DoDucAnh commited on
Commit
ddab52d
·
verified ·
1 Parent(s): 62ae033

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -8,164 +8,142 @@ tags:
8
  - loss:MultipleNegativesRankingLoss
9
  base_model: intfloat/multilingual-e5-large-instruct
10
  widget:
11
- - source_sentence: '''আমি'' শব্দটি কোন লিঙ্গ?
12
-
13
- A. উভয় লিঙ্গ
14
-
15
- B. ক্লীব লিঙ্গ
16
-
17
- C. পুংলিঙ্গ
18
-
19
- D. স্ত্রী লিঙ্গ'
 
 
 
20
  sentences:
21
- - F.P. Dobroslavin, tibbin müxtəlif sahələri üzrə tanınmış bir alimdir, ancaq daha
22
- çox baş vermiş tədqiqatlara görə seçilir. Onun əməyinin sanitar-gigiyenik sahəyə
23
- təsiri əhəmiyyətlidir.
24
- - 'বাংলা ভাষায় শব্দগুলোর লিঙ্গ সাধারণত তিনটি মূল শ্রেণিতে ভাগ হয়: পুংলিঙ্গ (পুরুষ),
25
- স্ত্রী লিঙ্গ (মহিলা), এবং ক্লীব লিঙ্গ (যার কোনো লিঙ্গ নেই)।'
26
- - Waves are disturbances that transfer energy from one place to another without
27
- transferring matter. Think of a ripple on a pond – the water molecules don't travel
28
- across the pond with the ripple; they mostly move up and down as the energy passes
29
- through them.
30
- - source_sentence: '企业产品组合中所拥有的产品线数目是
31
-
32
- A. 产品组合的宽度
33
-
34
- B. 产品组合的相关性
35
-
36
- C. 产品组合的深度
37
-
38
- D. 产品组合的长度'
 
