This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. 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: BAAI/bge-small-en-v1.5
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, 'architecture': '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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("andreeaiaconi/bge-small-tcr-finetuned")
# Run inference
sentences = [
    'Housing Discrimination Ordinance Based on Source of Income.',
    "I'm in a number of conversations with Councilwoman Ortega, and the question comes up from communities are we really building an integrated city where folks with different incomes live throughout the city, or do we just put affordable housing in one neighborhood and we put it in that neighborhood over and over again?",
    'And I have, eh,',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.5551, 0.2274],
#         [0.5551, 1.0000, 0.1910],
#         [0.2274, 0.1910, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 858,828 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 3 tokens
mean: 14.15 tokens
max: 172 tokens
min: 4 tokens
mean: 33.63 tokens
max: 191 tokens

Samples:

sentence_0	sentence_1
`Selection of Preferred Project Team for Civic Center Public-Private-Partnership.`	`However, I do not. I'm going to just be very clear. I do not want to compromise those services, the quality of life for the rest of the city to get this project done.`
`Zoning Change for 4201 East Arkansas Avenue and Adjacent Properties.`	`The State Office of Information Technology and C++ are currently in negotiations around that issue of what can be built on the site. Now, I have no knowledge of those negotiations. They could limit building height or agreement could be made to pave the way to allow for the full 12 storeys permitted under campus zoning.`
`Censure of Councilmember Jeannine Pearce for Misconduct.`	`Good afternoon. My name is Andrew L.A. and I am.`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false 5}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 1
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: 32
per_device_eval_batch_size: 32
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: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: None
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
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}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
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
hub_revision: None
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: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0186	500	2.6987
0.0373	1000	2.4209
0.0559	1500	2.3033
0.0745	2000	2.2567
0.0931	2500	2.2035
0.1118	3000	2.1569
0.1304	3500	2.1311
0.1490	4000	2.0865
0.1677	4500	2.0609
0.1863	5000	2.0464
0.2049	5500	2.0345
0.2236	6000	2.0106
0.2422	6500	2.0006
0.2608	7000	1.9731
0.2794	7500	1.9502
0.2981	8000	1.9466
0.3167	8500	1.916
0.3353	9000	1.9119
0.3540	9500	1.9052
0.3726	10000	1.8764
0.3912	10500	1.883
0.4099	11000	1.8582
0.4285	11500	1.8695
0.4471	12000	1.8533
0.4657	12500	1.8454
0.4844	13000	1.8309
0.5030	13500	1.8235
0.5216	14000	1.8093
0.5403	14500	1.8005
0.5589	15000	1.8071
0.5775	15500	1.7777
0.5961	16000	1.7791
0.6148	16500	1.7613
0.6334	17000	1.7704
0.6520	17500	1.7749
0.6707	18000	1.7877
0.6893	18500	1.7414
0.7079	19000	1.761
0.7266	19500	1.7502
0.7452	20000	1.7268
0.7638	20500	1.735
0.7824	21000	1.749
0.8011	21500	1.7375
0.8197	22000	1.7196
0.8383	22500	1.7238
0.8570	23000	1.7278
0.8756	23500	1.7145
0.8942	24000	1.7158
0.9129	24500	1.7046
0.9315	25000	1.7095
0.9501	25500	1.7329
0.9687	26000	1.7218
0.9874	26500	1.7148

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.2.2
Transformers: 4.57.6
PyTorch: 2.9.0+cu126
Accelerate: 1.12.0
Datasets: 4.5.0
Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

bibtex

@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

bibtex

@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}
}

	sentence_0	sentence_1
type	string	string
details	min: 3 tokens mean: 14.15 tokens max: 172 tokens	min: 4 tokens mean: 33.63 tokens max: 191 tokens