Hugging Face's logo

mertcobanov
/
mpnet-base-all-nli-triplet-turkish-v3

Sentence Similarity
sentence-transformers
Safetensors
English
mpnet
feature-extraction
Generated from Trainer
dataset_size:13842
loss:MultipleNegativesRankingLoss
Eval Results
Inference Endpoints
Model card Files Community
mertcobanov commited on
Commit
e1e0c18
•
1 Parent(s): 25a022e

Add new SentenceTransformer model

Browse files
Files changed (11) hide show
  1. 1_Pooling/config.json +10 -0
  2. README.md +519 -0
  3. config.json +24 -0
  4. config_sentence_transformers.json +10 -0
  5. model.safetensors +3 -0
  6. modules.json +14 -0
  7. sentence_bert_config.json +4 -0
  8. special_tokens_map.json +51 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +65 -0
  11. vocab.txt +0 -0
1_Pooling/config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "word_embedding_dimension": 768,
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 +1,519 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ license: apache-2.0
5
+ tags:
6
+ - sentence-transformers
7
+ - sentence-similarity
8
+ - feature-extraction
9
+ - generated_from_trainer
10
+ - dataset_size:13842
11
+ - loss:MultipleNegativesRankingLoss
12
+ base_model: microsoft/mpnet-base
13
+ widget:
14
+ - source_sentence: Bir adam bir elinde kahve fincanı, diğer elinde tuvalet fırçası
15
+ ile tuvaletin önünde duruyor.
16
+ sentences:
17
+ - Şef ve orkestra oturmuyor.
18
+ - Bir adam bir banyoda duruyor.
19
+ - Bir adam kahve demlemeye çalışıyor.
20
+ - source_sentence: Sarı ceketli ve siyah pantolonlu iki adam madalyalara sahip.
21
+ sentences:
22
+ - Erkeklere bir noktada bir ödül verilmiştir.
23
+ - Başlangıçtaki net ölçek faydası, ücret primleri olsun ya da olmasın, pozitiftir.
24
+ - Adamlar düz kırmızı ceketler ve mavi pantolonlar giymiş.
25
+ - source_sentence: 'Restoran zinciri içi: Planet Hollywood, çeşitli film hatıraları
26
+ mekânı süslüyor.'
27
+ sentences:
28
+ - Kadın bir şey tutuyor.
29
+ - Bir restoranın içi.
30
+ - Yeni gümüş makinelerin bulunduğu bir çamaşırhane içi.
31
+ - source_sentence: İki çocuk, binanın yakınındaki kaldırımda sokakta koşuyor.
32
+ sentences:
33
+ - Çocuklar dışarıda.
34
+ - Bazı odaların dışına balkonları vardır.
35
+ - Çocuklar içeride.
36
+ - source_sentence: Ağaçlarla çevrili bulvar denize üç bloktan daha az uzanıyor.
37
+ sentences:
38
+ - Deniz üç sokak bile uzakta değil.
39
+ - Çocuk başını duvardaki bir delikten geçiriyor.
40
+ - Denize ulaşmak için caddeden iki mil yol almanız gerekiyor.
41
+ datasets:
42
+ - mertcobanov/all-nli-triplets-turkish
43
+ pipeline_tag: sentence-similarity
44
+ library_name: sentence-transformers
45
+ metrics:
46
+ - cosine_accuracy
47
+ model-index:
48
+ - name: MPNet base trained on AllNLI-turkish triplets
49
+ results:
50
+ - task:
51
+ type: triplet
52
+ name: Triplet
53
+ dataset:
54
+ name: all nli dev turkish
55
+ type: all-nli-dev-turkish
56
+ metrics:
57
+ - type: cosine_accuracy
58
+ value: 0.7422539489671932
59
+ name: Cosine Accuracy
60
+ - task:
61
+ type: triplet
62
+ name: Triplet
63
+ dataset:
64
+ name: all nli test turkish
65
+ type: all-nli-test-turkish
66
+ metrics:
67
+ - type: cosine_accuracy
68
+ value: 0.7503404448479346
69
+ name: Cosine Accuracy
70
+ ---
71
+
72
+ # MPNet base trained on AllNLI-turkish triplets
73
+
74
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [microsoft/mpnet-base](https://huggingface.co/microsoft/mpnet-base) on the [all-nli-triplets-turkish](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish) dataset. 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.
