Hugging Face's logo

ve88ifz2
/
bge-base-en-v1.5-klej-dyk-v0.2

Sentence Similarity
sentence-transformers
Safetensors
English
bert
feature-extraction
dataset_size:1K<n<10K
loss:MatryoshkaLoss
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
ve88ifz2 commited on
Commit
2f989d7
•
1 Parent(s): d9a8c58

Add new SentenceTransformer model.

Browse files
Files changed (11) hide show
  1. 1_Pooling/config.json +10 -0
  2. README.md +924 -0
  3. config.json +32 -0
  4. config_sentence_transformers.json +10 -0
  5. model.safetensors +3 -0
  6. modules.json +20 -0
  7. sentence_bert_config.json +4 -0
  8. special_tokens_map.json +37 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +64 -0
  11. vocab.txt +0 -0
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
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 +1,924 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: BAAI/bge-base-en-v1.5
3
+ language:
4
+ - en
5
+ library_name: sentence-transformers
6
+ license: apache-2.0
7
+ metrics:
8
+ - cosine_accuracy@1
9
+ - cosine_accuracy@3
10
+ - cosine_accuracy@5
11
+ - cosine_accuracy@10
12
+ - cosine_precision@1
13
+ - cosine_precision@3
14
+ - cosine_precision@5
15
+ - cosine_precision@10
16
+ - cosine_recall@1
17
+ - cosine_recall@3
18
+ - cosine_recall@5
19
+ - cosine_recall@10
20
+ - cosine_ndcg@10
21
+ - cosine_mrr@10
22
+ - cosine_map@100
23
+ pipeline_tag: sentence-similarity
24
+ tags:
25
+ - sentence-transformers
26
+ - sentence-similarity
27
+ - feature-extraction
28
+ - dataset_size:1K<n<10K
29
+ - loss:MatryoshkaLoss
30
+ - loss:MultipleNegativesRankingLoss
31
+ widget:
32
+ - source_sentence: USS Conyngham (DD-58)
33
+ sentences:
34
+ - '"w jakich patrolach uczestniczył USS ""Conyngham"" (DD-58)?"'
35
+ - Jest ona najstarszą skoczkinią w kadrze norweskiej.
36
+ - kto uczył malarstwa olimpijczyka Bronisława Czecha?
37
+ - source_sentence: Danae (obraz Tycjana)
38
+ sentences:
39
+ - jakie różnice występują pomiędzy kolejnymi wersjami obrazu Tycjana Danae?
40
+ - z czego wykonana jest rzeźba Robotnik i kołchoźnica?
41
+ - z jakiego powodu zwołano synod w Whitby?
42
+ - source_sentence: dlaczego zapominamy?
43
+ sentences:
44
+ - Zamek w Haapsalu
45
+ - kto był tłumaczem języka angielskiego u Mao Zedonga?
46
+ - Najstarszy z trzech synów Hong Xiuquana; jego matką była Lai Lianying.
47
+ - source_sentence: kim był Steve Yzerman?
48
+ sentences:
49
+ - która hala ma najmniejszą widownię w NHL?
50
+ - za co krytykowany był papieski wykład ratyzboński?
51
+ - ' W 1867 oddano do użytku Kolej Warszawsko-Terespolską (całą linię).'
52
+ - source_sentence: Herkules na rozstajach
53
+ sentences:
54
+ - jak zinterpretować wymowę obrazu Herkules na rozstajach?
55
+ - Dowódcą grupy był Wiaczesław Razumowicz ps. „Chmara”.
56
+ - z jakiego powodu zwołano synod w Whitby?
57
+ model-index:
58
+ - name: bge-base-en-v1.5-klej-dyk
59
+ results:
60
+ - task:
61
+ type: information-retrieval
62
+ name: Information Retrieval
63
+ dataset:
64
+ name: dim 768
65
+ type: dim_768
66
+ metrics:
67
+ - type: cosine_accuracy@1
68
+ value: 0.17307692307692307
69
+ name: Cosine Accuracy@1
70
+ - type: cosine_accuracy@3
71
+ value: 0.46153846153846156
72
+ name: Cosine Accuracy@3
73
+ - type: cosine_accuracy@5
74
+ value: 0.6225961538461539
75
+ name: Cosine Accuracy@5
76
+ - type: cosine_accuracy@10
77
+ value: 0.7355769230769231
78
+ name: Cosine Accuracy@10
79
+ - type: cosine_precision@1
80
+ value: 0.17307692307692307
81
+ name: Cosine Precision@1
82
+ - type: cosine_precision@3
83
+ value: 0.15384615384615385
84
+ name: Cosine Precision@3
85
+ - type: cosine_precision@5
86
+ value: 0.12451923076923076
87
+ name: Cosine Precision@5
88
+ - type: cosine_precision@10
89
+ value: 0.0735576923076923
90
+ name: Cosine Precision@10
91
+ - type: cosine_recall@1
92
+ value: 0.17307692307692307
93
+ name: Cosine Recall@1
94
+ - type: cosine_recall@3
95
+ value: 0.46153846153846156
96
+ name: Cosine Recall@3
97
+ - type: cosine_recall@5
98
+ value: 0.6225961538461539
99
+ name: Cosine Recall@5
100
+ - type: cosine_recall@10
101
+ value: 0.7355769230769231
102
+ name: Cosine Recall@10
103
+ - type: cosine_ndcg@10
104
+ value: 0.4433646681639308
105
+ name: Cosine Ndcg@10
106
+ - type: cosine_mrr@10
107
+ value: 0.35053323412698395
108
+ name: Cosine Mrr@10
109
+ - type: cosine_map@100
