | - | - | - | - | - |
1081
+ | 1.7739 | 5100 | 0.1293 | 0.1391 | 0.8803 | 0.8702 | 0.8748 | 0.8790 | 0.8648 | 0.8803 | - | - | - | - | - | - |
1082
+ | 1.8087 | 5200 | 0.1233 | 0.1391 | 0.8793 | 0.8692 | 0.8739 | 0.8777 | 0.8639 | 0.8791 | - | - | - | - | - | - |
1083
+ | 1.8435 | 5300 | 0.1239 | 0.1394 | 0.8805 | 0.8705 | 0.8748 | 0.8788 | 0.8656 | 0.8802 | - | - | - | - | - | - |
1084
+ | 1.8783 | 5400 | 0.124 | 0.1392 | 0.8795 | 0.8692 | 0.8742 | 0.8780 | 0.8640 | 0.8792 | - | - | - | - | - | - |
1085
+ | 1.9130 | 5500 | 0.1245 | 0.1390 | 0.8797 | 0.8697 | 0.8744 | 0.8782 | 0.8645 | 0.8794 | - | - | - | - | - | - |
1086
+ | 1.9478 | 5600 | 0.1257 | 0.1391 | 0.8794 | 0.8689 | 0.8741 | 0.8778 | 0.8637 | 0.8791 | - | - | - | - | - | - |
1087
+ | 1.9826 | 5700 | 0.1231 | 0.1389 | 0.8807 | 0.8708 | 0.8756 | 0.8793 | 0.8663 | 0.8804 | - | - | - | - | - | - |
1088
+ | 2.0174 | 5800 | 0.1216 | 0.1390 | 0.8781 | 0.8678 | 0.8733 | 0.8768 | 0.8630 | 0.8779 | - | - | - | - | - | - |
1089
+ | 2.0522 | 5900 | 0.1252 | 0.1387 | 0.8795 | 0.8695 | 0.8745 | 0.8784 | 0.8639 | 0.8794 | - | - | - | - | - | - |
1090
+ | 2.0870 | 6000 | 0.1242 | 0.1387 | 0.8799 | 0.8703 | 0.8749 | 0.8787 | 0.8652 | 0.8798 | - | - | - | - | - | - |
1091
+ | 2.1217 | 6100 | 0.1231 | 0.1392 | 0.8796 | 0.8702 | 0.8748 | 0.8784 | 0.8653 | 0.8795 | - | - | - | - | - | - |
1092
+ | 2.1565 | 6200 | 0.1217 | 0.1391 | 0.8797 | 0.8704 | 0.8746 | 0.8784 | 0.8655 | 0.8794 | - | - | - | - | - | - |
1093
+ | 2.1913 | 6300 | 0.1259 | 0.1389 | 0.8803 | 0.8710 | 0.8756 | 0.8789 | 0.8664 | 0.8800 | - | - | - | - | - | - |
1094
+ | 2.2261 | 6400 | 0.1262 | 0.1386 | 0.8813 | 0.8714 | 0.8762 | 0.8796 | 0.8667 | 0.8809 | - | - | - | - | - | - |
1095
+ | 2.2609 | 6500 | 0.127 | 0.1392 | 0.8793 | 0.8701 | 0.8743 | 0.8778 | 0.8652 | 0.8792 | - | - | - | - | - | - |
1096
+ | 2.2957 | 6600 | 0.1275 | 0.1391 | 0.8806 | 0.8710 | 0.8755 | 0.8788 | 0.8663 | 0.8803 | - | - | - | - | - | - |
1097
+ | 2.3304 | 6700 | 0.1228 | 0.1394 | 0.8795 | 0.8693 | 0.8741 | 0.8774 | 0.8646 | 0.8791 | - | - | - | - | - | - |
1098
+ | 2.3652 | 6800 | 0.1243 | 0.1390 | 0.8803 | 0.8700 | 0.8747 | 0.8783 | 0.8655 | 0.8797 | - | - | - | - | - | - |
1099
+ | 2.4 | 6900 | 0.1292 | 0.1389 | 0.8795 | 0.8697 | 0.8743 | 0.8778 | 0.8650 | 0.8791 | - | - | - | - | - | - |
1100
+ | 2.4348 | 7000 | 0.1238 | 0.1390 | 0.8799 | 0.8697 | 0.8744 | 0.8782 | 0.8648 | 0.8795 | - | - | - | - | - | - |
1101
+ | 2.4696 | 7100 | 0.1246 | 0.1389 | 0.8800 | 0.8695 | 0.8743 | 0.8780 | 0.8649 | 0.8795 | - | - | - | - | - | - |
1102
+ | 2.5043 | 7200 | 0.1265 | 0.1396 | 0.8802 | 0.8695 | 0.8743 | 0.8781 | 0.8647 | 0.8796 | - | - | - | - | - | - |
1103
+ | 2.5391 | 7300 | 0.1229 | 0.1390 | 0.8813 | 0.8708 | 0.8753 | 0.8796 | 0.8665 | 0.8809 | - | - | - | - | - | - |
1104
+ | 2.5739 | 7400 | 0.1244 | 0.1389 | 0.8808 | 0.8706 | 0.8749 | 0.8790 | 0.8665 | 0.8803 | - | - | - | - | - | - |
1105
+ | 2.6087 | 7500 | 0.1223 | 0.1389 | 0.8813 | 0.8709 | 0.8753 | 0.8797 | 0.8662 | 0.8807 | - | - | - | - | - | - |