75
+
76
+ ## Model Details
77
+
78
+ ### Model Description
79
+ - **Model Type:** Sentence Transformer
80
+ - **Base model:** [microsoft/mpnet-base](https://huggingface.co/microsoft/mpnet-base) <!-- at revision 6996ce1e91bd2a9c7d7f61daec37463394f73f09 -->
81
+ - **Maximum Sequence Length:** 512 tokens
82
+ - **Output Dimensionality:** 768 dimensions
83
+ - **Similarity Function:** Cosine Similarity
84
+ - **Training Dataset:**
85
+ - [all-nli-triplets-turkish](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish)
86
+ - **Language:** en
87
+ - **License:** apache-2.0
88
+
89
+ ### Model Sources
90
+
91
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
92
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
93
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
94
+
95
+ ### Full Model Architecture
96
+
97
+ ```
98
+ SentenceTransformer(
99
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel
100
+ (1): Pooling({'word_embedding_dimension': 768, '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})
101
+ )
102
+ ```
103
+
104
+ ## Usage
105
+
106
+ ### Direct Usage (Sentence Transformers)
107
+
108
+ First install the Sentence Transformers library:
109
+
110
+ ```bash
111
+ pip install -U sentence-transformers
112
+ ```
113
+
114
+ Then you can load this model and run inference.
115
+ ```python
116
+ from sentence_transformers import SentenceTransformer
117
+
118
+ # Download from the 🤗 Hub
119
+ model = SentenceTransformer("mertcobanov/mpnet-base-all-nli-triplet-turkish-v3")
120
+ # Run inference
121
+ sentences = [
122
+ 'Ağaçlarla çevrili bulvar denize üç bloktan daha az uzanıyor.',
123
+ 'Deniz üç sokak bile uzakta değil.',
124
+ 'Denize ulaşmak için caddeden iki mil yol almanız gerekiyor.',
125
+ ]
126
+ embeddings = model.encode(sentences)
127
+ print(embeddings.shape)
128
+ # [3, 768]
129
+
130
+ # Get the similarity scores for the embeddings
131
+ similarities = model.similarity(embeddings, embeddings)
132
+ print(similarities.shape)
133
+ # [3, 3]
134
+ ```
135
+
136
+ <!--
137
+ ### Direct Usage (Transformers)
138
+
139
+ <details><summary>Click to see the direct usage in Transformers</summary>
140
+
141
+ </details>
142
+ -->
143
+
144
+ <!--
145
+ ### Downstream Usage (Sentence Transformers)
146
+
147
+ You can finetune this model on your own dataset.
148
+
149
+ <details><summary>Click to expand</summary>
150
+
151
+ </details>
152
+ -->
153
+
154
+ <!--
155
+ ### Out-of-Scope Use
156
+
157
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
158
+ -->
159
+
160
+ ## Evaluation
161
+
162
+ ### Metrics
163
+
164
+ #### Triplet
165
+
166
+ * Datasets: `all-nli-dev-turkish` and `all-nli-test-turkish`
167
+ * Evaluated with [<code>TripletEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.TripletEvaluator)
168
+
169
+ | Metric | all-nli-dev-turkish | all-nli-test-turkish |
170
+ |:--------------------|:--------------------|:---------------------|
171
+ | **cosine_accuracy** | **0.7423** | **0.7503** |
172
+
173
+ <!--
174
+ ## Bias, Risks and Limitations
175
+
176
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
177
+ -->
178
+
179
+ <!--
180
+ ### Recommendations
181
+
182
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
183
+ -->
184
+
185
+ ## Training Details
186
+
187
+ ### Training Dataset
188
+
189
