SentenceTransformer based on thenlper/gte-small

This is a sentence-transformers model finetuned from thenlper/gte-small. 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: thenlper/gte-small
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': False}) with Transformer model: BertModel 
  (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})
)

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("Alexhuou/embedder_model")
# Run inference
sentences = [
    'Gulde’s tax basis in Chyme Partnership was $26,000 at the time Gulde received a liquidating distribution of $12,000 cash and land with an adjusted basis to Chyme of $10,000 and a fair market value of $30,000. Chyme did not have unrealized receivables, appreciated inventory, or properties that had been contributed by its partners. What was the amount of Gulde’s basis in the land?',
    'A company exchanged land with an appraised value of $50,000 and an original cost of $20,000 for machinery with a fair value of $55,000. Assuming that the transaction has commercial substance, what is the gain on the exchange?',
    'The percentage of children in Ethiopia (age 8) who reported physical punishment by teachers in the past week in 2009 was about what?',
]
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: 5,700 training samples
Columns: sentence_0, sentence_1, and sentence_2

Approximate statistics based on the first 1000 samples:

	sentence_0	sentence_1	sentence_2
type	string	string	string
details	min: 5 tokens mean: 49.22 tokens max: 512 tokens	min: 5 tokens mean: 48.59 tokens max: 440 tokens	min: 5 tokens mean: 41.92 tokens max: 512 tokens

Samples:

sentence_0	sentence_1	sentence_2
This question refers to the following information. "The spontaneous forces of capitalism have been steadily growing in the countryside in recent years, with new rich peasants springing up everywhere and many well-to-do middle peasants striving to become rich peasants. On the other hand, many poor peasants are still living in poverty for lack of sufficient means of production, with some in debt and others selling or renting out their land. If this tendency goes unchecked, the polarization in the countryside will inevitably be aggravated day by day. Those peasants who lose their land and those who remain in poverty will complain that we are doing nothing to save them from ruin or to help them overcome their difficulties. Nor will the well-to-do middle peasants who are heading in the capitalist direction be pleased with us, for we shall never be able to satisfy their demands unless we intend to take the capitalist road. Can the worker-peasant alliance continue to stand in these circumstan...	This question refers to the following information. Woman, wake up; the bell of reason is being heard throughout the whole universe; discover your rights. Enslaved man has multiplied his strength, [but] having become free, he has become unjust to his companion. Oh, women, women! When will you cease to be blind? What advantage have you received from the Revolution? A more pronounced scorn, a more marked disdain. If our leaders persist, courageously oppose the force of reason to their empty pretentions of superiority. Regardless of what barriers confront you, it is in your power to free yourselves! Olympe de Gouges, "Declaration of the Rights of Woman and the Female Citizen," 1791 The independence? Nothing of what I hoped for was achieved. I had expected that my children would be able to have an education, but they did not get it. We were poor peasants then, we are poor peasants now. Nothing has changed. Everything is the same. The only thing is that we are free, the war is over, we work ...	`Which of the following most likely explains why Venus does not have a strong magnetic field?`
`In conducting international market research, there are three types of equivalence. Which of the following is NOT one of the equivalences?`	`Economic—marketing should encourage long-term economic development as opposed to short-term economic development.`	`The domain of the function $h(x) = \sqrt{25-x^2}+\sqrt{-(x-2)}$ is an interval of what width?`
`Which value is the most reasonable estimate of the volume of air an adult breathes in one day?`	`By what nickname is the Federal National Mortgage Association known?`	`If technology makes production less expensive and at the same time exports decrease which of the following will result with certainty?`

Loss: TripletLoss with these parameters:

{
    "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
    "triplet_margin": 5
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 30
multi_dataset_batch_sampler: round_robin

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
num_train_epochs: 30
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
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: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: False
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
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: round_robin

Training Logs

Epoch	Step	Training Loss
1.4006	500	1.7342
2.8011	1000	0.8812
1.4006	500	0.5667
2.8011	1000	0.3886
4.2017	1500	0.2434
5.6022	2000	0.1532
7.0028	2500	0.1159
8.4034	3000	0.079
9.8039	3500	0.0524
11.2045	4000	0.0442
12.6050	4500	0.03
14.0056	5000	0.0246
15.4062	5500	0.0196
16.8067	6000	0.0137
18.2073	6500	0.0161
19.6078	7000	0.0093
21.0084	7500	0.0109
22.4090	8000	0.0055
23.8095	8500	0.0047
25.2101	9000	0.0044
26.6106	9500	0.0033
28.0112	10000	0.0043
29.4118	10500	0.0027

Framework Versions

Python: 3.11.13
Sentence Transformers: 4.1.0
Transformers: 4.52.4
PyTorch: 2.6.0+cu124
Accelerate: 1.7.0
Datasets: 3.6.0
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",
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Alexhuou
/

embedder_model