 
39
  sentences:
40
- - 产品组合的宽度(Width)是指企业拥有的产品线数目。
41
- - This fluid is produced by the walls of the vagina and the Bartholin's glands.
42
- - "### Assumption of Risk Defined \nAssumption of risk is a legal doctrine used\
43
- \ in tort law that can limit or bar recovery in negligence claims. This doctrine\
44
- \ suggests that if a person voluntarily engages in a risky activity, knowing the\
45
- \ risks involved, they cannot hold another party responsible for resulting injuries.\
46
- \ Common scenarios where this applies include contact sports and recreational\
47
- \ activities, where participants understand the inherent hazards. \n\n### Elements\
48
- \ of Assumption of Risk \nTo successfully argue assumption of risk, certain elements\
49
- \ must be established: \n1. **Knowledge of the Risk**: The individual must have\
50
- \ actual or constructive knowledge of the risk involved. \n2. **Voluntary Exposure**:\
51
- \ The individual must voluntarily choose to expose themselves to that risk. \n\
52
- 3. **Informed Consent**: The individual must have consented to take that risk\
53
- \ despite being aware of it. \n\n### Contributory Negligence \nContributory\
54
- \ negligence is a legal concept that exists in some jurisdictions where a plaintiff's\
55
- \ own negligence contributes to their injury. Under this doctrine, if the plaintiff\
56
- \ is found to have played any part in their injury, they may be barred from recovering\
57
- \ damages, or the damage award could be reduced. It emphasizes the responsibility\
58
- \ of the injured party to exercise reasonable care for their own safety. \n\n\
59
- ### Interaction of Assumption of Risk and Contributory Negligence \nIn many jurisdictions,\
60
- \ both assumption of risk and contributory negligence can coexist as defenses.\
61
- \ However, some legal systems assert that if a plaintiff is found contributorily\
62
- \ negligent, they cannot also claim assumed risk for the same incident. This overlap\
63
- \ can complicate cases since the determination of the plaintiff's awareness and\
64
- \ behavior prior to the accident can alter the outcome. \n\n### Legal Standard\
65
- \ for Warnings and Liability \nIn negligence cases, the adequacy of warnings\
66
- \ provided is crucial. Courts often assess whether the warnings were sufficient\
67
- \ to inform the individual of the specific hazards present. A simple sign may\
68
- \ not meet the threshold if it fails to clearly communicate the danger involved,\
69
- \ especially if the harm is not immediately obvious or if the context (e.g., a\
70
- \ crowded street) suggests additional risks."
71
- - source_sentence: "While shopping at a grocery store, a customer tripped over a broken\
72
- \ tile, fell, and suffered a concussion. A few months after the accident, the\
73
- \ customer's attorney deposed a store employee. In the deposition, the employee\
74
- \ testified, \"I'd been telling the store manager for years to get that broken\
75
- \ tile fixed, but he wouldn't do it. \" The employee died in an automobile accident\
76
- \ after being deposed. At trial, the deposition should be\nA. admitted, as a dying\
77
- \ declaration. \nB. admitted, as former testimony. \nC. not admitted, because\
78
- \ it is hearsay not within any exception. \nD. not admitted, because the employee\
79
- \ is not available for cross-examination. "
80
  sentences:
81
- - In the context of human evolution, brain size is often compared to body size in
82
- a measurement called the encephalization quotient (EQ). This measure assesses
83
- the expected brain size for an animal of a given body size compared to actual
84
- brain size. An increase in EQ among hominins is often linked to advancements in
85
- cognitive abilities, such as problem-solving and social interaction.
86
- - Another exception to the hearsay rule, though often with specific requirements
87
- related to the declarant's belief of impending death, is the dying declaration.
88
- - In assessing moral actions, it is also essential to consider societal norms. In
89
- the U.S. context in 2020, moral standards often emphasize community well-being
90
- and individual rights. An action like diverting emergency supplies would likely
91
- be condemned in most social circles, while stepping out of rhythm during a line
92
- dancewould not commonly qualify as a serious moral offense. Thus, moral wrongness
93
- is often context-dependent and tied closely to consequences for individuals and
94
- society.
95
- - source_sentence: 'Recent research on hominid species dating from the Middle Pliocene
96
- indicates there was (as of 2020):
97
-
98
- A. multiple hominid species but with limited diversity.
99
-
100
- B. a single species with no diversity.
101
-
102
- C. decreased species diversity but increased numbers of hammerstones and flakes,
103
- indicating stone tool manufacture.
104
-
105
- D. a single dominant species that outcompeted all others, leading to decreased
106
- diversity.
107
-
108
- E. increased species diversity due to a prolonged ice age followed by a severe
109
- drought.
110
-
111
- F. decreased species diversity due to a prolonged ice age followed by a severe
112
- drought.
113
-
114
- G. a great amount of species diversity, or a single species that exhibited a lot
115
- of diversity.
116
-
117
- H. increased species diversity but with decreased population numbers due to harsh
118
- climate conditions.
119
-
120
- I. increased species diversity but decreased numbers of hammerstones and flakes,
121
- indicating less stone tool manufacture.
122
-
123
- J. very little species diversity during this period and very few hominids.'
124
  sentences:
125
- - Hammerstones and flakes are artifacts associated with early stone tool technology.
126
- Hammerstones are hard rocks used to strike other stones, while flakes are the
127
- sharp pieces produced from such strikes, which could be utilized for tasks like
128
- cutting or scraping, indicating early cognitive and manual skills in tool-making
129
- among certain species.
130
- - The Doppler effect is a phenomenon that occurs when the source of a wave and the
131
- observer are moving relative to each other. It results in a change in the observed
132
- frequency of the wave compared to the source frequency.
133
- - Counseling and therapeutic interventions can play a role in addressing student
134
- behavioral issues, but they should be considered within a broader context of classroom
135
- dynamics and educational strategies. Counseling might help the child develop coping
136
- mechanisms, social skills, and emotional regulation strategies. However, the effectiveness
137
- of counseling is often maximized when the child is supported in the classroom
138
- environment as well, suggesting that changes to the teacher's approach could lead
139
- to improved outcomes.
140
- - source_sentence: 'Hipotalamusi NUK kontrollon sekretimin e hormoneve:
141
-
142
- A. FSH dhe LH
143
-
144
- B. te rritjes(GH)
145
-
146
- C. ACTH
147
-
148
- D. te pankreasit'
 
 
 
 
 
 
149
  sentences:
150
- - In the context of estate planning and inheritance law, a will serves as a legal
151
- document outlining how a person's property and assets will be distributed after
152
- their death. The interpretation of a will often hinges on the intent of the testator,
153
- or the person who made the will, which can affect how property interests are determined.
154
- - State laws that regulate matters of legitimate local concern but have an incidental
155
- effect on interstate commerce are subject to a less strict balancing test. Under
156
- this test, a state law will be upheld unless the burden imposed on interstate
157
- commerce is clearly excessive in relation to the putative local benefits.
158
- - Hipotalamusi është një pjesë e trurit ndodhet nën talamusin. Ai luan një rol
159
- kryesor në lidhjen e sistemit nervor me sistemin endokrin përmes gjëndrës së hipofizës.
160
  datasets:
161
- - DoDucAnh/mcqa-rag-finetune
162
  pipeline_tag: sentence-similarity
163
  library_name: sentence-transformers
164
  ---
165
 
166
  # SentenceTransformer based on intfloat/multilingual-e5-large-instruct
167
 
168
- This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) on the [mcqa-rag-finetune](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune) dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
169
 
170
  ## Model Details
171
 
@@ -176,7 +154,7 @@ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [i
176
  - **Output Dimensionality:** 1024 dimensions
177
  - **Similarity Function:** Cosine Similarity
178
  - **Training Dataset:**
179
- - [mcqa-rag-finetune](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune)
180
  <!-- - **Language:** Unknown -->
181
  <!-- - **License:** Unknown -->
182
 
@@ -214,9 +192,9 @@ from sentence_transformers import SentenceTransformer
214
  model = SentenceTransformer("sentence_transformers_model_id")
215
  # Run inference
216
  sentences = [
217
- 'Hipotalamusi NUK kontrollon sekretimin e hormoneve:\nA. FSH dhe LH\nB. te rritjes(GH)\nC. ACTH\nD. te pankreasit',
218
- 'Hipotalamusi është një pjesë e trurit ndodhet nën talamusin. Ai luan një rol kryesor lidhjen e sistemit nervor me sistemin endokrin përmes gjëndrës hipofizës.',
219
- 'State laws that regulate matters of legitimate local concern but have an incidental effect on interstate commerce are subject to a less strict balancing test. Under this test, a state law will be upheld unless the burden imposed on interstate commerce is clearly excessive in relation to the putative local benefits.',
220
  ]
221
  embeddings = model.encode(sentences)
222
  print(embeddings.shape)
@@ -268,22 +246,22 @@ You can finetune this model on your own dataset.
268
 
269
  ### Training Dataset
270
 
271
- #### mcqa-rag-finetune
272
 
273
- * Dataset: [mcqa-rag-finetune](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune) at [d1f5446](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune/tree/d1f5446a80c070fb8e1abfffef8a9dace426026b)
274
  * Size: 594,028 training samples
275
  * Columns: <code>anchor</code> and <code>positive</code>
276
  * Approximate statistics based on the first 1000 samples:
277
- | | anchor | positive |
278
- |:--------|:-------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
279
- | type | string | string |
280
- | details | <ul><li>min: 22 tokens</li><li>mean: 105.96 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 12 tokens</li><li>mean: 70.95 tokens</li><li>max: 478 tokens</li></ul> |
281
  * Samples:
282
- | anchor | positive |
283
- |:------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
284
- | <code>Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field.<br>A. 0<br>B. 1<br>C. 2<br>D. 3</code> | <code>The notation Z_3 refers to the finite field with three elements, often denoted as {0, 1, 2}. This field operates under modular arithmetic, specifically modulo 3. Elements in Z_3 can be added and multiplied according to the rules of modulo 3, where any number can wrap around upon reaching 3.</code> |
285
- | <code>Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field.<br>A. 0<br>B. 1<br>C. 2<br>D. 3</code> | <code>A field is a set equipped with two operations, addition and multiplication, satisfying certain properties: associativity, commutativity, distributivity, the existence of additive and multiplicative identities, and the existence of additive inverses and multiplicative inverses (for all elements except the zero element). In order for Z_3[x]/(f(x)) to be a field, the polynomial f(x) must be irreducible over Z_3.</code> |
286
- | <code>Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field.<br>A. 0<br>B. 1<br>C. 2<br>D. 3</code> | <code>The expression Z_3[x] indicates the set of all polynomials with coefficients in Z_3. A polynomial is said to be irreducible over Z_3 if it cannot be factored into the product of two non-constant polynomials with coefficients in Z_3. In the case of quadratic polynomials like x^2 + c, irreducibility depends on whether it has any roots in the field Z_3.</code> |
287
  * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
288
  ```json
289
  {
@@ -294,10 +272,10 @@ You can finetune this model on your own dataset.
294
 
295
  ### Evaluation Dataset
296
 
297
- #### mcqa-rag-finetune
298
 
299
- * Dataset: [mcqa-rag-finetune](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune) at [d1f5446](https://huggingface.co/datasets/DoDucAnh/mcqa-rag-finetune/tree/d1f5446a80c070fb8e1abfffef8a9dace426026b)
300
- * Size: 1,000 evaluation samples
301
  * Columns: <code>anchor</code> and <code>positive</code>
302
  * Approximate statistics based on the first 1000 samples:
303
  | | anchor | positive |
@@ -322,11 +300,11 @@ You can finetune this model on your own dataset.
322
  #### Non-Default Hyperparameters
323
 
324
  - `eval_strategy`: steps
325
- - `per_device_train_batch_size`: 12
326
- - `per_device_eval_batch_size`: 12
327
- - `learning_rate`: 3e-05
328
- - `num_train_epochs`: 1
329
- - `warmup_steps`: 5000
330
  - `fp16`: True
331
  - `load_best_model_at_end`: True
332
 
@@ -337,25 +315,25 @@ You can finetune this model on your own dataset.
337
  - `do_predict`: False
338
  - `eval_strategy`: steps
339
  - `prediction_loss_only`: True
340
- - `per_device_train_batch_size`: 12
341
- - `per_device_eval_batch_size`: 12
342
  - `per_gpu_train_batch_size`: None
343
  - `per_gpu_eval_batch_size`: None
344
- - `gradient_accumulation_steps`: 1
345
  - `eval_accumulation_steps`: None
346
  - `torch_empty_cache_steps`: None
347
- - `learning_rate`: 3e-05
348
  - `weight_decay`: 0.0
349
  - `adam_beta1`: 0.9
350
  - `adam_beta2`: 0.999
351
  - `adam_epsilon`: 1e-08
352
  - `max_grad_norm`: 1.0
353
- - `num_train_epochs`: 1
354
  - `max_steps`: -1
355
  - `lr_scheduler_type`: linear
356
  - `lr_scheduler_kwargs`: {}
357
  - `warmup_ratio`: 0.0
358
- - `warmup_steps`: 5000
359
  - `log_level`: passive
360
  - `log_level_replica`: warning
361
  - `log_on_each_node`: True
@@ -452,33 +430,33 @@ You can finetune this model on your own dataset.
452
  ### Training Logs
453
  | Epoch | Step | Training Loss | Validation Loss |
454
  |:--------:|:--------:|:-------------:|:---------------:|
455
- | **0.05** | **2476** | **0.1209** | **0.0347** |
456
- | 0.1000 | 4952 | 0.0737 | 0.0459 |
457
- | 0.1501 | 7428 | 0.087 | 0.0732 |
458
- | 0.2001 | 9904 | 0.0825 | 0.1209 |
459
- | 0.2501 | 12380 | 0.0783 | 0.0934 |
460
- | 0.3001 | 14856 | 0.071 | 0.0793 |
461
- | 0.3501 | 17332 | 0.0661 | 0.0855 |
462
- | 0.4001 | 19808 | 0.0652 | 0.0964 |
463
- | 0.4502 | 22284 | 0.063 | 0.0892 |
464
- | 0.5002 | 24760 | 0.056 | 0.0923 |
465
- | 0.5502 | 27236 | 0.0509 | 0.1016 |
466
- | 0.6002 | 29712 | 0.045 | 0.0918 |
467
- | 0.6502 | 32188 | 0.0472 | 0.0896 |
468
- | 0.7002 | 34664 | 0.0396 | 0.0959 |
469
- | 0.7503 | 37140 | 0.0371 | 0.0819 |
470
- | 0.8003 | 39616 | 0.0341 | 0.0845 |
471
- | 0.8503 | 42092 | 0.0344 | 0.0790 |
472
- | 0.9003 | 44568 | 0.0288 | 0.0863 |
473
- | 0.9503 | 47044 | 0.03 | 0.0767 |
474
-
475
- * The bold row denotes the saved checkpoint.
476
 
477
  ### Framework Versions
478
- - Python: 3.11.9
479
  - Sentence Transformers: 4.1.0
480
- - Transformers: 4.52.3
481
- - PyTorch: 2.7.0+cu126
482
  - Accelerate: 1.7.0
483
  - Datasets: 3.6.0
484
  - Tokenizers: 0.21.1
 
8
  - loss:MultipleNegativesRankingLoss
9
  base_model: intfloat/multilingual-e5-large-instruct
10
  widget:
11
+ - source_sentence: "While driving her company vehicle near a pedestrian mall, a woman\
12
+ \ came upon the scene of a three-car accident. She was so busy gawking at the\
13
+ \ damaged vehicles that she failed to see one of the victims lying on the road\
14
+ \ in front of her car. She hit and ran over the victim, who survived and sued\
15
+ \ the woman's company. The victim offers the testimony of a witness to the incident.\
16
+ \ Referring to the woman, the witness stated, \"The driver of that car ran over\
17
+ \ the victim as he was lying on the ground awaiting an ambulance, and said \x80\
18
+ \x98It is all my fault, I should have been paying more attention to my driving.\
19
+ \ \" Assume for this question that the woman is available to testify. The trial\
20
+ \ judge should rule that the testimony is\nA. admissible as a declaration against\
21
+ \ interest.\nB. admissible as a present sense impression.\nC. admissible as an\
22
+ \ admission.\nD. inadmissible as hearsay not within any recognized exception."
23
  sentences:
24
+ - A present sense impression is a statement describing or explaining an event or
25
+ condition made while the declarant was perceiving the event or condition, or immediately
26
+ thereafter. This is an exception to the hearsay rule.
27
+ - Corporate managers are professionals within a business environment who handle
28
+ various aspects of management, including planning, organizing, leading, and controlling
29
+ resources. Their roles often draw from established management theories, such as
30
+ those by Henri Fayol, which emphasize functions like forecasting, commanding,
31
+ and coordinating to support organizational success.
32
+ - Ο ποιοτικός έλεγχος είναι μια διαδικασία που εφαρμόζεται στη βιομηχανία και σε
33
+ άλλους τομείς για να διασφαλιστεί ότι τα προϊόντα ή οι υπηρεσίες πληρούν συγκεκριμένες
34
+ προδιαγραφές και πρότυπα ποιότητας.
35
+ - source_sentence: '‘বিপরীত বৈষম্য’-এর নীতিটি প্রয়োগ করা হয়-
36
+
37
+ A. পিছিয়ে পড়া জনগােষ্ঠীর ক্ষেত্রে
38
+
39
+ B. . নারীদের ক্ষেত্রে
40
+
41
+ C. প্রতিবন্ধীদের ক্ষেত্রে
42
+
43
+ D. সংখ্যালঘুদের ক্ষেত্রে'
44
  sentences:
45
+ - বিপরীত বৈষম্য সাধারণত পিছিয়ে পড়া জনগোষ্ঠী বা সংখ্যালঘুদের মতো গোষ্ঠীর ক্ষেত্রে
46
+ প্রয়োগ করা হয় যারা ঐতিহাসিকভাবে বৈষম্যের শিকার হয়েছেন।
47
+ - Hummingbirds reproduce by laying eggs, usually in small nests that they build
48
+ on branches. The female is responsible for incubating the eggs and caring for
49
+ the young, which necessitates energy management to ensure survival and growth.
50
+ - In the Mughal Empire, zamindars were initially indigenous local chiefs of towns
51
+ and villages in rural areas. Later, they became landholders who could collect
52
+ taxes from peasants and tenants, transmitting a tenth or eleventh of their produce
53
+ to the imperial treasury. In contrast to the jagirdars, who were given land grants
54
+ as part of their service to the Mughal government, the zamindar tenure was hereditary.
55
+ The zamindars performed the functions of the ancient rajas (kings) or chieftains.
56
+ They were landowners who were expected to pay a fixed tribute to the Mughal emperor.
57
+ - source_sentence: 'In a global context, many companies have significant ______ power
58
+ due to their ability to threaten governments, in the face of ________ with relocation
59
+ to other territories, which Beck (1998) describes as ''corporate power of _______.
60
+
61
+ A. Economic, Commercial competition, Social sanction
62
+
63
+ B. Political, Undesirable regulation, Transnational withdrawal
64
+
65
+ C. Social, Commercial competition, Social sanction
66
+
67
+ D. Social, Undesirable regulation, Transnational withdrawal'
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
  sentences:
69
+ - L'anémie est une condition caractérisée par une diminution du nombre de globules
70
+ rouges ou de la quantité d'hémoglobine dans le sang, entraînant une réduction
71
+ du transport de l'oxygène.
72
+ - Another critical method for evaluating internal controls is to focus on risk identification
73
+ and the specific potential losses associated with those risks. Organizations often
74
+ start with a thorough risk analysis to understand vulnerabilities, which can then
75
+ inform the development or enhancement of control activities intended to mitigate
76
+ those risks.
77
+ - The concept of 'transnational withdrawal' refers to the phenomenon where companies
78
+ threaten to relocate their operations to countries with more favorable conditions.
79
+ This can include lighter regulations, lower taxes, or more lenient labor standards.
80
+ The threat of relocation can compel governments to modify their policies or regulations
81
+ to keep corporations within their jurisdictions, thereby illustrating the leverage
82
+ that global companies hold.
83
+ - source_sentence: 'Can armed violence perpetrated by non-State actors ever amount
84
+ to an armed attack under Article 51 UN Charter?
85
+
86
+ A. The conduct of non-State actors can never amount to an armed attack
87
+
88
+ B. The Caroline case serves as precedent that non-State actors can under particular
89
+ circumstances commit an armed attack
90
+
91
+ C. There is no precedent in international law for the proposition that non-State
92
+ actors can commit an armed attack
93
+
94
+ D. Non-State can both commit an armed attack and possess a right of self-defence
95
+ under international law'
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
96
  sentences:
97
+ - In international law, the concept of an armed attack typically refers to the use
98
+ of force by one state against another, which is significant under the UN Charter
99
+ as it may trigger the right of self-defense. This term is often discussed in the
100
+ context of customary international law and the interpretations by bodies like
101
+ the International Court of Justice.
102
+ - '2. **Force and Motion**: According to Newton''s second law, the acceleration
103
+ of an object is directly proportional to the net force acting on it and inversely
104
+ proportional to its mass (F = ma). If an object can accelerate in response to
105
+ a force, this indicates that the force applied contributes to the net work done
106
+ on the object, thereby altering its kinetic energy.'
107
+ - 委託に伴って個人データを提供する場合、委託先は「第三者」に該当しないとみなされることがあります。この場合、原則として本人の同意��不要です。
108
+ - source_sentence: 'A builder had a contract to build a swimming pool for a residential
109
+ customer. That customer''s next door neighbor went to the builder and paid him
110
+ extra to break the contract with the customer and instead to build a swimming
111
+ pool on the neighbor''s premises. The builder commenced building a swimming pool
112
+ for the neighbor and breached his contract with the original customer. The original
113
+ customer sued his neighbor in a tort claim for damages. Does the original customer
114
+ have a valid claim against his neighbor?
115
+
116
+ A. Yes, the neighbor committed the tort of interference with contract relations
117
+ by intentionally interfering with an existing contract.
118
+
119
+ B. No, people cannot be held in slavery
120
+
121
+ C. they have the right to contract with whomever they please.
122
+
123
+ D. No, the only remedy for the original customer is to sue the builder for breach
124
+ of contract.
125
+
126
+ E. Yes, the neighbor committed the tort of interference with prospective advantage.'
127
  sentences:
128
+ - Ներքին գործերի նախարար - Պաշտոն, որը պատասխանատու է երկրի ներքին անվտանգության,
129
+ հասարակական կարգի և օրենքի պահպանման համար։
130
+ - A tort is a civil wrong that causes harm or loss to another person, resulting
131
+ in legal liability for the person who commits the tort. Tort law allows individuals
132
+ to seek compensation for injuries or damages caused by the wrongful acts of others,
133
+ distinct from breaches of contract.
134
+ - Substance use, such as alcohol and tobacco, during pregnancy can lead to various
135
+ complications including low birth weight, developmental issues, and increased
136
+ risk of infections, highlighting the importance of cessation and support for affected
137
+ mothers.
138
  datasets:
139
+ - DoDucAnh/MNLP_M3_rag_dataset
140
  pipeline_tag: sentence-similarity
141
  library_name: sentence-transformers
142
  ---
143
 
144
  # SentenceTransformer based on intfloat/multilingual-e5-large-instruct
145
 
146
+ This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) on the [mnlp_m3_rag_dataset](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset) dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
147
 
148
  ## Model Details
149
 
 
154
  - **Output Dimensionality:** 1024 dimensions
155
  - **Similarity Function:** Cosine Similarity
156
  - **Training Dataset:**
157
+ - [mnlp_m3_rag_dataset](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset)
158
  <!-- - **Language:** Unknown -->
159
  <!-- - **License:** Unknown -->
160
 
 
192
  model = SentenceTransformer("sentence_transformers_model_id")
193
  # Run inference
194
  sentences = [
195
+ "A builder had a contract to build a swimming pool for a residential customer. That customer's next door neighbor went to the builder and paid him extra to break the contract with the customer and instead to build a swimming pool on the neighbor's premises. The builder commenced building a swimming pool for the neighbor and breached his contract with the original customer. The original customer sued his neighbor in a tort claim for damages. Does the original customer have a valid claim against his neighbor?\nA. Yes, the neighbor committed the tort of interference with contract relations by intentionally interfering with an existing contract.\nB. No, people cannot be held in slavery\nC. they have the right to contract with whomever they please.\nD. No, the only remedy for the original customer is to sue the builder for breach of contract.\nE. Yes, the neighbor committed the tort of interference with prospective advantage.",
196
+ 'A tort is a civil wrong that causes harm or loss to another person, resulting in legal liability for the person who commits the tort. Tort law allows individuals to seek compensation for injuries or damages caused by the wrongful acts of others, distinct from breaches of contract.',
197
+ 'Substance use, such as alcohol and tobacco, during pregnancy can lead to various complications including low birth weight, developmental issues, and increased risk of infections, highlighting the importance of cessation and support for affected mothers.',
198
  ]
199
  embeddings = model.encode(sentences)
200
  print(embeddings.shape)
 
246
 
247
  ### Training Dataset
248
 
249
+ #### mnlp_m3_rag_dataset
250
 
251
+ * Dataset: [mnlp_m3_rag_dataset](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset) at [e16d937](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset/tree/e16d937f0bb9981bb081e8c16a3eda5b3fbbc68a)
252
  * Size: 594,028 training samples
253
  * Columns: <code>anchor</code> and <code>positive</code>
254
  * Approximate statistics based on the first 1000 samples:
255
+ | | anchor | positive |
256
+ |:--------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|
257
+ | type | string | string |
258
+ | details | <ul><li>min: 21 tokens</li><li>mean: 359.4 tokens</li><li>max: 512 tokens</li></ul> | <ul><li>min: 4 tokens</li><li>mean: 56.63 tokens</li><li>max: 433 tokens</li></ul> |
259
  * Samples:
260
+ | anchor | positive |
261
+ |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
262
+ | <code>Little Lopsy fluttered into our home and our hearts one Saturday morning this summer. My husband went out to do something, and when he opened the door there was a great flutter on the ground and something came into the living room. It was clear that whatever it was was hurt. I was in a bit of a shock and didn't know what to do next. Fortunately it calmed down and tried to hide itself in a corner. I realized it was a sparrow chick . There are a few sparrow nests under the roof of our apartment, and this little fellow must have fallen out and hurt itself. It was also very young, and obviously far from ready to leave the safety of the nest. I ran to the place and found a box. Having read somewhere that one shouldn't touch a baby bird with one's hands, I picked the chick up with a hand towel and put it in the box. I placed the box outside the front door in the hope that the parents would try to feed it. They never came near it and I brought it inside. I placed the box on a table and it sl...</code> | <code>Having read somewhere that one shouldn't touch a baby bird with one's hands, I picked the chick up with a hand towel and put it in the box.</code> |
263
+ | <code>A thermal conductor is made of<br>A. types of rubber<br>B. types of wire<br>C. electrodes<br>D. that which conducts</code> | <code>A thermal conductor is a material that allows heat to flow through it easily. Common examples of thermal conductors include metals such as copper and aluminum, known for their high thermal conductivity due to their free-flowing electrons. Heat transfer occurs via conduction when heat energy moves from the hotter part of a conductor to the cooler part, often described by Fourier's Law of heat conduction.</code> |
264
+ | <code>A good example of increased demand may equal increased production is<br>A. soldiers eat beans, so beans are planted when there is war<br>B. dogs eat kibble, so stores sell it<br>C. cats eat mice, so mice are afraid of cats<br>D. people have babies, so baby clothes are made</code> | <code>Supply is the total amount of a specific good or service that is available to consumers. Supply can relate to the amount available at a specific price or the amount available across a range of prices if displayed on a graph.</code> |
265
  * Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
266
  ```json
267
  {
 
272
 
273
  ### Evaluation Dataset
274
 
275
+ #### mnlp_m3_rag_dataset
276
 
277
+ * Dataset: [mnlp_m3_rag_dataset](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset) at [e16d937](https://huggingface.co/datasets/DoDucAnh/MNLP_M3_rag_dataset/tree/e16d937f0bb9981bb081e8c16a3eda5b3fbbc68a)
278
+ * Size: 5,920 evaluation samples
279
  * Columns: <code>anchor</code> and <code>positive</code>
280
  * Approximate statistics based on the first 1000 samples:
281
  | | anchor | positive |
 
300
  #### Non-Default Hyperparameters
301
 
302
  - `eval_strategy`: steps
303
+ - `per_device_train_batch_size`: 16
304
+ - `per_device_eval_batch_size`: 16
305
+ - `gradient_accumulation_steps`: 2
306
+ - `learning_rate`: 2e-05
307
+ - `warmup_steps`: 5569
308
  - `fp16`: True
309
  - `load_best_model_at_end`: True
310
 
 
315
  - `do_predict`: False
316
  - `eval_strategy`: steps
317
  - `prediction_loss_only`: True
318
+ - `per_device_train_batch_size`: 16
319
+ - `per_device_eval_batch_size`: 16
320
  - `per_gpu_train_batch_size`: None
321
  - `per_gpu_eval_batch_size`: None
322
+ - `gradient_accumulation_steps`: 2
323
  - `eval_accumulation_steps`: None
324
  - `torch_empty_cache_steps`: None
325
+ - `learning_rate`: 2e-05
326
  - `weight_decay`: 0.0
327
  - `adam_beta1`: 0.9
328
  - `adam_beta2`: 0.999
329
  - `adam_epsilon`: 1e-08
330
  - `max_grad_norm`: 1.0
331
+ - `num_train_epochs`: 3
332
  - `max_steps`: -1
333
  - `lr_scheduler_type`: linear
334
  - `lr_scheduler_kwargs`: {}
335
  - `warmup_ratio`: 0.0
336
+ - `warmup_steps`: 5569
337
  - `log_level`: passive
338
  - `log_level_replica`: warning
339
  - `log_on_each_node`: True
 
430
  ### Training Logs
431
  | Epoch | Step | Training Loss | Validation Loss |
432
  |:--------:|:--------:|:-------------:|:---------------:|
433
+ | **0.15** | **2785** | **0.2684** | **0.0411** |
434
+ | 0.3001 | 5570 | 0.1112 | 0.0541 |
435
+ | 0.4501 | 8355 | 0.1153 | 0.0633 |
436
+ | 0.6001 | 11140 | 0.1045 | 0.0582 |
437
+ | 0.7501 | 13925 | 0.0943 | 0.0606 |
438
+ | 0.9002 | 16710 | 0.0883 | 0.0563 |
439
+ | 1.0502 | 19495 | 0.0744 | 0.0505 |
440
+ | 1.2002 | 22280 | 0.0592 | 0.0523 |
441
+ | 1.3502 | 25065 | 0.059 | 0.0516 |
442
+ | 1.5002 | 27850 | 0.0544 | 0.0617 |
443
+ | 1.6503 | 30635 | 0.0521 | 0.0549 |
444
+ | 1.8003 | 33420 | 0.0502 | 0.0589 |
445
+ | 1.9503 | 36205 | 0.0449 | 0.0550 |
446
+ | 2.1003 | 38990 | 0.0369 | 0.0619 |
447
+ | 2.2503 | 41775 | 0.0331 | 0.0604 |
448
+ | 2.4004 | 44560 | 0.0308 | 0.0566 |
449
+ | 2.5504 | 47345 | 0.0294 | 0.0533 |
450
+ | 2.7004 | 50130 | 0.0286 | 0.0531 |
451
+ | 2.8504 | 52915 | 0.0266 | 0.0537 |
452
+
453
+ * The bold row denotes the saved checkpoint. The training took 6h52m on a RTX5090
454
 
455
  ### Framework Versions
456
+ - Python: 3.12.3
457
  - Sentence Transformers: 4.1.0
458
+ - Transformers: 4.52.4
459
+ - PyTorch: 2.7.0+cu128
460
  - Accelerate: 1.7.0
461
  - Datasets: 3.6.0
462
  - Tokenizers: 0.21.1
config.json CHANGED
@@ -20,7 +20,7 @@
20
  "pad_token_id": 1,
21
  "position_embedding_type": "absolute",
22
  "torch_dtype": "float32",
23
- "transformers_version": "4.52.3",
24
  "type_vocab_size": 1,
25
  "use_cache": true,
26
  "vocab_size": 250002
 
20
  "pad_token_id": 1,
21
  "position_embedding_type": "absolute",
22
  "torch_dtype": "float32",
23
+ "transformers_version": "4.52.4",
24
  "type_vocab_size": 1,
25
  "use_cache": true,
26
  "vocab_size": 250002
config_sentence_transformers.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "__version__": {
3
  "sentence_transformers": "4.1.0",
4
- "transformers": "4.52.3",
5
- "pytorch": "2.7.0+cu126"
6
  },
7
  "prompts": {},
8
  "default_prompt_name": null,
 
1
  {
2
  "__version__": {
3
  "sentence_transformers": "4.1.0",
4
+ "transformers": "4.52.4",
5
+ "pytorch": "2.7.0+cu128"
6
  },
7
  "prompts": {},
8
  "default_prompt_name": null,
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1e2f752905c4511972f69bfbb0dc95468f97d108dab1f69093614fc6e79fe1a1
3
  size 2239607176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2cb5ff432aafe9465077d3be0c2750508fcebf23fcc965e95b4c30a355b20ed
3
  size 2239607176
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc67c94763bb3d5ae4066c0d59f3305d8199398cf64391c50b68b21528394025
3
  size 5905
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba8eb39843c6da8e2a874d163673a0ac9df77631f42e8c92e768d75bead0d471
3
  size 5905