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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'show me list of Employee department details',
    'Update the employee dependent marital status to Married',
    'Display the average performance rating of employees in the marketing department',
]
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]

Evaluation

Metrics

Information Retrieval

Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.9259
cosine_accuracy@3	0.963
cosine_accuracy@5	0.963
cosine_accuracy@10	0.963
cosine_precision@1	0.9259
cosine_precision@3	0.9136
cosine_precision@5	0.9185
cosine_precision@10	0.9111
cosine_recall@1	0.0492
cosine_recall@3	0.1401
cosine_recall@5	0.234
cosine_recall@10	0.4538
cosine_ndcg@10	0.9289
cosine_mrr@10	0.9444
cosine_map@100	0.9023

Training Details

Training Dataset

Unnamed Dataset

Size: 2,441 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: 7 tokens
mean: 7.37 tokens
max: 10 tokens
min: 5 tokens
mean: 9.75 tokens
max: 18 tokens
Samples:
sentence_0 sentence_1
Show me list of applicant Show applicant rating
Show me list of applicant Update my objective statement
Show me list of applicant Update my job title preferences
Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim" 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 30
per_device_eval_batch_size: 30
num_train_epochs: 111
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 30
per_device_eval_batch_size: 30
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: 111
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: False
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: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	cosine_ndcg@10
0.6098	50	-	0.5941
1.0	82	-	0.6506
1.2195	100	-	0.6689
1.8293	150	-	0.6894
2.0	164	-	0.6963
2.4390	200	-	0.7201
3.0	246	-	0.7488
3.0488	250	-	0.7519
3.6585	300	-	0.7863
4.0	328	-	0.8176
4.2683	350	-	0.8238
4.8780	400	-	0.8320
5.0	410	-	0.8471
5.4878	450	-	0.8525
6.0	492	-	0.8568
6.0976	500	1.6965	0.8533
6.7073	550	-	0.8646
7.0	574	-	0.8631
7.3171	600	-	0.8676
7.9268	650	-	0.8718
8.0	656	-	0.8765
8.5366	700	-	0.8764
9.0	738	-	0.8845
9.1463	750	-	0.8778
9.7561	800	-	0.8894
10.0	820	-	0.8848
10.3659	850	-	0.9048
10.9756	900	-	0.9029
11.0	902	-	0.9026
11.5854	950	-	0.8995
12.0	984	-	0.8956
12.1951	1000	1.0614	0.8922
12.8049	1050	-	0.9043
13.0	1066	-	0.9103
13.4146	1100	-	0.9057
14.0	1148	-	0.9097
14.0244	1150	-	0.9096
14.6341	1200	-	0.9223
15.0	1230	-	0.9258
15.2439	1250	-	0.9118
15.8537	1300	-	0.9207
16.0	1312	-	0.9239
16.4634	1350	-	0.9250
17.0	1394	-	0.9161
17.0732	1400	-	0.9203
17.6829	1450	-	0.9146
18.0	1476	-	0.9198
18.2927	1500	0.9705	0.9197
18.9024	1550	-	0.9250
19.0	1558	-	0.9248
19.5122	1600	-	0.9289

Framework Versions

Python: 3.12.7
Sentence Transformers: 3.3.0
Transformers: 4.46.2
PyTorch: 2.5.1
Accelerate: 0.26.0
Datasets: 3.1.0
Tokenizers: 0.20.3

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