scr17 commited on
Commit
b7cb643
·
1 Parent(s): 7826451

Upload fine-tuned model

Browse files
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 +1,440 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - sentence-transformers
4
+ - sentence-similarity
5
+ - feature-extraction
6
+ - generated_from_trainer
7
+ - dataset_size:5000
8
+ - loss:CosineSimilarityLoss
9
+ base_model: sentence-transformers/all-MiniLM-L6-v2
10
+ widget:
11
+ - source_sentence: Machine Learning Engineer. We are looking for a Machine Learning
12
+ Engineer to join our growing team and work on exciting projects.. Energy officer
13
+ later analysis.; Represent doctor must amount first new.; Standard store herself
14
+ buy.; Attorney later road drive high could new.; Public near program language..
15
+ Docker; Azure; Linux; JavaScript; React. SQL; DevOps; Linux; Python; TensorFlow;
16
+ C#
17
+ sentences:
18
+ - Backend Developer. Baby consider fall go. Year role financial firm physical prepare
19
+ wear financial. Mission training research me mouth home partner.. Article bad
20
+ style.; On see attorney traditional price reflect tough.; Pay training I.; President
21
+ the these.; Mouth close debate world nor sport security.. Docker; C#; Flask; NoSQL;
22
+ Django; Node.js; Cybersecurity. Operation low rich drive.; Receive middle likely.
23
+ - Cybersecurity Analyst. What light amount modern security receive. Build book street
24
+ challenge.. Light choice TV.; Beat piece usually day bar cost.; True government
25
+ sea training door.; Popular who also situation step.. React; TypeScript; NoSQL;
26
+ Deep Learning; C#; TensorFlow; Django. Hand behavior market religious four might
27
+ size.; Middle central knowledge fast rise all really.; Become fight argue.; Able
28
+ hundred force response.
29
+ - AI Researcher. Rather well administration police seat stand. Red produce yeah
30
+ site run fly purpose face.. Yourself address might expect his budget bill.; Later
31
+ top focus guess occur hour.; Have turn quickly help well its.; However research
32
+ visit.; Commercial building especially capital system each.. Python; Machine Learning;
33
+ Terraform; Deep Learning; Java; Cybersecurity; Linux. At exactly story letter
34
+ dream.; Paper experience control author like president girl education.; Education
35
+ fund hear side mother.; Who then more start various draw along.
36
+ - source_sentence: Full Stack Developer. We are looking for a Full Stack Developer
37
+ to join our growing team and work on exciting projects.. Threat store center scene
38
+ country can quite.; Campaign today degree.; Data when risk citizen common.; Current
39
+ few environment social about page.. Penetration Testing; Java; Node.js; Docker;
40
+ JavaScript. SQL; AWS; CI/CD; JavaScript; Machine Learning; C#
41
+ sentences:
42
+ - Data Scientist. Protect usually song treat front he. Thought style successful
43
+ suddenly role voice also. When federal hear eat investment.. Campaign environmental
44
+ none federal.; These poor conference cause capital.; Start rule third ok.; Network
45
+ age job charge benefit various.; Go almost cost great.. DevOps; React; Cybersecurity;
46
+ SQL; Linux; Penetration Testing; Docker. Rise it interest try else attorney.;
47
+ Always everybody fight actually.; Nearly west score go.; Its if less system during.
48
+ - Frontend Developer. Student knowledge catch trip specific structure activity be.
49
+ Listen reveal member.. Nature radio serve into.; Speak old side green second travel
50
+ clear.; Family instead chance entire despite site.; Approach form wonder wrong
51
+ billion four blood source.. Flask; NoSQL; Azure; Cybersecurity; Node.js; Java;
52
+ C#. Recently how hot.; Push yourself step word they.; Forward per difficult chance
53
+ general ten.
54
+ - AI Researcher. Start ball civil set although. Or environmental place boy because
55
+ chance.. Within value ahead.; Class democratic candidate arm.; Region represent
56
+ great note nothing recently low.; Way live according follow walk doctor loss..
57
+ Azure; TensorFlow; JavaScript; Machine Learning; CI/CD; Kubernetes; Terraform.
58
+ Here other next over down seem yourself model.; Discover natural generation traditional
59
+ suddenly management.; Discuss food majority professor.
60
+ - source_sentence: AI Researcher. We are looking for a AI Researcher to join our growing
61
+ team and work on exciting projects.. Improve hard street ask anyone accept history.;
62
+ Heavy a through old nothing various.; Fight clearly safe available similar hot.;
63
+ Movie body accept society heavy six.; Note close bad detail cell.. NoSQL; Azure;
64
+ Terraform; Flask; Django. Deep Learning; NoSQL; Terraform; Python; CI/CD; Flask
65
+ sentences:
66
+ - Frontend Developer. Model purpose most maintain price guess Republican. Manager
67
+ sure stuff beyond win. Wall type process.. Pattern million task so approach simple.;
68
+ Letter as tell tough price.; Tree ahead person building report likely see have.;
69
+ Thought name current hair avoid.. TensorFlow; Azure; DevOps; Machine Learning;
70
+ C#; React; NoSQL. Possible say son sister.; Nothing good later pressure board
71
+ stand.; Fly beat green picture stage.; Look sell same off else nature second.
72
+ - IT Project Manager. South although pass final number pick while. Others run contain
73
+ book. Bag single mission try true.. Power water determine go step common.; Student
74
+ people mission author stay.; Cup here father age age food.. React; Flask; Terraform;
75
+ DevOps; SQL; Docker; NoSQL. Marriage free security his before wear concern.; Future
76
+ great east use.; Senior plan require bit court often.
77
+ - Mobile App Developer. Until player time big design ten. Out billion money follow
78
+ bill so technology. Thousand north particularly difficult. Check social into decade
79
+ thing minute ahead.. Send memory ago full director although morning.; Relationship
80
+ sign front actually forget personal cold name.; Near debate notice their.. SQL;
81
+ AWS; C#; Node.js; Cybersecurity; Machine Learning; React. Produce fly sea.; Middle
82
+ race risk.; Land foot often action brother dinner.; Sign administration use book
83
+ section memory tree.
84
+ - source_sentence: Machine Learning Engineer. We are looking for a Machine Learning
85
+ Engineer to join our growing team and work on exciting projects.. Talk serious
86
+ or mouth night measure.; Article ahead capital no development.; Do minute chance
87
+ employee.; Account impact product land never military main show.. Cybersecurity;
88
+ Terraform; Deep Learning; Python; Linux. Azure; Django; Docker; NoSQL; TypeScript;
89
+ SQL
90
+ sentences:
91
+ - Data Scientist. Next smile gun course six. Performance month bar let expect everything.
92
+ Whom great heart college people million computer.. Probably we determine information
93
+ century.; Step heavy animal notice foot police.; True soldier one business car..
94
+ AWS; JavaScript; Machine Learning; React; CI/CD; Linux; Cybersecurity. Feel field
95
+ behind matter hair.; Tonight give Mrs organization.
96
+ - Data Scientist. Among easy indicate statement. Sit natural change strategy start
97
+ party.. Do hand star modern.; Eat hear for will picture hotel.; Build parent true
98
+ discover carry involve exactly.. Python; TypeScript; Docker; NoSQL; C#; Linux;
99
+ SQL. Sound former during way suffer bag want.; History it school look.; Phone
100
+ into notice piece wait show.
101
+ - IT Project Manager. Least trade these voice. Choose letter than. Do model effort
102
+ they not.. Reflect development forward hand.; Investment fall what guess.; Green
103
+ new instead language board.. Kubernetes; TypeScript; Django; TensorFlow; AWS;
104
+ C#; Deep Learning. Lay tax group message work statement ago.; Can try heart city.;
105
+ Positive social increase throw seat share standard.; Front far prepare.
106
+ - source_sentence: Software Engineer. We are looking for a Software Engineer to join
107
+ our growing team and work on exciting projects.. Suffer class note resource.;
108
+ Guess really character and right scientist behavior election.; Seat force cultural
109
+ arm while.; Single maintain from recently.; Not thing wife focus road.. CI/CD;
110
+ Terraform; DevOps; JavaScript; TypeScript. Docker; Java; Azure; Deep Learning;
111
+ AWS; Node.js
112
+ sentences:
113
+ - Full Stack Developer. Skin direction civil. Toward sure house stay sure if mouth
114
+ smile.. Range weight foreign.; Safe car at rest speech agency.; Her avoid her
115
+ heart three behind.; Deal goal send way power.. Azure; NoSQL; TypeScript; Java;
116
+ Kubernetes; Django; AWS. Environmental entire have charge state require artist.;
117
+ Among various instead our team.
118
+ - Software Engineer. Report ahead relate. Among employee that them.. Night continue
119
+ surface reduce instead education from.; None we forward notice miss wrong few.;
120
+ Business recently strategy else other recently environment.. Linux; NoSQL; Cybersecurity;
121
+ Machine Learning; Python; CI/CD; AWS. Social hot pay task commercial.; I throughout
122
+ participant sense.; Him station low happen available woman parent.; Measure recent
123
+ rock say city indeed allow value.
124
+ - Data Scientist. Standard defense clearly project.. Single always argue offer water
125
+ war.; Meeting certainly leader party heavy mind authority nearly.; Sister certain
126
+ any itself.; Paper top at area provide.. Cybersecurity; React; C#; TensorFlow;
127
+ Deep Learning; Penetration Testing; DevOps. Food safe wide key.; Word identify
128
+ cup life clear.
129
+ pipeline_tag: sentence-similarity
130
+ library_name: sentence-transformers
131
+ ---
132
+
133
+ # SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
134
+
135
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). 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.
136
+
137
+ ## Model Details
138
+
139
+ ### Model Description
140
+ - **Model Type:** Sentence Transformer
141
+ - **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
142
+ - **Maximum Sequence Length:** 256 tokens
143
+ - **Output Dimensionality:** 384 dimensions
144
+ - **Similarity Function:** Cosine Similarity
145
+ <!-- - **Training Dataset:** Unknown -->
146
+ <!-- - **Language:** Unknown -->
147
+ <!-- - **License:** Unknown -->
148
+
149
+ ### Model Sources
150
+
151
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
152
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
153
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
154
+
155
+ ### Full Model Architecture
156
+
157
+ ```
158
+ SentenceTransformer(
159
+ (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
160
+ (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})
161
+ (2): Normalize()
162
+ )
163
+ ```
164
+
165
+ ## Usage
166
+
167
+ ### Direct Usage (Sentence Transformers)
168
+
169
+ First install the Sentence Transformers library:
170
+
171
+ ```bash
172
+ pip install -U sentence-transformers
173
+ ```
174
+
175
+ Then you can load this model and run inference.
176
+ ```python
177
+ from sentence_transformers import SentenceTransformer
178
+
179
+ # Download from the 🤗 Hub
180
+ model = SentenceTransformer("sentence_transformers_model_id")
181
+ # Run inference
182
+ sentences = [
183
+ 'Software Engineer. We are looking for a Software Engineer to join our growing team and work on exciting projects.. Suffer class note resource.; Guess really character and right scientist behavior election.; Seat force cultural arm while.; Single maintain from recently.; Not thing wife focus road.. CI/CD; Terraform; DevOps; JavaScript; TypeScript. Docker; Java; Azure; Deep Learning; AWS; Node.js',
184
+ 'Data Scientist. Standard defense clearly project.. Single always argue offer water war.; Meeting certainly leader party heavy mind authority nearly.; Sister certain any itself.; Paper top at area provide.. Cybersecurity; React; C#; TensorFlow; Deep Learning; Penetration Testing; DevOps. Food safe wide key.; Word identify cup life clear.',
185
+ 'Software Engineer. Report ahead relate. Among employee that them.. Night continue surface reduce instead education from.; None we forward notice miss wrong few.; Business recently strategy else other recently environment.. Linux; NoSQL; Cybersecurity; Machine Learning; Python; CI/CD; AWS. Social hot pay task commercial.; I throughout participant sense.; Him station low happen available woman parent.; Measure recent rock say city indeed allow value.',
186
+ ]
187
+ embeddings = model.encode(sentences)
188
+ print(embeddings.shape)
189
+ # [3, 384]
190
+
191
+ # Get the similarity scores for the embeddings
192
+ similarities = model.similarity(embeddings, embeddings)
193
+ print(similarities.shape)
194
+ # [3, 3]
195
+ ```
196
+
197
+ <!--
198
+ ### Direct Usage (Transformers)
199
+
200
+ <details><summary>Click to see the direct usage in Transformers</summary>
201
+
202
+ </details>
203
+ -->
204
+
205
+ <!--
206
+ ### Downstream Usage (Sentence Transformers)
207
+
208
+ You can finetune this model on your own dataset.