110
+ value: 0.3573926265146405
111
+ name: Cosine Map@100
112
+ - task:
113
+ type: information-retrieval
114
+ name: Information Retrieval
115
+ dataset:
116
+ name: dim 512
117
+ type: dim_512
118
+ metrics:
119
+ - type: cosine_accuracy@1
120
+ value: 0.16826923076923078
121
+ name: Cosine Accuracy@1
122
+ - type: cosine_accuracy@3
123
+ value: 0.4519230769230769
124
+ name: Cosine Accuracy@3
125
+ - type: cosine_accuracy@5
126
+ value: 0.6009615384615384
127
+ name: Cosine Accuracy@5
128
+ - type: cosine_accuracy@10
129
+ value: 0.7091346153846154
130
+ name: Cosine Accuracy@10
131
+ - type: cosine_precision@1
132
+ value: 0.16826923076923078
133
+ name: Cosine Precision@1
134
+ - type: cosine_precision@3
135
+ value: 0.15064102564102563
136
+ name: Cosine Precision@3
137
+ - type: cosine_precision@5
138
+ value: 0.1201923076923077
139
+ name: Cosine Precision@5
140
+ - type: cosine_precision@10
141
+ value: 0.07091346153846154
142
+ name: Cosine Precision@10
143
+ - type: cosine_recall@1
144
+ value: 0.16826923076923078
145
+ name: Cosine Recall@1
146
+ - type: cosine_recall@3
147
+ value: 0.4519230769230769
148
+ name: Cosine Recall@3
149
+ - type: cosine_recall@5
150
+ value: 0.6009615384615384
151
+ name: Cosine Recall@5
152
+ - type: cosine_recall@10
153
+ value: 0.7091346153846154
154
+ name: Cosine Recall@10
155
+ - type: cosine_ndcg@10
156
+ value: 0.42955891948336516
157
+ name: Cosine Ndcg@10
158
+ - type: cosine_mrr@10
159
+ value: 0.3405992445054941
160
+ name: Cosine Mrr@10
161
+ - type: cosine_map@100
162
+ value: 0.3484580834493777
163
+ name: Cosine Map@100
164
+ - task:
165
+ type: information-retrieval
166
+ name: Information Retrieval
167
+ dataset:
168
+ name: dim 256
169
+ type: dim_256
170
+ metrics:
171
+ - type: cosine_accuracy@1
172
+ value: 0.19230769230769232
173
+ name: Cosine Accuracy@1
174
+ - type: cosine_accuracy@3
175
+ value: 0.4543269230769231
176
+ name: Cosine Accuracy@3
177
+ - type: cosine_accuracy@5
178
+ value: 0.5913461538461539
179
+ name: Cosine Accuracy@5
180
+ - type: cosine_accuracy@10
181
+ value: 0.6899038461538461
182
+ name: Cosine Accuracy@10
183
+ - type: cosine_precision@1
184
+ value: 0.19230769230769232
185
+ name: Cosine Precision@1
186
+ - type: cosine_precision@3
187
+ value: 0.15144230769230768
188
+ name: Cosine Precision@3
189
+ - type: cosine_precision@5
190
+ value: 0.11826923076923078
191
+ name: Cosine Precision@5
192
+ - type: cosine_precision@10
193
+ value: 0.0689903846153846
194
+ name: Cosine Precision@10
195
+ - type: cosine_recall@1
196
+ value: 0.19230769230769232
197
+ name: Cosine Recall@1
198
+ - type: cosine_recall@3
199
+ value: 0.4543269230769231
200
+ name: Cosine Recall@3
201
+ - type: cosine_recall@5
202
+ value: 0.5913461538461539
203
+ name: Cosine Recall@5
204
+ - type: cosine_recall@10
205
+ value: 0.6899038461538461
206
+ name: Cosine Recall@10
207
+ - type: cosine_ndcg@10
208
+ value: 0.4311008111471328
209
+ name: Cosine Ndcg@10
210
+ - type: cosine_mrr@10
211
+ value: 0.3488247863247859
212
+ name: Cosine Mrr@10
213
+ - type: cosine_map@100
214
+ value: 0.3560982492053804
215
+ name: Cosine Map@100
216
+ - task:
217
+ type: information-retrieval
218
+ name: Information Retrieval
219
+ dataset:
220
+ name: dim 128
221
+ type: dim_128
222
+ metrics:
223
+ - type: cosine_accuracy@1
224
+ value: 0.16346153846153846
225
+ name: Cosine Accuracy@1
226
+ - type: cosine_accuracy@3
227
+ value: 0.41586538461538464
228
+ name: Cosine Accuracy@3
229
+ - type: cosine_accuracy@5
230
+ value: 0.5168269230769231
231
+ name: Cosine Accuracy@5
232
+ - type: cosine_accuracy@10
233
+ value: 0.5985576923076923
234
+ name: Cosine Accuracy@10
235
+ - type: cosine_precision@1
236
+ value: 0.16346153846153846
237
+ name: Cosine Precision@1
238
+ - type: cosine_precision@3
239
+ value: 0.13862179487179488
240
+ name: Cosine Precision@3
241
+ - type: cosine_precision@5
242
+ value: 0.10336538461538461
243
+ name: Cosine Precision@5
244
+ - type: cosine_precision@10
245
+ value: 0.059855769230769226
246
+ name: Cosine Precision@10
247
+ - type: cosine_recall@1
248
+ value: 0.16346153846153846
249
+ name: Cosine Recall@1
250
+ - type: cosine_recall@3
251
+ value: 0.41586538461538464
252
+ name: Cosine Recall@3
253
+ - type: cosine_recall@5
254
+ value: 0.5168269230769231
255
