You can deploy this model and start earning money today!

tomaarsen/
splade-distilbert-base-uncased-gooaq-peft-r64

Name: splade-distilbert-base-uncased-gooaq-peft-r64
Availability: InStock

Views

No views yet

splade-distilbert-base-uncased trained on GooAQ

This is a SPLADE Sparse Encoder model finetuned from distilbert/distilbert-base-uncased on the gooaq dataset using the sentence-transformers library. It maps sentences & paragraphs to a 30522-dimensional sparse vector space and can be used for semantic search and sparse retrieval.

Model Details

Model Description

Model Type: SPLADE Sparse Encoder
Base model: distilbert/distilbert-base-uncased
Maximum Sequence Length: 256 tokens
Output Dimensionality: 30522 dimensions
Similarity Function: Dot Product
Training Dataset:
- gooaq
Language: en
License: apache-2.0

Model Sources

Documentation: Sentence Transformers Documentation
Documentation: Sparse Encoder Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sparse Encoders on Hugging Face

Full Model Architecture

SparseEncoder(
  (0): MLMTransformer({'max_seq_length': 256, 'do_lower_case': False, 'architecture': 'DistilBertForMaskedLM'})
  (1): SpladePooling({'pooling_strategy': 'max', 'activation_function': 'relu', 'word_embedding_dimension': 30522})
)

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 SparseEncoder

# Download from the 🤗 Hub
model = SparseEncoder("tomaarsen/splade-distilbert-base-uncased-gooaq-peft-r64")
# Run inference
queries = [
    "how many days for doxycycline to work on sinus infection?",
]
documents = [
    'Treatment of suspected bacterial infection is with antibiotics, such as amoxicillin/clavulanate or doxycycline, given for 5 to 7 days for acute sinusitis and for up to 6 weeks for chronic sinusitis.',
    'Most engagements typically have a cocktail dress code, calling for dresses at, or slightly above, knee-length and high heels. If your party states a different dress code, however, such as semi-formal or dressy-casual, you may need to dress up or down accordingly.',
    'The average service life of a gas furnace is about 15 years, but the actual life span of an individual unit can vary greatly. There are a number of contributing factors that determine the age a furnace reaches: The quality of the equipment.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 30522] [3, 30522]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[90.6936, 25.1699, 31.3902]])

Evaluation

Metrics

Sparse Information Retrieval

Datasets: NanoMSMARCO, NanoNFCorpus, NanoNQ, NanoClimateFEVER, NanoDBPedia, NanoFEVER, NanoFiQA2018, NanoHotpotQA, NanoMSMARCO, NanoNFCorpus, NanoNQ, NanoQuoraRetrieval, NanoSCIDOCS, NanoArguAna, NanoSciFact and NanoTouche2020
Evaluated with SparseInformationRetrievalEvaluator

Metric	NanoMSMARCO	NanoNFCorpus	NanoNQ	NanoClimateFEVER	NanoDBPedia	NanoFEVER	NanoFiQA2018	NanoHotpotQA	NanoQuoraRetrieval	NanoSCIDOCS	NanoArguAna	NanoSciFact	NanoTouche2020
dot_accuracy@1	0.32	0.32	0.28	0.18	0.54	0.54	0.18	0.66	0.4	0.34	0.02	0.44	0.449
dot_accuracy@3	0.48	0.42	0.48	0.32	0.74	0.68	0.36	0.72	0.58	0.52	0.16	0.58	0.8367
dot_accuracy@5	0.62	0.48	0.56	0.4	0.82	0.76	0.42	0.78	0.72	0.64	0.22	0.64	0.898
dot_accuracy@10	0.7	0.54	0.72	0.52	0.9	0.84	0.52	0.9	0.82	0.78	0.24	0.68	1.0
dot_precision@1	0.32	0.32	0.28	0.18	0.54	0.54	0.18	0.66	0.4	0.34	0.02	0.44	0.449
dot_precision@3	0.16	0.26	0.16	0.12	0.4667	0.2267	0.1533	0.3333	0.1933	0.22	0.0533	0.2067	0.5782
dot_precision@5	0.124	0.224	0.116	0.088	0.432	0.156	0.128	0.232	0.152	0.18	0.044	0.14	0.502
dot_precision@10	0.07	0.184	0.074	0.062	0.412	0.086	0.086	0.138	0.086	0.14	0.024	0.078	0.4184
dot_recall@1	0.32	0.0112	0.26	0.0917	0.055	0.53	0.1093	0.33	0.3867	0.0727	0.02	0.415	0.0286
dot_recall@3	0.48	0.0305	0.45	0.1633	0.1158	0.6467	0.2357	0.5	0.5467	0.1367	0.16	0.555	0.1117
dot_recall@5	0.62	0.0414	0.52	0.19	0.1573	0.7267	0.2821	0.58	0.6907	0.1857	0.22	0.615	0.1682
dot_recall@10	0.7	0.0543	0.67	0.2473	0.2872	0.8067	0.3574	0.69	0.7907	0.2877	0.24	0.67	0.2684
dot_ndcg@10	0.4978	0.2109	0.4535	0.2066	0.491	0.6669	0.2732	0.6059	0.5774	0.2664	0.1309	0.5536	0.4607
dot_mrr@10	0.4339	0.3881	0.4033	0.2785	0.6685	0.6359	0.2872	0.7181	0.5223	0.4536	0.0955	0.5219	0.6524
dot_map@100	0.4468	0.0714	0.3886	0.1647	0.3419	0.6247	0.2211	0.5211	0.5159	0.1799	0.1059	0.5201	0.3433
query_active_dims	121.6	199.74	97.54	243.54	180.3	255.34	88.84	147.4	52.78	197.98	798.38	270.48	37.9184
query_sparsity_ratio	0.996	0.9935	0.9968	0.992	0.9941	0.9916	0.9971	0.9952	0.9983	0.9935	0.9738	0.9911	0.9988
corpus_active_dims	397.7708	879.7701	625.5005	704.9307	549.1514	861.5084	414.4174	622.5403	59.613	729.089	696.6201	766.0603	507.7182
corpus_sparsity_ratio	0.987	0.9712	0.9795	0.9769	0.982	0.9718	0.9864	0.9796	0.998	0.9761	0.9772	0.9749	0.9834

Sparse Nano BEIR

Dataset: NanoBEIR_mean
Evaluated with SparseNanoBEIREvaluator with these parameters:
json
1{ 2 "dataset_names": [ 3 "msmarco", 4 "nfcorpus", 5 "nq" 6 ] 7}

Metric	Value
dot_accuracy@1	0.3
dot_accuracy@3	0.4733
dot_accuracy@5	0.5467
dot_accuracy@10	0.6533
dot_precision@1	0.3
dot_precision@3	0.2
dot_precision@5	0.1507
dot_precision@10	0.108
dot_recall@1	0.1822
dot_recall@3	0.3237
dot_recall@5	0.3932
dot_recall@10	0.4714
dot_ndcg@10	0.3812
dot_mrr@10	0.4072
dot_map@100	0.2943
query_active_dims	151.4533
query_sparsity_ratio	0.995
corpus_active_dims	576.6101
corpus_sparsity_ratio	0.9811

Sparse Nano BEIR

Dataset: NanoBEIR_mean
Evaluated with SparseNanoBEIREvaluator with these parameters:
json
1{ 2 "dataset_names": [ 3 "climatefever", 4 "dbpedia", 5 "fever", 6 "fiqa2018", 7 "hotpotqa", 8 "msmarco", 9 "nfcorpus", 10 "nq", 11 "quoraretrieval", 12 "scidocs", 13 "arguana", 14 "scifact", 15 "touche2020" 16 ] 17}

Metric	Value
dot_accuracy@1	0.3592
dot_accuracy@3	0.529
dot_accuracy@5	0.6122
dot_accuracy@10	0.7046
dot_precision@1	0.3592
dot_precision@3	0.2409
dot_precision@5	0.1937
dot_precision@10	0.143
dot_recall@1	0.2023
dot_recall@3	0.3179
dot_recall@5	0.3844
dot_recall@10	0.4669
dot_ndcg@10	0.415
dot_mrr@10	0.4661
dot_map@100	0.3419
query_active_dims	207.3251
query_sparsity_ratio	0.9932
corpus_active_dims	571.212
corpus_sparsity_ratio	0.9813

Training Details

Training Dataset

gooaq

Dataset: gooaq at b089f72
Size: 99,000 training samples
Columns: question and answer
Approximate statistics based on the first 1000 samples:
question answer
type string string
details
min: 8 tokens
mean: 11.79 tokens
max: 24 tokens
min: 14 tokens
mean: 60.02 tokens
max: 153 tokens

Samples:

question	answer
`what are the 5 characteristics of a star?`	`Key Concept: Characteristics used to classify stars include color, temperature, size, composition, and brightness.`
`are copic markers alcohol ink?`	`Copic Ink is alcohol-based and flammable. Keep away from direct sunlight and extreme temperatures.`
`what is the difference between appellate term and appellate division?`	`Appellate terms An appellate term is an intermediate appellate court that hears appeals from the inferior courts within their designated counties or judicial districts, and are intended to ease the workload on the Appellate Division and provide a less expensive forum closer to the people.`

Loss: SpladeLoss with these parameters:
json
1{ 2 "loss": "SparseMultipleNegativesRankingLoss(scale=1.0, similarity_fct='dot_score')", 3 "document_regularizer_weight": 3e-05, 4 "query_regularizer_weight": 5e-05 5}

Evaluation Dataset

gooaq

Dataset: gooaq at b089f72
Size: 1,000 evaluation samples
Columns: question and answer
Approximate statistics based on the first 1000 samples:
question answer
type string string
details
min: 8 tokens
mean: 11.93 tokens
max: 25 tokens
min: 14 tokens
mean: 60.84 tokens
max: 127 tokens

Samples:

question	answer
`should you take ibuprofen with high blood pressure?`	`In general, people with high blood pressure should use acetaminophen or possibly aspirin for over-the-counter pain relief. Unless your health care provider has said it's OK, you should not use ibuprofen, ketoprofen, or naproxen sodium. If aspirin or acetaminophen doesn't help with your pain, call your doctor.`
`how old do you have to be to work in sc?`	`The general minimum age of employment for South Carolina youth is 14, although the state allows younger children who are performers to work in show business. If their families are agricultural workers, children younger than age 14 may also participate in farm labor.`
`how to write a topic proposal for a research paper?`	`['Write down the main topic of your paper. ... ', 'Write two or three short sentences under the main topic that explain why you chose that topic. ... ', 'Write a thesis sentence that states the angle and purpose of your research paper. ... ', 'List the items you will cover in the body of the paper that support your thesis statement.']`

Loss: SpladeLoss with these parameters:
json
1{ 2 "loss": "SparseMultipleNegativesRankingLoss(scale=1.0, similarity_fct='dot_score')", 3 "document_regularizer_weight": 3e-05, 4 "query_regularizer_weight": 5e-05 5}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
learning_rate: 2e-05
num_train_epochs: 1
bf16: True
load_best_model_at_end: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
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: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
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: True
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: True
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: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	NanoMSMARCO_dot_ndcg@10	NanoNFCorpus_dot_ndcg@10	NanoNQ_dot_ndcg@10	NanoBEIR_mean_dot_ndcg@10	NanoClimateFEVER_dot_ndcg@10	NanoDBPedia_dot_ndcg@10	NanoFEVER_dot_ndcg@10	NanoFiQA2018_dot_ndcg@10	NanoHotpotQA_dot_ndcg@10	NanoQuoraRetrieval_dot_ndcg@10	NanoSCIDOCS_dot_ndcg@10	NanoArguAna_dot_ndcg@10	NanoSciFact_dot_ndcg@10	NanoTouche2020_dot_ndcg@10
0.0323	100	132.3714	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.0646	200	17.6927	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.0970	300	2.8291	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1293	400	0.9365	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1616	500	0.5878	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1939	600	0.4041	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.1972	610	-	0.2847	0.3951	0.1965	0.3212	0.3043	-	-	-	-	-	-	-	-	-	-
0.2262	700	0.3624	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2586	800	0.295	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.2909	900	0.282	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3232	1000	0.2947	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3555	1100	0.2664	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3878	1200	0.2387	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.3943	1220	-	0.2030	0.4818	0.2086	0.4084	0.3663	-	-	-	-	-	-	-	-	-	-
0.4202	1300	0.2512	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4525	1400	0.2117	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.4848	1500	0.2367	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5171	1600	0.2085	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5495	1700	0.1745	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5818	1800	0.1958	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.5915	1830	-	0.1449	0.4978	0.2109	0.4535	0.3874	-	-	-	-	-	-	-	-	-	-
0.6141	1900	0.1946	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6464	2000	0.1801	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.6787	2100	0.1854	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7111	2200	0.232	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7434	2300	0.1798	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7757	2400	0.1725	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.7886	2440	-	0.1437	0.4885	0.2206	0.4230	0.3774	-	-	-	-	-	-	-	-	-	-
0.8080	2500	0.1598	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8403	2600	0.1586	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.8727	2700	0.172	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9050	2800	0.1875	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9373	2900	0.1691	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9696	3000	0.1473	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-
0.9858	3050	-	0.1358	0.4923	0.2159	0.4354	0.3812	-	-	-	-	-	-	-	-	-	-
-1	-1	-	-	0.4978	0.2109	0.4535	0.4150	0.2066	0.4910	0.6669	0.2732	0.6059	0.5774	0.2664	0.1309	0.5536	0.4607

The bold row denotes the saved checkpoint.

Environmental Impact

Carbon emissions were measured using CodeCarbon.

Energy Consumed: 0.034 kWh
Carbon Emitted: 0.013 kg of CO2
Hours Used: 0.142 hours

Training Hardware

On Cloud: No
GPU Model: 1 x NVIDIA GeForce RTX 3090
CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
RAM Size: 31.78 GB

Framework Versions

Python: 3.11.6
Sentence Transformers: 4.2.0.dev0
Transformers: 4.52.4
PyTorch: 2.7.1+cu126
Accelerate: 1.5.1
Datasets: 2.21.0
Tokenizers: 0.21.1

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",
}

SpladeLoss

bibtex

@misc{formal2022distillationhardnegativesampling,
      title={From Distillation to Hard Negative Sampling: Making Sparse Neural IR Models More Effective},
      author={Thibault Formal and Carlos Lassance and Benjamin Piwowarski and Stéphane Clinchant},
      year={2022},
      eprint={2205.04733},
      archivePrefix={arXiv},
      primaryClass={cs.IR},
      url={https://arxiv.org/abs/2205.04733},
}

SparseMultipleNegativesRankingLoss

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}
}

FlopsLoss

bibtex

@article{paria2020minimizing,
    title={Minimizing flops to learn efficient sparse representations},
    author={Paria, Biswajit and Yeh, Chih-Kuan and Yen, Ian EH and Xu, Ning and Ravikumar, Pradeep and P{'o}czos, Barnab{'a}s},
    journal={arXiv preprint arXiv:2004.05665},
    year={2020}
}

	question	answer
type	string	string
details	min: 8 tokens mean: 11.79 tokens max: 24 tokens	min: 14 tokens mean: 60.02 tokens max: 153 tokens