Hugging Face's logo

boltuix
/
bert-mid

Fill-Mask
Transformers
Safetensors
English
bert
BERT
MNLI
NLI
transformer
pre-training
NLP
MIT-NLP-v1
Model card Files Community

License: MIT Model Size Type Performance

Model Card for boltuix/bert-mid

The boltuix/bert-mid model is a compact BERT variant designed for natural language processing tasks requiring well-rounded performance with moderate resource demands. Pretrained on English text using masked language modeling (MLM) and next sentence prediction (NSP) objectives, it is optimized for fine-tuning on a variety of NLP tasks, including sequence classification, token classification, and question answering. With a size of ~50 MB, it offers a balanced solution for applications needing solid accuracy and efficiency, ideal for mid-tier deployments.

Model Details

Model Description

The boltuix/bert-mid model is a PyTorch-based transformer model derived from TensorFlow checkpoints in the Google BERT repository. It builds on research from On the Importance of Pre-training Compact Models (arXiv) and Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics (arXiv). Ported to Hugging Face, this uncased model (~50 MB) is engineered for mid-tier NLP applications, such as sentiment analysis, named entity recognition, and natural language inference, making it suitable for developers and researchers seeking a cost-effective, balanced model.

  • Developed by: BoltUIX
  • Funded by: BoltUIX Research Fund
  • Shared by: Hugging Face
  • Model type: Transformer (BERT)
  • Language(s) (NLP): English (en)
  • License: MIT
  • Finetuned from model: google-bert/bert-base-uncased

Model Sources

Model Variants

BoltUIX offers a range of BERT-based models tailored to different performance and resource requirements. The boltuix/bert-mid model is a well-rounded mid-tier option, ideal for applications needing balanced accuracy and efficiency. Below is a summary of available models:

Tier Model ID Size (MB) Notes
Micro boltuix/bert-micro ~15 MB Smallest, blazing-fast, moderate accuracy
Mini boltuix/bert-mini ~17 MB Ultra-compact, fast, slightly better accuracy
Tinyplus boltuix/bert-tinyplus ~20 MB Slightly bigger, better capacity
Small boltuix/bert-small ~45 MB Good compact/accuracy balance
Mid boltuix/bert-mid ~50 MB Well-rounded mid-tier performance
Medium boltuix/bert-medium ~160 MB Strong general-purpose model
Large boltuix/bert-large ~365 MB Top performer below full-BERT
Pro boltuix/bert-pro ~420 MB Use only if max accuracy is mandatory
Mobile boltuix/bert-mobile ~140 MB Mobile-optimized; quantize to ~25 MB with no major loss

For more details on each variant, visit the BoltUIX Model Hub.

Uses

Direct Use

The model can be used directly for masked language modeling or next sentence prediction tasks, such as predicting missing words in sentences or determining sentence coherence, delivering balanced accuracy in these core tasks.

Downstream Use

The model is designed for fine-tuning on a range of downstream NLP tasks, including:

  • Sequence classification (e.g., sentiment analysis, intent detection)
  • Token classification (e.g., named entity recognition, part-of-speech tagging)
  • Question answering (e.g., extractive QA, reading comprehension)
  • Natural language inference (e.g., MNLI, RTE) It is recommended for developers, researchers, and small-to-medium enterprises seeking a mid-tier NLP model with solid performance and efficient resource usage.

Out-of-Scope Use

The model is not suitable for:

  • Text generation tasks (use generative models like GPT-3 instead).
  • Non-English language tasks without significant fine-tuning.
  • High-performance applications requiring maximum accuracy (use boltuix/bert-large or boltuix/bert-pro instead).

Bias, Risks, and Limitations

The model may inherit biases from its training data (BookCorpus and English Wikipedia), potentially reinforcing stereotypes, such as gender or occupational biases. For example:

from transformers import pipeline
unmasker = pipeline('fill-mask', model='boltuix/bert-mid')
unmasker("The man worked as a [MASK].")

Output:

[
  {'sequence': '[CLS] the man worked as a engineer. [SEP]', 'token_str': 'engineer'},
  {'sequence': '[CLS] the man worked as a doctor. [SEP]', 'token_str': 'doctor'},
  ...
]

unmasker("The woman worked as a [MASK].")

Output:

[
  {'sequence': '[CLS] the woman worked as a teacher. [SEP]', 'token_str': 'teacher'},
  {'sequence': '[CLS] the woman worked as a nurse. [SEP]', 'token_str': 'nurse'},
  ...
]

These biases may propagate to downstream tasks. Due to its size (~50 MB), the model is suitable for many devices but may still require optimization for ultra-constrained environments.

Recommendations

Users should:

  • Conduct bias audits tailored to their application.
  • Fine-tune with diverse, representative datasets to reduce bias.
  • Apply model compression techniques (e.g., quantization, pruning) for deployment on resource-constrained devices.

How to Get Started with the Model

Use the code below to get started with the model.

from transformers import pipeline, BertTokenizer, BertModel

# Masked Language Modeling
unmasker = pipeline('fill-mask', model='boltuix/bert-mid')
result = unmasker("Hello I'm a [MASK] model.")
print(result)

# Feature Extraction (PyTorch)
tokenizer = BertTokenizer.from_pretrained('boltuix/bert-mid')
model = BertModel.from_pretrained('boltuix/bert-mid')
text = "Replace me by any text you'd like."
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)

Training Details

Training Data

The model was pretrained on:

  • BookCorpus: ~11,038 unpublished books, providing diverse narrative text.
  • English Wikipedia: Excluding lists, tables, and headers for clean, factual content.

See the BoltUIX Dataset Card for more details.

Training Procedure

Preprocessing

  • Texts are lowercased and tokenized using WordPiece with a vocabulary size of 30,000.
  • Inputs are formatted as: [CLS] Sentence A [SEP] Sentence B [SEP].
  • 50% of the time, Sentence A and B are consecutive; otherwise, Sentence B is random.
  • Masking:
    • 15% of tokens are masked.
    • 80% of masked tokens are replaced with [MASK].
    • 10% are replaced with a random token.
    • 10% are left unchanged.

Training Hyperparameters

  • Training regime: fp16 mixed precision
  • Optimizer: Adam (learning rate 1e-4, β1=0.9, β2=0.999, weight decay 0.01)
  • Batch size: 128
  • Steps: 800,000
  • Sequence length: 128 tokens (95% of steps), 512 tokens (5% of steps)
  • Warmup: 8,000 steps with linear learning rate decay

Speeds, Sizes, Times

  • Training time: Approximately 120 hours
  • Checkpoint size: ~50 MB
  • Throughput: ~120 sentences/second on TPU infrastructure

Evaluation

Testing Data, Factors & Metrics

Testing Data

Evaluated on the GLUE benchmark, including tasks like MNLI, QQP, QNLI, SST-2, CoLA, STS-B, MRPC, and RTE.

Factors

  • Subpopulations: General English text, academic, and professional domains
  • Domains: News, books, Wikipedia, scientific articles

Metrics

  • Accuracy: For classification tasks (e.g., MNLI, SST-2)
  • F1 Score: For tasks like QQP, MRPC
  • Pearson/Spearman Correlation: For STS-B

Results

GLUE test results (fine-tuned):

Task MNLI-(m/mm) QQP QNLI SST-2 CoLA STS-B MRPC RTE Average
Score 83.5/82.3 70.9 89.4 92.1 50.7 84.6 87.5 65.3 78.3

Summary

The model delivers balanced performance across GLUE tasks, with solid results in SST-2 and QNLI. It outperforms smaller BERT variants like boltuix/bert-small in tasks such as RTE and CoLA, making it a well-rounded mid-tier option.

Model Examination

The model’s attention mechanisms were analyzed to ensure effective contextual understanding, with no significant overfitting observed during pretraining. Ablation studies confirmed the suitability of the training configuration for mid-tier performance.

Environmental Impact

Carbon emissions estimated using the Machine Learning Impact calculator from Lacoste et al. (2019).

  • Hardware Type: 2 cloud TPUs (8 TPU chips)
  • Hours used: 120 hours
  • Cloud Provider: Google Cloud
  • Compute Region: us-central1
  • Carbon Emitted: ~80 kg CO2eq (estimated based on TPU energy consumption and regional grid carbon intensity)

Technical Specifications

Model Architecture and Objective

  • Architecture: BERT (transformer-based, bidirectional)
  • Objective: Masked Language Modeling (MLM) and Next Sentence Prediction (NSP)
  • Layers: 6
  • Hidden Size: 512
  • Attention Heads: 8

Compute Infrastructure

Hardware

  • 2 cloud TPUs in Pod configuration (8 TPU chips total)

Software

  • PyTorch
  • Transformers library (Hugging Face)

Citation

BibTeX:

@article{DBLP:journals/corr/abs-1810-04805,
  author    = {Jacob Devlin and Ming{-}Wei Chang and Kenton Lee and Kristina Toutanova},
  title     = {{BERT:} Pre-training of Deep Bidirectional Transformers for Language Understanding},
  journal   = {CoRR},
  volume    = {abs/1810.04805},
  year      = {2018},
  url       = {http://arxiv.org/abs/1810.04805},
  archivePrefix = {arXiv},
  eprint    = {1810.04805}
}

APA: Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. CoRR, abs/1810.04805. http://arxiv.org/abs/1810.04805

Glossary

  • MLM: Masked Language Modeling, where 15% of tokens are masked for prediction.
  • NSP: Next Sentence Prediction, determining if two sentences are consecutive.
  • WordPiece: Tokenization method splitting words into subword units.

More Information

Model Card Authors

  • Hugging Face team
  • BoltUIX contributors

Model Card Contact

For questions, please contact [email protected] or open an issue on the model repository.

Downloads last month
0
Safetensors
Model size
12.8M params
Tensor type
F32
·
Inference Providers NEW
Fill-Mask
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support