+ name: Cosine Recall@5
256
+ - type: cosine_recall@10
257
+ value: 0.5985576923076923
258
+ name: Cosine Recall@10
259
+ - type: cosine_ndcg@10
260
+ value: 0.37641559536404157
261
+ name: Cosine Ndcg@10
262
+ - type: cosine_mrr@10
263
+ value: 0.3052140567765567
264
+ name: Cosine Mrr@10
265
+ - type: cosine_map@100
266
+ value: 0.3151839890893904
267
+ name: Cosine Map@100
268
+ - task:
269
+ type: information-retrieval
270
+ name: Information Retrieval
271
+ dataset:
272
+ name: dim 64
273
+ type: dim_64
274
+ metrics:
275
+ - type: cosine_accuracy@1
276
+ value: 0.1658653846153846
277
+ name: Cosine Accuracy@1
278
+ - type: cosine_accuracy@3
279
+ value: 0.35096153846153844
280
+ name: Cosine Accuracy@3
281
+ - type: cosine_accuracy@5
282
+ value: 0.43990384615384615
283
+ name: Cosine Accuracy@5
284
+ - type: cosine_accuracy@10
285
+ value: 0.5288461538461539
286
+ name: Cosine Accuracy@10
287
+ - type: cosine_precision@1
288
+ value: 0.1658653846153846
289
+ name: Cosine Precision@1
290
+ - type: cosine_precision@3
291
+ value: 0.11698717948717949
292
+ name: Cosine Precision@3
293
+ - type: cosine_precision@5
294
+ value: 0.08798076923076924
295
+ name: Cosine Precision@5
296
+ - type: cosine_precision@10
297
+ value: 0.052884615384615384
298
+ name: Cosine Precision@10
299
+ - type: cosine_recall@1
300
+ value: 0.1658653846153846
301
+ name: Cosine Recall@1
302
+ - type: cosine_recall@3
303
+ value: 0.35096153846153844
304
+ name: Cosine Recall@3
305
+ - type: cosine_recall@5
306
+ value: 0.43990384615384615
307
+ name: Cosine Recall@5
308
+ - type: cosine_recall@10
309
+ value: 0.5288461538461539
310
+ name: Cosine Recall@10
311
+ - type: cosine_ndcg@10
312
+ value: 0.33823482580826353
313
+ name: Cosine Ndcg@10
314
+ - type: cosine_mrr@10
315
+ value: 0.27800194597069605
316
+ name: Cosine Mrr@10
317
+ - type: cosine_map@100
318
+ value: 0.2876731521968676
319
+ name: Cosine Map@100
320
+ ---
321
+
322
+ # bge-base-en-v1.5-klej-dyk
323
+
324
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5). It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
325
+
326
+ ## Model Details
327
+
328
+ ### Model Description
329
+ - **Model Type:** Sentence Transformer
330
+ - **Base model:** [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) <!-- at revision a5beb1e3e68b9ab74eb54cfd186867f64f240e1a -->
331
+ - **Maximum Sequence Length:** 512 tokens
332
+ - **Output Dimensionality:** 768 tokens
333
+ - **Similarity Function:** Cosine Similarity
334
+ <!-- - **Training Dataset:** Unknown -->
335
+ - **Language:** en
336
+ - **License:** apache-2.0
337
+
338
+ ### Model Sources
339
+
340
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
341
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
342
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
343
+
344
+ ### Full Model Architecture
345
+
346
+ ```
347
+ SentenceTransformer(
348
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel
349
+ (1): Pooling({'word_embedding_dimension': 768, '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})
350
+ (2): Normalize()
351
+ )
352
+ ```
353
+
354
+ ## Usage
355
+
356
+ ### Direct Usage (Sentence Transformers)
357
+
358
+ First install the Sentence Transformers library:
359
+
360
+ ```bash
361
+ pip install -U sentence-transformers
362
+ ```
363
+
364
+ Then you can load this model and run inference.
365
+ ```python
366
+ from sentence_transformers import SentenceTransformer
367
+
368
+ # Download from the 🤗 Hub
369
+ model = SentenceTransformer("sentence_transformers_model_id")
370
+ # Run inference
371
+ sentences = [
372
+ 'Herkules na rozstajach',
373
+ 'jak zinterpretować wymowę obrazu Herkules na rozstajach?',
374
+ 'Dowódcą grupy był Wiaczesław Razumowicz ps. „Chmara”.',
375
+ ]
376
+ embeddings = model.encode(sentences)
377
+ print(embeddings.shape)
378
+ # [3, 768]
379
+
380
+ # Get the similarity scores for the embeddings
381
+ similarities = model.similarity(embeddings, embeddings)
382
+ print(similarities.shape)
383
+ # [3, 3]
384
+ ```
385
+
386
+ <!--
387
+ ### Direct Usage (Transformers)
388
+
389
+ <details><summary>Click to see the direct usage in Transformers</summary>
390
+
391
+ </details>
392
+ -->
393
+
394
+ <!--
395
+ ### Downstream Usage (Sentence Transformers)
396
+
397
+ You can finetune this model on your own dataset.