+ #### all-nli-triplets-turkish
190
+
191
+ * Dataset: [all-nli-triplets-turkish](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish) at [bff203b](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish/tree/bff203b01bbf5b818f7ad85be0adbe8d64eba9ee)
192
+ * Size: 13,842 training samples
193
+ * Columns: <code>anchor_translated</code>, <code>positive_translated</code>, and <code>negative_translated</code>
194
+ * Approximate statistics based on the first 1000 samples:
195
+ | | anchor_translated | positive_translated | negative_translated |
196
+ |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
197
+ | type | string | string | string |
198
+ | details | <ul><li>min: 8 tokens</li><li>mean: 13.42 tokens</li><li>max: 95 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 31.64 tokens</li><li>max: 93 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 32.03 tokens</li><li>max: 89 tokens</li></ul> |
199
+ * Samples:
200
+ | anchor_translated | positive_translated | negative_translated |
201
+ |:-----------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------|
202
+ | <code>Asyalı okul çocukları birbirlerinin omuzlarında oturuyor.</code> | <code>Okul çocukları bir arada</code> | <code>Asyalı fabrika işçileri oturuyor.</code> |
203
+ | <code>İnsanlar dışarıda.</code> | <code>Arka planda resmi kıyafetler giymiş bir grup insan var ve beyaz gömlekli, haki pantolonlu bir adam toprak yoldan yeşil çimenlere atlıyor.</code> | <code>Bir odada üç kişiyle birlikte büyük bir kamera tutan bir adam.</code> |
204
+ | <code>Bir adam dışarıda.</code> | <code>Adam yarış sırasında yan sepetten bir su birikintisine düşer.</code> | <code>Beyaz bir sarık sarmış gömleksiz bir adam bir ağaç gövdesine tırmanıyor.</code> |
205
+ * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
206
+ ```json
207
+ {
208
+ "scale": 20.0,
209
+ "similarity_fct": "cos_sim"
210
+ }
211
+ ```
212
+
213
+ ### Evaluation Dataset
214
+
215
+ #### all-nli-triplets-turkish
216
+
217
+ * Dataset: [all-nli-triplets-turkish](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish) at [bff203b](https://huggingface.co/datasets/mertcobanov/all-nli-triplets-turkish/tree/bff203b01bbf5b818f7ad85be0adbe8d64eba9ee)
218
+ * Size: 6,584 evaluation samples
219
+ * Columns: <code>anchor_translated</code>, <code>positive_translated</code>, and <code>negative_translated</code>
220
+ * Approximate statistics based on the first 1000 samples:
221
+ | | anchor_translated | positive_translated | negative_translated |
222
+ |:--------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
223
+ | type | string | string | string |
224
+ | details | <ul><li>min: 5 tokens</li><li>mean: 42.62 tokens</li><li>max: 192 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 22.58 tokens</li><li>max: 77 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 22.07 tokens</li><li>max: 65 tokens</li></ul> |
225
+ * Samples:
226
+ | anchor_translated | positive_translated | negative_translated |
227
+ |:--------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------|
228
+ | <code>Ayrıca, bu özel tüketim vergileri, diğer vergiler gibi, hükümetin ödeme zorunluluğunu sağlama yetkisini kullanarak belirlenir.</code> | <code>Hükümetin ödeme zorlaması, özel tüketim vergilerinin nasıl hesaplandığını belirler.</code> | <code>Özel tüketim vergileri genel kuralın bir istisnasıdır ve aslında GSYİH payına dayalı olarak belirlenir.</code> |
