enochlev/xlm-similarity-large

SentenceTransformer based on sentence-transformers/all-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2. 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/all-mpnet-base-v2
• Maximum Sequence Length: 384 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity

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': 384, 'do_lower_case': False}) with Transformer model: MPNetModel 
  (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})
  (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("enochlev/xlm-similarity-large")
# Run inference
sentences = [
    'Oh, okay, so just the iPhone ## only.',
    "Yes, that's correct for know. Our price is £ and then it won't go down to £ after you apply the discount.",
    "I'm now going to send you a one time code. The first message is a warning to not give the code to scammers pretending to work for O2. The second message is the code to continue with your request.",
]
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

Semantic Similarity

• Dataset: sts_dev
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5907
spearman_cosine	0.2789
pearson_manhattan	0.6309
spearman_manhattan	0.2781
pearson_euclidean	0.6349
spearman_euclidean	0.2789
pearson_dot	0.5907
spearman_dot	0.2789
pearson_max	0.6349
spearman_max	0.2789

Training Details

Training Dataset

Unnamed Dataset

• Size: 7,960 training samples
• Columns: text1, text2, and label
• Approximate statistics based on the first 1000 samples:

  	text1	text2	label
  type	string	string	float
  details	min: 5 tokens mean: 20.94 tokens max: 66 tokens	min: 13 tokens mean: 28.35 tokens max: 71 tokens	min: 0.2 mean: 0.22 max: 1.0

• Samples:

  text1	text2	label
  Hello, welcome to O2. My name is __ How can I help you today?	Thank you for calling over to my name is how can I help you.	1.0
  Hello, welcome to O2. My name is __ How can I help you today?	I was about to ask us to confirm the email address that we have on the account or on your file. So what I can you tell me your email address.	0.2
  Hello, welcome to O2. My name is __ How can I help you today?	Are you planning to get a new sim only plan or a new phone?	0.2

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Evaluation Dataset

Unnamed Dataset

• Size: 1,980 evaluation samples
• Columns: text1, text2, and label
• Approximate statistics based on the first 1000 samples:

  	text1	text2	label
  type	string	string	float
  details	min: 8 tokens mean: 36.02 tokens max: 241 tokens	min: 13 tokens mean: 28.35 tokens max: 71 tokens	min: 0.2 mean: 0.22 max: 1.0

• Samples:

  text1	text2	label
  So for example, since this is for the 2nd line bro more. So if you have any family that you want to add on your account. Yeah, we do have a same offer plan. This offer promo today.	The same discounts you can have been added as an additional line and do into your account. It needs be entitled to % discount off of the costs.	1.0
  So for example, since this is for the 2nd line bro more. So if you have any family that you want to add on your account. Yeah, we do have a same offer plan. This offer promo today.	I was about to ask us to confirm the email address that we have on the account or on your file. So what I can you tell me your email address.	0.2
  So for example, since this is for the 2nd line bro more. So if you have any family that you want to add on your account. Yeah, we do have a same offer plan. This offer promo today.	Are you planning to get a new sim only plan or a new phone?	0.2

• Loss: CoSENTLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "pairwise_cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 50
• per_device_eval_batch_size: 50
• learning_rate: 2e-05
• num_train_epochs: 1
• warmup_ratio: 0.1
• 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: 50
• per_device_eval_batch_size: 50
• 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: 0.1
• 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: 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: 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}
• 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Validation Loss	sts_dev_spearman_max
1.0	160	0.1772	0.2789

Framework Versions

• Python: 3.11.9
• Sentence Transformers: 3.2.1
• Transformers: 4.45.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.1.1
• Datasets: 3.1.0
• Tokenizers: 0.20.1

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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}