398
+
399
+ <details><summary>Click to expand</summary>
400
+
401
+ </details>
402
+ -->
403
+
404
+ <!--
405
+ ### Out-of-Scope Use
406
+
407
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
408
+ -->
409
+
410
+ ## Evaluation
411
+
412
+ ### Metrics
413
+
414
+ #### Information Retrieval
415
+ * Dataset: `dim_768`
416
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
417
+
418
+ | Metric | Value |
419
+ |:--------------------|:-----------|
420
+ | cosine_accuracy@1 | 0.1731 |
421
+ | cosine_accuracy@3 | 0.4615 |
422
+ | cosine_accuracy@5 | 0.6226 |
423
+ | cosine_accuracy@10 | 0.7356 |
424
+ | cosine_precision@1 | 0.1731 |
425
+ | cosine_precision@3 | 0.1538 |
426
+ | cosine_precision@5 | 0.1245 |
427
+ | cosine_precision@10 | 0.0736 |
428
+ | cosine_recall@1 | 0.1731 |
429
+ | cosine_recall@3 | 0.4615 |
430
+ | cosine_recall@5 | 0.6226 |
431
+ | cosine_recall@10 | 0.7356 |
432
+ | cosine_ndcg@10 | 0.4434 |
433
+ | cosine_mrr@10 | 0.3505 |
434
+ | **cosine_map@100** | **0.3574** |
435
+
436
+ #### Information Retrieval
437
+ * Dataset: `dim_512`
438
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
439
+
440
+ | Metric | Value |
441
+ |:--------------------|:-----------|
442
+ | cosine_accuracy@1 | 0.1683 |
443
+ | cosine_accuracy@3 | 0.4519 |
444
+ | cosine_accuracy@5 | 0.601 |
445
+ | cosine_accuracy@10 | 0.7091 |
446
+ | cosine_precision@1 | 0.1683 |
447
+ | cosine_precision@3 | 0.1506 |
448
+ | cosine_precision@5 | 0.1202 |
449
+ | cosine_precision@10 | 0.0709 |
450
+ | cosine_recall@1 | 0.1683 |
451
+ | cosine_recall@3 | 0.4519 |
452
+ | cosine_recall@5 | 0.601 |
453
+ | cosine_recall@10 | 0.7091 |
454
+ | cosine_ndcg@10 | 0.4296 |
455
+ | cosine_mrr@10 | 0.3406 |
456
+ | **cosine_map@100** | **0.3485** |
457
+
458
+ #### Information Retrieval
459
+ * Dataset: `dim_256`
460
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
461
+
462
+ | Metric | Value |
463
+ |:--------------------|:-----------|
464
+ | cosine_accuracy@1 | 0.1923 |
465
+ | cosine_accuracy@3 | 0.4543 |
466
+ | cosine_accuracy@5 | 0.5913 |
467
+ | cosine_accuracy@10 | 0.6899 |
468
+ | cosine_precision@1 | 0.1923 |
469
+ | cosine_precision@3 | 0.1514 |
470
+ | cosine_precision@5 | 0.1183 |
471
+ | cosine_precision@10 | 0.069 |
472
+ | cosine_recall@1 | 0.1923 |
473
+ | cosine_recall@3 | 0.4543 |
474
+ | cosine_recall@5 | 0.5913 |
475
+ | cosine_recall@10 | 0.6899 |
476
+ | cosine_ndcg@10 | 0.4311 |
477
+ | cosine_mrr@10 | 0.3488 |
478
+ | **cosine_map@100** | **0.3561** |
479
+
480
+ #### Information Retrieval
481
+ * Dataset: `dim_128`
482
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
483
+
484
+ | Metric | Value |
485
+ |:--------------------|:-----------|
486
+ | cosine_accuracy@1 | 0.1635 |
487
+ | cosine_accuracy@3 | 0.4159 |
488
+ | cosine_accuracy@5 | 0.5168 |
489
+ | cosine_accuracy@10 | 0.5986 |
490
+ | cosine_precision@1 | 0.1635 |
491
+ | cosine_precision@3 | 0.1386 |
492
+ | cosine_precision@5 | 0.1034 |
493
+ | cosine_precision@10 | 0.0599 |
494
+ | cosine_recall@1 | 0.1635 |
495
+ | cosine_recall@3 | 0.4159 |
496
+ | cosine_recall@5 | 0.5168 |
497
+ | cosine_recall@10 | 0.5986 |
498
+ | cosine_ndcg@10 | 0.3764 |
499
+ | cosine_mrr@10 | 0.3052 |
500
+ | **cosine_map@100** | **0.3152** |
501
+
502
+ #### Information Retrieval
503
+ * Dataset: `dim_64`
504
+ * Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
505
+
506
+ | Metric | Value |
507
+ |:--------------------|:-----------|
508
+ | cosine_accuracy@1 | 0.1659 |
509
+ | cosine_accuracy@3 | 0.351 |
510
+ | cosine_accuracy@5 | 0.4399 |
511
+ | cosine_accuracy@10 | 0.5288 |
512
+ | cosine_precision@1 | 0.1659 |
513
+ | cosine_precision@3 | 0.117 |
514
+ | cosine_precision@5 | 0.088 |
515
+ | cosine_precision@10 | 0.0529 |
516
+ | cosine_recall@1 | 0.1659 |
517
+ | cosine_recall@3 | 0.351 |
518
+ | cosine_recall@5 | 0.4399 |
519
+ | cosine_recall@10 | 0.5288 |
520
+ | cosine_ndcg@10 | 0.3382 |
521
+ | cosine_mrr@10 | 0.278 |
522
+ | **cosine_map@100** | **0.2877** |
523
+
524
+ <!--
525
+ ## Bias, Risks and Limitations
526
+
527
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
528
+ -->
529
+
530
+ <!--
531
+ ### Recommendations
532
+
533
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
534
+ -->
535
+
536
+ ## Training Details
537
+
538
+ ### Training Dataset
539
+
540
+ #### Unnamed Dataset
541
+
542
+
543
+ * Size: 3,738 training samples
544
+ * Columns: <code>positive</code> and <code>anchor</code>
545
+ * Approximate statistics based on the first 1000 samples:
546
+ | | positive | anchor |
547
+ |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
548
+ | type | string | string |
549
+ | details | <ul><li>min: 6 tokens</li><li>mean: 90.01 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 10 tokens</li><li>mean: 30.82 tokens</li><li>max: 76 tokens</li></ul> |
550
+ * Samples:
551
+ | positive | anchor |
552
+ |:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------|
553
+ | <code>Londyńska premiera w Ambassadors Theatre na londyńskim West Endzie miała miejsce 25 listopada 1952 roku, a przedstawione grane jest do dziś (od 1974 r.) w sąsiednim St Martin's Theatre. W Polsce była wystawiana m.in. w Teatrze Nowym w Zabrzu.</code> | <code>w którym londyńskim muzeum wystawiana była instalacja My Bed?</code> |
554
+ | <code>Theridion grallator osiąga długość 5 mm. U niektórych postaci na żółtym odwłoku występuje wzór przypominający uśmiechniętą lub śmiejącą się twarz klowna.</code> | <code>które pająki noszą na grzbiecie wzór przypominający uśmiechniętego klauna?</code> |
555
+ | <code>W 1998 w wyniku sporów o wytyczenie granicy między dwoma państwami wybuchła wojna erytrejsko-etiopska. Zakończyła się porozumieniem zawartym w Algierze 12 grudnia 2000. Od tego czasu strefa graniczna jest patrolowana przez siły pokojowe ONZ.</code> | <code>jakie były skutki wojny erytrejsko-etiopskiej?</code> |
556
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
557
+ ```json
558
+ {
559
+ "loss": "MultipleNegativesRankingLoss",
560
+ "matryoshka_dims": [
561
+ 768,
562
+ 512,
563
+ 256,
564
+ 128,
565
+ 64
566
+ ],
567
+ "matryoshka_weights": [
568
+ 1,
569
+ 1,
570
+ 1,
571
+ 1,
572
+ 1
573
+ ],
574
+ "n_dims_per_step": -1
575
+ }
576
+ ```
577
+
578
+ ### Training Hyperparameters
579
+ #### Non-Default Hyperparameters
580
+
581
+ - `eval_strategy`: epoch
582
+ - `per_device_train_batch_size`: 16
583
+ - `per_device_eval_batch_size`: 16
584
+ - `gradient_accumulation_steps`: 16
585
+ - `learning_rate`: 2e-05
586
+ - `num_train_epochs`: 10
587
+ - `lr_scheduler_type`: cosine
588
+ - `warmup_ratio`: 0.1
589
+ - `bf16`: True
590
+ - `tf32`: True
591
+ - `load_best_model_at_end`: True
592
+ - `optim`: adamw_torch_fused
593
+ - `batch_sampler`: no_duplicates
594
+
595
+ #### All Hyperparameters
596
+ <details><summary>Click to expand</summary>
597
+
598
+ - `overwrite_output_dir`: False
599
+ - `do_predict`: False
600
+ - `eval_strategy`: epoch
601
+ - `prediction_loss_only`: True
602
+ - `per_device_train_batch_size`: 16
603
+ - `per_device_eval_batch_size`: 16
604
+ - `per_gpu_train_batch_size`: None
605
+ - `per_gpu_eval_batch_size`: None
606
+ - `gradient_accumulation_steps`: 16
607
+ - `eval_accumulation_steps`: None
608
+ - `learning_rate`: 2e-05
609
+ - `weight_decay`: 0.0
610
+ - `adam_beta1`: 0.9
611
+ - `adam_beta2`: 0.999
612
+ - `adam_epsilon`: 1e-08
613
+ - `max_grad_norm`: 1.0
614
+ - `num_train_epochs`: 10
615
+ - `max_steps`: -1
616
+ - `lr_scheduler_type`: cosine
617
+ - `lr_scheduler_kwargs`: {}
618
+ - `warmup_ratio`: 0.1
619
+ - `warmup_steps`: 0
620
+ - `log_level`: passive
621
+ - `log_level_replica`: warning
622
+ - `log_on_each_node`: True
623
+ - `logging_nan_inf_filter`: True
624
+ - `save_safetensors`: True
625
+ - `save_on_each_node`: False
626
+ - `save_only_model`: False
627
+ - `restore_callback_states_from_checkpoint`: False
628
+ - `no_cuda`: False
629
+ - `use_cpu`: False
630
+ - `use_mps_device`: False
631
+ - `seed`: 42
632
+ - `data_seed`: None
633
+ - `jit_mode_eval`: False
634
+ - `use_ipex`: False
635
+ - `bf16`: True
636
+ - `fp16`: False
637
+ - `fp16_opt_level`: O1
638
+ - `half_precision_backend`: auto
639
+ - `bf16_full_eval`: False
640
+ - `fp16_full_eval`: False
641
+ - `tf32`: True
642
+ - `local_rank`: 0
643
+ - `ddp_backend`: None
644
+ - `tpu_num_cores`: None
645
+ - `tpu_metrics_debug`: False
646
+ - `debug`: []
647
+ - `dataloader_drop_last`: False
648
+ - `dataloader_num_workers`: 0
649
+ - `dataloader_prefetch_factor`: None
650
+ - `past_index`: -1
651
+ - `disable_tqdm`: False
652
+ - `remove_unused_columns`: True
653
+ - `label_names`: None
654
+ - `load_best_model_at_end`: True
655
+ - `ignore_data_skip`: False
656