1106
+ | 2.6435 | 7600 | 0.1268 | 0.1387 | 0.8810 | 0.8704 | 0.8752 | 0.8793 | 0.8659 | 0.8805 | - | - | - | - | - | - |
1107
+ | 2.6783 | 7700 | 0.1218 | 0.1387 | 0.8817 | 0.8710 | 0.8755 | 0.8798 | 0.8665 | 0.8809 | - | - | - | - | - | - |
1108
+ | 2.7130 | 7800 | 0.1225 | 0.1388 | 0.8804 | 0.8700 | 0.8745 | 0.8787 | 0.8653 | 0.8799 | - | - | - | - | - | - |
1109
+ | 2.7478 | 7900 | 0.1263 | 0.1391 | 0.8807 | 0.8703 | 0.8745 | 0.8788 | 0.8654 | 0.8801 | - | - | - | - | - | - |
1110
+ | 2.7826 | 8000 | 0.1261 | 0.1388 | 0.8804 | 0.8698 | 0.8743 | 0.8787 | 0.8652 | 0.8799 | - | - | - | - | - | - |
1111
+ | 2.8174 | 8100 | 0.1267 | 0.1386 | 0.8814 | 0.8707 | 0.8750 | 0.8795 | 0.8658 | 0.8807 | - | - | - | - | - | - |
1112
+ | 2.8522 | 8200 | 0.1236 | 0.1387 | 0.8809 | 0.8703 | 0.8747 | 0.8792 | 0.8659 | 0.8803 | - | - | - | - | - | - |
1113
+ | 2.8870 | 8300 | 0.1222 | 0.1390 | 0.8802 | 0.8696 | 0.8741 | 0.8786 | 0.8649 | 0.8799 | - | - | - | - | - | - |
1114
+ | 2.9217 | 8400 | 0.1236 | 0.1388 | 0.8807 | 0.8700 | 0.8747 | 0.8790 | 0.8653 | 0.8802 | - | - | - | - | - | - |
1115
+ | 2.9565 | 8500 | 0.1233 | 0.1389 | 0.8808 | 0.8705 | 0.8752 | 0.8791 | 0.8659 | 0.8806 | - | - | - | - | - | - |
1116
+ | 2.9913 | 8600 | 0.1262 | 0.1388 | 0.8808 | 0.8704 | 0.8750 | 0.8792 | 0.8658 | 0.8805 | - | - | - | - | - | - |
1117
+ | 3.0261 | 8700 | 0.1277 | 0.1388 | 0.8795 | 0.8690 | 0.8737 | 0.8778 | 0.8640 | 0.8791 | - | - | - | - | - | - |
1118
+ | 3.0609 | 8800 | 0.1243 | 0.1387 | 0.8809 | 0.8705 | 0.8751 | 0.8791 | 0.8656 | 0.8803 | - | - | - | - | - | - |
1119
+ | 3.0957 | 8900 | 0.1206 | 0.1387 | 0.8813 | 0.8709 | 0.8754 | 0.8796 | 0.8661 | 0.8807 | - | - | - | - | - | - |
1120
+ | 3.1304 | 9000 | 0.1217 | 0.1388 | 0.8815 | 0.8716 | 0.8758 | 0.8797 | 0.8670 | 0.8810 | - | - | - | - | - | - |
1121
+ | 3.1652 | 9100 | 0.1236 | 0.1390 | 0.8803 | 0.8702 | 0.8744 | 0.8785 | 0.8653 | 0.8798 | - | - | - | - | - | - |
1122
+ | 3.2 | 9200 | 0.1244 | 0.1389 | 0.8799 | 0.8697 | 0.8741 | 0.8783 | 0.8647 | 0.8795 | - | - | - | - | - | - |
1123
+ | 3.2348 | 9300 | 0.1247 | 0.1388 | 0.8802 | 0.8698 | 0.8743 | 0.8785 | 0.8650 | 0.8798 | - | - | - | - | - | - |
1124
+ | 3.2696 | 9400 | 0.1214 | 0.1388 | 0.8810 | 0.8710 | 0.8751 | 0.8793 | 0.8663 | 0.8806 | - | - | - | - | - | - |
1125
+ | 3.3043 | 9500 | 0.121 | 0.1386 | 0.8808 | 0.8709 | 0.8749 | 0.8791 | 0.8662 | 0.8803 | - | - | - | - | - | - |
1126
+ | 3.3391 | 9600 | 0.1205 | 0.1387 | 0.8804 | 0.8705 | 0.8746 | 0.8789 | 0.8655 | 0.8800 | - | - | - | - | - | - |
1127
+ | 3.3739 | 9700 | 0.1203 | 0.1387 | 0.8807 | 0.8708 | 0.8750 | 0.8790 | 0.8661 | 0.8802 | - | - | - | - | - | - |
1128
+ | 3.4087 | 9800 | 0.1239 | 0.1386 | 0.8811 | 0.8711 | 0.8752 | 0.8794 | 0.8663 | 0.8805 | - | - | - | - | - | - |
1129
+ | 3.4435 | 9900 | 0.1197 | 0.1387 | 0.8808 | 0.8709 | 0.8750 | 0.8792 | 0.8662 | 0.8804 | - | - | - | - | - | - |
1130
+ | 3.4783 | 10000 | 0.1252 | 0.1388 | 0.8805 | 0.8704 | 0.8746 | 0.8787 | 0.8657 | 0.8800 | - | - | - | - | - | - |
