metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1340
  - loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-base-en-v1.5
widget:
  - source_sentence: Can you tell me about the origin of the word 'Shehnai'?
    sentences:
      - >-
        Krishan Kant (28 February 1927 – 27 July 2002) was the tenth Vice
        President of India from 1997 until his death. Previously, he was
        Governor of Andhra Pradesh from 1990 to 1997.
      - >-
        Acherontia lachesis is a large (up to 13 cm wingspan) Sphingid moth
        found in India and much of the Oriental region, one of the three species
        of Death's-head Hawkmoth, also known as the "Bee Robber".
      - >-
        A Shehnai is a South Asian music instrument which is normally played at
        marriages and other ceremonies, rites and rituals. The word itself is of
        Muslim/Turkish origin, combining 'Sheh' (or 'Shah') 'Royal' and '-Nai'
        or 'Ney', a type of Flute. A version of the "Shehnai", the "Surnai", is
        also played in the Northern and North-western areas of India and
        Pakistan, in particular at traditional Polo matches.
  - source_sentence: How do scammers typically operate in these scams?
    sentences:
      - >-
        The Singapore strategy was a strategy about defending the British Empire
        in the Asian Far East, mainly against the Empire of Japan. The strategy
        involved a number of different plans and stages, developed between 1919
        and 1941. The basic idea was to base a fleet of ships in the Far East.
        This fleet could then be used to stop and defeat a Japanese force
        heading towards India or Australia. In 1919, Singapore was chosen
        because of its strategic location at the end of the Strait of Malacca.
      - >-
        The Non-Cooperation Movement was a significant phase of the Indian
        independence movement from British rule. It was led by Mohandas
        Karamchand Gandhi after the Jallianwala Bagh Massacre. It aimed to
        resist British rule in India through non-violent means or "satyagraha".
        Protestors would refuse to buy British goods, adopt nihal use of local
        handicrafts and picket and liquor shops. The ideas of Ahimsa and
        nonviolence, and Gandhi's ability to rally hundreds of thousands of
        common citizens towards the cause of Indian independence, were first
        seen on a large scale in this movement through the summer 1920. Gandhi
        feared that the movement might lead to popular violence. The
        non-cooperation movement was launched on 12th August, 1921.
      - >-
        A technical support scam is a form of telephone fraud that tricks people
        by pretending that they are a service which helps people fix their
        computers. In most cases they convince the victim they have a computer
        problem that does not actually exist. A common type is when someone gets
        a call from someone (usually from places like India or Pakistan)
        pretending to be from a company that sounds real such as "Microsoft" or
        "Windows" support. Often the caller tries to gain the victim's trust.
        They may use confusing and very technical language to sound authentic.
        They may ask the victim to perform several tasks on their computer.
        Often they target legitimate files on the victim's computer saying these
        are viruses. These tactics are designed to scare people into letting the
        scammer fix the problem (that does not really exist). The caller may
        have the victim install malicious software that could capture sensitive
        data, such as online banking passwords or credit card information.
  - source_sentence: How is Northeast India connected to the rest of India?
    sentences:
      - >-
        Air India Flight 182 was a passenger plane which, on June 23, 1985,
        exploded from a bomb that was placed on the plane. The aircraft was
        going between Montréal-Mirabel International Airport, Montreal, Quebec,
        and New Delhi, India. It was an Air India Boeing 747-237B, registration
        VT-EFO. The bombing was called the largest mass murder in modern
        Canadian history, and the deadliest act of air terrorism before 9/11.
      - >-
        Hinduism is not only a religion but also a way of life. Hinduism is
        widely practiced in South Asia mainly in India and Nepal. Hinduism is
        the oldest religion in the world, and Hindus refer to it as "", "the
        eternal tradition," or the "eternal way," beyond human history. Scholars
        regard Hinduism as a combination of different Indian cultures and
        traditions, with diverse roots. Hinduism has no founder and origins of
        Hinduism is unknown. What we now call Hinduism have roots in cave
        paintings that have been preserved from Mesolithic sites dating from c.
        30,000 BCE in Bhimbetka, near present-day Bhopal, in the Vindhya
        Mountains in the Madhya Pradesh." There was no concept of religion in
        India and Hinduism was not a religion. Hinduism as a religion started to
        develop between 500 BCE and 300 CE, after the Vedic period (1500 BCE to
        500 BCE).
      - >-
        Various groups are involved in the Insurgency in Northeast India,
        India's northeast states, which are connected to the rest of the
        Republic of India by a narrow strip of land known as the Siliguri
        Corridor. In the region several armed factions operate. Some groups call
        for a separate state, others for regional autonomy, while some extreme
        groups demand complete independence.
  - source_sentence: How many songs did Rafi sing during his career?
    sentences:
      - "Inder Kumar Gujral (4 December 1919\_– 30 November 2012) was an Indian politician. He was the 12th Prime Minister of India from April 1997 to March 1998. Gujral was the third Prime Minister to be from the Rajya Sabha."
      - >-
        Mohammed Rafi (, , December 24, 1924 – July 31, 1980) was a popular
        Bollywood playback singer. In a career of over 40 years, Rafi sang more
        than 26,000 songs in the national languages of India and sometimes in
        other languages.
      - >-
        The University of Calcutta (informally known as Calcutta University or
        CU) is a public state university located in Kolkata (formerly
        "Calcutta"), West Bengal, India. It was created on 24 January 1857.
        Within India it is recognized as a "Five-Star University" and a "Centre
        with Potential for Excellence" by the University Grants Commission and
        the National Assessment and Accreditation Council.
  - source_sentence: Who was Lal Bahadur Shastri?
    sentences:
      - >-
        The Bharatiya Janata Party (abbreviated BJP) is one of the two major
        political parties in India. (The second being the Indian National
        Congress). Since the Indian elections in 2014, the BJP has 303 of the
        542 seats in the Lok Sabha, the lower house of the Parliament of India
        and 78 of the 238 seats in Rajya Sabha, the upper house of the
        Parliament of India. Amit Shah is the national president of BJP since
        2014.
      - >-
        Rex Vernon Whitehead (26 October 1948 – 26 June 2014) was an Australian
        Test cricket match umpire and cricketer. He umpired four Test matches
        between 1981 and 1982. His first match was between Australia and India
        in Sydney on 2 January to 4 January 1981. Altogether, he umpired 15
        first-class matches in his career between 1979 and 1983.
      - "Lal Bahadur Shastri (, , 2 October 1904\_– 11 January 1966) was an Indian politician. He was the 2nd Prime Minister of India from 1964 to 1966. He was a senior leader of the Indian National Congress political party."
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on BAAI/bge-base-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. 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.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sunjupskilling/sunj-bge-base-en-v1.5")
# Run inference
sentences = [
    'Who was Lal Bahadur Shastri?',
    'Lal Bahadur Shastri (, , 2 October 1904\xa0– 11 January 1966) was an Indian politician. He was the 2nd Prime Minister of India from 1964 to 1966. He was a senior leader of the Indian National Congress political party.',
    'Rex Vernon Whitehead (26 October 1948 – 26 June 2014) was an Australian Test cricket match umpire and cricketer. He umpired four Test matches between 1981 and 1982. His first match was between Australia and India in Sydney on 2 January to 4 January 1981. Altogether, he umpired 15 first-class matches in his career between 1979 and 1983.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

Unnamed Dataset

Size: 1,340 training samples
Columns: question and context
Approximate statistics based on the first 1000 samples:
question context
type string string
details
min: 6 tokens
mean: 12.39 tokens
max: 24 tokens

min: 9 tokens
mean: 83.99 tokens
max: 510 tokens

	question	context
type	string	string
details	min: 6 tokens mean: 12.39 tokens max: 24 tokens	min: 9 tokens mean: 83.99 tokens max: 510 tokens

Samples:

question	context
`What is Basil commonly known as?`	Basil ("Ocimum basilicum") ( or ) is a plant of the Family Lamiaceae. It is also known as Sweet Basil or Tulsi. It is a tender low-growing herb that is grown as a perennial in warm, tropical climates. Basil is originally native to India and other tropical regions of Asia. It has been cultivated there for more than 5,000 years. It is prominently featured in many cuisines throughout the world. Some of them are Italian, Thai, Vietnamese and Laotian cuisines. It grows to between 30–60 cm tall. It has light green, silky leaves 3–5 cm long and 1–3 cm broad. The leaves are opposite each other. The flowers are quite big. They are white in color and arranged as a spike.
`Where is Basil originally native to?`	Basil ("Ocimum basilicum") ( or ) is a plant of the Family Lamiaceae. It is also known as Sweet Basil or Tulsi. It is a tender low-growing herb that is grown as a perennial in warm, tropical climates. Basil is originally native to India and other tropical regions of Asia. It has been cultivated there for more than 5,000 years. It is prominently featured in many cuisines throughout the world. Some of them are Italian, Thai, Vietnamese and Laotian cuisines. It grows to between 30–60 cm tall. It has light green, silky leaves 3–5 cm long and 1–3 cm broad. The leaves are opposite each other. The flowers are quite big. They are white in color and arranged as a spike.
`What is the significance of the Roerich Pact?`	The Roerich Pact is a treaty on Protection of Artistic and Scientific Institutions and Historic Monuments, signed by the representatives of 21 states in the Oval Office of the White House on 15 April 1935. As of January 1, 1990, the Roerich Pact had been ratified by ten nations: Brazil, Chile, Colombia, Cuba, the Dominican Republic, El Salvador, Guatemala, Mexico, the United States, and Venezuela. It went into effect on 26 August 1935. The Government of India approved the Treaty in 1948, but did not take any further formal action. The Roerich Pact is also known as "Pax Cultura" ("Cultural Peace" or "Peace through Culture"). The most important part of the Roerich Pact is the legal recognition that the protection of culture is always more important than any military necessity.

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Evaluation Dataset

Unnamed Dataset

Size: 100 evaluation samples
Columns: question and context
Approximate statistics based on the first 100 samples:
question context
type string string
details
min: 8 tokens
mean: 12.2 tokens
max: 20 tokens

min: 22 tokens
mean: 85.76 tokens
max: 510 tokens

	question	context
type	string	string
details	min: 8 tokens mean: 12.2 tokens max: 20 tokens	min: 22 tokens mean: 85.76 tokens max: 510 tokens

Samples:

question	context
`What are the bases of political relations between India and Ireland?`	`Indo-Irish relations between the Republic of Ireland and the Republic of India picked up steam during the freedom struggles of the respective countries against a common imperial empire in the United Kingdom. Political relations between the two states have largely been based on socio-cultural ties, although political and economic ties have also helped build relations. Indians recognise Northern Ireland as part of its country.`
`When did Rex Whitehead umpire his first Test match?`	`Rex Vernon Whitehead (26 October 1948 – 26 June 2014) was an Australian Test cricket match umpire and cricketer. He umpired four Test matches between 1981 and 1982. His first match was between Australia and India in Sydney on 2 January to 4 January 1981. Altogether, he umpired 15 first-class matches in his career between 1979 and 1983.`
`What can you tell me about Nayaganj?`	`Nayaganj is a village in Vaishali District, Bihar, India. It is very close to the river Ganga. It is also a postal office of India`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 20.0,
    "similarity_fct": "cos_sim"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 3e-06
weight_decay: 0.03
max_steps: 332
warmup_ratio: 0.1
warmup_steps: 1
fp16: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 3e-06
weight_decay: 0.03
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 3.0
max_steps: 332
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 1
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss
0.1190	10	0.0998	-
0.2381	20	0.082	0.0253
0.3571	30	0.0843	-
0.4762	40	0.0496	0.0138
0.5952	50	0.0731	-
0.7143	60	0.0244	0.0093
0.8333	70	0.0338	-
0.9524	80	0.0484	0.0075
1.0714	90	0.0258	-
1.1905	100	0.0226	0.0067
1.3095	110	0.0331	-
1.4286	120	0.0193	0.0061
1.5476	130	0.0299	-
1.6667	140	0.0146	0.0055
1.7857	150	0.0228	-
1.9048	160	0.0543	0.0035
2.0238	170	0.0368	-
2.1429	180	0.025	0.0031
2.2619	190	0.0113	-
2.3810	200	0.0123	0.0029
2.5	210	0.0301	-
2.6190	220	0.0358	0.0027
2.7381	230	0.009	-
2.8571	240	0.01	0.0024
2.9762	250	0.0152	-
3.0952	260	0.013	0.0021
3.2143	270	0.0121	-
3.3333	280	0.012	0.0020
3.4524	290	0.0168	-
3.5714	300	0.0292	0.0019
3.6905	310	0.054	-
3.8095	320	0.0227	0.0019
3.9286	330	0.0144	-

Framework Versions

Python: 3.11.11
Sentence Transformers: 3.4.1
Transformers: 4.48.3
PyTorch: 2.5.1+cu124
Accelerate: 1.3.0
Datasets: 3.3.0
Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}