229
+ | <code>Gri bir sweatshirt giymiş bir sanatçı, canlı renklerde bir kasaba tablosu üzerinde çalışıyor.</code> | <code>Bir ressam gri giysiler içinde bir kasabanın resmini yapıyor.</code> | <code>Bir kişi bir beyzbol sopası tutuyor ve gelen bir atış için planda bekliyor.</code> |
230
+ | <code>İmkansız.</code> | <code>Yapılamaz.</code> | <code>Tamamen mümkün.</code> |
231
+ * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
232
+ ```json
233
+ {
234
+ "scale": 20.0,
235
+ "similarity_fct": "cos_sim"
236
+ }
237
+ ```
238
+
239
+ ### Training Hyperparameters
240
+ #### Non-Default Hyperparameters
241
+
242
+ - `eval_strategy`: steps
243
+ - `per_device_train_batch_size`: 16
244
+ - `per_device_eval_batch_size`: 16
245
+ - `learning_rate`: 2e-05
246
+ - `num_train_epochs`: 10
247
+ - `warmup_ratio`: 0.1
248
+ - `fp16`: True
249
+ - `batch_sampler`: no_duplicates
250
+
251
+ #### All Hyperparameters
252
+ <details><summary>Click to expand</summary>
253
+
254
+ - `overwrite_output_dir`: False
255
+ - `do_predict`: False
256
+ - `eval_strategy`: steps
257
+ - `prediction_loss_only`: True
258
+ - `per_device_train_batch_size`: 16
259
+ - `per_device_eval_batch_size`: 16
260
+ - `per_gpu_train_batch_size`: None
261
+ - `per_gpu_eval_batch_size`: None
262
+ - `gradient_accumulation_steps`: 1
263
+ - `eval_accumulation_steps`: None
264
+ - `torch_empty_cache_steps`: None
265
+ - `learning_rate`: 2e-05
266
+ - `weight_decay`: 0.0
267
+ - `adam_beta1`: 0.9
268
+ - `adam_beta2`: 0.999
269
+ - `adam_epsilon`: 1e-08
270
+ - `max_grad_norm`: 1.0
271
+ - `num_train_epochs`: 10
272
+ - `max_steps`: -1
273
+ - `lr_scheduler_type`: linear
274
+ - `lr_scheduler_kwargs`: {}
275
+ - `warmup_ratio`: 0.1
276
+ - `warmup_steps`: 0
277
+ - `log_level`: passive
278
+ - `log_level_replica`: warning
279
+ - `log_on_each_node`: True
280
+ - `logging_nan_inf_filter`: True
281
+ - `save_safetensors`: True
282
+ - `save_on_each_node`: False
283
+ - `save_only_model`: False
284
+ - `restore_callback_states_from_checkpoint`: False
285
+ - `no_cuda`: False
286
+ - `use_cpu`: False
287
+ - `use_mps_device`: False
288
+ - `seed`: 42
289
+ - `data_seed`: None
290
+ - `jit_mode_eval`: False
291
+ - `use_ipex`: False
292
+ - `bf16`: False
293
+ - `fp16`: True
294
+ - `fp16_opt_level`: O1
295
+ - `half_precision_backend`: auto
296
+ - `bf16_full_eval`: False
297
+ - `fp16_full_eval`: False
298
+ - `tf32`: None
299
+ - `local_rank`: 0
300
+ - `ddp_backend`: None
301
+ - `tpu_num_cores`: None
302
+ - `tpu_metrics_debug`: False
303
+ - `debug`: []
304
+ - `dataloader_drop_last`: False
305
+ - `dataloader_num_workers`: 0
306
+ - `dataloader_prefetch_factor`: None
307
+ - `past_index`: -1
308
+ - `disable_tqdm`: False
309
+ - `remove_unused_columns`: True
310
+ - `label_names`: None
311
+ - `load_best_model_at_end`: False
312
+ - `ignore_data_skip`: False
313
+ - `fsdp`: []
314
+ - `fsdp_min_num_params`: 0
315
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
316
+ - `fsdp_transformer_layer_cls_to_wrap`: None
317
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
318
+ - `deepspeed`: None
319
+ - `label_smoothing_factor`: 0.0
320
+ - `optim`: adamw_torch
321
