metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:200
  - loss:CoSENTLoss
base_model: avsolatorio/GIST-small-Embedding-v0
widget:
  - source_sentence: who is imf chief economist?
    sentences:
      - >-
        Metoprolol succinate is also known by the brand name Toprol XL. It is
        the extended-release form of metoprolol. Metoprolol succinate is
        approved to treat high blood pressure, chronic chest pain, and
        congestive heart failure.
      - >-
        He wants to confirm if he is talking to Priya or Angel Priya (I.e., if
        he is really talking to a girl or just a guy with fake profile) They are
        talking to you and want to see how you look. I found it normal but would
        say, be careful about whom do you share your picture with as they might
        misuse it. I hate this one.
      - >-
        A Dependent Care Flexible Spending Account, or “FSA,” is a pre-tax
        benefit account used to pay for dependent care services while you are at
        work. The money you contribute to a Dependent Care FSA is not subject to
        payroll taxes, so you end up paying less in taxes and taking home more
        of your paycheck.
  - source_sentence: is it possible to get a false negative flu test?
    sentences:
      - >-
        The saying "a piece of cake" means something that's simple to
        accomplish. If a school assignment is a piece of cake, it's so easy that
        you will barely have to think about it. Other ways to say "it's a piece
        of cake" include no problem or it's a breeze.
      - >-
        This variation in ability to detect viruses can result in some people
        who are infected with the flu having a negative rapid test result. (This
        situation is called a false negative test result.)
      - >-
        Unstable Wi-Fi is often caused by wireless congestion. Congestion
        problems are common in apartment complexes or densely packed
        neighborhoods. The more people using the internet, the greater the
        instability. When many people in the same area are working from home,
        connectivity suffers.
  - source_sentence: what are the requirements to become a health inspector?
    sentences:
      - >-
        You'll need an accredited health and safety qualification to become a
        health and safety inspector. Many recruiters ask for a NEBOSH diploma as
        it's accredited by the Institution of Occupational Health and Safety.
        This is a degree-level course that you can study at a variety of
        institutions, as well as online.
      - >-
        ['Open a PDF file in Acrobat DC.', 'Click on the “Export PDF” tool in
        the right pane.', 'Choose Microsoft Word as your export format, and then
        choose “Word Document.”', 'Click “Export.” If your PDF contains scanned
        text, the Acrobat Word converter will run text recognition
        automatically.']
      - >-
        ['Remain calm. ... ', "Don't take it personally. ... ", 'Use your best
        listening skills. ... ', 'Actively sympathize. ... ', 'Apologize
        gracefully. ... ', 'Find a solution. ... ', 'Take a few minutes on your
        own.']
  - source_sentence: is toprol xl the same as metoprolol?
    sentences:
      - >-
        Carbs: 35 grams. Fiber: 11 grams. Folate: 88% of the DV. Copper: 50% of
        the DV.
      - >-
        A Dependent Care Flexible Spending Account, or “FSA,” is a pre-tax
        benefit account used to pay for dependent care services while you are at
        work. The money you contribute to a Dependent Care FSA is not subject to
        payroll taxes, so you end up paying less in taxes and taking home more
        of your paycheck.
      - >-
        Metoprolol succinate is also known by the brand name Toprol XL. It is
        the extended-release form of metoprolol. Metoprolol succinate is
        approved to treat high blood pressure, chronic chest pain, and
        congestive heart failure.
  - source_sentence: how can i get copy of marriage license?
    sentences:
      - >-
        Probiotics can help with digestion Without probiotics, antibiotics can
        sometimes wipe out the protective gut bacteria, which is no good for
        your digestive system. Probiotics are thought to directly kill or
        inhibit the growth of harmful bacteria, stopping them from producing
        toxic substances that can make you ill.
      - >-
        Order in person You can order a certificate in person from Monday to
        Friday between 9am and 5pm. Please come to the register office at 45
        Beavor Lane, Hammersmith, London W6 9AR.
      - >-
        Worms and ants are more related because spiders contain hair and ants do
        not. Worms do not contain hair as well.
pipeline_tag: sentence-similarity
library_name: sentence-transformers

SentenceTransformer based on avsolatorio/GIST-small-Embedding-v0

This is a sentence-transformers model finetuned from avsolatorio/GIST-small-Embedding-v0. 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.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: avsolatorio/GIST-small-Embedding-v0
Maximum Sequence Length: 512 tokens
Output Dimensionality: 384 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': 384, '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("moshew/gist_small_ft_gooaq")
# Run inference
sentences = [
    'how can i get copy of marriage license?',
    'Order in person You can order a certificate in person from Monday to Friday between 9am and 5pm. Please come to the register office at 45 Beavor Lane, Hammersmith, London W6 9AR.',
    'Worms and ants are more related because spiders contain hair and ants do not. Worms do not contain hair as well.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

Training Details

Training Dataset

Unnamed Dataset

Size: 200 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 200 samples:
sentence1 sentence2 label
type string string float
details
min: 8 tokens
mean: 11.8 tokens
max: 23 tokens

min: 18 tokens
mean: 61.8 tokens
max: 125 tokens

min: 0.0
mean: 0.5
max: 1.0

	sentence1	sentence2	label
type	string	string	float
details	min: 8 tokens mean: 11.8 tokens max: 23 tokens	min: 18 tokens mean: 61.8 tokens max: 125 tokens	min: 0.0 mean: 0.5 max: 1.0

Samples:

sentence1	sentence2	label
`how many days can i drive my car without mot?`	`If your car fails its MOT you can only continue to drive it if the previous year's MOT is still valid - which might occur if you submitted the car for its test two weeks early. You can still drive it away from the testing centre or garage if no 'dangerous' problems were identified during the MOT.`	`1.0`
`how many days can i drive my car without mot?`	`Low-FODMAP vegetables include: Bean sprouts, capsicum, carrot, choy sum, eggplant, kale, tomato, spinach and zucchini ( 7 , 8 ). Summary: Vegetables contain a diverse range of FODMAPs. However, many vegetables are naturally low in FODMAPs.`	`0.0`
`what are underlying shares of stock?`	`Underlying Shares means the shares of Common Stock issued and issuable upon conversion of the Preferred Stock, upon exercise of the Warrants and issued and issuable in lieu of the cash payment of dividends on the Preferred Stock in accordance with the terms of the Certificate of Designation.`	`1.0`

Loss: CoSENTLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 1
warmup_ratio: 0.1
seed: 12
bf16: True
half_precision_backend: cpu_amp
dataloader_num_workers: 4

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
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: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
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: 12
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: True
fp16: False
fp16_opt_level: O1
half_precision_backend: cpu_amp
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: 4
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}
tp_size: 0
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
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: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss
0.0769	1	0.2709

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.1.0
Transformers: 4.51.1
PyTorch: 2.6.0+cu124
Accelerate: 1.5.2
Datasets: 3.5.0
Tokenizers: 0.21.1

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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}