--- tags: - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:18281 - loss:MultipleNegativesRankingLoss base_model: intfloat/multilingual-e5-large-instruct widget: - source_sentence: 'query: Travel restrictions showed up after the fact, but physical distancing, case isolation, & testing did the trick for #FlattenTheCurve. Our global crew''s analysis (co-led by @user & @user) of data from >30,000 #COVID19 cases is out in @ScienceMagazine. 1/12' sentences: - 'passage: title: The effect of human mobility and control measures on the COVID-19 epidemic in China abstract: The ongoing coronavirus disease 2019 (COVID-19) outbreak expanded rapidly throughout China. Major behavioral, clinical, and state interventions were undertaken to mitigate the epidemic and prevent the persistence of the virus in human populations in China and worldwide. It remains unclear how these unprecedented interventions, including travel restrictions, affected COVID-19 spread in China. We used real-time mobility data from Wuhan and detailed case data including travel history to elucidate the role of case importation in transmission in cities across China and to ascertain the impact of control measures. Early on, the spatial distribution of COVID-19 cases in China was explained well by human mobility data. After the implementation of control measures, this correlation dropped and growth rates became negative in most locations, although shifts in the demographics of reported cases were still indicative of local chains of transmission outside of Wuhan. This study shows that the drastic control measures implemented in China substantially mitigated the spread of COVID-19.' - 'passage: title: Reverse-transcribed SARS-CoV-2 RNA can integrate into the genome of cultured human cells and can be expressed in patient-derived tissues abstract: Prolonged detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA and recurrence of PCR-positive tests have been widely reported in patients after recovery from COVID-19, but some of these patients do not appear to shed infectious virus. We investigated the possibility that SARS-CoV-2 RNAs can be reverse-transcribed and integrated into the DNA of human cells in culture and that transcription of the integrated sequences might account for some of the positive PCR tests seen in patients. In support of this hypothesis, we found that DNA copies of SARS-CoV-2 sequences can be integrated into the genome of infected human cells.' - 'passage: title: Covid-19 and alcohol—a dangerous cocktail abstract: The principal aim of this work was to better understand how regenerating muscle fibers become innervated in adult animals. To induce muscle regeneration, individual identified muscle fibers in a mouse were damaged with a laser focused through a microscope. The muscle fiber that degenerated and the muscle fiber that was formed in its place were followed by viewing the same site repeatedly over a period of 2 d to 40 weeks. Commonly, the nerve terminal innervating the irradiated muscle fiber partially retracted during muscle fiber degeneration, and then sprouted to innervate the regenerating muscle fiber at the same site it had previously innervated the muscle fiber that was damaged. During the early phase of muscle regeneration we also observed sprouts originating from nerve terminals on adjacent muscle fibers. The new nerve growth was a response to the regenerating muscle fiber rather than to the degenerated fiber it replaced because repeated damage of the same site every 2–3 d over a 10 d period (to prevent regeneration) did not cause any sprouting.' - source_sentence: 'query: Check out this #preprint on @researchsquare: Vaccine‑driven immunity offers stronger cross‑type immunity versus natural infection against emerging SARS‑CoV‑2 variants of concern.' sentences: - 'passage: title: Vaccine-induced immunity provides more robust heterotypic immunity than natural infection to emerging SARS-CoV-2 variants of concern. abstract: Abstract Both natural infection with SARS-CoV-2 and immunization with a number of vaccines induce protective immunity. However, the ability of such immune responses to recognize and therefore protect against emerging variants is a matter of increasing importance. Such variants of concern (VOC) include isolates of lineage B1.1.7, first identified in the UK, and B1.351, first identified in South Africa. Our data confirm that VOC, particularly those with substitutions at residues 484 and 417 escape neutralization by antibodies directed to the ACE2-binding Class 1 and the adjacent Class 2 epitopes but are susceptible to neutralization by the generally less potent antibodies directed to Class 3 and 4 epitopes on the flanks RBD.' - 'passage: title: Optic neuritis following COVID-19 vaccination: Coincidence or side-effect? - A case series abstract: The whole world waiting for the elimination of COVID-19. This is a short series of three cases that presented with optic neuritis. On further inquiry, all had received the Covishield vaccine within 5-12 days just before the presentation, with no history of COVID-19 positive RT-PCR. The range of age was 27-48 years. All patients improved after pulse steroid therapy and are still under follow-up. After being plagued by COVID-19 for nearly 2 years, the whole world wishes for little more than complete eradication of the disease. Our country commenced the much-awaited vaccination drive from Jan 2021. Ophthalmic manifestations have appeared in many forms post-COVID-19, among which neuro-ophthalmic manifestations are infrequent. To the best of our knowledge, this is the first report of a short case series from our country presenting with optic neuritis after COVID-19 vaccination, without any sign of active infection.' - 'passage: title: Circulating Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Vaccine Antigen Detected in the Plasma of mRNA-1273 Vaccine Recipients abstract: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) proteins were measured in longitudinal plasma samples collected from 13 participants who received two doses of mRNA-1273 vaccine. Eleven of 13 participants showed detectable levels of SARS-CoV-2 protein as early as day 1 after first vaccine injection. Clearance of detectable SARS-CoV-2 protein correlated with production of immunoglobulin G (IgG) and immunoglobulin A (IgA).' - source_sentence: 'query: WHO is missing? Embarek Obviously this daft narrow‑minded team of scientific puppets complaining about "time windows" are irate about his recent Danish interview Origins of SARS‑CoV‑2: window is closing for key scientific studies' sentences: - 'passage: -2 cycle threshold values during infection (i.e. lower viral burden), and less frequently reported any symptoms. title: Anti-spike antibody response to natural SARS-CoV-2 infection in the general population Among those who seroconverted, using Bayesian linear mixed models, the estimated anti-spike IgG peak level was 7.3-fold higher than the level previously associated with 50% protection against reinfection, with higher peak levels in older participants and those of non-white ethnicity. The estimated anti-spike IgG half-life was 184 days, being longer in females and those of white ethnicity. We estimated antibody levels associated with protection against reinfection likely last 1.5-2 years on average, with levels associated with protection from severe infection present for several years. These estimates could inform planning for vaccination booster strategies.' - 'passage: title: Trends in Income From 1975 to 2018 abstract: For two decades after World War II, incomes grew at a rate close to the U.S. economy-wide growth rate. Anemic growth from 1969 to 1974 kept inequality in check. But since then, the benefits of growth have not been evenly distributed. Racial and gender inequality is also manifested in income inequality.' - 'passage: title: Origins of SARS-CoV-2: window is closing for key scientific studies abstract: Authors of the March WHO report into how COVID-19 emerged warn that further delay makes crucial inquiry biologically difficult. Authors of the March WHO report into how COVID-19 emerged warn that further delay makes crucial inquiry biologically difficult.' - source_sentence: 'query: Hartklachten en vaccinaties The chance of myocarditis after vaccination is steadily higher in younger men, especially after a 2nd dose of RNA mRNA-1273 vaccine Chance of myocarditis after sequential COVID-19 vaccinations by age and gender' sentences: - 'passage: title: Risk of myocarditis following sequential COVID-19 vaccinations by age and sex abstract: ABSTRACT In an updated self-controlled case series analysis of 42,200,614 people aged 13 years or more, we evaluate the association between COVID-19 vaccination and myocarditis, stratified by age and sex, including 10,978,507 people receiving a third vaccine dose. Myocarditis risk was increased during 1-28 days following a third dose of BNT162b2 (IRR 2.02, 95%CI 1.40, 2.91).' - 'passage: in) and faster viral clearance by PCR. Viral clearance was treatment dose- and duration-dependent. title: Meta-analysis of randomized trials of ivermectin to treat SARS-CoV-2 infection In six randomized trials of moderate or severe infection, there was a 75% reduction in mortality (Relative Risk=0.25 [95%CI 0.12-0.52]; p=0.0002); 14/650 (2.1%) deaths on ivermectin; 57/597 (9.5%) deaths in controls) with favorable clinical recovery and reduced hospitalization. Many studies included were not peer reviewed and meta-analyses are prone to confounding issues. Ivermectin should be validated in larger, appropriately controlled randomized trials before the results are sufficient for review by regulatory authorities.' - 'passage: title: Asymptomatic transmission of covid-19 abstract: The UK''s £100bn "Operation Moonshot" to roll out mass testing for covid-19 to cities and universities around the country raises two key questions.How infectious are people who test positive but have no symptoms?And, what is their contribution to transmission of live virus?' - source_sentence: 'query: @user That’s not what some of the data suggests. 25% of those who had a light infection the first time required an ER visit the 2nd time. And we know there is a mounting load with reinfections.' sentences: - 'passage: title: Ce que les sondages font à l''opinion publique abstract: Ce que les sondages font à l''opinion. Loïc Blondiaux [117-136]. Cet article se propose de revisiter la controverse récurrente autour des sondages et de l''opinion qui traverse les sciences sociales et divise en particulier la science politique. Il commence par recenser les principales critiques adressées aux sondages d''opinion dans la sociologie et la science politique française et anglo-saxonne. Il tente ensuite de reconstituer une brève histoire des usages du concept d''opinion dans le discours savant. La conclusion de ce double inventaire apparaît sans ambiguïté : les sondages ne mesurent pas l''opinion publique au sens où les sciences sociales et le discours politique savant entendent habituellement cette notion. La troisième et dernière partie discute plusieurs hypothèses susceptibles de rendre compte de l''extraordinaire réussite de cette étrange mesure de l''opinion publique.' - 'passage: randomization to hospital discharge. title: Effect of a Single High Dose of Vitamin D3 on Hospital Length of Stay in Patients With Moderate to Severe COVID-19 Prespecified secondary outcomes included mortality during hospitalization; the number of patients admitted to the intensive care unit; the number of patients who required mechanical ventilation and the duration of mechanical ventilation; and serum levels of 25-hydroxyvitamin D, total calcium, creatinine, and C-reactive protein.Of 240 randomized patients, 237 were included in the primary analysis (mean [SD] age, 56.2 [14.4] years; 104 [43.9%] women; mean [SD] baseline 25-hydroxyvitamin D level, 20.9 [9.2] ng/mL). Median (interquartile range) length of stay was not significantly different between the vitamin D3 (7.0 [4.0-10.0] days) and placebo groups (7.0 [5.0-13.0] days) (log-rank P = .59; unadjusted hazard ratio for hospital discharge, 1.07 [95% CI, 0.82-1.39]; P = .62).' - 'passage: diagnoses occur closer to the index date for infection or reinfection in the Omicron BA epoch. title: SARS-CoV-2 Reinfection is Preceded by Unique Biomarkers and Related to Initial Infection Timing and Severity: an N3C RECOVER EHR-Based Cohort Study We report lower albumin levels leading up to reinfection and a statistically significant association of severity between first infection and reinfection (chi-squared value: 9446.2, p-value: 0) with a medium effect size (Cramer''s V: 0.18, DoF = 4).' pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - cosine_accuracy@1 - cosine_accuracy@3 - cosine_accuracy@5 - cosine_accuracy@10 - cosine_precision@1 - cosine_precision@3 - cosine_precision@5 - cosine_precision@10 - cosine_recall@1 - cosine_recall@3 - cosine_recall@5 - cosine_recall@10 - cosine_ndcg@10 - cosine_mrr@10 - cosine_map@100 model-index: - name: SentenceTransformer based on intfloat/multilingual-e5-large-instruct results: - task: type: information-retrieval name: Information Retrieval dataset: name: CT26 dev split type: CT26-dev-split metrics: - type: cosine_accuracy@1 value: 0.6926272066458983 name: Cosine Accuracy@1 - type: cosine_accuracy@3 value: 0.857736240913811 name: Cosine Accuracy@3 - type: cosine_accuracy@5 value: 0.8982346832814122 name: Cosine Accuracy@5 - type: cosine_accuracy@10 value: 0.9273104880581516 name: Cosine Accuracy@10 - type: cosine_precision@1 value: 0.6926272066458983 name: Cosine Precision@1 - type: cosine_precision@3 value: 0.28591208030460363 name: Cosine Precision@3 - type: cosine_precision@5 value: 0.17964693665628245 name: Cosine Precision@5 - type: cosine_precision@10 value: 0.09273104880581517 name: Cosine Precision@10 - type: cosine_recall@1 value: 0.6926272066458983 name: Cosine Recall@1 - type: cosine_recall@3 value: 0.857736240913811 name: Cosine Recall@3 - type: cosine_recall@5 value: 0.8982346832814122 name: Cosine Recall@5 - type: cosine_recall@10 value: 0.9273104880581516 name: Cosine Recall@10 - type: cosine_ndcg@10 value: 0.8179831495116502 name: Cosine Ndcg@10 - type: cosine_mrr@10 value: 0.7820204717400984 name: Cosine Mrr@10 - type: cosine_map@100 value: 0.7847045028904378 name: Cosine Map@100 --- # SentenceTransformer based on intfloat/multilingual-e5-large-instruct This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) on the clef-me5-mined-pairs-train-pairs dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for retrieval. ## Model Details ### Model Description - **Model Type:** Sentence Transformer - **Base model:** [intfloat/multilingual-e5-large-instruct](https://huggingface.co/intfloat/multilingual-e5-large-instruct) - **Maximum Sequence Length:** 256 tokens - **Output Dimensionality:** 1024 dimensions - **Similarity Function:** Cosine Similarity - **Supported Modality:** Text - **Training Dataset:** - clef-me5-mined-pairs-train-pairs ### Model Sources - **Documentation:** [Sentence Transformers Documentation](https://sbert.net) - **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers) - **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) ### Full Model Architecture ``` SentenceTransformer( (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'XLMRobertaModel'}) (1): Pooling({'embedding_dimension': 1024, 'pooling_mode': 'mean', 'include_prompt': True}) (2): Normalize({}) ) ``` ## Usage ### Direct Usage (Sentence Transformers) First install the Sentence Transformers library: ```bash 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("MinhPhuc0804/me5-256-kiem-tra-di-t1-v2.2-epoch-10") # Run inference sentences = [ 'query: @user That’s not what some of the data suggests. 25% of those who had a light infection the first time required an ER visit the 2nd time. And we know there is a mounting load with reinfections.', "passage: diagnoses occur closer to the index date for infection or reinfection in the Omicron BA epoch.\n\ntitle: SARS-CoV-2 Reinfection is Preceded by Unique Biomarkers and Related to Initial Infection Timing and Severity: an N3C RECOVER EHR-Based Cohort Study\nWe report lower albumin levels leading up to reinfection and a statistically significant association of severity between first infection and reinfection (chi-squared value: 9446.2, p-value: 0) with a medium effect size (Cramer's V: 0.18, DoF = 4).", 'passage: randomization to hospital discharge.\n\ntitle: Effect of a Single High Dose of Vitamin D3 on Hospital Length of Stay in Patients With Moderate to Severe COVID-19\nPrespecified secondary outcomes included mortality during hospitalization; the number of patients admitted to the intensive care unit; the number of patients who required mechanical ventilation and the duration of mechanical ventilation; and serum levels of 25-hydroxyvitamin D, total calcium, creatinine, and C-reactive protein.Of 240 randomized patients, 237 were included in the primary analysis (mean [SD] age, 56.2 [14.4] years; 104 [43.9%] women; mean [SD] baseline 25-hydroxyvitamin D level, 20.9 [9.2] ng/mL). Median (interquartile range) length of stay was not significantly different between the vitamin D3 (7.0 [4.0-10.0] days) and placebo groups (7.0 [5.0-13.0] days) (log-rank P = .59; unadjusted hazard ratio for hospital discharge, 1.07 [95% CI, 0.82-1.39]; P = .62).', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 1024] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities) # tensor([[1.0000, 0.5041, 0.1662], # [0.5041, 1.0000, 0.0334], # [0.1662, 0.0334, 1.0000]]) ``` ## Evaluation ### Metrics #### Information Retrieval * Dataset: `CT26-dev-split` * Evaluated with [InformationRetrievalEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.sentence_transformer.evaluation.InformationRetrievalEvaluator) | Metric | Value | |:--------------------|:----------| | cosine_accuracy@1 | 0.6926 | | cosine_accuracy@3 | 0.8577 | | cosine_accuracy@5 | 0.8982 | | cosine_accuracy@10 | 0.9273 | | cosine_precision@1 | 0.6926 | | cosine_precision@3 | 0.2859 | | cosine_precision@5 | 0.1796 | | cosine_precision@10 | 0.0927 | | cosine_recall@1 | 0.6926 | | cosine_recall@3 | 0.8577 | | cosine_recall@5 | 0.8982 | | cosine_recall@10 | 0.9273 | | **cosine_ndcg@10** | **0.818** | | cosine_mrr@10 | 0.782 | | cosine_map@100 | 0.7847 | ## Training Details ### Training Dataset #### clef-me5-mined-pairs-train-pairs * Dataset: clef-me5-mined-pairs-train-pairs * Size: 18,281 training samples * Columns: anchor and positive * Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | query: Peer-reviewed scientific studies project universal mask use would spare nearly 130,000 U.S. lives by February. But Utah carries on with a baffling, lax patchwork of feeble mask rules. If elected, I’ll shield your family. #PetersonProtects #utpol | passage: 469,578–578,347) lives could be lost to COVID-19 across the United States by 28 February 2021.

title: Modeling COVID-19 scenarios for the United States
We find that achieving universal mask use (95% mask use in public) could be sufficient to ameliorate the worst effects of epidemic resurgences in many states. Universal mask use could save an additional 129,574 (85,284–170,867) lives from September 22, 2020 through the end of February 2021, or an additional 95,814 (60,731–133,077) lives assuming a lesser adoption of mask wearing (85%), when compared to the reference scenario. | | query: "Seroprevalence studies may fail to detect people who have had mild covid-19. Consideration should be given to [...] calibration of assay thresholds, the breadth of the antibody response, and the role of mucosal antibodies" | passage: was used to analyse the data.

title: Are we underestimating seroprevalence of SARS-CoV-2?

Results

The study indicates a fairly high knowledge level of STI/HIV (89.4%) with more males (87%) and younger participants (88%) possessing good knowledge of STI/HIV. Majority of the participants are sexually active (63.3%) and of this are more males (61.3%) and younger participants (60%). Findings show that age (β = 0.025; t = 0.04; p > 0.05) of the three predictor variables was not a predictor of attitude towards STI/HIV. However, knowledge of STI/HIV (β = 0.459; t = 5.032; p < 0.05) and sexual behaviour (β = 0.341; t = 4.278; p < 0.05) were strong predictors of attitude towards STI/HIV.

Conclusion

This study shows the need for strong advocacy, enlightenment and community mobilisation for improved awareness of STI/HIV. | | query: Rebuilding of October #Arctic sea ice volume spanning over the past 100 years... [Side-by-side look between PIOMAS-20C and PIOMAS data sets now refreshed through October 2021. Model details available at | passage: title: Arctic Sea Ice Volume Variability over 1901–2010: A Model-Based Reconstruction abstract: Abstract PIOMAS-20C, an Arctic sea ice reconstruction for 1901–2010, is produced by forcing the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) with ERA-20C atmospheric data. ERA-20C performance over Arctic sea ice is assessed by comparisons with measurements and data from other reanalyses. ERA-20C performs similarly with respect to the annual cycle of downwelling radiation, air temperature, and wind speed compared to reanalyses with more extensive data assimilation such as ERA-Interim and MERRA. PIOMAS-20C sea ice thickness and volume are then compared with in situ and aircraft remote sensing observations for the period of ~1950–2010. Error statistics are similar to those for PIOMAS. We compare the magnitude and patterns of sea ice variability between the first half of the twentieth century (1901–40) and the more recent period (1980–2010), both marked by sea ice decl... | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim", "gather_across_devices": false, "directions": [ "query_to_doc" ], "partition_mode": "joint", "hardness_mode": null, "hardness_strength": 0.0 } ``` ### Evaluation Dataset #### clef-me5-mined-pairs-train-pairs * Dataset: clef-me5-mined-pairs-train-pairs * Size: 963 evaluation samples * Columns: anchor and positive * Approximate statistics based on the first 963 samples: | | anchor | positive | |:--------|:------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------| | type | string | string | | details | | | * Samples: | anchor | positive | |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | query: I reckon we’ll be hearing more about plitidepsin, which (in vitro, at least) is 27.5 times stronger than remdesivir #COVID19 | passage: title: Plitidepsin has potent preclinical efficacy against SARS-CoV-2 by targeting the host protein eEF1A
abstract: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral proteins interact with the eukaryotic translation machinery, and inhibitors of translation have potent antiviral effects. We found that the drug plitidepsin (aplidin), which has limited clinical approval, possesses antiviral activity (90% inhibitory concentration = 0.88 nM) that is more potent than remdesivir against SARS-CoV-2 in vitro by a factor of 27.5, with limited toxicity in cell culture. Through the use of a drug-resistant mutant, we show that the antiviral activity of plitidepsin against SARS-CoV-2 is mediated through inhibition of the known target eEF1A (eukaryotic translation elongation factor 1A). | | query: 2020 research on #LongCovid. Pre‑vaccine era. “Young, low risk patients with ongoing symptoms of #covid19 had signs of damage to multiple organs four months after initially being infected.” | passage: title: Long covid: Damage to multiple organs presents in young, low risk patients
abstract: Young, low risk patients with ongoing symptoms of covid-19 had signs of damage to multiple organs four months after initially being infected, a preprint study has suggested. | | query: L'inflammation indépendante provoquée par les macrophages encourage-t-elle les lésions alvéolaires dans la COVID-19 ? | passage: title: Does autonomous macrophage-driven inflammation promote alveolar damage in COVID-19?
abstract: The editorial reviews an ERJ publication which shows direct viral replication is rare in the alveolar space due to rare ACE2 expression. Instead it posits that autonomous macrophage inflammation occurs and drives lung injury.https://bit.ly/3CqjwiT | * Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters: ```json { "scale": 20.0, "similarity_fct": "cos_sim", "gather_across_devices": false, "directions": [ "query_to_doc" ], "partition_mode": "joint", "hardness_mode": null, "hardness_strength": 0.0 } ``` ### Training Hyperparameters #### Non-Default Hyperparameters - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `learning_rate`: 1.6e-05 - `num_train_epochs`: 10 - `warmup_ratio`: 0.1 - `fp16`: True - `dataloader_num_workers`: 8 #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `prediction_loss_only`: True - `per_device_train_batch_size`: 64 - `per_device_eval_batch_size`: 64 - `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`: 1.6e-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`: 10 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.1 - `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`: True - `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`: 8 - `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`: proportional - `router_mapping`: {} - `learning_rate_mapping`: {}
### Training Logs | Epoch | Step | Training Loss | Validation Loss | CT26-dev-split_cosine_ndcg@10 | |:------:|:----:|:-------------:|:---------------:|:-----------------------------:| | 0.3497 | 100 | 1.412 | - | - | | 0.6993 | 200 | 0.4583 | - | - | | 1.0 | 286 | - | 0.3687 | 0.8100 | | 1.0490 | 300 | 0.4552 | - | - | | 1.3986 | 400 | 0.3449 | - | - | | 1.7483 | 500 | 0.334 | - | - | | 2.0 | 572 | - | 0.3241 | 0.8166 | | 2.0979 | 600 | 0.2666 | - | - | | 2.4476 | 700 | 0.1872 | - | - | | 2.7972 | 800 | 0.2041 | - | - | | 3.0 | 858 | - | 0.3176 | 0.8194 | | 3.1469 | 900 | 0.1789 | - | - | | 3.4965 | 1000 | 0.1246 | - | - | | 3.8462 | 1100 | 0.1279 | - | - | | 4.0 | 1144 | - | 0.3149 | 0.8181 | | 4.1958 | 1200 | 0.1071 | - | - | | 4.5455 | 1300 | 0.0869 | - | - | | 4.8951 | 1400 | 0.0895 | - | - | | 5.0 | 1430 | - | 0.3100 | 0.8152 | | 5.2448 | 1500 | 0.0773 | - | - | | 5.5944 | 1600 | 0.0726 | - | - | | 5.9441 | 1700 | 0.0767 | - | - | | 6.0 | 1716 | - | 0.2971 | 0.8175 | | 6.2937 | 1800 | 0.0625 | - | - | | 6.6434 | 1900 | 0.06 | - | - | | 6.9930 | 2000 | 0.0667 | - | - | | 7.0 | 2002 | - | 0.2981 | 0.8210 | | 7.3427 | 2100 | 0.0609 | - | - | | 7.6923 | 2200 | 0.0549 | - | - | | 8.0 | 2288 | - | 0.3009 | 0.8222 | | 8.0420 | 2300 | 0.0503 | - | - | | 8.3916 | 2400 | 0.0487 | - | - | | 8.7413 | 2500 | 0.0498 | - | - | | 9.0 | 2574 | - | 0.3020 | 0.8210 | | 9.0909 | 2600 | 0.0456 | - | - | | 9.4406 | 2700 | 0.0496 | - | - | | 9.7902 | 2800 | 0.0521 | - | - | | 10.0 | 2860 | - | 0.2993 | 0.8180 | ### Training Time - **Training**: 21.5 minutes ### Framework Versions - Python: 3.12.6 - Sentence Transformers: 5.4.1 - Transformers: 4.56.0 - PyTorch: 2.8.0+cu129 - Accelerate: 1.10.1 - Datasets: 4.8.5 - Tokenizers: 0.22.0 ## 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{oord2019representationlearningcontrastivepredictive, title={Representation Learning with Contrastive Predictive Coding}, author={Aaron van den Oord and Yazhe Li and Oriol Vinyals}, year={2019}, eprint={1807.03748}, archivePrefix={arXiv}, primaryClass={cs.LG}, url={https://arxiv.org/abs/1807.03748}, } ```