1131
+ | 3.5130 | 10100 | 0.1229 | 0.1388 | 0.8803 | 0.8703 | 0.8745 | 0.8786 | 0.8654 | 0.8799 | - | - | - | - | - | - |
1132
+ | 3.5478 | 10200 | 0.1258 | 0.1387 | 0.8805 | 0.8704 | 0.8747 | 0.8787 | 0.8653 | 0.8801 | - | - | - | - | - | - |
1133
+ | 3.5826 | 10300 | 0.1232 | 0.1387 | 0.8806 | 0.8706 | 0.8750 | 0.8790 | 0.8656 | 0.8802 | - | - | - | - | - | - |
1134
+ | 3.6174 | 10400 | 0.1286 | 0.1388 | 0.8807 | 0.8706 | 0.8749 | 0.8790 | 0.8656 | 0.8802 | - | - | - | - | - | - |
1135
+ | 3.6522 | 10500 | 0.1248 | 0.1387 | 0.8806 | 0.8706 | 0.8748 | 0.8789 | 0.8653 | 0.8802 | - | - | - | - | - | - |
1136
+ | 3.6870 | 10600 | 0.1277 | 0.1389 | 0.8800 | 0.8699 | 0.8742 | 0.8782 | 0.8647 | 0.8796 | - | - | - | - | - | - |
1137
+ | 3.7217 | 10700 | 0.1219 | 0.1388 | 0.8799 | 0.8697 | 0.8740 | 0.8780 | 0.8645 | 0.8794 | - | - | - | - | - | - |
1138
+ | 3.7565 | 10800 | 0.1269 | 0.1388 | 0.8803 | 0.8702 | 0.8745 | 0.8785 | 0.8649 | 0.8798 | - | - | - | - | - | - |
1139
+ | 3.7913 | 10900 | 0.1289 | 0.1387 | 0.8805 | 0.8703 | 0.8746 | 0.8787 | 0.8651 | 0.8800 | - | - | - | - | - | - |
1140
+ | 3.8261 | 11000 | 0.1234 | 0.1387 | 0.8806 | 0.8704 | 0.8749 | 0.8789 | 0.8653 | 0.8801 | - | - | - | - | - | - |
1141
+ | 3.8609 | 11100 | 0.1229 | 0.1387 | 0.8806 | 0.8706 | 0.8749 | 0.8788 | 0.8654 | 0.8802 | - | - | - | - | - | - |
1142
+ | 3.8957 | 11200 | 0.1266 | 0.1387 | 0.8806 | 0.8706 | 0.8749 | 0.8789 | 0.8655 | 0.8801 | - | - | - | - | - | - |
1143
+ | 3.9304 | 11300 | 0.1253 | 0.1387 | 0.8804 | 0.8704 | 0.8747 | 0.8787 | 0.8653 | 0.8800 | - | - | - | - | - | - |
1144
+ | 3.9652 | 11400 | 0.1279 | 0.1388 | 0.8804 | 0.8704 | 0.8747 | 0.8787 | 0.8653 | 0.8799 | - | - | - | - | - | - |
1145
+ | 4.0 | 11500 | 0.1195 | 0.1388 | 0.8804 | 0.8704 | 0.8747 | 0.8786 | 0.8652 | 0.8799 | 0.8623 | 0.8546 | 0.8583 | 0.8624 | 0.8507 | 0.8632 |
1146
+
1147
+ </details>
1148
+
1149
+ ### Framework Versions
1150
+ - Python: 3.9.16
1151
+ - Sentence Transformers: 3.0.0
1152
+ - Transformers: 4.42.0.dev0
1153
+ - PyTorch: 2.2.2+cu118
1154
+ - Accelerate: 0.31.0
1155
+ - Datasets: 2.19.1
1156
+ - Tokenizers: 0.19.1
1157
+
1158
+ ## Citation
1159
+
1160
+ ### BibTeX
1161
+
1162
+ #### Sentence Transformers
1163
+ ```bibtex
1164
+ @inproceedings{reimers-2019-sentence-bert,
1165
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
1166
+ author = "Reimers, Nils and Gurevych, Iryna",
1167
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
1168
+ month = "11",
1169
+ year = "2019",
1170
+ publisher = "Association for Computational Linguistics",
1171
+ url = "https://arxiv.org/abs/1908.10084",
1172
+ }
1173
+ ```
1174
+
1175
+ #### MatryoshkaLoss
1176
+ ```bibtex
1177
+ @misc{kusupati2024matryoshka,
1178
+ title={Matryoshka Representation Learning},
1179
+ 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},
1180
+ year={2024},
1181
+ eprint={2205.13147},
1182
+ archivePrefix={arXiv},
1183
+ primaryClass={cs.LG}
1184
+ }
1185
+ ```
1186
+
1187
+ #### ContrastiveLoss
1188
+ ```bibtex
1189