+ - `optim_args`: None
322
+ - `adafactor`: False
323
+ - `group_by_length`: False
324
+ - `length_column_name`: length
325
+ - `ddp_find_unused_parameters`: None
326
+ - `ddp_bucket_cap_mb`: None
327
+ - `ddp_broadcast_buffers`: False
328
+ - `dataloader_pin_memory`: True
329
+ - `dataloader_persistent_workers`: False
330
+ - `skip_memory_metrics`: True
331
+ - `use_legacy_prediction_loop`: False
332
+ - `push_to_hub`: False
333
+ - `resume_from_checkpoint`: None
334
+ - `hub_model_id`: None
335
+ - `hub_strategy`: every_save
336
+ - `hub_private_repo`: False
337
+ - `hub_always_push`: False
338
+ - `gradient_checkpointing`: False
339
+ - `gradient_checkpointing_kwargs`: None
340
+ - `include_inputs_for_metrics`: False
341
+ - `include_for_metrics`: []
342
+ - `eval_do_concat_batches`: True
343
+ - `fp16_backend`: auto
344
+ - `push_to_hub_model_id`: None
345
+ - `push_to_hub_organization`: None
346
+ - `mp_parameters`:
347
+ - `auto_find_batch_size`: False
348
+ - `full_determinism`: False
349
+ - `torchdynamo`: None
350
+ - `ray_scope`: last
351
+ - `ddp_timeout`: 1800
352
+ - `torch_compile`: False
353
+ - `torch_compile_backend`: None
354
+ - `torch_compile_mode`: None
355
+ - `dispatch_batches`: None
356
+ - `split_batches`: None
357
+ - `include_tokens_per_second`: False
358
+ - `include_num_input_tokens_seen`: False
359
+ - `neftune_noise_alpha`: None
360
+ - `optim_target_modules`: None
361
+ - `batch_eval_metrics`: False
362
+ - `eval_on_start`: False
363
+ - `use_liger_kernel`: False
364
+ - `eval_use_gather_object`: False
365
+ - `average_tokens_across_devices`: False
366
+ - `prompts`: None
367
+ - `batch_sampler`: no_duplicates
368
+ - `multi_dataset_batch_sampler`: proportional
369
+
370
+ </details>
371
+
372
+ ### Training Logs
373
+ | Epoch | Step | Training Loss | Validation Loss | all-nli-dev-turkish_cosine_accuracy | all-nli-test-turkish_cosine_accuracy |
374
+ |:------:|:----:|:-------------:|:---------------:|:-----------------------------------:|:------------------------------------:|
375
+ | 0 | 0 | - | - | 0.6092 | - |
376
+ | 0.1155 | 100 | 3.3654 | 2.9084 | 0.6624 | - |
377
+ | 0.2309 | 200 | 2.6321 | 1.7277 | 0.7395 | - |
378
+ | 0.3464 | 300 | 1.9629 | 1.5000 | 0.7512 | - |
379
+ | 0.4619 | 400 | 1.6662 | 1.4965 | 0.7494 | - |
380
+ | 0.5774 | 500 | 1.4712 | 1.5374 | 0.7418 | - |
381
+ | 0.6928 | 600 | 1.0429 | 1.6301 | 0.7360 | - |
382
+ | 0.8083 | 700 | 0.8995 | 2.1626 | 0.7044 | - |
383
+ | 0.9238 | 800 | 0.7269 | 2.0440 | 0.6996 | - |
384
+ | 1.0381 | 900 | 1.0584 | 1.6714 | 0.7438 | - |
385
+ | 1.1536 | 1000 | 1.1864 | 1.5326 | 0.7495 | - |
386
+ | 1.2691 | 1100 | 1.0193 | 1.4498 | 0.7518 | - |
387
+ | 1.3845 | 1200 | 0.8237 | 1.5399 | 0.7506 | - |
388
+ | 1.5 | 1300 | 0.8279 | 1.6747 | 0.7521 | - |
389
+ | 1.6155 | 1400 | 0.626 | 1.5776 | 0.7453 | - |
390
+ | 1.7309 | 1500 | 0.5396 | 1.8877 | 0.7139 | - |
391
+ | 1.8464 | 1600 | 0.4294 | 2.2258 | 0.6947 | - |
392
+ | 1.9619 | 1700 | 0.4988 | 1.8753 | 0.7204 | - |
393
+ | 2.0762 | 1800 | 0.6987 | 1.5408 | 0.7524 | - |
394
+ | 2.1917 | 1900 | 0.6684 | 1.4434 | 0.7618 | - |
395
+ | 2.3072 | 2000 | 0.6072 | 1.4840 | 0.7520 | - |
396
+ | 2.4226 | 2100 | 0.5081 | 1.5225 | 0.7561 | - |
397
+ | 2.5381 | 2200 | 0.5216 | 1.5280 | 0.7514 | - |
398
+ | 2.6536 | 2300 | 0.2627 | 1.8830 | 0.7227 | - |
399
+ | 2.7691 | 2400 | 0.2585 | 1.9529 | 0.7221 | - |
400
+ | 2.8845 | 2500 | 0.129 | 2.2323 | 0.7047 | - |
401
+ | 3.0 | 2600 | 0.1698 | 2.2904 | 0.7063 | - |
402
+ | 3.1143 | 2700 | 0.5559 | 1.6110 | 0.7553 | - |
403
+ | 3.2298 | 2800 | 0.4356 | 1.5544 | 0.7508 | - |
404
+ | 3.3453 | 2900 | 0.3886 | 1.5437 | 0.7539 | - |
405
+ | 3.4607 | 3000 | 0.3573 | 1.6262 | 0.7539 | - |
406
+ | 3.5762 | 3100 | 0.2652 | 1.8391 | 0.7321 | - |
407
+ | 3.6917 | 3200 | 0.0765 | 2.0359 | 0.7186 | - |
408
+ | 3.8072 | 3300 | 0.0871 | 2.0946 | 0.7262 | - |
409
+ | 3.9226 | 3400 | 0.0586 | 2.2168 | 0.7093 | - |
410
+ | 4.0370 | 3500 | 0.1755 | 1.7567 | 0.7462 | - |
411
+ | 4.1524 | 3600 | 0.3397 | 1.7735 | 0.7442 | - |
412
+ | 4.2679 | 3700 | 0.3067 | 1.7475 | 0.7497 | - |
413
+ | 4.3834 | 3800 | 0.246 | 1.7075 | 0.7476 | - |
414
+ | 4.4988 | 3900 | 0.253 | 1.7648 | 0.7483 | - |
415
+ | 4.6143 | 4000 | 0.1223 | 1.9139 | 0.7246 | - |
416
+ | 4.7298 | 4100 | 0.0453 | 2.1138 | 0.7152 | - |
417
+ | 4.8453 | 4200 | 0.0241 | 2.2354 | 0.7240 | - |
418
+ | 4.9607 | 4300 | 0.0363 | 2.3080 | 0.7251 | - |
419
+ | 5.0751 | 4400 | 0.1897 | 1.7394 | 0.7494 | - |
420
+ | 5.1905 | 4500 | 0.2114 | 1.6929 | 0.7524 | - |
421
+ | 5.3060 | 4600 | 0.2101 | 1.7402 | 0.7556 | - |
422
+ | 5.4215 | 4700 | 0.1471 | 1.7990 | 0.7445 | - |
423
+ | 5.5370 | 4800 | 0.1783 | 1.8060 | 0.7456 | - |
424
+ | 5.6524 | 4900 | 0.0215 | 2.0118 | 0.7325 | - |
425
+ | 5.7679 | 5000 | 0.0083 | 2.0766 | 0.7265 | - |
426
+ | 5.8834 | 5100 | 0.0138 | 2.2054 | 0.7201 | - |
427
+ | 5.9988 | 5200 | 0.0144 | 2.1667 | 0.7164 | - |
428
+ | 6.1132 | 5300 | 0.2023 | 1.7309 | 0.7543 | - |
429
+ | 6.2286 | 5400 | 0.1356 | 1.6685 | 0.7622 | - |
430
+ | 6.3441 | 5500 | 0.1307 | 1.7292 | 0.7527 | - |
431
+ | 6.4596 | 5600 | 0.1222 | 1.8403 | 0.7435 | - |
432
+ | 6.5751 | 5700 | 0.1049 | 1.8456 | 0.7394 | - |
433
+ | 6.6905 | 5800 | 0.0051 | 1.9898 | 0.7362 | - |
434
+ | 6.8060 | 5900 | 0.0131 | 2.0532 | 0.7310 | - |
435
+ | 6.9215 | 6000 | 0.0132 | 2.2237 | 0.7186 | - |
436
+ | 7.0358 | 6100 | 0.0453 | 1.8965 | 0.7397 | - |
437
+ | 7.1513 | 6200 | 0.1109 | 1.7195 | 0.7550 | - |
438
+ | 7.2667 | 6300 | 0.1002 | 1.7547 | 0.7530 | - |
439
+ | 7.3822 | 6400 | 0.0768 | 1.7701 | 0.7433 | - |
440
+ | 7.4977 | 6500 | 0.0907 | 1.8472 | 0.7406 | - |
441
+ | 7.6132 | 6600 | 0.038 | 1.9162 | 0.7377 | - |
442
+ | 7.7286 | 6700 | 0.0151 | 1.9407 | 0.7312 | - |
443
+ | 7.8441 | 6800 | 0.0087 | 1.9657 | 0.7289 | - |
444
+ | 7.9596 | 6900 | 0.0104 | 2.0302 | 0.7227 | - |
445
+ | 8.0739 | 7000 | 0.0727 | 1.8692 | 0.7514 | - |
446
+ | 8.1894 | 7100 | 0.0733 | 1.8039 | 0.7520 | - |
447
+ | 8.3048 | 7200 | 0.0728 | 1.7400 | 0.7539 | - |
448
+ | 8.4203 | 7300 | 0.0537 | 1.8062 | 0.7461 | - |
449
+ | 8.5358 | 7400 | 0.059 | 1.8469 | 0.7489 | - |
450
+ | 8.6513 | 7500 | 0.0089 | 1.9033 | 0.7403 | - |
451
+ | 8.7667 | 7600 | 0.0034 | 1.9683 | 0.7354 | - |
452
+ | 8.8822 | 7700 | 0.0018 | 2.0075 | 0.7366 | - |
453
+ | 8.9977 | 7800 | 0.0023 | 2.0646 | 0.7322 | - |
454
+ | 9.1120 | 7900 | 0.0642 | 1.9063 | 0.7430 | - |
455
+ | 9.2275 | 8000 | 0.0596 | 1.8492 | 0.7468 | - |
456
+ | 9.3430 | 8100 | 0.0479 | 1.8180 | 0.7517 | - |
457
+ | 9.4584 | 8200 | 0.0561 | 1.8122 | 0.7468 | - |
458
+ | 9.5739 | 8300 | 0.0311 | 1.8528 | 0.7456 | - |
