ithasa-mmbert

SentenceTransformer based on jhu-clsp/mmBERT-base

This is a sentence-transformers model finetuned from jhu-clsp/mmBERT-base. 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: jhu-clsp/mmBERT-base <!-- at revision c5955035435e2bf121cde7f3c8863ef52ff35d82 -->
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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': 8192, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
  (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("sentence_transformers_model_id")
# Run inference
sentences = [
    '"""Starting position (Sthiti), Tadasana"""',
    'Sanskrit: """प्रारम्भिका स्थितिः, ताडासनम् ।"""\nEnglish: """Starting position (Sthiti), Tadasana"""',
    "Sanskrit: भोजनपचनानन्तरं यदा व्यक्ति: तस्य पार्श्वे भोजनार्थं गच्छति तदा तं भोजनपरिवेषणं कथ्यते ।\nEnglish: Once the food is cooked it is brought to the persons who will consume it which we call 'Serving'.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0003, 0.8767, 0.7860],
#         [0.8767, 1.0003, 0.9032],
#         [0.7860, 0.9032, 1.0003]])

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* -->

Evaluation

Metrics

Triplet

• Dataset: triplet-dev
• Evaluated with <code>TripletEvaluator</code>

<!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

Unnamed Dataset

• Size: 256,886 training samples
• Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	<ul><li>min: 12 tokens</li><li>mean: 31.26 tokens</li><li>max: 88 tokens</li></ul>	<ul><li>min: 34 tokens</li><li>mean: 102.95 tokens</li><li>max: 280 tokens</li></ul>	<ul><li>min: 12 tokens</li><li>mean: 81.61 tokens</li><li>max: 772 tokens</li></ul>

• Samples:

  anchor	positive	negative
  <code>There was no Mughal tradition of primogeniture, the systematic passing of rule, upon an emperor's death, to his eldest son.</code>	<code>Sanskrit: चक्रवर्तिनः मृत्योः अनन्तरं तस्य शासनस्य व्यवस्थितरूपेण सङ्क्रमणस्य, मुघलपरम्परायाः ज्येष्ठपुत्राधिकारपद्धतिः नासीत्।<br>English: There was no Mughal tradition of primogeniture, the systematic passing of rule, upon an emperor's death, to his eldest son.</code>	<code>Sanskrit: F6 नोदनेन व्यवस्थां रन् कुर्मः ।<br>English: Let's now run the application by pressing F6 on your keyboard.</code>
  <code>The four sons of Shah Jahan all held governorships during their father's reign.</code>	<code>Sanskrit: शाह्-जहाँ-नामकस्य चत्वारः पुत्राः, सर्वे पितुः शासनकाले शासकपदम् अधारयन्।<br>English: The four sons of Shah Jahan all held governorships during their father's reign.</code>	<code>Sanskrit: यदा च पृथिवीं सर्वां यजमानोऽनुपर्यगाः। शम्याक्षेपेण विधिना तदाऽऽसीत् किं तु ते हृदि॥ न ते पश्यामि भृङ्गारं न च्छत्रं व्यजने न च। ब्रह्मदत्तां च ते मालां न पश्याम्यसुराधिप॥<br>English: I do not now see that golden jar of yours, nor that umbrella of yours, nor those fans. I behold not also, O king of the Asuras, that garland which was given given to you by Grandfather.'</code>
  <code>In this regard he discusses the correlation between social opportunities of education and health and how both of these complement economic and political freedoms as a healthy and well-educated person is better suited to make informed economic decisions and be involved in fruitful political demonstrations etc.</code>	<code>Sanskrit: अस्मिन् विषये सः शिक्षणस्य स्वास्थ्यस्य च सामाजिकावकाशानाम् अन्योन्य-सम्बन्धस्य, तथा च एतद्द्वयम् अपि आर्थिक-राजनैतिक-स्वातन्त्र्ययोः कथं पूरकं भवतः इति च चर्चां करोति, यतोहि स्वस्था सुशिक्षिता च व्यक्तिः ज्ञानपूर्वम् आर्थिकविषयान् निर्णेतुं तथा फलप्रदेषु राजनैतिकेषु प्रतिपादनादिषु संलग्नः भवितुं च अधिकारी भवति इति।<br>English: In this regard he discusses the correlation between social opportunities of education and health and how both of these complement economic and political freedoms as a healthy and well-educated person is better suited to make informed economic decisions and be involved in fruitful political demonstrations etc.</code>	<code>Sanskrit: पदं देवस्य मीढुषोऽनाधृष्टाभिरूतिभिः । भद्रा सूर्य इवोपदृक्<br>English: O agni! You are light, admirable, enemyless and patient. Seeing you is as beneficial as seeing the sun.</code>

• Loss: <code>TripletLoss</code> with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 1.0
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 2,000 evaluation samples
• Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	<ul><li>min: 4 tokens</li><li>mean: 18.39 tokens</li><li>max: 64 tokens</li></ul>	<ul><li>min: 12 tokens</li><li>mean: 56.57 tokens</li><li>max: 171 tokens</li></ul>	<ul><li>min: 11 tokens</li><li>mean: 56.98 tokens</li><li>max: 169 tokens</li></ul>

• Samples:

  anchor	positive	negative
  <code>You two were eloquent speakers.</code>	<code>Sanskrit: युवां वाक्पटू आस्तम् ।<br>English: You two were eloquent speakers.</code>	<code>Sanskrit: Mozilla Firefox - परिचय: इत्यस्मिन् Spoken tutorial मध्ये स्वागतम् |<br>English: Welcome to the Spoken tutorial on Introduction to Mozilla Firefox.</code>
  <code>"""And James the son of Zebedee, and John the brother of James; and he surnamed them Boanerges, which is, The sons of thunder:"""</code>	<code>Sanskrit: """याकूब् तस्य भ्राता योहन् च आन्द्रियः फिलिपो बर्थलमयः,"""<br>English: """And James the son of Zebedee, and John the brother of James; and he surnamed them Boanerges, which is, The sons of thunder:"""</code>	<code>Sanskrit: पुनः एक्सिक्यूट् कुर्मः । अप् एरो कीलकं नुत्वा Enter नुदन्तु ।<br>English: "Let's execute again. Press the up arrow key, press Enter."</code>
  <code>The heart of a healthy adult person beats 60-80 times per minute</code>	<code>Sanskrit: स्वस्थस्य कस्यचन प्रौढस्य प्रति मिनट्-काले षष्टिवारात अधिकाधिकतया अशीतिवारं कम्पते।<br>English: The heart of a healthy adult person beats 60-80 times per minute</code>	<code>Sanskrit: यत्र कुत्रापि नुदन्तु । पश्यन्तु यत् द्वेऽपि नोड्स् नीले अभवतामिति ।<br>English: Click on any segment. Notice both the nodes turn blue.</code>

• Loss: <code>TripletLoss</code> with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 1.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 128
• learning_rate: 2e-05
• num_train_epochs: 10
• warmup_ratio: 0.05
• bf16: True
• load_best_model_at_end: True

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 2
• per_device_eval_batch_size: 128
• 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: 10
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.05
• 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
• 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}
• 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
• project: huggingface
• trackio_space_id: trackio
• 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: no
• 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: True
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

<details><summary>Click to expand</summary>

The model card is truncated. Read the rest on Hugging Face.