vazish/all-Mini-fine-tuned

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

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2. 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

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}) with Transformer model: 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("vazish/all-Mini-fine-tuned")
# Run inference
sentences = [
    'Crunchyroll: Anime Trends 2023',
    'Search Hotels in Montreal',
    'Online Shopping for Vacuum Cleaners',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

• Dataset: dev-eval
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.9644
spearman_cosine	0.6245

Training Details

Training Dataset

Unnamed Dataset

• Size: 133,380 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	min: 3 tokens mean: 8.75 tokens max: 21 tokens	min: 3 tokens mean: 8.55 tokens max: 19 tokens	min: 0.0 mean: 0.16 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  Military Times	Financial Analyst Resume Sample	0.0
  Outdoor Music Festivals for Adventurers	Balancing Mental Health with Outdoor Adventures	0.0
  The Rise of Artificial Intelligence in Video Games	Winter Deals on Streaming Equipment	0.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• num_train_epochs: 1
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• 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: 5e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1
• 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
• 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}
• tp_size: 0
• 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: 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: 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
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	dev-eval_spearman_cosine
0.1199	500	0.0841	-
0.2399	1000	0.0769	-
0.3598	1500	0.0671	-
0.4797	2000	0.0623	-
0.5997	2500	0.0558	-
0.7196	3000	0.0502	-
0.8395	3500	0.046	-
0.9595	4000	0.0433	-
-1	-1	-	0.6101
0.1199	500	0.0362	-
0.2399	1000	0.0353	-
0.3598	1500	0.0337	-
0.4797	2000	0.0332	-
0.5997	2500	0.0327	-
0.7196	3000	0.0312	-
0.8395	3500	0.0287	-
0.9595	4000	0.0286	-
-1	-1	-	0.6196
0.1199	500	0.0253	-
0.2399	1000	0.0232	-
0.3598	1500	0.0207	-
0.4797	2000	0.0195	-
0.5997	2500	0.0182	-
0.7196	3000	0.0162	-
0.8395	3500	0.0139	-
0.9595	4000	0.0139	-
-1	-1	-	0.6221
0.1199	500	0.0195	-
0.2399	1000	0.0166	-
0.3598	1500	0.0136	-
0.4797	2000	0.012	-
0.5997	2500	0.0108	-
0.7196	3000	0.0087	-
0.8395	3500	0.0072	-
0.9595	4000	0.0069	-
-1	-1	-	0.6227
0.1199	500	0.0162	-
0.2399	1000	0.0127	-
0.3598	1500	0.0096	-
0.4797	2000	0.0075	-
0.5997	2500	0.0065	-
0.7196	3000	0.0049	-
0.8395	3500	0.0043	-
0.9595	4000	0.0043	-
-1	-1	-	0.6229
0.1199	500	0.0139	-
0.2399	1000	0.0099	-
0.3598	1500	0.0069	-
0.4797	2000	0.005	-
0.5997	2500	0.0042	-
0.7196	3000	0.0031	-
0.8395	3500	0.0027	-
0.9595	4000	0.0029	-
-1	-1	-	0.6234
0.1199	500	0.0125	-
0.2399	1000	0.0078	-
0.3598	1500	0.005	-
0.4797	2000	0.0036	-
0.5997	2500	0.0028	-
0.7196	3000	0.0022	-
0.8395	3500	0.002	-
0.9595	4000	0.0022	-
-1	-1	-	0.6248
0.1199	500	0.0114	-
0.2399	1000	0.0068	-
0.3598	1500	0.004	-
0.4797	2000	0.0027	-
0.5997	2500	0.0023	-
0.7196	3000	0.0014	-
0.8395	3500	0.0015	-
0.9595	4000	0.0015	-
-1	-1	-	0.6245
0.1199	500	0.0107	-
0.2399	1000	0.0058	-
0.3598	1500	0.0034	-
0.4797	2000	0.0021	-
0.5997	2500	0.0016	-
0.7196	3000	0.0011	-
0.8395	3500	0.0013	-
0.9595	4000	0.0011	-
-1	-1	-	0.6249
0.1199	500	0.0097	-
0.2399	1000	0.0048	-
0.3598	1500	0.0024	-
0.4797	2000	0.0015	-
0.5997	2500	0.0013	-
0.7196	3000	0.0009	-
0.8395	3500	0.001	-
0.9595	4000	0.0009	-
-1	-1	-	0.6245

Framework Versions

• Python: 3.11.12
• Sentence Transformers: 4.1.0
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.5.2
• Datasets: 3.5.0
• Tokenizers: 0.21.1

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",
}