+ - `fsdp`: []
657
+ - `fsdp_min_num_params`: 0
658
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
659
+ - `fsdp_transformer_layer_cls_to_wrap`: None
660
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
661
+ - `deepspeed`: None
662
+ - `label_smoothing_factor`: 0.0
663
+ - `optim`: adamw_torch_fused
664
+ - `optim_args`: None
665
+ - `adafactor`: False
666
+ - `group_by_length`: False
667
+ - `length_column_name`: length
668
+ - `ddp_find_unused_parameters`: None
669
+ - `ddp_bucket_cap_mb`: None
670
+ - `ddp_broadcast_buffers`: False
671
+ - `dataloader_pin_memory`: True
672
+ - `dataloader_persistent_workers`: False
673
+ - `skip_memory_metrics`: True
674
+ - `use_legacy_prediction_loop`: False
675
+ - `push_to_hub`: False
676
+ - `resume_from_checkpoint`: None
677
+ - `hub_model_id`: None
678
+ - `hub_strategy`: every_save
679
+ - `hub_private_repo`: False
680
+ - `hub_always_push`: False
681
+ - `gradient_checkpointing`: False
682
+ - `gradient_checkpointing_kwargs`: None
683
+ - `include_inputs_for_metrics`: False
684
+ - `eval_do_concat_batches`: True
685
+ - `fp16_backend`: auto
686
+ - `push_to_hub_model_id`: None
687
+ - `push_to_hub_organization`: None
688
+ - `mp_parameters`:
689
+ - `auto_find_batch_size`: False
690
+ - `full_determinism`: False
691
+ - `torchdynamo`: None
692
+ - `ray_scope`: last
693
+ - `ddp_timeout`: 1800
694
+ - `torch_compile`: False
695
+ - `torch_compile_backend`: None
696
+ - `torch_compile_mode`: None
697
+ - `dispatch_batches`: None
698
+ - `split_batches`: None
699
+ - `include_tokens_per_second`: False
700
+ - `include_num_input_tokens_seen`: False
701
+ - `neftune_noise_alpha`: None
702
+ - `optim_target_modules`: None
703
+ - `batch_eval_metrics`: False
704
+ - `batch_sampler`: no_duplicates
705
+ - `multi_dataset_batch_sampler`: proportional
706
+
707
+ </details>
708
+
709
+ ### Training Logs
710
+ <details><summary>Click to expand</summary>
711
+
712
+ | Epoch | Step | Training Loss | dim_128_cosine_map@100 | dim_256_cosine_map@100 | dim_512_cosine_map@100 | dim_64_cosine_map@100 | dim_768_cosine_map@100 |
713
+ |:----------:|:------:|:-------------:|:----------------------:|:----------------------:|:----------------------:|:---------------------:|:----------------------:|
714
+ | 0.0684 | 1 | 7.2706 | - | - | - | - | - |
715
+ | 0.1368 | 2 | 8.2776 | - | - | - | - | - |
716
+ | 0.2051 | 3 | 7.1399 | - | - | - | - | - |
717
+ | 0.2735 | 4 | 6.6905 | - | - | - | - | - |
718
+ | 0.3419 | 5 | 6.735 | - | - | - | - | - |
719
+ | 0.4103 | 6 | 7.0537 | - | - | - | - | - |
720
+ | 0.4786 | 7 | 6.871 | - | - | - | - | - |
721
+ | 0.5470 | 8 | 6.7277 | - | - | - | - | - |
722
+ | 0.6154 | 9 | 5.9853 | - | - | - | - | - |
723
+ | 0.6838 | 10 | 6.0518 | - | - | - | - | - |
724
+ | 0.7521 | 11 | 5.8291 | - | - | - | - | - |
725
+ | 0.8205 | 12 | 5.0064 | - | - | - | - | - |
726
+ | 0.8889 | 13 | 4.8572 | - | - | - | - | - |
727
+ | 0.9573 | 14 | 5.1899 | 0.2812 | 0.3335 | 0.3486 | 0.2115 | 0.3639 |
728
+ | 1.0256 | 15 | 4.2996 | - | - | - | - | - |
729
+ | 1.0940 | 16 | 4.1475 | - | - | - | - | - |
730
+ | 1.1624 | 17 | 4.6174 | - | - | - | - | - |
731
+ | 1.2308 | 18 | 4.394 | - | - | - | - | - |
732
+ | 1.2991 | 19 | 4.0255 | - | - | - | - | - |
733
+ | 1.3675 | 20 | 3.9722 | - | - | - | - | - |
734
+ | 1.4359 | 21 | 3.9509 | - | - | - | - | - |
735
+ | 1.5043 | 22 | 3.7674 | - | - | - | - | - |
736
+ | 1.5726 | 23 | 3.7572 | - | - | - | - | - |
737
+ | 1.6410 | 24 | 3.9463 | - | - | - | - | - |
738
+ | 1.7094 | 25 | 3.7151 | - | - | - | - | - |
739
+ | 1.7778 | 26 | 3.7771 | - | - | - | - | - |
740
+ | 1.8462 | 27 | 3.5228 | - | - | - | - | - |
741
+ | 1.9145 | 28 | 2.7906 | - | - | - | - | - |
742
+ | 1.9829 | 29 | 3.4555 | 0.3164 | 0.3529 | 0.3641 | 0.2636 | 0.3681 |
743
+ | 2.0513 | 30 | 2.737 | - | - | - | - | - |
744
+ | 2.1197 | 31 | 3.1976 | - | - | - | - | - |
745
+ | 2.1880 | 32 | 3.1363 | - | - | - | - | - |
746
+ | 2.2564 | 33 | 2.9706 | - | - | - | - | - |
747
+ | 2.3248 | 34 | 2.9629 | - | - | - | - | - |
748
+ | 2.3932 | 35 | 2.7226 | - | - | - | - | - |
749
+ | 2.4615 | 36 | 2.4378 | - | - | - | - | - |
750
+ | 2.5299 | 37 | 2.7201 | - | - | - | - | - |
751
+ | 2.5983 | 38 | 2.6802 | - | - | - | - | - |
752
+ | 2.6667 | 39 | 3.1613 | - | - | - | - | - |
753
+ | 2.7350 | 40 | 2.9344 | - | - | - | - | - |
754