+ @inproceedings{hadsell2006dimensionality,
1190
+ author={Hadsell, R. and Chopra, S. and LeCun, Y.},
1191
+ booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
1192
+ title={Dimensionality Reduction by Learning an Invariant Mapping},
1193
+ year={2006},
1194
+ volume={2},
1195
+ number={},
1196
+ pages={1735-1742},
1197
+ doi={10.1109/CVPR.2006.100}
1198
+ }
1199
+ ```
1200
+
1201
+ <!--
1202
+ ## Glossary
1203
+
1204
+ *Clearly define terms in order to be accessible across audiences.*
1205
+ -->
1206
+
1207
+ <!--
1208
+ ## Model Card Authors
1209
+
1210
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
1211
+ -->
1212
+
1213
+ <!--
1214
+ ## Model Card Contact
1215
+
1216
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
1217
+ -->
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "aari1995/gbert-large-nli_mix",
3
+ "architectures": [
4
+ "JinaBertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "attn_implementation": null,
8
+ "auto_map": {
9
+ "AutoConfig": "aari1995/gbert-large-alibi--configuration_bert.JinaBertConfig",
10
+ "AutoModel": "aari1995/gbert-large-alibi--modeling_bert.JinaBertModel",
11
+ "AutoModelForMaskedLM": "aari1995/gbert-large-alibi--modeling_bert.JinaBertForMaskedLM",
12
+ "AutoModelForSequenceClassification": "aari1995/gbert-large-alibi--modeling_bert.JinaBertForSequenceClassification"
13
+ },
14
+ "classifier_dropout": null,
15
+ "emb_pooler": null,
16
+ "feed_forward_type": "original",
17
+ "hidden_act": "gelu",
18
+ "hidden_dropout_prob": 0.1,
19
+ "hidden_size": 1024,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 4096,
22
+ "layer_norm_eps": 1e-12,
23
+ "max_position_embeddings": 8192,
24
+ "model_type": "bert",
25
+ "num_attention_heads": 16,
26
+ "num_hidden_layers": 24,
27
+ "pad_token_id": 0,
28
+ "position_embedding_type": "alibi",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.42.0.dev0",
31
+ "type_vocab_size": 2,
32
+ "use_cache": true,
33
+ "vocab_size": 31102
34
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.0",
4
+ "transformers": "4.42.0.dev0",
5
+ "pytorch": "2.2.2+cu118"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:473567809c2a05e488169e164709b9813a7dd1cd34c0c9367e9d5a8cf2018ff4
3
+ size 1340890848
modules.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ }
14
+ ]
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 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "101": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "102": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "103": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "104": {
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_basic_tokenize": true,
47
+ "do_lower_case": false,
48
+ "mask_token": "[MASK]",
49
+ "max_len": 9999999999,
50
+ "max_length": 8192,
51
+ "model_max_length": 8192,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 0,
59
+ "strip_accents": false,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "longest_first",
64
+ "unk_token": "[UNK]"
65
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff