mehularora/scrabble-embed-v2

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the csv dataset. It maps sentences & paragraphs to a 384-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/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • csv

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': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (2): Normalize()
)

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("mehularora/scrabble-embed-v2")
# Run inference
sentences = [
    'TRAUMATOLOGY',
    "'the study of wounds and their effects [n TRAUMATOLOGIES]'",
    "'FELLATRIX, a female who fellates [n]'",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7472, 0.0801],
#         [0.7472, 1.0000, 0.2525],
#         [0.0801, 0.2525, 1.0000]])

Evaluation

Metrics

Information Retrieval

• Dataset: dictionary-test
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.6825
cosine_accuracy@3	0.8121
cosine_accuracy@5	0.8311
cosine_accuracy@10	0.8506
cosine_precision@1	0.6825
cosine_precision@3	0.2707
cosine_precision@5	0.1662
cosine_precision@10	0.0851
cosine_recall@1	0.6825
cosine_recall@3	0.8121
cosine_recall@5	0.8311
cosine_recall@10	0.8506
cosine_ndcg@10	0.7751
cosine_mrr@10	0.75
cosine_map@100	0.7522

Training Details

Training Dataset

csv

• Dataset: csv
• Size: 222,635 training samples
• Columns: word and definition
• Approximate statistics based on the first 1000 samples:

  	word	definition
  type	string	string
  details	min: 3 tokens mean: 4.87 tokens max: 9 tokens	min: 10 tokens mean: 20.32 tokens max: 98 tokens

• Samples:

  word	definition
  LICHGATES	'LICHGATE, the roofed gate of a churchyard, also LYCHGATE [n]'
  MOULDING	'a long, narrow strip used to decorate a surface, also MOLDING [n -S]'
  PARABAPTISM	'uncanonical baptism [n -S]'

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          384,
          256
      ],
      "matryoshka_weights": [
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• learning_rate: 2e-05
• num_train_epochs: 1
• fp16: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 8
• 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
• 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: 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
• 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: {}

Training Logs

Epoch	Step	Training Loss	dictionary-test_cosine_ndcg@10
0.0287	100	1.0186	0.7180
0.0575	200	0.7633	0.7274
0.0862	300	0.75	0.7398
0.1150	400	0.7503	0.7456
0.1437	500	0.7271	0.7496
0.1725	600	0.6531	0.7508
0.2012	700	0.6586	0.7560
0.2300	800	0.6559	0.7591
0.2587	900	0.6116	0.7572
0.2874	1000	0.615	0.7625
0.3162	1100	0.5926	0.7596
0.3449	1200	0.6414	0.7623
0.3737	1300	0.6143	0.7641
0.4024	1400	0.6464	0.7655
0.4312	1500	0.6039	0.7676
0.4599	1600	0.514	0.7643
0.4886	1700	0.5719	0.7675
0.5174	1800	0.612	0.7675
0.5461	1900	0.5639	0.7698
0.5749	2000	0.6025	0.7672
0.6036	2100	0.5623	0.7719
0.6324	2200	0.5484	0.7698
0.6611	2300	0.5799	0.7730
0.6899	2400	0.5253	0.7716
0.7186	2500	0.5134	0.7732
0.7473	2600	0.5543	0.7721
0.7761	2700	0.5342	0.7736
0.8048	2800	0.5507	0.7746
0.8336	2900	0.5176	0.7737
0.8623	3000	0.5067	0.7751
0.8911	3100	0.548	0.7749
0.9198	3200	0.5443	0.7751
0.9485	3300	0.5603	0.7751
0.9773	3400	0.5774	0.7751

Framework Versions

• Python: 3.11.4
• Sentence Transformers: 5.1.2
• Transformers: 4.57.3
• PyTorch: 2.9.1+cpu
• Accelerate: 1.12.0
• Datasets: 4.4.1
• Tokenizers: 0.22.1

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",
}

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}
}