Hugging Face's logo

Omartificial-Intelligence-Space
/
E5-all-nli-triplet-Matryoshka

Sentence Similarity
sentence-transformers
Safetensors
bert
feature-extraction
Generated from Trainer
dataset_size:557850
loss:MatryoshkaLoss
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
Omartificial-Intelligence-Space commited on
Commit
371ff09
•
1 Parent(s): 2f7a59f

Add new SentenceTransformer model.

Browse files
Files changed (12) hide show
  1. .gitattributes +1 -0
  2. 1_Pooling/config.json +10 -0
  3. README.md +739 -0
  4. config.json +26 -0
  5. config_sentence_transformers.json +10 -0
  6. model.safetensors +3 -0
  7. modules.json +20 -0
  8. sentence_bert_config.json +4 -0
  9. sentencepiece.bpe.model +3 -0
  10. special_tokens_map.json +51 -0
  11. tokenizer.json +3 -0
  12. tokenizer_config.json +55 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 384,
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 +1,739 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: []
3
+ library_name: sentence-transformers
4
+ tags:
5
+ - sentence-transformers
6
+ - sentence-similarity
7
+ - feature-extraction
8
+ - generated_from_trainer
9
+ - dataset_size:557850
10
+ - loss:MatryoshkaLoss
11
+ - loss:MultipleNegativesRankingLoss
12
+ base_model: intfloat/multilingual-e5-small
13
+ datasets: []
14
+ metrics:
15
+ - pearson_cosine
16
+ - spearman_cosine
17
+ - pearson_manhattan
18
+ - spearman_manhattan
19
+ - pearson_euclidean
20
+ - spearman_euclidean
21
+ - pearson_dot
22
+ - spearman_dot
23
+ - pearson_max
24
+ - spearman_max
25
+ widget:
26
+ - source_sentence: ذكر متوازن بعناية يقف على قدم واحدة بالقرب من منطقة شاطئ المحيط
27
+ النظيفة
28
+ sentences:
29
+ - رجل يقدم عرضاً
30
+ - هناك رجل بالخارج قرب الشاطئ
31
+ - رجل يجلس على أريكه
32
+ - source_sentence: رجل يقفز إلى سريره القذر
33
+ sentences:
34
+ - السرير قذر.
35
+ - رجل يضحك أثناء غسيل الملابس
36
+ - الرجل على القمر
37
+ - source_sentence: الفتيات بالخارج
38
+ sentences:
39
+ - امرأة تلف الخيط إلى كرات بجانب كومة من الكرات
40
+ - فتيان يركبان في جولة متعة
41
+ - ثلاث فتيات يقفون سوية في غرفة واحدة تستمع وواحدة تكتب على الحائط والثالثة تتحدث
42
+ إليهن
43
+ - source_sentence: الرجل يرتدي قميصاً أزرق.
44
+ sentences:
45
+ - رجل يرتدي قميصاً أزرق يميل إلى الجدار بجانب الطريق مع شاحنة زرقاء وسيارة حمراء
46
+ مع الماء في الخلفية.
47
+ - كتاب القصص مفتوح
48
+ - رجل يرتدي قميص أسود يعزف على الجيتار.
49
+ - source_sentence: يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة
50
+ شابة.
51
+ sentences:
52
+ - ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه
53
+ - رجل يستلقي على وجهه على مقعد في الحديقة.
54
+ - الشاب نائم بينما الأم تقود ابنتها إلى الحديقة
55
+ pipeline_tag: sentence-similarity
56
+ model-index:
57
+ - name: SentenceTransformer based on intfloat/multilingual-e5-small
58
+ results:
59
+ - task:
60
+ type: semantic-similarity
61
+ name: Semantic Similarity
62
+ dataset:
63
+ name: sts test 384
64
+ type: sts-test-384
65
+ metrics:
66
+ - type: pearson_cosine
67
+ value: 0.7883137447514015
68
+ name: Pearson Cosine
69
+ - type: spearman_cosine
70
+ value: 0.7971624317482785
71
+ name: Spearman Cosine
72
+ - type: pearson_manhattan
73
+ value: 0.7845904338398069
74
+ name: Pearson Manhattan
75
+ - type: spearman_manhattan
76
+ value: 0.7939541836133244
77
+ name: Spearman Manhattan
78
+ - type: pearson_euclidean
79
+ value: 0.7882887522003604
80
+ name: Pearson Euclidean
81
+ - type: spearman_euclidean
82
+ value: 0.7971601260546269
83
+ name: Spearman Euclidean
84
+ - type: pearson_dot
85
+ value: 0.7883137483129774
86
+ name: Pearson Dot
87
+ - type: spearman_dot
88
+ value: 0.7971605875966696
89
+ name: Spearman Dot
90
+ - type: pearson_max
91
+ value: 0.7883137483129774
92
+ name: Pearson Max
93
+ - type: spearman_max
94
+ value: 0.7971624317482785
95
+ name: Spearman Max
96
+ - task:
97
+ type: semantic-similarity
98
+ name: Semantic Similarity
99
+ dataset:
100
+ name: sts test 256
101
+ type: sts-test-256
102
+ metrics:
103
+ - type: pearson_cosine
104
+ value: 0.7851969391652749
105
+ name: Pearson Cosine
106
+ - type: spearman_cosine
107
+ value: 0.7968026743946358
108
+ name: Spearman Cosine
109
+ - type: pearson_manhattan
110
+ value: 0.7852783784725356
111
+ name: Pearson Manhattan
112
+ - type: spearman_manhattan
113
+ value: 0.7935883492889713
114
+ name: Spearman Manhattan
115
+ - type: pearson_euclidean
116
+ value: 0.7882018230746569
117
+ name: Pearson Euclidean
118
+ - type: spearman_euclidean
119
+ value: 0.7963116553267949
120
+ name: Spearman Euclidean
121
+ - type: pearson_dot
122
+ value: 0.7786421988393841
123
+ name: Pearson Dot
124
+ - type: spearman_dot
125
+ value: 0.7867782644180616
126
+ name: Spearman Dot
127
+ - type: pearson_max
128
+ value: 0.7882018230746569
129
+ name: Pearson Max
130
+ - type: spearman_max
131
+ value: 0.7968026743946358
132
+ name: Spearman Max
133
+ - task:
134
+ type: semantic-similarity
135
+ name: Semantic Similarity
136
+ dataset:
137
+ name: sts test 128
138
+ type: sts-test-128
139
+ metrics:
140
+ - type: pearson_cosine
141
+ value: 0.7754967709350954
142
+ name: Pearson Cosine
143
+ - type: spearman_cosine
144
+ value: 0.7933453885370457
145
+ name: Spearman Cosine
146
+ - type: pearson_manhattan
147
+ value: 0.7832834632297865
148
+ name: Pearson Manhattan
149
+ - type: spearman_manhattan
150
+ value: 0.7907589269176767
151
+ name: Spearman Manhattan
152
+ - type: pearson_euclidean
153
+ value: 0.7867583047946054
154
+ name: Pearson Euclidean
155
+ - type: spearman_euclidean
156
+ value: 0.7935816990844704
157
+ name: Spearman Euclidean
158
+ - type: pearson_dot
159
+ value: 0.7317253736607925
160
+ name: Pearson Dot
161
+ - type: spearman_dot
162
+ value: 0.7335574962775742
163
+ name: Spearman Dot
164
+ - type: pearson_max
165
+ value: 0.7867583047946054
166
+ name: Pearson Max
167
+ - type: spearman_max
168
+ value: 0.7935816990844704
169
+ name: Spearman Max
170
+ - task:
171
+ type: semantic-similarity
172
+ name: Semantic Similarity
173
+ dataset:
174
+ name: sts test 64
175
+ type: sts-test-64
176
+ metrics:
177
+ - type: pearson_cosine
178
+ value: 0.7625204599039478
179
+ name: Pearson Cosine
180
+ - type: spearman_cosine
181
+ value: 0.7837078735068292
182
+ name: Spearman Cosine
183
+ - type: pearson_manhattan
184
+ value: 0.7752889433866854
185
+ name: Pearson Manhattan
186
+ - type: spearman_manhattan
187
+ value: 0.7790888579029828
188
+ name: Spearman Manhattan
189
+ - type: pearson_euclidean
190
+ value: 0.777961287133872
191
+ name: Pearson Euclidean
192
+ - type: spearman_euclidean
193
+ value: 0.7815940757356076
194
+ name: Spearman Euclidean
195
+ - type: pearson_dot
196
+ value: 0.6685094830550401
197
+ name: Pearson Dot
198
+ - type: spearman_dot
199
+ value: 0.6621206899696827
200
+ name: Spearman Dot
201
+ - type: pearson_max
202
+ value: 0.777961287133872
203
+ name: Pearson Max
204
+ - type: spearman_max
205
+ value: 0.7837078735068292
206
+ name: Spearman Max
207
+ ---
208
+
209
+ # SentenceTransformer based on intfloat/multilingual-e5-small
210
+
211
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-small](https://huggingface.co/intfloat/multilingual-e5-small) on the Omartificial-Intelligence-Space/arabic-n_li-triplet dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
212
+
213
+ ## Model Details
214
+
215
+ ### Model Description
216
+ - **Model Type:** Sentence Transformer
217
+ - **Base model:** [intfloat/multilingual-e5-small](https://huggingface.co/intfloat/multilingual-e5-small) <!-- at revision 0a68dcd3dad5b4962a78daa930087728292b241d -->
218
+ - **Maximum Sequence Length:** 512 tokens
219
+ - **Output Dimensionality:** 384 tokens
220
+ - **Similarity Function:** Cosine Similarity
221
+ - **Training Dataset:**
222
+ - Omartificial-Intelligence-Space/arabic-n_li-triplet
223
+ <!-- - **Language:** Unknown -->
224
+ <!-- - **License:** Unknown -->
225
+
226
+ ### Model Sources
227
+
228
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
229
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
230
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
231
+
232
+ ### Full Model Architecture
233
+
234
+ ```
235
+ SentenceTransformer(
236
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
237
+ (1): Pooling({'word_embedding_dimension': 384, '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})
238
+ (2): Normalize()
239
+ )
240
+ ```
241
+
242
+ ## Usage
243
+
244
+ ### Direct Usage (Sentence Transformers)
245
+
246
+ First install the Sentence Transformers library:
247
+
248
+ ```bash
249
+ pip install -U sentence-transformers
250
+ ```
251
+
252
+ Then you can load this model and run inference.
253
+ ```python
254
+ from sentence_transformers import SentenceTransformer
255
+
256
+ # Download from the 🤗 Hub
257
+ model = SentenceTransformer("Omartificial-Intelligence-Space/E5-Matro")
258
+ # Run inference
259
+ sentences = [
260
+ 'يجلس شاب ذو شعر أشقر على الحائط يقرأ جريدة بينما تمر امرأة وفتاة شابة.',
261
+ 'ذكر شاب ينظر إلى جريدة بينما تمر إمرأتان بجانبه',
262
+ 'الشاب نائم بينما الأم تقود ابنتها إلى الحديقة',
263
+ ]
264
+ embeddings = model.encode(sentences)
265
+ print(embeddings.shape)
266
+ # [3, 384]
267
+
268
+ # Get the similarity scores for the embeddings
269
+ similarities = model.similarity(embeddings, embeddings)
270
+ print(similarities.shape)
271
+ # [3, 3]
272
+ ```
273
+
274
+ <!--
275
+ ### Direct Usage (Transformers)
276
+
277
+ <details><summary>Click to see the direct usage in Transformers</summary>
278
+
279
+ </details>
280
+ -->
281
+
282
+ <!--
283
+ ### Downstream Usage (Sentence Transformers)
284
+
285
+ You can finetune this model on your own dataset.
286
+
287
+ <details><summary>Click to expand</summary>
288
+
289
+ </details>
290
+ -->
291
+
292
+ <!--
293
+ ### Out-of-Scope Use
294
+
295
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
296
+ -->
297
+
298
+ ## Evaluation
299
+
300
+ ### Metrics
301
+
302
+ #### Semantic Similarity
303
+ * Dataset: `sts-test-384`
304
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
305
+
306
+ | Metric | Value |
307
+ |:--------------------|:-----------|
308
+ | pearson_cosine | 0.7883 |
309
+ | **spearman_cosine** | **0.7972** |
310
+ | pearson_manhattan | 0.7846 |
311
+ | spearman_manhattan | 0.794 |
312
+ | pearson_euclidean | 0.7883 |
313
+ | spearman_euclidean | 0.7972 |
314
+ | pearson_dot | 0.7883 |
315
+ | spearman_dot | 0.7972 |
316
+ | pearson_max | 0.7883 |
317
+ | spearman_max | 0.7972 |
318
+
319
+ #### Semantic Similarity
320
+ * Dataset: `sts-test-256`
321
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
322
+
323
+ | Metric | Value |
324
+ |:--------------------|:-----------|
325
+ | pearson_cosine | 0.7852 |
326
+ | **spearman_cosine** | **0.7968** |
327
+ | pearson_manhattan | 0.7853 |
328
+ | spearman_manhattan | 0.7936 |
329
+ | pearson_euclidean | 0.7882 |
330
+ | spearman_euclidean | 0.7963 |
331
+ | pearson_dot | 0.7786 |
332
+ | spearman_dot | 0.7868 |
333
+ | pearson_max | 0.7882 |
334
+ | spearman_max | 0.7968 |
335
+
336
+ #### Semantic Similarity
337
+ * Dataset: `sts-test-128`
338
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
339
+
340
+ | Metric | Value |
341
+ |:--------------------|:-----------|
342
+ | pearson_cosine | 0.7755 |
343
+ | **spearman_cosine** | **0.7933** |
344
+ | pearson_manhattan | 0.7833 |
345
+ | spearman_manhattan | 0.7908 |
346
+ | pearson_euclidean | 0.7868 |
347
+ | spearman_euclidean | 0.7936 |
348
+ | pearson_dot | 0.7317 |
349
+ | spearman_dot | 0.7336 |
350
+ | pearson_max | 0.7868 |
351
+ | spearman_max | 0.7936 |
352
+
353
+ #### Semantic Similarity
354
+ * Dataset: `sts-test-64`
355
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
356
+
357
+ | Metric | Value |
358
+ |:--------------------|:-----------|
359
+ | pearson_cosine | 0.7625 |
360
+ | **spearman_cosine** | **0.7837** |
361
+ | pearson_manhattan | 0.7753 |
362
+ | spearman_manhattan | 0.7791 |
363
+ | pearson_euclidean | 0.778 |
364
+ | spearman_euclidean | 0.7816 |
365
+ | pearson_dot | 0.6685 |
366
+ | spearman_dot | 0.6621 |
367
+ | pearson_max | 0.778 |
368
+ | spearman_max | 0.7837 |
369
+
370
+ <!--
371
+ ## Bias, Risks and Limitations
372
+
373
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
374
+ -->
375
+
376
+ <!--
377
+ ### Recommendations
378
+
379
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
380
+ -->
381
+
382
+ ## Training Details
383
+
384
+ ### Training Dataset
385
+
386
+ #### Omartificial-Intelligence-Space/arabic-n_li-triplet
387
+
388
+ * Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet
389
+ * Size: 557,850 training samples
390
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
391
+ * Approximate statistics based on the first 1000 samples:
392
+ | | anchor | positive | negative |
393
+ |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
394
+ | type | string | string | string |
395
+ | details | <ul><li>min: 5 tokens</li><li>mean: 10.33 tokens</li><li>max: 52 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 13.21 tokens</li><li>max: 49 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 15.32 tokens</li><li>max: 53 tokens</li></ul> |
396
+ * Samples:
397
+ | anchor | positive | negative |
398
+ |:------------------------------------------------------------|:--------------------------------------------|:------------------------------------|
399
+ | <code>شخص على حصان يقفز فوق طائرة معطلة</code> | <code>شخص في الهواء الطلق، على حصان.</code> | <code>شخص في مطعم، يطلب عجة.</code> |
400
+ | <code>أطفال يبتسمون و يلوحون للكاميرا</code> | <code>هناك أطفال حاضرون</code> | <code>الاطفال يتجهمون</code> |
401
+ | <code>صبي يقفز على لوح التزلج في منتصف الجسر الأحمر.</code> | <code>الفتى يقوم بخدعة التزلج</code> | <code>الصبي يتزلج على الرصيف</code> |
402
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
403
+ ```json
404
+ {
405
+ "loss": "MultipleNegativesRankingLoss",
406
+ "matryoshka_dims": [
407
+ 384,
408
+ 256,
409
+ 128,
410
+ 64
411
+ ],
412
+ "matryoshka_weights": [
413
+ 1,
414
+ 1,
415
+ 1,
416
+ 1
417
+ ],
418
+ "n_dims_per_step": -1
419
+ }
420
+ ```
421
+
422
+ ### Evaluation Dataset
423
+
424
+ #### Omartificial-Intelligence-Space/arabic-n_li-triplet
425
+
426
+ * Dataset: Omartificial-Intelligence-Space/arabic-n_li-triplet
427
+ * Size: 6,584 evaluation samples
428
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
429
+ * Approximate statistics based on the first 1000 samples:
430
+ | | anchor | positive | negative |
431
+ |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|
432
+ | type | string | string | string |
433
+ | details | <ul><li>min: 5 tokens</li><li>mean: 21.86 tokens</li><li>max: 105 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 10.22 tokens</li><li>max: 49 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 11.2 tokens</li><li>max: 33 tokens</li></ul> |
434
+ * Samples:
435
+ | anchor | positive | negative |
436
+ |:-----------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------|:---------------------------------------------------|
437
+ | <code>امرأتان يتعانقان بينما يحملان حزمة</code> | <code>إمرأتان يحملان حزمة</code> | <code>الرجال يتشاجرون خارج مطعم</code> |
438
+ | <code>طفلين صغيرين يرتديان قميصاً أزرق، أحدهما يرتدي الرقم 9 والآخر يرتدي الرقم 2 يقفان على خطوات خشبية في الحمام ويغسلان أيديهما في المغسلة.</code> | <code>طفلين يرتديان قميصاً مرقماً يغسلون أيديهم</code> | <code>طفلين يرتديان سترة يذهبان إلى المدرسة</code> |
439
+ | <code>رجل يبيع الدونات لعميل خلال معرض عالمي أقيم في مدينة أنجليس</code> | <code>رجل يبيع الدونات لعميل</code> | <code>امرأة تشرب قهوتها في مقهى صغير</code> |
440
+ * Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
441
+ ```json
442
+ {
443
+ "loss": "MultipleNegativesRankingLoss",
444
+ "matryoshka_dims": [
445
+ 384,
446
+ 256,
447
+ 128,
448
+ 64
449
+ ],
450
+ "matryoshka_weights": [
451
+ 1,
452
+ 1,
453
+ 1,
454
+ 1
455
+ ],
456
+ "n_dims_per_step": -1
457
+ }
458
+ ```
459
+
460
+ ### Training Hyperparameters
461
+ #### Non-Default Hyperparameters
462
+
463
+ - `per_device_train_batch_size`: 32
464
+ - `per_device_eval_batch_size`: 32
465
+ - `warmup_ratio`: 0.1
466
+ - `fp16`: True
467
+ - `batch_sampler`: no_duplicates
468
+
469
+ #### All Hyperparameters
470
+ <details><summary>Click to expand</summary>
471
+
472
+ - `overwrite_output_dir`: False
473
+ - `do_predict`: False
474
+ - `prediction_loss_only`: True
475
+ - `per_device_train_batch_size`: 32
476
+ - `per_device_eval_batch_size`: 32
477
+ - `per_gpu_train_batch_size`: None
478
+ - `per_gpu_eval_batch_size`: None
479
+ - `gradient_accumulation_steps`: 1
480
+ - `eval_accumulation_steps`: None
481
+ - `learning_rate`: 5e-05
482
+ - `weight_decay`: 0.0
483
+ - `adam_beta1`: 0.9
484
+ - `adam_beta2`: 0.999
485
+ - `adam_epsilon`: 1e-08
486
+ - `max_grad_norm`: 1.0
487
+ - `num_train_epochs`: 3
488
+ - `max_steps`: -1
489
+ - `lr_scheduler_type`: linear
490
+ - `lr_scheduler_kwargs`: {}
491
+ - `warmup_ratio`: 0.1
492
+ - `warmup_steps`: 0
493
+ - `log_level`: passive
494
+ - `log_level_replica`: warning
495
+ - `log_on_each_node`: True
496
+ - `logging_nan_inf_filter`: True
497
+ - `save_safetensors`: True
498
+ - `save_on_each_node`: False
499
+ - `save_only_model`: False
500
+ - `no_cuda`: False
501
+ - `use_cpu`: False
502
+ - `use_mps_device`: False
503
+ - `seed`: 42
504
+ - `data_seed`: None
505
+ - `jit_mode_eval`: False
506
+ - `use_ipex`: False
507
+ - `bf16`: False
508
+ - `fp16`: True
509
+ - `fp16_opt_level`: O1
510
+ - `half_precision_backend`: auto
511
+ - `bf16_full_eval`: False
512
+ - `fp16_full_eval`: False
513
+ - `tf32`: None
514
+ - `local_rank`: 0
515
+ - `ddp_backend`: None
516
+ - `tpu_num_cores`: None
517
+ - `tpu_metrics_debug`: False
518
+ - `debug`: []
519
+ - `dataloader_drop_last`: False
520
+ - `dataloader_num_workers`: 0
521
+ - `dataloader_prefetch_factor`: None
522
+ - `past_index`: -1
523
+ - `disable_tqdm`: False
524
+ - `remove_unused_columns`: True
525
+ - `label_names`: None
526
+ - `load_best_model_at_end`: False
527
+ - `ignore_data_skip`: False
528
+ - `fsdp`: []
529
+ - `fsdp_min_num_params`: 0
530
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
531
+ - `fsdp_transformer_layer_cls_to_wrap`: None
532
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'gradient_accumulation_kwargs': None}
533
+ - `deepspeed`: None
534
+ - `label_smoothing_factor`: 0.0
535
+ - `optim`: adamw_torch
536
+ - `optim_args`: None
537
+ - `adafactor`: False
538
+ - `group_by_length`: False
539
+ - `length_column_name`: length
540
+ - `ddp_find_unused_parameters`: None
541
+ - `ddp_bucket_cap_mb`: None
542
+ - `ddp_broadcast_buffers`: False
543
+ - `dataloader_pin_memory`: True
544
+ - `dataloader_persistent_workers`: False
545
+ - `skip_memory_metrics`: True
546
+ - `use_legacy_prediction_loop`: False
547
+ - `push_to_hub`: False
548
+ - `resume_from_checkpoint`: None
549
+ - `hub_model_id`: None
550
+ - `hub_strategy`: every_save
551
+ - `hub_private_repo`: False
552
+ - `hub_always_push`: False
553
+ - `gradient_checkpointing`: False
554
+ - `gradient_checkpointing_kwargs`: None
555
+ - `include_inputs_for_metrics`: False
556
+ - `eval_do_concat_batches`: True
557
+ - `fp16_backend`: auto
558
+ - `push_to_hub_model_id`: None
559
+ - `push_to_hub_organization`: None
560
+ - `mp_parameters`:
561
+ - `auto_find_batch_size`: False
562
+ - `full_determinism`: False
563
+ - `torchdynamo`: None
564
+ - `ray_scope`: last
565
+ - `ddp_timeout`: 1800
566
+ - `torch_compile`: False
567
+ - `torch_compile_backend`: None
568
+ - `torch_compile_mode`: None
569
+ - `dispatch_batches`: None
570
+ - `split_batches`: None
571
+ - `include_tokens_per_second`: False
572
+ - `include_num_input_tokens_seen`: False
573
+ - `neftune_noise_alpha`: None
574
+ - `optim_target_modules`: None
575
+ - `batch_sampler`: no_duplicates
576
+ - `multi_dataset_batch_sampler`: proportional
577
+
578
+ </details>
579
+
580
+ ### Training Logs
581
+ | Epoch | Step | Training Loss | sts-test-128_spearman_cosine | sts-test-256_spearman_cosine | sts-test-384_spearman_cosine | sts-test-64_spearman_cosine |
582
+ |:------:|:-----:|:-------------:|:----------------------------:|:----------------------------:|:----------------------------:|:---------------------------:|
583
+ | 0.0344 | 200 | 13.1208 | - | - | - | - |
584
+ | 0.0688 | 400 | 9.1894 | - | - | - | - |
585
+ | 0.1033 | 600 | 8.0222 | - | - | - | - |
586
+ | 0.1377 | 800 | 7.2405 | - | - | - | - |
587
+ | 0.1721 | 1000 | 7.1622 | - | - | - | - |
588
+ | 0.2065 | 1200 | 6.4282 | - | - | - | - |
589
+ | 0.2409 | 1400 | 6.0936 | - | - | - | - |
590
+ | 0.2753 | 1600 | 5.99 | - | - | - | - |
591
+ | 0.3098 | 1800 | 5.6939 | - | - | - | - |
592
+ | 0.3442 | 2000 | 5.694 | - | - | - | - |
593
+ | 0.3786 | 2200 | 5.2366 | - | - | - | - |
594
+ | 0.4130 | 2400 | 5.2994 | - | - | - | - |
595
+ | 0.4474 | 2600 | 5.2079 | - | - | - | - |
596
+ | 0.4818 | 2800 | 5.0532 | - | - | - | - |
597
+ | 0.5163 | 3000 | 4.9978 | - | - | - | - |
598
+ | 0.5507 | 3200 | 5.1764 | - | - | - | - |
599
+ | 0.5851 | 3400 | 5.1315 | - | - | - | - |
600
+ | 0.6195 | 3600 | 5.0198 | - | - | - | - |
601
+ | 0.6539 | 3800 | 5.0308 | - | - | - | - |
602
+ | 0.6883 | 4000 | 5.1631 | - | - | - | - |
603
+ | 0.7228 | 4200 | 4.7916 | - | - | - | - |
604
+ | 0.7572 | 4400 | 4.363 | - | - | - | - |
605
+ | 0.7916 | 4600 | 3.2357 | - | - | - | - |
606
+ | 0.8260 | 4800 | 2.9915 | - | - | - | - |
607
+ | 0.8604 | 5000 | 2.8143 | - | - | - | - |
608
+ | 0.8949 | 5200 | 2.6125 | - | - | - | - |
609
+ | 0.9293 | 5400 | 2.5493 | - | - | - | - |
610
+ | 0.9637 | 5600 | 2.4991 | - | - | - | - |
611
+ | 0.9981 | 5800 | 2.163 | - | - | - | - |
612
+ | 1.0325 | 6000 | 0.0 | - | - | - | - |
613
+ | 1.0669 | 6200 | 0.0 | - | - | - | - |
614
+ | 1.1014 | 6400 | 0.0 | - | - | - | - |
615
+ | 1.1358 | 6600 | 0.0 | - | - | - | - |
616
+ | 1.1702 | 6800 | 0.0 | - | - | - | - |
617
+ | 1.2046 | 7000 | 0.0 | - | - | - | - |
618
+ | 1.2390 | 7200 | 0.0 | - | - | - | - |
619
+ | 1.2734 | 7400 | 0.0 | - | - | - | - |
620
+ | 1.3079 | 7600 | 0.0 | - | - | - | - |
621
+ | 1.3423 | 7800 | 0.0 | - | - | - | - |
622
+ | 1.3767 | 8000 | 0.0 | - | - | - | - |
623
+ | 1.4111 | 8200 | 0.0037 | - | - | - | - |
624
+ | 1.4455 | 8400 | 0.0372 | - | - | - | - |
625
+ | 1.4800 | 8600 | 0.0221 | - | - | - | - |
626
+ | 1.0229 | 8800 | 4.3738 | - | - | - | - |
627
+ | 1.0573 | 9000 | 6.338 | - | - | - | - |
628
+ | 1.0917 | 9200 | 6.2223 | - | - | - | - |
629
+ | 1.1261 | 9400 | 5.8673 | - | - | - | - |
630
+ | 1.1606 | 9600 | 5.5907 | - | - | - | - |
631
+ | 1.1950 | 9800 | 5.0307 | - | - | - | - |
632
+ | 1.2294 | 10000 | 4.9193 | - | - | - | - |
633
+ | 1.2638 | 10200 | 4.8798 | - | - | - | - |
634
+ | 1.2982 | 10400 | 4.401 | - | - | - | - |
635
+ | 1.3326 | 10600 | 4.2705 | - | - | - | - |
636
+ | 1.3671 | 10800 | 4.3023 | - | - | - | - |
637
+ | 1.4015 | 11000 | 4.1344 | - | - | - | - |
638
+ | 1.4359 | 11200 | 4.0464 | - | - | - | - |
639
+ | 1.4703 | 11400 | 4.0115 | - | - | - | - |
640
+ | 1.5047 | 11600 | 3.9206 | - | - | - | - |
641
+ | 1.5391 | 11800 | 4.0106 | - | - | - | - |
642
+ | 1.5736 | 12000 | 4.1365 | - | - | - | - |
643
+ | 1.6080 | 12200 | 4.0401 | - | - | - | - |
644
+ | 1.6424 | 12400 | 4.0602 | - | - | - | - |
645
+ | 1.6768 | 12600 | 4.076 | - | - | - | - |
646
+ | 1.7112 | 12800 | 3.97 | - | - | - | - |
647
+ | 1.7457 | 13000 | 3.7905 | - | - | - | - |
648
+ | 1.7801 | 13200 | 2.414 | - | - | - | - |
649
+ | 1.8145 | 13400 | 2.1811 | - | - | - | - |
650
+ | 1.8489 | 13600 | 2.1183 | - | - | - | - |
651
+ | 1.8833 | 13800 | 2.0578 | - | - | - | - |
652
+ | 1.9177 | 14000 | 2.0173 | - | - | - | - |
653
+ | 1.9522 | 14200 | 2.0093 | - | - | - | - |
654
+ | 1.9866 | 14400 | 1.9467 | - | - | - | - |
655
+ | 2.0210 | 14600 | 0.4674 | - | - | - | - |
656
+ | 2.0554 | 14800 | 0.0 | - | - | - | - |
657
+ | 2.0898 | 15000 | 0.0 | - | - | - | - |
658
+ | 2.1242 | 15200 | 0.0 | - | - | - | - |
659
+ | 2.1587 | 15400 | 0.0 | - | - | - | - |
660
+ | 2.1931 | 15600 | 0.0 | - | - | - | - |
661
+ | 2.2275 | 15800 | 0.0 | - | - | - | - |
662
+ | 2.2619 | 16000 | 0.0 | - | - | - | - |
663
+ | 2.2963 | 16200 | 0.0 | - | - | - | - |
664
+ | 2.3308 | 16400 | 0.0 | - | - | - | - |
665
+ | 2.3652 | 16600 | 0.0 | - | - | - | - |
666
+ | 2.3996 | 16800 | 0.0 | - | - | - | - |
667
+ | 2.4340 | 17000 | 0.0 | - | - | - | - |
668
+ | 2.4684 | 17200 | 0.0256 | - | - | - | - |
669
+ | 2.0114 | 17400 | 2.4155 | - | - | - | - |
670
+ | 2.0170 | 17433 | - | 0.7933 | 0.7968 | 0.7972 | 0.7837 |
671
+
672
+
673
+ ### Framework Versions
674
+ - Python: 3.9.18
675
+ - Sentence Transformers: 3.0.1
676
+ - Transformers: 4.40.0
677
+ - PyTorch: 2.2.2+cu121
678
+ - Accelerate: 0.26.1
679
+ - Datasets: 2.19.0
680
+ - Tokenizers: 0.19.1
681
+
682
+ ## Citation
683
+
684
+ ### BibTeX
685
+
686
+ #### Sentence Transformers
687
+ ```bibtex
688
+ @inproceedings{reimers-2019-sentence-bert,
689
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
690
+ author = "Reimers, Nils and Gurevych, Iryna",
691
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
692
+ month = "11",
693
+ year = "2019",
694
+ publisher = "Association for Computational Linguistics",
695
+ url = "https://arxiv.org/abs/1908.10084",
696
+ }
697
+ ```
698
+
699
+ #### MatryoshkaLoss
700
+ ```bibtex
701
+ @misc{kusupati2024matryoshka,
702
+ title={Matryoshka Representation Learning},
703
+ 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},
704
+ year={2024},
705
+ eprint={2205.13147},
706
+ archivePrefix={arXiv},
707
+ primaryClass={cs.LG}
708
+ }
709
+ ```
710
+
711
+ #### MultipleNegativesRankingLoss
712
+ ```bibtex
713
+ @misc{henderson2017efficient,
714
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
715
+ 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},
716
+ year={2017},
717
+ eprint={1705.00652},
718
+ archivePrefix={arXiv},
719
+ primaryClass={cs.CL}
720
+ }
721
+ ```
722
+
723
+ <!--
724
+ ## Glossary
725
+
726
+ *Clearly define terms in order to be accessible across audiences.*
727
+ -->
728
+
729
+ <!--
730
+ ## Model Card Authors
731
+
732
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
733
+ -->
734
+
735
+ <!--
736
+ ## Model Card Contact
737
+
738
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
739
+ -->
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "intfloat/multilingual-e5-small",
3
+ "architectures": [
4
+ "BertModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 384,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 1536,
13
+ "layer_norm_eps": 1e-12,
14
+ "max_position_embeddings": 512,
15
+ "model_type": "bert",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 12,
18
+ "pad_token_id": 0,
19
+ "position_embedding_type": "absolute",
20
+ "tokenizer_class": "XLMRobertaTokenizer",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.40.0",
23
+ "type_vocab_size": 2,
24
+ "use_cache": true,
25
+ "vocab_size": 250037
26
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.1",
4
+ "transformers": "4.40.0",
5
+ "pytorch": "2.2.2+cu121"
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 sha256:5fb3953ddeabff81ffe731bc0d2efc85427653853b4ef19aba90dbfb3bd9c3d0
3
+ size 470637416
modules.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "idx": 0,
4
+ "name": "0",
5
+ "path": "",
6
+ "type": "sentence_transformers.models.Transformer"
7
+ },
8
+ {
9
+ "idx": 1,
10
+ "name": "1",
11
+ "path": "1_Pooling",
12
+ "type": "sentence_transformers.models.Pooling"
13
+ },
14
+ {
15
+ "idx": 2,
16
+ "name": "2",
17
+ "path": "2_Normalize",
18
+ "type": "sentence_transformers.models.Normalize"
19
+ }
20
+ ]
sentence_bert_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "max_seq_length": 512,
3
+ "do_lower_case": false
4
+ }
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef04f2b385d1514f500e779207ace0f53e30895ce37563179e29f4022d28ca38
3
+ size 17083053
tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "<mask>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 512,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "sp_model_kwargs": {},
53
+ "tokenizer_class": "XLMRobertaTokenizer",
54
+ "unk_token": "<unk>"
55
+ }