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': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 768, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    '[SECTION HEADING] § 24.6 Performance appraisal system. The members of the Service shall be subject to a performance appraisal system that is designed to encourage excellence in performance and shall provide for periodic and systematic assessment of the performance of members.',
    '[SECTION HEADING] § 24.6 Performance appraisal system. The members of the Service shall be subject to a performance appraisal system that is designed to encourage excellence in performance and shall provide for periodic and systematic assessment of the performance of members.',
    '[SUBSECTION a] General rules.. (1) An HMO or CMP that has an APCRP (as determined under § 417.590) greater than its ACR (as determined under § 417.594) must elect one of the options specified in paragraph (b) of this section. [CLAUSE 2] The dollar value of the elected option must, over the course of a contract period, be at least equal to the difference between the APCRP and the proposed ACR.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 1.0000, 0.6782],
#         [1.0000, 1.0000, 0.6782],
#         [0.6782, 0.6782, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 24,880 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: 35 tokens
mean: 105.52 tokens
max: 335 tokens

min: 37 tokens
mean: 112.72 tokens
max: 335 tokens

	sentence_0	sentence_1
type	string	string
details	min: 35 tokens mean: 105.52 tokens max: 335 tokens	min: 37 tokens mean: 112.72 tokens max: 335 tokens

Samples:

sentence_0	sentence_1
[ITEM i] The name and TIN of the CJR collaborator and the name, TIN, and NPI of the collaboration agent. [ITEM ii] The start date and, if applicable, end date, for the distribution arrangement between the CJR collaborator and the collaboration agent. [ENUM Downstream collaboration agents.] (3) For each physician, nonphysician practitioner, or therapist who is a downstream collaboration agent during the period of the CJR performance year specified by CMS— [ITEM i] The name and TIN of the CJR collaborator and the [MASK] TIN, and NPI of the downstream collaboration	[ITEM i] The name and TIN of the CJR collaborator and the name, TIN, and NPI of the collaboration agent. [ITEM ii] The start date and, if applicable, end date, for the distribution arrangement between the CJR collaborator and the collaboration agent. [ENUM Downstream collaboration agents.] (3) For each physician, nonphysician practitioner, or therapist who is a downstream collaboration agent during the period of the CJR performance year specified by CMS— [ITEM i] The name and TIN of the CJR collaborator and the name and TIN of the collaboration agent and the name, TIN, and NPI of the downstream collaboration
[SUBSECTION a] Termination of agreements.. (1) CMS may terminate any approved agreement if it finds, after the procedures described in this paragraph are followed that the State system does not satisfactorily meet the requirements of section 1886(c) of the Act or the regulations in this subpart. A termination must be effective on the last day of a calendar quarter. [CLAUSE 2] CMS will give the State reasonable notice of the proposed termination of an agreement [MASK] days before the effective date of the termination. [CLAUSE 3] CMS will give the State the opportunity to present evidence to refute the finding.	[SUBSECTION a] Termination of agreements.. (1) CMS may terminate any approved agreement if it finds, after the procedures described in this paragraph are followed that the State system does not satisfactorily meet the requirements of section 1886(c) of the Act or the regulations in this subpart. A termination must be effective on the last day of a calendar quarter. [CLAUSE 2] CMS will give the State reasonable notice of the proposed termination of an agreement and of the reasons for the termination at least 90 days before the effective date of the termination. [CLAUSE 3] CMS will give the State the opportunity to present evidence to refute the finding.
[CLAUSE 4] The amount of the post-TDAPA add-on payment adjustment is equal to 65 percent of the amount calculated in paragraph (g)(2) of this section, multiplied by the reduction factor specified in paragraph (g)(3) of this section, and multiplied by the latest available forecast of annual growth in the ESRD bundled market basket composite price proxy for pharmaceuticals. [CLAUSE 5] The post-TDAPA [MASK] ESRD PPS claim is adjsuted by any applicable patient-level case-mix adjustments under § 413.235. [CITATIONS]	[CLAUSE 4] The amount of the post-TDAPA add-on payment adjustment is equal to 65 percent of the amount calculated in paragraph (g)(2) of this section, multiplied by the reduction factor specified in paragraph (g)(3) of this section, and multiplied by the latest available forecast of annual growth in the ESRD bundled market basket composite price proxy for pharmaceuticals. [CLAUSE 5] The post-TDAPA add-on payment adjustment that is applied to an ESRD PPS claim is adjsuted by any applicable patient-level case-mix adjustments under § 413.235. [CITATIONS]

Loss: DenoisingAutoEncoderLoss

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
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: 3
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}
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
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: False
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: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.3215	500	5.7169
0.6431	1000	4.3196
0.9646	1500	3.8613
1.2862	2000	3.5443
1.6077	2500	3.357
1.9293	3000	3.2075
2.2508	3500	3.0466
2.5723	4000	2.9261
2.8939	4500	2.8525

Framework Versions

Python: 3.12.6
Sentence Transformers: 5.2.0
Transformers: 4.56.0
PyTorch: 2.8.0+cu129
Accelerate: 1.10.1
Datasets: 4.4.1
Tokenizers: 0.22.0

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",
}

DenoisingAutoEncoderLoss

@inproceedings{wang-2021-TSDAE,
    title = "TSDAE: Using Transformer-based Sequential Denoising Auto-Encoderfor Unsupervised Sentence Embedding Learning",
    author = "Wang, Kexin and Reimers, Nils and Gurevych, Iryna",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
    month = nov,
    year = "2021",
    address = "Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    pages = "671--688",
    url = "https://arxiv.org/abs/2104.06979",
}

Downloads last month: 12

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for atx-labs/bge-base-en-cfr

Base model

BAAI/bge-base-en-v1.5

Finetuned

(434)

this model