Shashwat13333/msmarco-distilbert-base-v4

BGE base Financial Matryoshka

This is a sentence-transformers model finetuned from sentence-transformers/msmarco-distilbert-base-v4. 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: sentence-transformers/msmarco-distilbert-base-v4
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Language: en
• License: apache-2.0

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: DistilBertModel 
  (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("Shashwat13333/msmarco-distilbert-base-v4")
# Run inference
sentences = [
    'What managed services does TechChefz provide ?',
    '  What we do\n\nDigital Strategy\nCreating digital frameworks that transform your digital enterprise and produce a return on investment.\n\nPlatform Selection\nHelping you select the optimal digital experience, commerce, cloud and marketing platform for your enterprise.\n\nPlatform Builds\nDeploying next-gen scalable and agile enterprise digital platforms, along with multi-platform integrations.\n\nProduct Builds\nHelp you ideate, strategize, and engineer your product with help of our enterprise frameworks \n\nTeam Augmentation\nHelp you scale up and augment your existing team to solve your hiring challenges with our easy to deploy staff augmentation offerings .\nManaged Services\nOperate and monitor your business-critical applications, data, and IT workloads, along with Application maintenance and operations\n',
    'In the Introducing the world of Global Insurance Firm, we crafted Effective Solutions for Complex Problems and delieverd a comprehensive Website Development, Production Support & Managed Services, we optimized customer journeys, integrate analytics, CRM, ERP, and third-party applications, and implement cutting-edge technologies for enhanced performance and efficiency\nand achievied 200% Reduction in operational time & effort managing content & experience, 70% Reduction in Deployment Errors and Downtime, 2.5X Customer Engagement, Conversion & Retention',
]
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

• Datasets: dim_768, dim_512, dim_256, dim_128 and dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	dim_768	dim_512	dim_256	dim_128	dim_64
cosine_accuracy@1	0.1067	0.1067	0.1467	0.12	0.16
cosine_accuracy@3	0.4933	0.4667	0.4533	0.4533	0.3867
cosine_accuracy@5	0.5333	0.5333	0.4933	0.4933	0.4667
cosine_accuracy@10	0.6267	0.6133	0.6	0.6	0.5467
cosine_precision@1	0.1067	0.1067	0.1467	0.12	0.16
cosine_precision@3	0.1644	0.1556	0.1511	0.1511	0.1289
cosine_precision@5	0.1067	0.1067	0.0987	0.0987	0.0933
cosine_precision@10	0.0627	0.0613	0.06	0.06	0.0547
cosine_recall@1	0.1067	0.1067	0.1467	0.12	0.16
cosine_recall@3	0.4933	0.4667	0.4533	0.4533	0.3867
cosine_recall@5	0.5333	0.5333	0.4933	0.4933	0.4667
cosine_recall@10	0.6267	0.6133	0.6	0.6	0.5467
cosine_ndcg@10	0.3697	0.3703	0.3732	0.3495	0.3449
cosine_mrr@10	0.2865	0.2909	0.3006	0.2696	0.281
cosine_map@100	0.2993	0.3047	0.3135	0.2815	0.2953

Training Details

Training Dataset

Unnamed Dataset

• Size: 150 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 150 samples:

  	anchor	positive
  type	string	string
  details	min: 7 tokens mean: 12.45 tokens max: 18 tokens	min: 20 tokens mean: 126.17 tokens max: 378 tokens

• Samples:

  anchor	positive
  How can digital transformation enhance customer interactions across multiple channels?	We offer custom software development, digital marketing strategies, and tailored solutions to drive tangible results for your business. Our expert team combines technical prowess with industry insights to propel your business forward in the digital landscape. "Engage, analyze & target your customers Digital transformation enables you to interact with customers across multiple channels, providing personalized experiences. This could include social media engagement, interactive websites, and mobile apps." "Empower your employees & partners The push for digital transformation has led many companies to embrace cloud solutions. However, the migration and integration of legacy systems into the cloud often present challenges." "Optimize & automate your operations The push for digital transformation has led many companies to embrace cloud solutions. However, the migration and integration of legacy systems into the cloud often present challenges." "Transform your products The push for digi...
  How does a CRM system improve customer retention?	Our MarTech capabilities Personalization Involves tailoring marketing messages and experiences to individual customers. It enhances customer engagement, loyalty, and ultimately, conversion rates. Marketing Automation Marketing automation streamlines repetitive tasks such as email marketing, lead nurturing, and social media posting. It improves efficiency, saves time, and ensures timely communication with customers. Customer Relationship Management CRM systems help manage interactions with current and potential customers. They store customer data, track interactions, and facilitate communication, improving customer retention.
  How can your recommendation engines improve our business?	How can your recommendation engines improve our business? Our recommendation engines are designed to analyze customer behavior and preferences to deliver personalized suggestions, enhancing user experience, increasing sales, and boosting customer retention.

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• gradient_accumulation_steps: 4
• learning_rate: 1e-05
• weight_decay: 0.01
• num_train_epochs: 4
• lr_scheduler_type: cosine
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: True
• optim: adamw_torch_fused
• push_to_hub: True
• hub_model_id: Shashwat13333/msmarco-distilbert-base-v4
• push_to_hub_model_id: msmarco-distilbert-base-v4
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 8
• per_device_eval_batch_size: 8
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 4
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 1e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: cosine
• 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: 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_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: True
• resume_from_checkpoint: None
• hub_model_id: Shashwat13333/msmarco-distilbert-base-v4
• 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: msmarco-distilbert-base-v4
• 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
0.2105	1	3.5757	-	-	-	-	-
0.8421	4	-	0.3563	0.3543	0.3378	0.3681	0.3077
1.2105	5	4.4031	-	-	-	-	-
1.8421	8	-	0.3652	0.3547	0.3574	0.3542	0.3579
2.4211	10	3.3423	-	-	-	-	-
2.8421	12	-	0.3783	0.3680	0.3558	0.3807	0.3408
3.6316	15	2.3695	-	-	-	-	-
3.8421	16	-	0.3697	0.3703	0.3732	0.3495	0.3449

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.3.1
• Transformers: 4.47.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• Datasets: 3.2.0
• Tokenizers: 0.21.0

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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

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}
}