+ | 2.8034 | 41 | 2.5254 | - | - | - | - | - |
755
+ | 2.8718 | 42 | 2.5617 | - | - | - | - | - |
756
+ | 2.9402 | 43 | 2.459 | 0.3197 | 0.3571 | 0.3640 | 0.2739 | 0.3733 |
757
+ | 3.0085 | 44 | 2.3785 | - | - | - | - | - |
758
+ | 3.0769 | 45 | 1.9408 | - | - | - | - | - |
759
+ | 3.1453 | 46 | 2.7095 | - | - | - | - | - |
760
+ | 3.2137 | 47 | 2.4774 | - | - | - | - | - |
761
+ | 3.2821 | 48 | 2.2178 | - | - | - | - | - |
762
+ | 3.3504 | 49 | 2.0884 | - | - | - | - | - |
763
+ | 3.4188 | 50 | 2.1044 | - | - | - | - | - |
764
+ | 3.4872 | 51 | 2.1504 | - | - | - | - | - |
765
+ | 3.5556 | 52 | 2.1177 | - | - | - | - | - |
766
+ | 3.6239 | 53 | 2.2283 | - | - | - | - | - |
767
+ | 3.6923 | 54 | 2.3964 | - | - | - | - | - |
768
+ | 3.7607 | 55 | 2.0972 | - | - | - | - | - |
769
+ | 3.8291 | 56 | 2.0961 | - | - | - | - | - |
770
+ | 3.8974 | 57 | 1.783 | - | - | - | - | - |
771
+ | **3.9658** | **58** | **2.1031** | **0.3246** | **0.3533** | **0.3603** | **0.2829** | **0.3687** |
772
+ | 4.0342 | 59 | 1.6699 | - | - | - | - | - |
773
+ | 4.1026 | 60 | 1.6675 | - | - | - | - | - |
774
+ | 4.1709 | 61 | 2.1672 | - | - | - | - | - |
775
+ | 4.2393 | 62 | 1.8881 | - | - | - | - | - |
776
+ | 4.3077 | 63 | 1.701 | - | - | - | - | - |
777
+ | 4.3761 | 64 | 1.9154 | - | - | - | - | - |
778
+ | 4.4444 | 65 | 1.4549 | - | - | - | - | - |
779
+ | 4.5128 | 66 | 1.5444 | - | - | - | - | - |
780
+ | 4.5812 | 67 | 1.8352 | - | - | - | - | - |
781
+ | 4.6496 | 68 | 1.7908 | - | - | - | - | - |
782
+ | 4.7179 | 69 | 1.6876 | - | - | - | - | - |
783
+ | 4.7863 | 70 | 1.7366 | - | - | - | - | - |
784
+ | 4.8547 | 71 | 1.8689 | - | - | - | - | - |
785
+ | 4.9231 | 72 | 1.4676 | - | - | - | - | - |
786
+ | 4.9915 | 73 | 1.5045 | 0.3170 | 0.3538 | 0.3606 | 0.2829 | 0.3675 |
787
+ | 5.0598 | 74 | 1.2155 | - | - | - | - | - |
788
+ | 5.1282 | 75 | 1.4365 | - | - | - | - | - |
789
+ | 5.1966 | 76 | 1.7451 | - | - | - | - | - |
790
+ | 5.2650 | 77 | 1.4537 | - | - | - | - | - |
791
+ | 5.3333 | 78 | 1.3813 | - | - | - | - | - |
792
+ | 5.4017 | 79 | 1.4035 | - | - | - | - | - |
793
+ | 5.4701 | 80 | 1.3912 | - | - | - | - | - |
794
+ | 5.5385 | 81 | 1.3286 | - | - | - | - | - |
795
+ | 5.6068 | 82 | 1.5153 | - | - | - | - | - |
796
+ | 5.6752 | 83 | 1.6745 | - | - | - | - | - |
797
+ | 5.7436 | 84 | 1.4323 | - | - | - | - | - |
798
+ | 5.8120 | 85 | 1.5299 | - | - | - | - | - |
799
+ | 5.8803 | 86 | 1.488 | - | - | - | - | - |
800
+ | 5.9487 | 87 | 1.5195 | 0.3206 | 0.3556 | 0.3530 | 0.2878 | 0.3605 |
801
+ | 6.0171 | 88 | 1.2999 | - | - | - | - | - |
802
+ | 6.0855 | 89 | 1.1511 | - | - | - | - | - |
803
+ | 6.1538 | 90 | 1.552 | - | - | - | - | - |
804
+ | 6.2222 | 91 | 1.35 | - | - | - | - | - |
805
+ | 6.2906 | 92 | 1.218 | - | - | - | - | - |
806
+ | 6.3590 | 93 | 1.1712 | - | - | - | - | - |
807
+ | 6.4274 | 94 | 1.3381 | - | - | - | - | - |
808
+ | 6.4957 | 95 | 1.1716 | - | - | - | - | - |
809
+ | 6.5641 | 96 | 1.2117 | - | - | - | - | - |
810
+ | 6.6325 | 97 | 1.5349 | - | - | - | - | - |
811
+ | 6.7009 | 98 | 1.4564 | - | - | - | - | - |
812
+ | 6.7692 | 99 | 1.3541 | - | - | - | - | - |
813
+ | 6.8376 | 100 | 1.2468 | - | - | - | - | - |
814
+ | 6.9060 | 101 | 1.1519 | - | - | - | - | - |
815
+ | 6.9744 | 102 | 1.2421 | 0.3150 | 0.3555 | 0.3501 | 0.2858 | 0.3575 |
816
+ | 7.0427 | 103 | 1.0096 | - | - | - | - | - |
817
+ | 7.1111 | 104 | 1.1405 | - | - | - | - | - |
818
+ | 7.1795 | 105 | 1.2958 | - | - | - | - | - |
819
+ | 7.2479 | 106 | 1.35 | - | - | - | - | - |
820
+ | 7.3162 | 107 | 1.1291 | - | - | - | - | - |
821
+ | 7.3846 | 108 | 0.9968 | - | - | - | - | - |
822
+ | 7.4530 | 109 | 1.0454 | - | - | - | - | - |
823
+ | 7.5214 | 110 | 1.102 | - | - | - | - | - |
824
+ | 7.5897 | 111 | 1.1328 | - | - | - | - | - |
825
+ | 7.6581 | 112 | 1.5988 | - | - | - | - | - |
826
+ | 7.7265 | 113 | 1.2992 | - | - | - | - | - |
827
+ | 7.7949 | 114 | 1.2572 | - | - | - | - | - |
828
+ | 7.8632 | 115 | 1.1414 | - | - | - | - | - |
829
+ | 7.9316 | 116 | 1.1432 | - | - | - | - | - |
830
+ | 8.0 | 117 | 1.1181 | 0.3154 | 0.3545 | 0.3509 | 0.2884 | 0.3578 |
831
+ | 8.0684 | 118 | 0.9365 | - | - | - | - | - |
832
+ | 8.1368 | 119 | 1.3286 | - | - | - | - | - |
833
+ | 8.2051 | 120 | 1.3711 | - | - | - | - | - |
834
+ | 8.2735 | 121 | 1.2001 | - | - | - | - | - |
835
+ | 8.3419 | 122 | 1.165 | - | - | - | - | - |
