tomaarsen HF staff commited on
Commit
f6111f0
1 Parent(s): c10b169

Add new SentenceTransformer model.

Browse files
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 +1,535 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ library_name: sentence-transformers
5
+ tags:
6
+ - sentence-transformers
7
+ - sentence-similarity
8
+ - feature-extraction
9
+ - loss:GISTEmbedLoss
10
+ base_model: distilbert/distilroberta-base
11
+ metrics:
12
+ - pearson_cosine
13
+ - spearman_cosine
14
+ - pearson_manhattan
15
+ - spearman_manhattan
16
+ - pearson_euclidean
17
+ - spearman_euclidean
18
+ - pearson_dot
19
+ - spearman_dot
20
+ - pearson_max
21
+ - spearman_max
22
+ widget:
23
+ - source_sentence: A woman sings.
24
+ sentences:
25
+ - The woman is singing.
26
+ - A story book is open.
27
+ - The men have blonde hair.
28
+ - source_sentence: a baby smiling
29
+ sentences:
30
+ - A baby is unhappy.
31
+ - a fireman on a ladder
32
+ - Five men stand on chairs.
33
+ - source_sentence: The boy scowls
34
+ sentences:
35
+ - A boy is outdoors.
36
+ - a man is wearing blue
37
+ - Two women are sleeping.
38
+ - source_sentence: There's a dock
39
+ sentences:
40
+ - A boat docked on a river.
41
+ - He is playing a song.
42
+ - The baby is in the crib.
43
+ - source_sentence: an eagle flies
44
+ sentences:
45
+ - A bird flying.
46
+ - The woman is outside.
47
+ - The people are sleeping.
48
+ pipeline_tag: sentence-similarity
49
+ co2_eq_emissions:
50
+ emissions: 1.6492452883656235
51
+ energy_consumed: 0.004242955498982829
52
+ source: codecarbon
53
+ training_type: fine-tuning
54
+ on_cloud: false
55
+ cpu_model: 13th Gen Intel(R) Core(TM) i7-13700K
56
+ ram_total_size: 31.777088165283203
57
+ hours_used: 0.021
58
+ hardware_used: 1 x NVIDIA GeForce RTX 3090
59
+ model-index:
60
+ - name: SentenceTransformer based on distilbert/distilroberta-base
61
+ results:
62
+ - task:
63
+ type: semantic-similarity
64
+ name: Semantic Similarity
65
+ dataset:
66
+ name: sts dev
67
+ type: sts-dev
68
+ metrics:
69
+ - type: pearson_cosine
70
+ value: 0.7695103533338594
71
+ name: Pearson Cosine
72
+ - type: spearman_cosine
73
+ value: 0.8046160770503588
74
+ name: Spearman Cosine
75
+ - type: pearson_manhattan
76
+ value: 0.7673329964610834
77
+ name: Pearson Manhattan
78
+ - type: spearman_manhattan
79
+ value: 0.7756781613323356
80
+ name: Spearman Manhattan
81
+ - type: pearson_euclidean
82
+ value: 0.7718833134570839
83
+ name: Pearson Euclidean
84
+ - type: spearman_euclidean
85
+ value: 0.7784941712509205
86
+ name: Spearman Euclidean
87
+ - type: pearson_dot
88
+ value: 0.22148844887336572
89
+ name: Pearson Dot
90
+ - type: spearman_dot
91
+ value: 0.2092109979282621
92
+ name: Spearman Dot
93
+ - type: pearson_max
94
+ value: 0.7718833134570839
95
+ name: Pearson Max
96
+ - type: spearman_max
97
+ value: 0.8046160770503588
98
+ name: Spearman Max
99
+ - task:
100
+ type: semantic-similarity
101
+ name: Semantic Similarity
102
+ dataset:
103
+ name: sts test
104
+ type: sts-test
105
+ metrics:
106
+ - type: pearson_cosine
107
+ value: 0.7270251484636511
108
+ name: Pearson Cosine
109
+ - type: spearman_cosine
110
+ value: 0.7463390012771995
111
+ name: Spearman Cosine
112
+ - type: pearson_manhattan
113
+ value: 0.7295418823252019
114
+ name: Pearson Manhattan
115
+ - type: spearman_manhattan
116
+ value: 0.7198414342133578
117
+ name: Spearman Manhattan
118
+ - type: pearson_euclidean
119
+ value: 0.7347198114628469
120
+ name: Pearson Euclidean
121
+ - type: spearman_euclidean
122
+ value: 0.724025904164009
123
+ name: Spearman Euclidean
124
+ - type: pearson_dot
125
+ value: 0.19404927455056548
126
+ name: Pearson Dot
127
+ - type: spearman_dot
128
+ value: 0.1791431711812991
129
+ name: Spearman Dot
130
+ - type: pearson_max
131
+ value: 0.7347198114628469
132
+ name: Pearson Max
133
+ - type: spearman_max
134
+ value: 0.7463390012771995
135
+ name: Spearman Max
136
+ ---
137
+
138
+ # SentenceTransformer based on distilbert/distilroberta-base
139
+
140
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [distilbert/distilroberta-base](https://huggingface.co/distilbert/distilroberta-base) on the [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) 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.
141
+
142
+ ## Model Details
143
+
144
+ ### Model Description
145
+ - **Model Type:** Sentence Transformer
146
+ - **Base model:** [distilbert/distilroberta-base](https://huggingface.co/distilbert/distilroberta-base) <!-- at revision fb53ab8802853c8e4fbdbcd0529f21fc6f459b2b -->
147
+ - **Maximum Sequence Length:** 512 tokens
148
+ - **Output Dimensionality:** 768 tokens
149
+ - **Similarity Function:** Cosine Similarity
150
+ - **Training Dataset:**
151
+ - [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli)
152
+ - **Language:** en
153
+ <!-- - **License:** Unknown -->
154
+
155
+ ### Model Sources
156
+
157
+ - **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
158
+ - **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
159
+ - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
160
+
161
+ ### Full Model Architecture
162
+
163
+ ```
164
+ SentenceTransformer(
165
+ (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: RobertaModel
166
+ (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})
167
+ )
168
+ ```
169
+
170
+ ## Usage
171
+
172
+ ### Direct Usage (Sentence Transformers)
173
+
174
+ First install the Sentence Transformers library:
175
+
176
+ ```bash
177
+ pip install -U sentence-transformers
178
+ ```
179
+
180
+ Then you can load this model and run inference.
181
+ ```python
182
+ from sentence_transformers import SentenceTransformer
183
+
184
+ # Download from the 🤗 Hub
185
+ model = SentenceTransformer("tomaarsen/distilroberta-base-nli-v3")
186
+ # Run inference
187
+ sentences = [
188
+ 'an eagle flies',
189
+ 'A bird flying.',
190
+ 'The woman is outside.',
191
+ ]
192
+ embeddings = model.encode(sentences)
193
+ print(embeddings.shape)
194
+ # [3, 768]
195
+
196
+ # Get the similarity scores for the embeddings
197
+ similarities = model.similarity(embeddings)
198
+ print(similarities.shape)
199
+ # [3, 3]
200
+ ```
201
+
202
+ <!--
203
+ ### Direct Usage (Transformers)
204
+
205
+ <details><summary>Click to see the direct usage in Transformers</summary>
206
+
207
+ </details>
208
+ -->
209
+
210
+ <!--
211
+ ### Downstream Usage (Sentence Transformers)
212
+
213
+ You can finetune this model on your own dataset.
214
+
215
+ <details><summary>Click to expand</summary>
216
+
217
+ </details>
218
+ -->
219
+
220
+ <!--
221
+ ### Out-of-Scope Use
222
+
223
+ *List how the model may foreseeably be misused and address what users ought not to do with the model.*
224
+ -->
225
+
226
+ ## Evaluation
227
+
228
+ ### Metrics
229
+
230
+ #### Semantic Similarity
231
+ * Dataset: `sts-dev`
232
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
233
+
234
+ | Metric | Value |
235
+ |:--------------------|:-----------|
236
+ | pearson_cosine | 0.7695 |
237
+ | **spearman_cosine** | **0.8046** |
238
+ | pearson_manhattan | 0.7673 |
239
+ | spearman_manhattan | 0.7757 |
240
+ | pearson_euclidean | 0.7719 |
241
+ | spearman_euclidean | 0.7785 |
242
+ | pearson_dot | 0.2215 |
243
+ | spearman_dot | 0.2092 |
244
+ | pearson_max | 0.7719 |
245
+ | spearman_max | 0.8046 |
246
+
247
+ #### Semantic Similarity
248
+ * Dataset: `sts-test`
249
+ * Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
250
+
251
+ | Metric | Value |
252
+ |:--------------------|:-----------|
253
+ | pearson_cosine | 0.727 |
254
+ | **spearman_cosine** | **0.7463** |
255
+ | pearson_manhattan | 0.7295 |
256
+ | spearman_manhattan | 0.7198 |
257
+ | pearson_euclidean | 0.7347 |
258
+ | spearman_euclidean | 0.724 |
259
+ | pearson_dot | 0.194 |
260
+ | spearman_dot | 0.1791 |
261
+ | pearson_max | 0.7347 |
262
+ | spearman_max | 0.7463 |
263
+
264
+ <!--
265
+ ## Bias, Risks and Limitations
266
+
267
+ *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
268
+ -->
269
+
270
+ <!--
271
+ ### Recommendations
272
+
273
+ *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
274
+ -->
275
+
276
+ ## Training Details
277
+
278
+ ### Training Dataset
279
+
280
+ #### sentence-transformers/all-nli
281
+
282
+ * Dataset: [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [cc6c526](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/cc6c526380e29912b5c6fa03682da4daf773c013)
283
+ * Size: 10,000 training samples
284
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
285
+ * Approximate statistics based on the first 1000 samples:
286
+ | | anchor | positive | negative |
287
+ |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|
288
+ | type | string | string | string |
289
+ | details | <ul><li>min: 7 tokens</li><li>mean: 10.38 tokens</li><li>max: 45 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 12.8 tokens</li><li>max: 39 tokens</li></ul> | <ul><li>min: 6 tokens</li><li>mean: 13.4 tokens</li><li>max: 50 tokens</li></ul> |
290
+ * Samples:
291
+ | anchor | positive | negative |
292
+ |:---------------------------------------------------------------------------|:-------------------------------------------------|:-----------------------------------------------------------|
293
+ | <code>A person on a horse jumps over a broken down airplane.</code> | <code>A person is outdoors, on a horse.</code> | <code>A person is at a diner, ordering an omelette.</code> |
294
+ | <code>Children smiling and waving at camera</code> | <code>There are children present</code> | <code>The kids are frowning</code> |
295
+ | <code>A boy is jumping on skateboard in the middle of a red bridge.</code> | <code>The boy does a skateboarding trick.</code> | <code>The boy skates down the sidewalk.</code> |
296
+ * Loss: [<code>GISTEmbedLoss</code>](https://sbert.net/docs/package_reference/losses.html#gistembedloss) with these parameters:
297
+ ```json
298
+ {'guide': SentenceTransformer(
299
+ (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
300
+ (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})
301
+ (2): Normalize()
302
+ ), 'temperature': 0.01}
303
+ ```
304
+
305
+ ### Evaluation Dataset
306
+
307
+ #### sentence-transformers/all-nli
308
+
309
+ * Dataset: [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [cc6c526](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/cc6c526380e29912b5c6fa03682da4daf773c013)
310
+ * Size: 1,000 evaluation samples
311
+ * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
312
+ * Approximate statistics based on the first 1000 samples:
313
+ | | anchor | positive | negative |
314
+ |:--------|:----------------------------------------------------------------------------------|:---------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
315
+ | type | string | string | string |
316
+ | details | <ul><li>min: 6 tokens</li><li>mean: 18.02 tokens</li><li>max: 66 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 9.81 tokens</li><li>max: 29 tokens</li></ul> | <ul><li>min: 5 tokens</li><li>mean: 10.37 tokens</li><li>max: 29 tokens</li></ul> |
317
+ * Samples:
318
+ | anchor | positive | negative |
319
+ |:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------|:--------------------------------------------------------|
320
+ | <code>Two women are embracing while holding to go packages.</code> | <code>Two woman are holding packages.</code> | <code>The men are fighting outside a deli.</code> |
321
+ | <code>Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.</code> | <code>Two kids in numbered jerseys wash their hands.</code> | <code>Two kids in jackets walk to school.</code> |
322
+ | <code>A man selling donuts to a customer during a world exhibition event held in the city of Angeles</code> | <code>A man selling donuts to a customer.</code> | <code>A woman drinks her coffee in a small cafe.</code> |
323
+ * Loss: [<code>GISTEmbedLoss</code>](https://sbert.net/docs/package_reference/losses.html#gistembedloss) with these parameters:
324
+ ```json
325
+ {'guide': SentenceTransformer(
326
+ (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel
327
+ (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})
328
+ (2): Normalize()
329
+ ), 'temperature': 0.01}
330
+ ```
331
+
332
+ ### Training Hyperparameters
333
+ #### Non-Default Hyperparameters
334
+
335
+ - `eval_strategy`: steps
336
+ - `per_device_train_batch_size`: 128
337
+ - `per_device_eval_batch_size`: 128
338
+ - `num_train_epochs`: 1
339
+ - `warmup_ratio`: 0.1
340
+ - `fp16`: True
341
+ - `batch_sampler`: no_duplicates
342
+
343
+ #### All Hyperparameters
344
+ <details><summary>Click to expand</summary>
345
+
346
+ - `overwrite_output_dir`: False
347
+ - `do_predict`: False
348
+ - `eval_strategy`: steps
349
+ - `prediction_loss_only`: False
350
+ - `per_device_train_batch_size`: 128
351
+ - `per_device_eval_batch_size`: 128
352
+ - `per_gpu_train_batch_size`: None
353
+ - `per_gpu_eval_batch_size`: None
354
+ - `gradient_accumulation_steps`: 1
355
+ - `eval_accumulation_steps`: None
356
+ - `learning_rate`: 5e-05
357
+ - `weight_decay`: 0.0
358
+ - `adam_beta1`: 0.9
359
+ - `adam_beta2`: 0.999
360
+ - `adam_epsilon`: 1e-08
361
+ - `max_grad_norm`: 1.0
362
+ - `num_train_epochs`: 1
363
+ - `max_steps`: -1
364
+ - `lr_scheduler_type`: linear
365
+ - `lr_scheduler_kwargs`: {}
366
+ - `warmup_ratio`: 0.1
367
+ - `warmup_steps`: 0
368
+ - `log_level`: passive
369
+ - `log_level_replica`: warning
370
+ - `log_on_each_node`: True
371
+ - `logging_nan_inf_filter`: True
372
+ - `save_safetensors`: True
373
+ - `save_on_each_node`: False
374
+ - `save_only_model`: False
375
+ - `no_cuda`: False
376
+ - `use_cpu`: False
377
+ - `use_mps_device`: False
378
+ - `seed`: 42
379
+ - `data_seed`: None
380
+ - `jit_mode_eval`: False
381
+ - `use_ipex`: False
382
+ - `bf16`: False
383
+ - `fp16`: True
384
+ - `fp16_opt_level`: O1
385
+ - `half_precision_backend`: auto
386
+ - `bf16_full_eval`: False
387
+ - `fp16_full_eval`: False
388
+ - `tf32`: None
389
+ - `local_rank`: 0
390
+ - `ddp_backend`: None
391
+ - `tpu_num_cores`: None
392
+ - `tpu_metrics_debug`: False
393
+ - `debug`: []
394
+ - `dataloader_drop_last`: False
395
+ - `dataloader_num_workers`: 0
396
+ - `dataloader_prefetch_factor`: None
397
+ - `past_index`: -1
398
+ - `disable_tqdm`: False
399
+ - `remove_unused_columns`: True
400
+ - `label_names`: None
401
+ - `load_best_model_at_end`: False
402
+ - `ignore_data_skip`: False
403
+ - `fsdp`: []
404
+ - `fsdp_min_num_params`: 0
405
+ - `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
406
+ - `fsdp_transformer_layer_cls_to_wrap`: None
407
+ - `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
408
+ - `deepspeed`: None
409
+ - `label_smoothing_factor`: 0.0
410
+ - `optim`: adamw_torch
411
+ - `optim_args`: None
412
+ - `adafactor`: False
413
+ - `group_by_length`: False
414
+ - `length_column_name`: length
415
+ - `ddp_find_unused_parameters`: None
416
+ - `ddp_bucket_cap_mb`: None
417
+ - `ddp_broadcast_buffers`: None
418
+ - `dataloader_pin_memory`: True
419
+ - `dataloader_persistent_workers`: False
420
+ - `skip_memory_metrics`: True
421
+ - `use_legacy_prediction_loop`: False
422
+ - `push_to_hub`: False
423
+ - `resume_from_checkpoint`: None
424
+ - `hub_model_id`: None
425
+ - `hub_strategy`: every_save
426
+ - `hub_private_repo`: False
427
+ - `hub_always_push`: False
428
+ - `gradient_checkpointing`: False
429
+ - `gradient_checkpointing_kwargs`: None
430
+ - `include_inputs_for_metrics`: False
431
+ - `eval_do_concat_batches`: True
432
+ - `fp16_backend`: auto
433
+ - `push_to_hub_model_id`: None
434
+ - `push_to_hub_organization`: None
435
+ - `mp_parameters`:
436
+ - `auto_find_batch_size`: False
437
+ - `full_determinism`: False
438
+ - `torchdynamo`: None
439
+ - `ray_scope`: last
440
+ - `ddp_timeout`: 1800
441
+ - `torch_compile`: False
442
+ - `torch_compile_backend`: None
443
+ - `torch_compile_mode`: None
444
+ - `dispatch_batches`: None
445
+ - `split_batches`: None
446
+ - `include_tokens_per_second`: False
447
+ - `include_num_input_tokens_seen`: False
448
+ - `neftune_noise_alpha`: None
449
+ - `optim_target_modules`: None
450
+ - `batch_sampler`: no_duplicates
451
+ - `multi_dataset_batch_sampler`: proportional
452
+
453
+ </details>
454
+
455
+ ### Training Logs
456
+ | Epoch | Step | loss | sts-dev_spearman_cosine | sts-test_spearman_cosine |
457
+ |:------:|:----:|:------:|:-----------------------:|:------------------------:|
458
+ | 0 | 0 | - | 0.6375 | - |
459
+ | 0.1266 | 10 | 2.5172 | 0.7944 | - |
460
+ | 0.2532 | 20 | 1.8059 | 0.8061 | - |
461
+ | 0.3797 | 30 | 1.6805 | 0.8163 | - |
462
+ | 0.5063 | 40 | 1.8153 | 0.8167 | - |
463
+ | 0.6329 | 50 | 1.7177 | 0.8121 | - |
464
+ | 0.7595 | 60 | 1.8622 | 0.8031 | - |
465
+ | 0.8861 | 70 | 1.8056 | 0.8046 | - |
466
+ | 1.0 | 79 | - | - | 0.7463 |
467
+
468
+
469
+ ### Environmental Impact
470
+ Carbon emissions were measured using [CodeCarbon](https://github.com/mlco2/codecarbon).
471
+ - **Energy Consumed**: 0.004 kWh
472
+ - **Carbon Emitted**: 0.002 kg of CO2
473
+ - **Hours Used**: 0.021 hours
474
+
475
+ ### Training Hardware
476
+ - **On Cloud**: No
477
+ - **GPU Model**: 1 x NVIDIA GeForce RTX 3090
478
+ - **CPU Model**: 13th Gen Intel(R) Core(TM) i7-13700K
479
+ - **RAM Size**: 31.78 GB
480
+
481
+ ### Framework Versions
482
+ - Python: 3.11.6
483
+ - Sentence Transformers: 3.0.0.dev0
484
+ - Transformers: 4.41.0.dev0
485
+ - PyTorch: 2.3.0+cu121
486
+ - Accelerate: 0.26.1
487
+ - Datasets: 2.18.0
488
+ - Tokenizers: 0.19.1
489
+
490
+ ## Citation
491
+
492
+ ### BibTeX
493
+
494
+ #### Sentence Transformers
495
+ ```bibtex
496
+ @inproceedings{reimers-2019-sentence-bert,
497
+ title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
498
+ author = "Reimers, Nils and Gurevych, Iryna",
499
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
500
+ month = "11",
501
+ year = "2019",
502
+ publisher = "Association for Computational Linguistics",
503
+ url = "https://arxiv.org/abs/1908.10084",
504
+ }
505
+ ```
506
+
507
+ #### GISTEmbedLoss
508
+ ```bibtex
509
+ @misc{solatorio2024gistembed,
510
+ title={GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuning},
511
+ author={Aivin V. Solatorio},
512
+ year={2024},
513
+ eprint={2402.16829},
514
+ archivePrefix={arXiv},
515
+ primaryClass={cs.LG}
516
+ }
517
+ ```
518
+
519
+ <!--
520
+ ## Glossary
521
+
522
+ *Clearly define terms in order to be accessible across audiences.*
523
+ -->
524
+
525
+ <!--
526
+ ## Model Card Authors
527
+
528
+ *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
529
+ -->
530
+
531
+ <!--
532
+ ## Model Card Contact
533
+
534
+ *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
535
+ -->
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilroberta-base",
3
+ "architectures": [
4
+ "RobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 6,
20
+ "pad_token_id": 1,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.41.0.dev0",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 50265
27
+ }
config_sentence_transformers.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "__version__": {
3
+ "sentence_transformers": "3.0.0.dev0",
4
+ "transformers": "4.41.0.dev0",
5
+ "pytorch": "2.3.0+cu121"
6
+ },
7
+ "prompts": {},
8
+ "default_prompt_name": null,
9
+ "similarity_fn_name": null
10
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:48a0145f07d9271b1d30e6130c312f64950ef20ec8908fc2b8b87ca4fa82f74b
3
+ size 328485128
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 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 512,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "RobertaTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff