SentenceTransformer based on jinaai/jina-embedding-b-en-v1
This is a sentence-transformers model finetuned from jinaai/jina-embedding-b-en-v1. 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: jinaai/jina-embedding-b-en-v1
- 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}) with Transformer model: T5EncoderModel
(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 ๐ค Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'Sort my investment portfolio by ESG rating, please.',
'Show my investments sorted by ESG rating.',
'Can you show my worst performing holdings',
]
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
- Dataset:
test-eval
- Evaluated with
InformationRetrievalEvaluator
Metric | Value |
---|---|
cosine_accuracy@1 | 0.8636 |
cosine_accuracy@3 | 0.9924 |
cosine_accuracy@5 | 1.0 |
cosine_accuracy@10 | 1.0 |
cosine_precision@1 | 0.8636 |
cosine_precision@3 | 0.3308 |
cosine_precision@5 | 0.2 |
cosine_precision@10 | 0.1 |
cosine_recall@1 | 0.8636 |
cosine_recall@3 | 0.9924 |
cosine_recall@5 | 1.0 |
cosine_recall@10 | 1.0 |
cosine_ndcg@10 | 0.9437 |
cosine_mrr@10 | 0.9242 |
cosine_map@100 | 0.9242 |
Training Details
Training Datasets
Unnamed Dataset
- Size: 1,320 training samples
- Columns:
sentence_0
,sentence_1
, andlabel
- Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 label type string string float details - min: 4 tokens
- mean: 10.62 tokens
- max: 20 tokens
- min: 4 tokens
- mean: 9.06 tokens
- max: 17 tokens
- min: 1.0
- mean: 1.0
- max: 1.0
- Samples:
sentence_0 sentence_1 label How does my portfolio score look?
What is my portfolio score?
1.0
Show me the risk profile of my portfolio.
Details on my portfolio risk
1.0
Which of my shares are the most erratic?
Which of my stocks are most volatile?
1.0
- Loss:
MultipleNegativesRankingLoss
with these parameters:{ "scale": 20.0, "similarity_fct": "cos_sim" }
Unnamed Dataset
- Size: 1,320 training samples
- Columns:
sentence_0
,sentence_1
, andlabel
- Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 label type string string float details - min: 4 tokens
- mean: 10.62 tokens
- max: 22 tokens
- min: 4 tokens
- mean: 9.05 tokens
- max: 17 tokens
- min: 1.0
- mean: 1.0
- max: 1.0
- Samples:
sentence_0 sentence_1 label What holdings carry the least risk in my portfolio?
What are the least risky holdings in my portfolio?
1.0
How have my investments fared over the previous year?
How has my portfolio performed over the last year?
1.0
How well is my portfolio performing?
How is my portfolio performing
1.0
- Loss:
CosineSimilarityLoss
with these parameters:{ "loss_fct": "torch.nn.modules.loss.MSELoss" }
Training Hyperparameters
Non-Default Hyperparameters
eval_strategy
: stepsper_device_train_batch_size
: 32per_device_eval_batch_size
: 32num_train_epochs
: 15multi_dataset_batch_sampler
: round_robin
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: stepsprediction_loss_only
: Trueper_device_train_batch_size
: 32per_device_eval_batch_size
: 32per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonetorch_empty_cache_steps
: Nonelearning_rate
: 5e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1num_train_epochs
: 15max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.0warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 42data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Falsefp16
: Falsefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}tp_size
: 0fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Nonehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseinclude_for_metrics
: []eval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falseeval_on_start
: Falseuse_liger_kernel
: Falseeval_use_gather_object
: Falseaverage_tokens_across_devices
: Falseprompts
: Nonebatch_sampler
: batch_samplermulti_dataset_batch_sampler
: round_robin
Training Logs
Epoch | Step | Training Loss | test-eval_cosine_ndcg@10 |
---|---|---|---|
1.0 | 84 | - | 0.8877 |
2.0 | 168 | - | 0.8944 |
3.0 | 252 | - | 0.9042 |
4.0 | 336 | - | 0.9123 |
5.0 | 420 | - | 0.9241 |
5.9524 | 500 | 0.2478 | 0.9209 |
6.0 | 504 | - | 0.9209 |
7.0 | 588 | - | 0.9261 |
8.0 | 672 | - | 0.9327 |
9.0 | 756 | - | 0.9364 |
10.0 | 840 | - | 0.9370 |
11.0 | 924 | - | 0.9437 |
Framework Versions
- Python: 3.10.16
- Sentence Transformers: 4.1.0
- Transformers: 4.51.3
- PyTorch: 2.7.0
- Accelerate: 1.6.0
- 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",
}
MultipleNegativesRankingLoss
@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}
}
- Downloads last month
- 73
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
Model tree for 1shoomun/pq_cache_3
Base model
jinaai/jina-embedding-b-en-v1Evaluation results
- Cosine Accuracy@1 on test evalself-reported0.864
- Cosine Accuracy@3 on test evalself-reported0.992
- Cosine Accuracy@5 on test evalself-reported1.000
- Cosine Accuracy@10 on test evalself-reported1.000
- Cosine Precision@1 on test evalself-reported0.864
- Cosine Precision@3 on test evalself-reported0.331
- Cosine Precision@5 on test evalself-reported0.200
- Cosine Precision@10 on test evalself-reported0.100
- Cosine Recall@1 on test evalself-reported0.864
- Cosine Recall@3 on test evalself-reported0.992