836
+ | 8.4103 | 123 | 1.0575 | - | - | - | - | - |
837
+ | 8.4786 | 124 | 1.105 | - | - | - | - | - |
838
+ | 8.5470 | 125 | 1.077 | - | - | - | - | - |
839
+ | 8.6154 | 126 | 1.2217 | - | - | - | - | - |
840
+ | 8.6838 | 127 | 1.3254 | - | - | - | - | - |
841
+ | 8.7521 | 128 | 1.2165 | - | - | - | - | - |
842
+ | 8.8205 | 129 | 1.3021 | - | - | - | - | - |
843
+ | 8.8889 | 130 | 1.0927 | - | - | - | - | - |
844
+ | 8.9573 | 131 | 1.3961 | 0.3150 | 0.3540 | 0.3490 | 0.2882 | 0.3588 |
845
+ | 9.0256 | 132 | 1.0779 | - | - | - | - | - |
846
+ | 9.0940 | 133 | 0.901 | - | - | - | - | - |
847
+ | 9.1624 | 134 | 1.313 | - | - | - | - | - |
848
+ | 9.2308 | 135 | 1.1409 | - | - | - | - | - |
849
+ | 9.2991 | 136 | 1.1635 | - | - | - | - | - |
850
+ | 9.3675 | 137 | 1.0244 | - | - | - | - | - |
851
+ | 9.4359 | 138 | 1.0576 | - | - | - | - | - |
852
+ | 9.5043 | 139 | 1.0101 | - | - | - | - | - |
853
+ | 9.5726 | 140 | 1.1516 | 0.3152 | 0.3561 | 0.3485 | 0.2877 | 0.3574 |
854
+
855
+ * The bold row denotes the saved checkpoint.
856
+ </details>
857
+
858
+ ### Framework Versions
859
+ - Python: 3.12.2
860
+ - Sentence Transformers: 3.0.0
861
+ - Transformers: 4.41.2
862
+ - PyTorch: 2.3.1
863
+ - Accelerate: 0.27.2
864
+ - Datasets: 2.19.1
865
+ - Tokenizers: 0.19.1
866
+
867
+ ## Citation
868
+
869
+ ### BibTeX
870
+
871
+ #### Sentence Transformers
872
+ ```bibtex
873
+ @inproceedings{reimers-2019-sentence-bert,
874
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
875
+ author = "Reimers, Nils and Gurevych, Iryna",
876
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
877
+ month = "11",
878
+ year = "2019",
879
+ publisher = "Association for Computational Linguistics",
880
+ url = "https://arxiv.org/abs/1908.10084",
881
+ }
882
+ ```
883
+
884
+ #### MatryoshkaLoss
885
+ ```bibtex
886
+ @misc{kusupati2024matryoshka,
887
+ title={Matryoshka Representation Learning},
888
+ 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},
889
+ year={2024},
890
+ eprint={2205.13147},
891
+ archivePrefix={arXiv},
892
+ primaryClass={cs.LG}
893
+ }
894
+ ```
895
+
896
+ #### MultipleNegativesRankingLoss
897
+ ```bibtex
898
+ @misc{henderson2017efficient,
899
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
900
+ 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},
901
+ year={2017},
902
+ eprint={1705.00652},
903
+ archivePrefix={arXiv},
904
+ primaryClass={cs.CL}
905
+ }
906
+ ```
907
+
908
+ <!--
909
+ ## Glossary
910
+
911
+ *Clearly define terms in order to be accessible across audiences.*
912
+ -->
913
+
914
+ <!--
915
+ ## Model Card Authors
916
+
917
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
918
+ -->
919
+
920
+ <!--
921
+ ## Model Card Contact
922
+
923
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
924
+ -->
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "models/bge-base-en-v1.5-klej-dyk-v0.2",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "id2label": {
13
+ "0": "LABEL_0"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 3072,
17
+ "label2id": {
18
+ "LABEL_0": 0
19
+ },
20
+ "layer_norm_eps": 1e-12,
21
+ "max_position_embeddings": 512,
22
+ "model_type": "bert",
23
+ "num_attention_heads": 12,
24
+ "num_hidden_layers": 12,
25
+ "pad_token_id": 0,
26
+ "position_embedding_type": "absolute",
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.41.2",
29
+ "type_vocab_size": 2,
30
+ "use_cache": true,
31
+ "vocab_size": 30522
32
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "2.2.2",
4
+ "transformers": "4.28.1",
5
+ "pytorch": "1.13.0+cu117"
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 sha256:97ba14b47ac8f31d9cab93f843ecac6409cede066f6162518ef699553b44c12e
3
+ size 437951328
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": true
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 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "mask_token": "[MASK]",
49
+ "max_length": 512,
50
+ "model_max_length": 512,
51
+ "never_split": null,
52
+ "pad_to_multiple_of": null,
53
+ "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
+ "sep_token": "[SEP]",
57
+ "stride": 0,
58
+ "strip_accents": null,
59
+ "tokenize_chinese_chars": true,
60
+ "tokenizer_class": "BertTokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
+ "unk_token": "[UNK]"
64
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff