Hugging Face's logo

yahyaabd
/
allstats-search-miniLM-v1-6

Sentence Similarity
sentence-transformers
Safetensors
bert
feature-extraction
Generated from Trainer
dataset_size:25580
loss:OnlineContrastiveLoss
Eval Results
text-embeddings-inference
Model card Files Community
yahyaabd commited on
Commit
e0b9383
·
verified ·
1 Parent(s): 4fe1bcf

Add new SentenceTransformer model

Browse files
Files changed (12) hide show
  1. .gitattributes +2 -0
  2. 1_Pooling/config.json +10 -0
  3. README.md +599 -0
  4. config.json +26 -0
  5. config_sentence_transformers.json +10 -0
  6. model.safetensors +3 -0
  7. modules.json +14 -0
  8. sentence_bert_config.json +4 -0
  9. special_tokens_map.json +51 -0
  10. tokenizer.json +3 -0
  11. tokenizer_config.json +65 -0
  12. unigram.json +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ 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
37
+ unigram.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 +1,599 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - generated_from_trainer
7
+ - dataset_size:25580
8
+ - loss:OnlineContrastiveLoss
9
+ base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
10
+ widget:
11
+ - source_sentence: ikhtisar arus kas triwulan 1, 2004 (miliar)
12
+ sentences:
13
+ - Balita (0-59 Bulan) Menurut Status Gizi, Tahun 1998-2005
14
+ - Perbandingan Indeks dan Tingkat Inflasi Desember 2023 Kota-kota di Luar Pulau
15
+ Jawa dan Sumatera dengan Nasional (2018=100)
16
+ - Rata-rata Konsumsi dan Pengeluaran Perkapita Seminggu Menurut Komoditi Makanan
17
+ dan Golongan Pengeluaran per Kapita Seminggu di Provinsi Sulawesi Tengah, 2018-2023
18
+ - source_sentence: BaIgaimana gambaran neraca arus dana dUi Indonesia pada kuartal
19
+ kedua tahun 2015?
20
+ sentences:
21
+ - Jumlah Sekolah, Guru, dan Murid Sekolah Menengah Pertama (SMP) di Bawah Kementrian
22
+ Pendidikan dan Kebudayaan Menurut Provinsi 2011/2012-2015/2016
23
+ - Ringkasan Neraca Arus Dana Triwulan III Tahun 2003 (Miliar Rupiah)
24
+ - Rata-rata Konsumsi dan Pengeluaran Perkapita Seminggu Menurut Komoditi Makanan
25
+ dan Golongan Pengeluaran per Kapita Seminggu di Provinsi Sulawesi Tenggara, 2018-2023
26
+ - source_sentence: Berapa persen pengeluaran orang di kotaa untuk makanan vs non-makanan,
27
+ per provinsi, 2018?
28
+ sentences:
29
+ - Ekspor Tanaman Obat, Aromatik, dan Rempah-Rempah menurut Negara Tujuan Utama,
30
+ 2012-2023
31
+ - Rata-rata Pendapatan Bersih Pekerja Bebas Menurut Provinsi dan Pendidikan Tertinggi
32
+ yang Ditamatkan (ribu rupiah), 2017
33
+ - IHK dan Rata-rata Upah per Bulan Buruh Industri di Bawah Mandor (Supervisor),
34
+ 1996-2014 (1996=100)
35
+ - source_sentence: Negara-negara asal impor crude oil dan produk turunannya tahun
36
+ 2002-2023
37
+ sentences:
38
+ - Persentase Pengeluaran Rata-rata per Kapita Sebulan Menurut Kelompok Barang, Indonesia,
39
+ 1999, 2002-2023
40
+ - Rata-rata Pendapatan Bersih Berusaha Sendiri menurut Provinsi dan Pendidikan yang
41
+ Ditamatkan (ribu rupiah), 2016
42
+ - Perkembangan Beberapa Agregat Pendapatan dan Pendapatan per Kapita Atas Dasar
43
+ Harga Berlaku, 2010-2016
44
+ - source_sentence: Arus dana Q3 2006
45
+ sentences:
46
+ - Posisi Simpanan Berjangka Rupiah pada Bank Umum dan BPR Menurut Golongan Pemilik
47
+ (miliar rupiah), 2005-2018
48
+ - Ringkasan Neraca Arus Dana, Triwulan III, 2006, (Miliar Rupiah)
49
+ - Rata-Rata Pengeluaran per Kapita Sebulan di Daerah Perkotaan Menurut Kelompok
50
+ Barang dan Golongan Pengeluaran per Kapita Sebulan, 2000-2012
51
+ datasets:
52
+ - yahyaabd/query-hard-pos-neg-doc-pairs-statictable
53
+ pipeline_tag: sentence-similarity
54
+ library_name: sentence-transformers
55
+ metrics:
56
+ - cosine_accuracy
57
+ - cosine_accuracy_threshold
58
+ - cosine_f1
59
+ - cosine_f1_threshold
60
+ - cosine_precision
61
+ - cosine_recall
62
+ - cosine_ap
63
+ - cosine_mcc
64
+ model-index:
65
+ - name: SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
66
+ results:
67
+ - task:
68
+ type: binary-classification
69
+ name: Binary Classification
70
+ dataset:
71
+ name: allstats semantic mini v1 test
72
+ type: allstats-semantic-mini-v1_test
73
+ metrics:
74
+ - type: cosine_accuracy
75
+ value: 0.9808882417182381
76
+ name: Cosine Accuracy
77
+ - type: cosine_accuracy_threshold
78
+ value: 0.7710697650909424
79
+ name: Cosine Accuracy Threshold
80
+ - type: cosine_f1
81
+ value: 0.9706129303106633
82
+ name: Cosine F1
83
+ - type: cosine_f1_threshold
84
+ value: 0.7710697650909424
85
+ name: Cosine F1 Threshold
86
+ - type: cosine_precision
87
+ value: 0.9725182277061133
88
+ name: Cosine Precision
89
+ - type: cosine_recall
90
+ value: 0.9687150837988827
91
+ name: Cosine Recall
92
+ - type: cosine_ap
93
+ value: 0.995703716321768
94
+ name: Cosine Ap
95
+ - type: cosine_mcc
96
+ value: 0.9564565560407811
97
+ name: Cosine Mcc
98
+ - task:
99
+ type: binary-classification
100
+ name: Binary Classification
101
+ dataset:
102
+ name: allstats semantic mini v1 dev
103
+ type: allstats-semantic-mini-v1_dev
104
+ metrics:
105
+ - type: cosine_accuracy
106
+ value: 0.9753604672385472
107
+ name: Cosine Accuracy
108
+ - type: cosine_accuracy_threshold
109
+ value: 0.773653507232666
110
+ name: Cosine Accuracy Threshold
111
+ - type: cosine_f1
112
+ value: 0.9622799664710814
113
+ name: Cosine F1
114
+ - type: cosine_f1_threshold
115
+ value: 0.773653507232666
116
+ name: Cosine F1 Threshold
117
+ - type: cosine_precision
118
+ value: 0.9545454545454546
119
+ name: Cosine Precision
120
+ - type: cosine_recall
121
+ value: 0.9701408450704225
122
+ name: Cosine Recall
123
+ - type: cosine_ap
124
+ value: 0.9927278433062661
125
+ name: Cosine Ap
126
+ - type: cosine_mcc
127
+ value: 0.9440549314838564
128
+ name: Cosine Mcc
129
+ ---
130
+
131
+ # SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
132
+
133
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) on the [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) 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.
134
+
135
+ ## Model Details
136
+
137
+ ### Model Description
138
+ - **Model Type:** Sentence Transformer
139
+ - **Base model:** [sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2) <!-- at revision 8d6b950845285729817bf8e1af1861502c2fed0c -->
140
+ - **Maximum Sequence Length:** 128 tokens
141
+ - **Output Dimensionality:** 384 dimensions
142
+ - **Similarity Function:** Cosine Similarity
143
+ - **Training Dataset:**
144
+ - [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable)
145
+ <!-- - **Language:** Unknown -->
146
+ <!-- - **License:** Unknown -->
147
+
148
+ ### Model Sources
149
+
150
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
151
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
152
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
153
+
154
+ ### Full Model Architecture
155
+
156
+ ```
157
+ SentenceTransformer(
158
+ (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel
159
+ (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})
160
+ )
161
+ ```
162
+
163
+ ## Usage
164
+
165
+ ### Direct Usage (Sentence Transformers)
166
+
167
+ First install the Sentence Transformers library:
168
+
169
+ ```bash
170
+ pip install -U sentence-transformers
171
+ ```
172
+
173
+ Then you can load this model and run inference.
174
+ ```python
175
+ from sentence_transformers import SentenceTransformer
176
+
177
+ # Download from the 🤗 Hub
178
+ model = SentenceTransformer("yahyaabd/allstats-search-miniLM-v1-6")
179
+ # Run inference
180
+ sentences = [
181
+ 'Arus dana Q3 2006',
182
+ 'Ringkasan Neraca Arus Dana, Triwulan III, 2006, (Miliar Rupiah)',
183
+ 'Rata-Rata Pengeluaran per Kapita Sebulan di Daerah Perkotaan Menurut Kelompok Barang dan Golongan Pengeluaran per Kapita Sebulan, 2000-2012',
184
+ ]
185
+ embeddings = model.encode(sentences)
186
+ print(embeddings.shape)
187
+ # [3, 384]
188
+
189
+ # Get the similarity scores for the embeddings
190
+ similarities = model.similarity(embeddings, embeddings)
191
+ print(similarities.shape)
192
+ # [3, 3]
193
+ ```
194
+
195
+ <!--
196
+ ### Direct Usage (Transformers)
197
+
198
+ <details><summary>Click to see the direct usage in Transformers</summary>
199
+
200
+ </details>
201
+ -->
202
+
203
+ <!--
204
+ ### Downstream Usage (Sentence Transformers)
205
+
206
+ You can finetune this model on your own dataset.
207
+
208
+ <details><summary>Click to expand</summary>
209
+
210
+ </details>
211
+ -->
212
+
213
+ <!--
214
+ ### Out-of-Scope Use
215
+
216
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
217
+ -->
218
+
219
+ ## Evaluation
220
+
221
+ ### Metrics
222
+
223
+ #### Binary Classification
224
+
225
+ * Datasets: `allstats-semantic-mini-v1_test` and `allstats-semantic-mini-v1_dev`
226
+ * Evaluated with [<code>BinaryClassificationEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator)
227
+
228
+ | Metric | allstats-semantic-mini-v1_test | allstats-semantic-mini-v1_dev |
229
+ |:--------------------------|:-------------------------------|:------------------------------|
230
+ | cosine_accuracy | 0.9809 | 0.9754 |
231
+ | cosine_accuracy_threshold | 0.7711 | 0.7737 |
232
+ | cosine_f1 | 0.9706 | 0.9623 |
233
+ | cosine_f1_threshold | 0.7711 | 0.7737 |
234
+ | cosine_precision | 0.9725 | 0.9545 |
235
+ | cosine_recall | 0.9687 | 0.9701 |
236
+ | **cosine_ap** | **0.9957** | **0.9927** |
237
+ | cosine_mcc | 0.9565 | 0.9441 |
238
+
239
+ <!--
240
+ ## Bias, Risks and Limitations
241
+
242
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
243
+ -->
244
+
245
+ <!--
246
+ ### Recommendations
247
+
248
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
249
+ -->
250
+
251
+ ## Training Details
252
+
253
+ ### Training Dataset
254
+
255
+ #### query-hard-pos-neg-doc-pairs-statictable
256
+
257
+ * Dataset: [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) at [7b28b96](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable/tree/7b28b964daa3073a4d012d1ffca46ecd4f26bb5f)
258
+ * Size: 25,580 training samples
259
+ * Columns: <code>query</code>, <code>doc</code>, and <code>label</code>
260
+ * Approximate statistics based on the first 1000 samples:
261
+ | | query | doc | label |
262
+ |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:------------------------------------------------|
263
+ | type | string | string | int |
264
+ | details | <ul><li>min: 7 tokens</li><li>mean: 20.14 tokens</li><li>max: 55 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 24.9 tokens</li><li>max: 47 tokens</li></ul> | <ul><li>0: ~70.80%</li><li>1: ~29.20%</li></ul> |
265
+ * Samples:
266
+ | query | doc | label |
267
+ |:-------------------------------------------------------------------------|:----------------------------------------------|:---------------|
268
+ | <code>Status pekerjaan utama penduduk usia 15+ yang bekerja, 2020</code> | <code>Jumlah Penghuni Lapas per Kanwil</code> | <code>0</code> |
269
+ | <code>status pekerjaan utama penduduk usia 15+ yang bekerja, 2020</code> | <code>Jumlah Penghuni Lapas per Kanwil</code> | <code>0</code> |
270
+ | <code>STATUS PEKERJAAN UTAMA PENDUDUK USIA 15+ YANG BEKERJA, 2020</code> | <code>Jumlah Penghuni Lapas per Kanwil</code> | <code>0</code> |
271
+ * Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
272
+
273
+ ### Evaluation Dataset
274
+
275
+ #### query-hard-pos-neg-doc-pairs-statictable
276
+
277
+ * Dataset: [query-hard-pos-neg-doc-pairs-statictable](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable) at [7b28b96](https://huggingface.co/datasets/yahyaabd/query-hard-pos-neg-doc-pairs-statictable/tree/7b28b964daa3073a4d012d1ffca46ecd4f26bb5f)
278
+ * Size: 5,479 evaluation samples
279
+ * Columns: <code>query</code>, <code>doc</code>, and <code>label</code>
280
+ * Approximate statistics based on the first 1000 samples:
281
+ | | query | doc | label |
282
+ |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------|
283
+ | type | string | string | int |
284
+ | details | <ul><li>min: 7 tokens</li><li>mean: 20.78 tokens</li><li>max: 52 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 26.28 tokens</li><li>max: 43 tokens</li></ul> | <ul><li>0: ~71.50%</li><li>1: ~28.50%</li></ul> |
285
+ * Samples:
286
+ | query | doc | label |
287
+ |:-----------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------|:---------------|
288
+ | <code>Bagaimana perbandingan PNS pria dan wanita di berbagai golongan tahun 2014?</code> | <code>Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017</code> | <code>0</code> |
289
+ | <code>bagaimana perbandingan pns pria dan wanita di berbagai golongan tahun 2014?</code> | <code>Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017</code> | <code>0</code> |
290
+ | <code>BAGAIMANA PERBANDINGAN PNS PRIA DAN WANITA DI BERBAGAI GOLONGAN TAHUN 2014?</code> | <code>Rata-rata Pendapatan Bersih Berusaha Sendiri Menurut Provinsi dan Lapangan Pekerjaan Utama (ribu rupiah), 2017</code> | <code>0</code> |
291
+ * Loss: [<code>OnlineContrastiveLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#onlinecontrastiveloss)
292
+
293
+ ### Training Hyperparameters
294
+ #### Non-Default Hyperparameters
295
+
296
+ - `eval_strategy`: steps
297
+ - `per_device_train_batch_size`: 32
298
+ - `per_device_eval_batch_size`: 32
299
+ - `warmup_ratio`: 0.2
300
+ - `fp16`: True
301
+ - `load_best_model_at_end`: True
302
+ - `eval_on_start`: True
303
+
304
+ #### All Hyperparameters
305
+ <details><summary>Click to expand</summary>
306
+
307
+ - `overwrite_output_dir`: False
308
+ - `do_predict`: False
309
+ - `eval_strategy`: steps
310
+ - `prediction_loss_only`: True
311
+ - `per_device_train_batch_size`: 32
312
+ - `per_device_eval_batch_size`: 32
313
+ - `per_gpu_train_batch_size`: None
314
+ - `per_gpu_eval_batch_size`: None
315
+ - `gradient_accumulation_steps`: 1
316
+ - `eval_accumulation_steps`: None
317
+ - `torch_empty_cache_steps`: None
318
+ - `learning_rate`: 5e-05
319
+ - `weight_decay`: 0.0
320
+ - `adam_beta1`: 0.9
321
+ - `adam_beta2`: 0.999
322
+ - `adam_epsilon`: 1e-08
323
+ - `max_grad_norm`: 1.0
324
+ - `num_train_epochs`: 3
325
+ - `max_steps`: -1
326
+ - `lr_scheduler_type`: linear
327
+ - `lr_scheduler_kwargs`: {}
328
+ - `warmup_ratio`: 0.2
329
+ - `warmup_steps`: 0
330
+ - `log_level`: passive
331
+ - `log_level_replica`: warning
332
+ - `log_on_each_node`: True
333
+ - `logging_nan_inf_filter`: True
334
+ - `save_safetensors`: True
335
+ - `save_on_each_node`: False
336
+ - `save_only_model`: False
337
+ - `restore_callback_states_from_checkpoint`: False
338
+ - `no_cuda`: False
339
+ - `use_cpu`: False
340
+ - `use_mps_device`: False
341
+ - `seed`: 42
342
+ - `data_seed`: None
343
+ - `jit_mode_eval`: False
344
+ - `use_ipex`: False
345
+ - `bf16`: False
346
+ - `fp16`: True
347
+ - `fp16_opt_level`: O1
348
+ - `half_precision_backend`: auto
349
+ - `bf16_full_eval`: False
350
+ - `fp16_full_eval`: False
351
+ - `tf32`: None
352
+ - `local_rank`: 0
353
+ - `ddp_backend`: None
354
+ - `tpu_num_cores`: None
355
+ - `tpu_metrics_debug`: False
356
+ - `debug`: []
357
+ - `dataloader_drop_last`: False
358
+ - `dataloader_num_workers`: 0
359
+ - `dataloader_prefetch_factor`: None
360
+ - `past_index`: -1
361
+ - `disable_tqdm`: False
362
+ - `remove_unused_columns`: True
363
+ - `label_names`: None
364
+ - `load_best_model_at_end`: True
365
+ - `ignore_data_skip`: False
366
+ - `fsdp`: []
367
+ - `fsdp_min_num_params`: 0
368
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
369
+ - `fsdp_transformer_layer_cls_to_wrap`: None
370
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
371
+ - `deepspeed`: None
372
+ - `label_smoothing_factor`: 0.0
373
+ - `optim`: adamw_torch
374
+ - `optim_args`: None
375
+ - `adafactor`: False
376
+ - `group_by_length`: False
377
+ - `length_column_name`: length
378
+ - `ddp_find_unused_parameters`: None
379
+ - `ddp_bucket_cap_mb`: None
380
+ - `ddp_broadcast_buffers`: False
381
+ - `dataloader_pin_memory`: True
382
+ - `dataloader_persistent_workers`: False
383
+ - `skip_memory_metrics`: True
384
+ - `use_legacy_prediction_loop`: False
385
+ - `push_to_hub`: False
386
+ - `resume_from_checkpoint`: None
387
+ - `hub_model_id`: None
388
+ - `hub_strategy`: every_save
389
+ - `hub_private_repo`: None
390
+ - `hub_always_push`: False
391
+ - `gradient_checkpointing`: False
392
+ - `gradient_checkpointing_kwargs`: None
393
+ - `include_inputs_for_metrics`: False
394
+ - `include_for_metrics`: []
395
+ - `eval_do_concat_batches`: True
396
+ - `fp16_backend`: auto
397
+ - `push_to_hub_model_id`: None
398
+ - `push_to_hub_organization`: None
399
+ - `mp_parameters`:
400
+ - `auto_find_batch_size`: False
401
+ - `full_determinism`: False
402
+ - `torchdynamo`: None
403
+ - `ray_scope`: last
404
+ - `ddp_timeout`: 1800
405
+ - `torch_compile`: False
406
+ - `torch_compile_backend`: None
407
+ - `torch_compile_mode`: None
408
+ - `dispatch_batches`: None
409
+ - `split_batches`: None
410
+ - `include_tokens_per_second`: False
411
+ - `include_num_input_tokens_seen`: False
412
+ - `neftune_noise_alpha`: None
413
+ - `optim_target_modules`: None
414
+ - `batch_eval_metrics`: False
415
+ - `eval_on_start`: True
416
+ - `use_liger_kernel`: False
417
+ - `eval_use_gather_object`: False
418
+ - `average_tokens_across_devices`: False
419
+ - `prompts`: None
420
+ - `batch_sampler`: batch_sampler
421
+ - `multi_dataset_batch_sampler`: proportional
422
+
423
+ </details>
424
+
425
+ ### Training Logs
426
+ <details><summary>Click to expand</summary>
427
+
428
+ | Epoch | Step | Training Loss | Validation Loss | allstats-semantic-mini-v1_test_cosine_ap | allstats-semantic-mini-v1_dev_cosine_ap |
429
+ |:---------:|:--------:|:-------------:|:---------------:|:----------------------------------------:|:---------------------------------------:|
430
+ | -1 | -1 | - | - | 0.8910 | - |
431
+ | 0 | 0 | - | 1.0484 | - | 0.8789 |
432
+ | 0.025 | 20 | 1.0003 | 0.9175 | - | 0.8856 |
433
+ | 0.05 | 40 | 0.6667 | 0.6433 | - | 0.9010 |
434
+ | 0.075 | 60 | 0.5982 | 0.5203 | - | 0.9145 |
435
+ | 0.1 | 80 | 0.4476 | 0.4175 | - | 0.9344 |
436
+ | 0.125 | 100 | 0.3489 | 0.3152 | - | 0.9540 |
437
+ | 0.15 | 120 | 0.1643 | 0.2726 | - | 0.9602 |
438
+ | 0.175 | 140 | 0.2126 | 0.2525 | - | 0.9631 |
439
+ | 0.2 | 160 | 0.1797 | 0.2151 | - | 0.9715 |
440
+ | 0.225 | 180 | 0.1304 | 0.1895 | - | 0.9756 |
441
+ | 0.25 | 200 | 0.1714 | 0.2142 | - | 0.9767 |
442
+ | 0.275 | 220 | 0.1758 | 0.1840 | - | 0.9791 |
443
+ | 0.3 | 240 | 0.0562 | 0.1723 | - | 0.9801 |
444
+ | 0.325 | 260 | 0.0863 | 0.1656 | - | 0.9773 |
445
+ | 0.35 | 280 | 0.12 | 0.1806 | - | 0.9788 |
446
+ | 0.375 | 300 | 0.0982 | 0.1792 | - | 0.9769 |
447
+ | 0.4 | 320 | 0.0421 | 0.1724 | - | 0.9783 |
448
+ | 0.425 | 340 | 0.1078 | 0.2158 | - | 0.9733 |
449
+ | 0.45 | 360 | 0.0882 | 0.1501 | - | 0.9822 |
450
+ | 0.475 | 380 | 0.0251 | 0.1334 | - | 0.9843 |
451
+ | 0.5 | 400 | 0.0267 | 0.1238 | - | 0.9855 |
452
+ | 0.525 | 420 | 0.0899 | 0.1404 | - | 0.9859 |
453
+ | 0.55 | 440 | 0.0782 | 0.1253 | - | 0.9852 |
454
+ | 0.575 | 460 | 0.1209 | 0.1772 | - | 0.9768 |
455
+ | 0.6 | 480 | 0.0643 | 0.1817 | - | 0.9763 |
456
+ | 0.625 | 500 | 0.1051 | 0.2030 | - | 0.9748 |
457
+ | 0.65 | 520 | 0.0494 | 0.1405 | - | 0.9814 |
458
+ | 0.675 | 540 | 0.0548 | 0.1175 | - | 0.9831 |
459
+ | 0.7 | 560 | 0.121 | 0.1597 | - | 0.9819 |
460
+ | 0.725 | 580 | 0.0642 | 0.1675 | - | 0.9811 |
461
+ | 0.75 | 600 | 0.0618 | 0.1539 | - | 0.9827 |
462
+ | 0.775 | 620 | 0.0745 | 0.1149 | - | 0.9845 |
463
+ | 0.8 | 640 | 0.0452 | 0.1562 | - | 0.9797 |
464
+ | 0.825 | 660 | 0.0816 | 0.1580 | - | 0.9816 |
465
+ | 0.85 | 680 | 0.0957 | 0.1192 | - | 0.9830 |
466
+ | 0.875 | 700 | 0.06 | 0.1100 | - | 0.9863 |
467
+ | 0.9 | 720 | 0.018 | 0.1300 | - | 0.9822 |
468
+ | 0.925 | 740 | 0.0213 | 0.1267 | - | 0.9843 |
469
+ | 0.95 | 760 | 0.0263 | 0.1687 | - | 0.9796 |
470
+ | 0.975 | 780 | 0.032 | 0.1250 | - | 0.9849 |
471
+ | 1.0 | 800 | 0.065 | 0.1363 | - | 0.9828 |
472
+ | 1.025 | 820 | 0.0174 | 0.1394 | - | 0.9835 |
473
+ | 1.05 | 840 | 0.0568 | 0.1124 | - | 0.9849 |
474
+ | 1.075 | 860 | 0.0464 | 0.1174 | - | 0.9826 |
475
+ | 1.1 | 880 | 0.013 | 0.1178 | - | 0.9814 |
476
+ | 1.125 | 900 | 0.0331 | 0.1239 | - | 0.9812 |
477
+ | 1.15 | 920 | 0.0416 | 0.1240 | - | 0.9817 |
478
+ | 1.175 | 940 | 0.0111 | 0.1303 | - | 0.9840 |
479
+ | 1.2 | 960 | 0.0441 | 0.1156 | - | 0.9854 |
480
+ | 1.225 | 980 | 0.0243 | 0.0972 | - | 0.9879 |
481
+ | 1.25 | 1000 | 0.0 | 0.0917 | - | 0.9877 |
482
+ | 1.275 | 1020 | 0.0477 | 0.0863 | - | 0.9885 |
483
+ | 1.3 | 1040 | 0.0108 | 0.1029 | - | 0.9877 |
484
+ | 1.325 | 1060 | 0.0 | 0.1103 | - | 0.9869 |
485
+ | 1.35 | 1080 | 0.0134 | 0.1113 | - | 0.9871 |
486
+ | 1.375 | 1100 | 0.0 | 0.1146 | - | 0.9870 |
487
+ | 1.4 | 1120 | 0.0132 | 0.1218 | - | 0.9862 |
488
+ | 1.425 | 1140 | 0.0223 | 0.0948 | - | 0.9883 |
489
+ | 1.45 | 1160 | 0.0183 | 0.0883 | - | 0.9883 |
490
+ | 1.475 | 1180 | 0.0378 | 0.0961 | - | 0.9881 |
491
+ | 1.5 | 1200 | 0.0114 | 0.0961 | - | 0.9882 |
492
+ | 1.525 | 1220 | 0.0143 | 0.1020 | - | 0.9861 |
493
+ | 1.55 | 1240 | 0.0183 | 0.0867 | - | 0.9888 |
494
+ | 1.575 | 1260 | 0.0 | 0.0858 | - | 0.9892 |
495
+ | 1.6 | 1280 | 0.0 | 0.0858 | - | 0.9892 |
496
+ | 1.625 | 1300 | 0.0 | 0.0858 | - | 0.9892 |
497
+ | 1.65 | 1320 | 0.0172 | 0.0846 | - | 0.9896 |
498
+ | 1.675 | 1340 | 0.0153 | 0.0754 | - | 0.9917 |
499
+ | 1.7 | 1360 | 0.0163 | 0.0770 | - | 0.9913 |
500
+ | 1.725 | 1380 | 0.0167 | 0.0943 | - | 0.9901 |
501
+ | 1.75 | 1400 | 0.0148 | 0.0964 | - | 0.9899 |
502
+ | 1.775 | 1420 | 0.0065 | 0.0930 | - | 0.9902 |
503
+ | 1.8 | 1440 | 0.0 | 0.0945 | - | 0.9904 |
504
+ | 1.825 | 1460 | 0.0067 | 0.0991 | - | 0.9895 |
505
+ | 1.85 | 1480 | 0.0194 | 0.0996 | - | 0.9894 |
506
+ | 1.875 | 1500 | 0.0 | 0.0953 | - | 0.9903 |
507
+ | 1.9 | 1520 | 0.0236 | 0.0883 | - | 0.9906 |
508
+ | 1.925 | 1540 | 0.0111 | 0.0858 | - | 0.9904 |
509
+ | 1.95 | 1560 | 0.0 | 0.0878 | - | 0.9903 |
510
+ | 1.975 | 1580 | 0.0147 | 0.0849 | - | 0.9906 |
511
+ | 2.0 | 1600 | 0.0154 | 0.0852 | - | 0.9902 |
512
+ | 2.025 | 1620 | 0.0067 | 0.0861 | - | 0.9903 |
513
+ | 2.05 | 1640 | 0.019 | 0.0859 | - | 0.9907 |
514
+ | 2.075 | 1660 | 0.0083 | 0.0875 | - | 0.9908 |
515
+ | 2.1 | 1680 | 0.0067 | 0.0771 | - | 0.9917 |
516
+ | 2.125 | 1700 | 0.0 | 0.0773 | - | 0.9917 |
517
+ | 2.15 | 1720 | 0.0071 | 0.0771 | - | 0.9919 |
518
+ | 2.175 | 1740 | 0.0064 | 0.0756 | - | 0.9916 |
519
+ | 2.2 | 1760 | 0.0 | 0.0772 | - | 0.9916 |
520
+ | 2.225 | 1780 | 0.0 | 0.0772 | - | 0.9915 |
521
+ | 2.25 | 1800 | 0.0158 | 0.0734 | - | 0.9920 |
522
+ | 2.275 | 1820 | 0.0 | 0.0730 | - | 0.9920 |
523
+ | 2.3 | 1840 | 0.0 | 0.0733 | - | 0.9920 |
524
+ | 2.325 | 1860 | 0.0161 | 0.0681 | - | 0.9922 |
525
+ | 2.35 | 1880 | 0.0 | 0.0713 | - | 0.9920 |
526
+ | 2.375 | 1900 | 0.0 | 0.0721 | - | 0.9920 |
527
+ | 2.4 | 1920 | 0.0 | 0.0722 | - | 0.9920 |
528
+ | 2.425 | 1940 | 0.0064 | 0.0648 | - | 0.9928 |
529
+ | 2.45 | 1960 | 0.0068 | 0.0641 | - | 0.9930 |
530
+ | **2.475** | **1980** | **0.0069** | **0.0635** | **-** | **0.9929** |
531
+ | 2.5 | 2000 | 0.0066 | 0.0657 | - | 0.9929 |
532
+ | 2.525 | 2020 | 0.0 | 0.0657 | - | 0.9930 |
533
+ | 2.55 | 2040 | 0.0139 | 0.0657 | - | 0.9931 |
534
+ | 2.575 | 2060 | 0.0 | 0.0667 | - | 0.9931 |
535
+ | 2.6 | 2080 | 0.0 | 0.0666 | - | 0.9931 |
536
+ | 2.625 | 2100 | 0.0 | 0.0666 | - | 0.9931 |
537
+ | 2.65 | 2120 | 0.0 | 0.0666 | - | 0.9931 |
538
+ | 2.675 | 2140 | 0.0 | 0.0667 | - | 0.9931 |
539
+ | 2.7 | 2160 | 0.0 | 0.0666 | - | 0.9931 |
540
+ | 2.725 | 2180 | 0.0 | 0.0666 | - | 0.9931 |
541
+ | 2.75 | 2200 | 0.0071 | 0.0665 | - | 0.9931 |
542
+ | 2.775 | 2220 | 0.0 | 0.0671 | - | 0.9931 |
543
+ | 2.8 | 2240 | 0.0071 | 0.0692 | - | 0.9928 |
544
+ | 2.825 | 2260 | 0.0 | 0.0700 | - | 0.9927 |
545
+ | 2.85 | 2280 | 0.0068 | 0.0688 | - | 0.9927 |
546
+ | 2.875 | 2300 | 0.0 | 0.0688 | - | 0.9927 |
547
+ | 2.9 | 2320 | 0.0 | 0.0688 | - | 0.9927 |
548
+ | 2.925 | 2340 | 0.0 | 0.0688 | - | 0.9927 |
549
+ | 2.95 | 2360 | 0.0 | 0.0688 | - | 0.9927 |
550
+ | 2.975 | 2380 | 0.0 | 0.0688 | - | 0.9927 |
551
+ | 3.0 | 2400 | 0.0 | 0.0688 | - | 0.9927 |
552
+ | -1 | -1 | - | - | 0.9957 | - |
553
+
554
+ * The bold row denotes the saved checkpoint.
555
+ </details>
556
+
557
+ ### Framework Versions
558
+ - Python: 3.10.12
559
+ - Sentence Transformers: 3.4.0
560
+ - Transformers: 4.48.1
561
+ - PyTorch: 2.5.1+cu124
562
+ - Accelerate: 1.3.0
563
+ - Datasets: 3.2.0
564
+ - Tokenizers: 0.21.0
565
+
566
+ ## Citation
567
+
568
+ ### BibTeX
569
+
570
+ #### Sentence Transformers
571
+ ```bibtex
572
+ @inproceedings{reimers-2019-sentence-bert,
573
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
574
+ author = "Reimers, Nils and Gurevych, Iryna",
575
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
576
+ month = "11",
577
+ year = "2019",
578
+ publisher = "Association for Computational Linguistics",
579
+ url = "https://arxiv.org/abs/1908.10084",
580
+ }
581
+ ```
582
+
583
+ <!--
584
+ ## Glossary
585
+
586
+ *Clearly define terms in order to be accessible across audiences.*
587
+ -->
588
+
589
+ <!--
590
+ ## Model Card Authors
591
+
592
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
593
+ -->
594
+
595
+ <!--
596
+ ## Model Card Contact
597
+
598
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
599
+ -->
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "sentence-transformers/paraphrase-multilingual-miniLM-L12-v2",
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": 384,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 1536,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.48.1",
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": "3.4.0",
4
+ "transformers": "4.48.1",
5
+ "pytorch": "2.5.1+cu124"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": "cosine"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8bb4b4e0f0a70669214912c3b144a8e3f8f61f09c9fc9bb1ea69cfcd07835d6
3
+ size 470637416
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": 128,
3
+ "do_lower_case": false
4
+ }
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": true,
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:cad551d5600a84242d0973327029452a1e3672ba6313c2a3c3d69c4310e12719
3
+ size 17082987
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": false,
46
+ "cls_token": "<s>",
47
+ "do_lower_case": true,
48
+ "eos_token": "</s>",
49
+ "extra_special_tokens": {},
50
+ "mask_token": "<mask>",
51
+ "max_length": 128,
52
+ "model_max_length": 128,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "<pad>",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "</s>",
58
+ "stride": 0,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "longest_first",
64
+ "unk_token": "<unk>"
65
+ }
unigram.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da145b5e7700ae40f16691ec32a0b1fdc1ee3298db22a31ea55f57a966c4a65d
3
+ size 14763260