README.md

---
language:
- en
license: apache-2.0
library_name: transformers
tags:
- transformers
- modernbert
- fill-mask
- masked-language-model
pipeline_tag: fill-mask
datasets:
- mjbommar/ogbert-v1-mlm
model-index:
- name: ogbert-110m-base
  results:
  - task:
      type: word-similarity
    dataset:
      name: SimLex-999
      type: simlex999
    metrics:
    - type: spearman
      value: 0.345
---

# OGBert-110M-Base

A 110M parameter ModernBERT-based masked language model trained on glossary and domain-specific text.

**Related models:**
- [mjbommar/ogbert-110m-sentence](https://huggingface.co/mjbommar/ogbert-110m-sentence) - Sentence embedding version with mean pooling + L2 normalization

## Model Details

| Property | Value |
|----------|-------|
| Architecture | ModernBERT |
| Parameters | 110M |
| Hidden size | 768 |
| Layers | 12 |
| Attention heads | 12 |
| Vocab size | 32,768 |
| Max sequence | 1,024 tokens |

## Training

- **Task**: Masked Language Modeling (MLM)
- **Dataset**: [mjbommar/ogbert-v1-mlm](https://huggingface.co/datasets/mjbommar/ogbert-v1-mlm) - derived from [OpenGloss](https://arxiv.org/abs/2511.18622), a synthetic encyclopedic dictionary with 537K senses across 150K lexemes
- **Masking**: Standard 15% token masking
- **Training steps**: 8,000 steps (selected for optimal downstream performance)
- **Tokens processed**: ~4.5B
- **Batch size**: 1,024
- **Peak learning rate**: 3e-4

## Performance

### Word Similarity (SimLex-999)

**SimLex-999** measures Spearman correlation between model cosine similarities and human judgments on 999 word pairs. Higher = better alignment with human perception of word similarity.

| Model | Params | SimLex-999 (ρ) |
|-------|--------|----------------|
| **OGBert-110M-Base** | **110M** | **0.345** |
| BERT-base | 110M | 0.070 |
| RoBERTa-base | 125M | -0.061 |

OGBert-110M-Base achieves **5x better** word similarity than BERT-base with the same parameter count.

### Document Clustering

Evaluated on 80 domain-specific documents across 10 categories using KMeans.

| Model | Params | ARI | Cluster Acc |
|-------|--------|-----|-------------|
| **OGBert-110M-Base** | **110M** | **0.941** | **0.975** |
| BERT-base | 110M | 0.896 | 0.950 |
| RoBERTa-base | 125M | 0.941 | 0.975 |

OGBert-110M-Base matches or exceeds RoBERTa-base on clustering tasks.

## Usage

### Fill-Mask Pipeline

```python
from transformers import pipeline

fill_mask = pipeline('fill-mask', model='mjbommar/ogbert-110m-base')
result = fill_mask('The financial <|mask|> was approved.')
```

### Direct Model Usage

```python
from transformers import AutoModelForMaskedLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained('mjbommar/ogbert-110m-base')
model = AutoModelForMaskedLM.from_pretrained('mjbommar/ogbert-110m-base')

inputs = tokenizer('The <|mask|> definition is clear.', return_tensors='pt')
outputs = model(**inputs)
```

### For Sentence Embeddings

Use [mjbommar/ogbert-110m-sentence](https://huggingface.co/mjbommar/ogbert-110m-sentence) instead, which includes mean pooling and L2 normalization for optimal similarity search.

## Citation

If you use this model, please cite the OpenGloss dataset:

```bibtex
@article{bommarito2025opengloss,
  title={OpenGloss: A Synthetic Encyclopedic Dictionary and Semantic Knowledge Graph},
  author={Bommarito II, Michael J.},
  journal={arXiv preprint arXiv:2511.18622},
  year={2025}
}
```

## License

Apache 2.0