209
+
210
+ <details><summary>Click to expand</summary>
211
+
212
+ </details>
213
+ -->
214
+
215
+ <!--
216
+ ### Out-of-Scope Use
217
+
218
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
219
+ -->
220
+
221
+ <!--
222
+ ## Bias, Risks and Limitations
223
+
224
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
225
+ -->
226
+
227
+ <!--
228
+ ### Recommendations
229
+
230
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
231
+ -->
232
+
233
+ ## Training Details
234
+
235
+ ### Training Dataset
236
+
237
+ #### Unnamed Dataset
238
+
239
+ * Size: 5,000 training samples
240
+ * Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
241
+ * Approximate statistics based on the first 1000 samples:
242
+ | | sentence_0 | sentence_1 | label |
243
+ |:--------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------|
244
+ | type | string | string | float |
245
+ | details | <ul><li>min: 71 tokens</li><li>mean: 88.58 tokens</li><li>max: 109 tokens</li></ul> | <ul><li>min: 67 tokens</li><li>mean: 96.1 tokens</li><li>max: 131 tokens</li></ul> | <ul><li>min: 0.2</li><li>mean: 0.48</li><li>max: 0.83</li></ul> |
246
+ * Samples:
247
+ | sentence_0 | sentence_1 | label |
248
+ |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------|
249
+ | <code>Machine Learning Engineer. We are looking for a Machine Learning Engineer to join our growing team and work on exciting projects.. Good kind capital special human.; Community great goal.; Reach discover wall blood black style after somebody.. Python; TensorFlow; React; CI/CD; AWS. DevOps; Django; Node.js; Azure; CI/CD; Kubernetes</code> | <code>Mobile App Developer. At do American than partner sound. Plan decade industry deep establish wide whole.. All recognize edge southern.; Home hope house develop major there.; Crime push local present thus.. Node.js; Penetration Testing; TensorFlow; Java; C#; Deep Learning; NoSQL. Above people everything.; Eat game left past pull range.; Letter create must including.</code> | <code>0.47</code> |
250
+ | <code>Mobile App Developer. We are looking for a Mobile App Developer to join our growing team and work on exciting projects.. Try hotel where catch reveal help.; Seat nor quality factor movie.; Good image realize respond possible.. Machine Learning; Terraform; JavaScript; Deep Learning; DevOps. Python; JavaScript; Cybersecurity; TensorFlow; Penetration Testing; DevOps</code> | <code>Mobile App Developer. Future large lead tree clear about building. Manage concern stuff shoulder.. Very star necessary military beautiful structure look.; Reveal something church particular instead special.; Than long series central.; Agent sister value.; Teacher production career more safe.. Penetration Testing; Node.js; SQL; TypeScript; Docker; React; Django. Ask song reveal.; Member top power certain pattern.; Trip away success.</code> | <code>0.42</code> |
251
+ | <code>Frontend Developer. We are looking for a Frontend Developer to join our growing team and work on exciting projects.. Per major government hotel population walk.; Suddenly artist century few research.; Exist to outside son onto member.. CI/CD; Python; Flask; Deep Learning; Java. SQL; Penetration Testing; AWS; Java; Linux; Node.js</code> | <code>Software Engineer. Down which want debate. Situation establish find cold that. Take Republican over set people.. Understand event image suffer.; Kind go alone consumer develop tonight star.; Page radio former imagine evidence pick girl budget.. TypeScript; Cybersecurity; Machine Learning; Azure; NoSQL; SQL; React. End bed stand whatever challenge.; West moment act management can second between.</code> | <code>0.46</code> |
252
+ * Loss: [<code>CosineSimilarityLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#cosinesimilarityloss) with these parameters:
253
+ ```json
254
+ {
255
+ "loss_fct": "torch.nn.modules.loss.MSELoss"
256
+ }
257
+ ```
258
+
259
+ ### Training Hyperparameters
260
+ #### Non-Default Hyperparameters
261
+
262
+ - `per_device_train_batch_size`: 16
263
+ - `per_device_eval_batch_size`: 16
264
+ - `num_train_epochs`: 10
265
+ - `multi_dataset_batch_sampler`: round_robin
266
+
267
+ #### All Hyperparameters
268
+ <details><summary>Click to expand</summary>
269
+
270
+ - `overwrite_output_dir`: False
271
+ - `do_predict`: False
272
+ - `eval_strategy`: no
273
+ - `prediction_loss_only`: True
274
+ - `per_device_train_batch_size`: 16
275
+ - `per_device_eval_batch_size`: 16
276
+ - `per_gpu_train_batch_size`: None
277
+ - `per_gpu_eval_batch_size`: None
278
+ - `gradient_accumulation_steps`: 1
279
+ - `eval_accumulation_steps`: None
280
+ - `torch_empty_cache_steps`: None
281
+ - `learning_rate`: 5e-05
282
+ - `weight_decay`: 0.0
283
+ - `adam_beta1`: 0.9
284
+ - `adam_beta2`: 0.999
285
+ - `adam_epsilon`: 1e-08
286
+ - `max_grad_norm`: 1
287
+ - `num_train_epochs`: 10
288
+ - `max_steps`: -1
289
+ - `lr_scheduler_type`: linear
290
+ - `lr_scheduler_kwargs`: {}
291
+ - `warmup_ratio`: 0.0
292
+ - `warmup_steps`: 0
293
+ - `log_level`: passive
294
+ - `log_level_replica`: warning
295
+ - `log_on_each_node`: True
296
+ - `logging_nan_inf_filter`: True
297
+ - `save_safetensors`: True
298
+ - `save_on_each_node`: False
299
+ - `save_only_model`: False
300
+ - `restore_callback_states_from_checkpoint`: False
301
+ - `no_cuda`: False
302
+ - `use_cpu`: False
303
+ - `use_mps_device`: False
304
+ - `seed`: 42
305
+ - `data_seed`: None
306
+ - `jit_mode_eval`: False
307
+ - `use_ipex`: False
308
+ - `bf16`: False
309
+ - `fp16`: False
310
+ - `fp16_opt_level`: O1
311
+ - `half_precision_backend`: auto
312
+ - `bf16_full_eval`: False
313
+ - `fp16_full_eval`: False
314
+ - `tf32`: None
315
+ - `local_rank`: 0
316
+ - `ddp_backend`: None
317
+ - `tpu_num_cores`: None
318
+ - `tpu_metrics_debug`: False
319
+ - `debug`: []
320
+ - `dataloader_drop_last`: False
321
+ - `dataloader_num_workers`: 0
322
+ - `dataloader_prefetch_factor`: None
323
+ - `past_index`: -1
324
+ - `disable_tqdm`: False
325
+ - `remove_unused_columns`: True
326
+ - `label_names`: None
327
+ - `load_best_model_at_end`: False
328
+ - `ignore_data_skip`: False
329
+ - `fsdp`: []
330
+ - `fsdp_min_num_params`: 0
331
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
332
+ - `tp_size`: 0
333
+ - `fsdp_transformer_layer_cls_to_wrap`: None
334
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
335
+ - `deepspeed`: None
336
+ - `label_smoothing_factor`: 0.0
337
+ - `optim`: adamw_torch
338
+ - `optim_args`: None
339
+ - `adafactor`: False
340
+ - `group_by_length`: False
341
+ - `length_column_name`: length
342
+ - `ddp_find_unused_parameters`: None
343
+ - `ddp_bucket_cap_mb`: None
344
+ - `ddp_broadcast_buffers`: False
345
+ - `dataloader_pin_memory`: True
346
+ - `dataloader_persistent_workers`: False
347
+ - `skip_memory_metrics`: True
348
+ - `use_legacy_prediction_loop`: False
349
+ - `push_to_hub`: False
350
+ - `resume_from_checkpoint`: None
351
+ - `hub_model_id`: None
352
+ - `hub_strategy`: every_save
353
+ - `hub_private_repo`: None
354
+ - `hub_always_push`: False
355
+ - `gradient_checkpointing`: False
356
+ - `gradient_checkpointing_kwargs`: None
357
+ - `include_inputs_for_metrics`: False
358
+ - `include_for_metrics`: []
359
+ - `eval_do_concat_batches`: True
360
+ - `fp16_backend`: auto
361
+ - `push_to_hub_model_id`: None
362
+ - `push_to_hub_organization`: None
363
+ - `mp_parameters`:
364
+ - `auto_find_batch_size`: False
365
+ - `full_determinism`: False
366
+ - `torchdynamo`: None
367
+ - `ray_scope`: last
368
+ - `ddp_timeout`: 1800
369
+ - `torch_compile`: False
370
+ - `torch_compile_backend`: None
371
+ - `torch_compile_mode`: None
372
+ - `include_tokens_per_second`: False
373
+ - `include_num_input_tokens_seen`: False
374
+ - `neftune_noise_alpha`: None
375
+ - `optim_target_modules`: None
376
+ - `batch_eval_metrics`: False
377
+ - `eval_on_start`: False
378
+ - `use_liger_kernel`: False
379
+ - `eval_use_gather_object`: False
380
+ - `average_tokens_across_devices`: False
381
+ - `prompts`: None
382
+ - `batch_sampler`: batch_sampler
383
+ - `multi_dataset_batch_sampler`: round_robin
384
+
385
+ </details>
386
+
387
+ ### Training Logs
388
+ | Epoch | Step | Training Loss |
389
+ |:------:|:----:|:-------------:|
390
+ | 1.5974 | 500 | 0.0052 |
391
+ | 3.1949 | 1000 | 0.0039 |
392
+ | 4.7923 | 1500 | 0.0033 |
393
+ | 6.3898 | 2000 | 0.0029 |
394
+ | 7.9872 | 2500 | 0.0026 |
395
+ | 9.5847 | 3000 | 0.0023 |
396
+
397
+
398
+ ### Framework Versions
399
+ - Python: 3.11.12
400
+ - Sentence Transformers: 3.4.1
401
+ - Transformers: 4.51.3
402
+ - PyTorch: 2.6.0+cu124
403
+ - Accelerate: 1.5.2
404
+ - Datasets: 3.5.0
405
+ - Tokenizers: 0.21.1
406
+
407
+ ## Citation
408
+
409
+ ### BibTeX
410
+
411
+ #### Sentence Transformers
412
+ ```bibtex
413
+ @inproceedings{reimers-2019-sentence-bert,
414
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
415
+ author = "Reimers, Nils and Gurevych, Iryna",
416
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
417
+ month = "11",
418
+ year = "2019",
419
+ publisher = "Association for Computational Linguistics",
420
+ url = "https://arxiv.org/abs/1908.10084",
421
+ }
422
+ ```
423
+
424
+ <!--
425
+ ## Glossary
426
+
427
+ *Clearly define terms in order to be accessible across audiences.*
428
+ -->
429
+
430
+ <!--
431
+ ## Model Card Authors
432
+
433
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
434
+ -->
435
+
436
+ <!--
437
+ ## Model Card Contact
438
+
439
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
440
+ -->
config.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertModel"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
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": 6,
18
+ "pad_token_id": 0,
19
+ "position_embedding_type": "absolute",
20
+ "torch_dtype": "float32",
21
+ "transformers_version": "4.51.3",
22
+ "type_vocab_size": 2,
23
+ "use_cache": true,
24
+ "vocab_size": 30522
25
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.4.1",
4
+ "transformers": "4.51.3",
5
+ "pytorch": "2.6.0+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 sha256:1f1da7461cd2e4b865d0f29ef7d24edee9784535b582e6a18f757c6a93c0da76
3
+ size 90864192
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": 256,
3
+ "do_lower_case": false
4
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "mask_token": {
10
+ "content": "[MASK]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "[PAD]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "sep_token": {
24
+ "content": "[SEP]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "unk_token": {
31
+ "content": "[UNK]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ }
37
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "mask_token": "[MASK]",
50
+ "max_length": 128,
51
+ "model_max_length": 256,
52
+ "never_split": null,
53
+ "pad_to_multiple_of": null,
54
+ "pad_token": "[PAD]",
55
+ "pad_token_type_id": 0,
56
+ "padding_side": "right",
57
+ "sep_token": "[SEP]",
58
+ "stride": 0,
59
+ "strip_accents": null,
60
+ "tokenize_chinese_chars": true,
61
+ "tokenizer_class": "BertTokenizer",
62
+ "truncation_side": "right",
63
+ "truncation_strategy": "longest_first",
64
+ "unk_token": "[UNK]"
65
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff