tomaarsen/TinyBERT_L-4_H-312_v2-distilled-from-stsb-roberta-base-v2-L2

SentenceTransformer based on nreimers/TinyBERT_L-4_H-312_v2

This is a sentence-transformers model finetuned from nreimers/TinyBERT_L-4_H-312_v2. It maps sentences & paragraphs to a 312-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: nreimers/TinyBERT_L-4_H-312_v2
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 312 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': 512, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 312, '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("tomaarsen/TinyBERT_L-4_H-312_v2-distilled-from-stsb-roberta-base-v2-L2")
# Run inference
sentences = [
    'A small group of children are standing in a classroom and one of them has a foot in a trashcan, which also has a rope leading out of it.',
    'Some men with jerseys are in a bar, watching a soccer match.',
    'There are two people running around a track in lane three and the one wearing a blue shirt with a green thing over the eyes is just barely ahead of the guy wearing an orange shirt and sunglasses.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 312]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.1114, 0.1211],
#         [0.1114, 1.0000, 0.3643],
#         [0.1211, 0.3643, 1.0000]])

Evaluation

Metrics

Semantic Similarity

• Datasets: sts-dev and sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	sts-dev	sts-test
pearson_cosine	0.7998	0.7475
spearman_cosine	0.8136	0.75

Knowledge Distillation

• Evaluated with MSEEvaluator

Metric	Value
negative_mse	-50.6792

Training Details

Training Dataset

Unnamed Dataset

• Size: 200,000 training samples
• Columns: sentence and label
• Approximate statistics based on the first 1000 samples:

  	sentence	label
  type	string	list
  details	min: 4 tokens mean: 12.24 tokens max: 52 tokens	size: 312 elements

• Samples:

  sentence	label
  A person on a horse jumps over a broken down airplane.	[0.029216572642326355, 0.7606895565986633, -2.639708995819092, 1.5909428596496582, 1.2897400856018066, ...]
  Children smiling and waving at camera	[-2.8437485694885254, 2.944169282913208, 7.142611503601074, 5.286141395568848, -2.155975341796875, ...]
  A boy is jumping on skateboard in the middle of a red bridge.	[2.7921977043151855, 3.262112617492676, 1.0734096765518188, 6.316248893737793, -1.0134869813919067, ...]

• Loss: MSELoss

Evaluation Dataset

Unnamed Dataset

• Size: 10,000 evaluation samples
• Columns: sentence and label
• Approximate statistics based on the first 1000 samples:

  	sentence	label
  type	string	list
  details	min: 5 tokens mean: 13.23 tokens max: 57 tokens	size: 312 elements

• Samples:

  sentence	label
  Two women are embracing while holding to go packages.	[-6.028911590576172, -2.3896284103393555, 2.2538001537323, -2.044457197189331, 1.669250726699829, ...]
  Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.	[-1.9341195821762085, 0.6948365569114685, 2.5149948596954346, 3.9593544006347656, -3.2706212997436523, ...]
  A man selling donuts to a customer during a world exhibition event held in the city of Angeles	[3.196415662765503, 3.1836068630218506, -0.26187384128570557, -2.4298267364501953, 3.2045652866363525, ...]

• Loss: MSELoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• learning_rate: 0.0001
• num_train_epochs: 1
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: 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: 64
• 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: 0.0001
• 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: 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
• 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
• 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: False
• 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: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	Validation Loss	sts-dev_spearman_cosine	negative_mse	sts-test_spearman_cosine
0.032	100	16.5376	-	-	-	-
0.064	200	15.6489	-	-	-	-
0.096	300	14.5959	-	-	-	-
0.128	400	13.7611	-	-	-	-
0.16	500	13.2159	13.9388	0.7559	-63.5636	-
0.192	600	12.8215	-	-	-	-
0.224	700	12.4917	-	-	-	-
0.256	800	12.2748	-	-	-	-
0.288	900	12.009	-	-	-	-
0.32	1000	11.7972	13.1279	0.7896	-57.2613	-
0.352	1100	11.6088	-	-	-	-
0.384	1200	11.459	-	-	-	-
0.416	1300	11.3065	-	-	-	-
0.448	1400	11.1917	-	-	-	-
0.48	1500	11.1288	12.6637	0.8021	-53.8425	-
0.512	1600	10.9378	-	-	-	-
0.544	1700	10.8963	-	-	-	-
0.576	1800	10.8034	-	-	-	-
0.608	1900	10.7124	-	-	-	-
0.64	2000	10.6427	12.4148	0.8092	-52.1864	-
0.672	2100	10.6062	-	-	-	-
0.704	2200	10.5628	-	-	-	-
0.736	2300	10.5185	-	-	-	-
0.768	2400	10.4376	-	-	-	-
0.8	2500	10.3779	12.2752	0.8115	-51.3012	-
0.832	2600	10.3268	-	-	-	-
0.864	2700	10.323	-	-	-	-
0.896	2800	10.2904	-	-	-	-
0.928	2900	10.2949	-	-	-	-
0.96	3000	10.2858	12.187	0.8136	-50.6792	-
0.992	3100	10.245	-	-	-	-
-1	-1	-	-	-	-	0.7500

• The bold row denotes the saved checkpoint.

Environmental Impact

Carbon emissions were measured using CodeCarbon.

• Energy Consumed: 0.011 kWh
• Carbon Emitted: 0.003 kg of CO2
• Hours Used: 0.061 hours

Training Hardware

• On Cloud: No
• GPU Model: 1 x NVIDIA GeForce RTX 3090
• CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
• RAM Size: 31.78 GB

Framework Versions

• Python: 3.11.6
• Sentence Transformers: 5.2.0.dev0
• Transformers: 4.53.3
• PyTorch: 2.8.0+cu128
• Accelerate: 1.6.0
• Datasets: 4.2.0
• Tokenizers: 0.21.4

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",
}

MSELoss

@inproceedings{reimers-2020-multilingual-sentence-bert,
    title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2020",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2004.09813",
}