459
+ | 9.6894 | 8400 | 0.0069 | 1.8778 | 0.7447 | - |
460
+ | 9.8048 | 8500 | 0.0027 | 1.8989 | 0.7423 | - |
461
+ | 9.9203 | 8600 | 0.0093 | 1.9089 | 0.7423 | - |
462
+ | 9.9896 | 8660 | - | - | - | 0.7503 |
463
+
464
+
465
+ ### Framework Versions
466
+ - Python: 3.10.14
467
+ - Sentence Transformers: 3.3.1
468
+ - Transformers: 4.46.3
469
+ - PyTorch: 2.3.0
470
+ - Accelerate: 1.1.1
471
+ - Datasets: 3.1.0
472
+ - Tokenizers: 0.20.3
473
+
474
+ ## Citation
475
+
476
+ ### BibTeX
477
+
478
+ #### Sentence Transformers
479
+ ```bibtex
480
+ @inproceedings{reimers-2019-sentence-bert,
481
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
482
+ author = "Reimers, Nils and Gurevych, Iryna",
483
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
484
+ month = "11",
485
+ year = "2019",
486
+ publisher = "Association for Computational Linguistics",
487
+ url = "https://arxiv.org/abs/1908.10084",
488
+ }
489
+ ```
490
+
491
+ #### MultipleNegativesRankingLoss
492
+ ```bibtex
493
+ @misc{henderson2017efficient,
494
+ title={Efficient Natural Language Response Suggestion for Smart Reply},
495
+ 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},
496
+ year={2017},
497
+ eprint={1705.00652},
498
+ archivePrefix={arXiv},
499
+ primaryClass={cs.CL}
500
+ }
501
+ ```
502
+
503
+ <!--
504
+ ## Glossary
505
+
506
+ *Clearly define terms in order to be accessible across audiences.*
507
+ -->
508
+
509
+ <!--
510
+ ## Model Card Authors
511
+
512
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
513
+ -->
514
+
515
+ <!--
516
+ ## Model Card Contact
517
+
518
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
519
+ -->
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/mpnet-base",
3
+ "architectures": [
4
+ "MPNetModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-05,
15
+ "max_position_embeddings": 514,
16
+ "model_type": "mpnet",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 1,
20
+ "relative_attention_num_buckets": 32,
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.46.3",
23
+ "vocab_size": 30527
24
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.3.1",
4
+ "transformers": "4.46.3",
5
+ "pytorch": "2.3.0"
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:c6e32d54b9d7b27ed3fa086155871fa2763982becf26a8bb4f58c25362089a07
3
+ size 437967672
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": 512,
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": true,
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": true,
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
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<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": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "104": {
36
+ "content": "[UNK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "30526": {
44
+ "content": "<mask>",
45
+ "lstrip": true,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "cls_token": "<s>",
55
+ "do_lower_case": true,
56
+ "eos_token": "</s>",
57
+ "mask_token": "<mask>",
58
+ "model_max_length": 512,
59
+ "pad_token": "<pad>",
60
+ "sep_token": "</s>",
61
+ "strip_accents": null,
62
+ "tokenize_chinese_chars": true,
63
+ "tokenizer_class": "MPNetTokenizer",
64
+ "unk_token": "[UNK]"
65
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff