File size: 1,558 Bytes
901c842 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 | ---
license: other
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- dense
- baseline
- ablation
- memory-matched
datasets:
- allenai/OLMoE-mix-0924
---
# Dense_1b_130B
An active-parameter-matched dense baseline released alongside [EMO: Pretraining Mixture of Experts for Emergent Modularity](https://arxiv.org/abs/2605.06663) — referred to as **"Dense @ 8"** in Figure 1 of the paper. Not midtrained.
1B parameter dense decoder-only Transformer (no MoE) pretrained from scratch on 130B tokens of the OLMoE pretraining mix. Provides an active-parameter-matched comparison point against 8-expert subsets carved out of the larger 1B/14B EMO models.
## Usage
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "allenai/Dense_1b_130B"
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
inputs = tokenizer(["Language modeling is "], return_tensors="pt", return_token_type_ids=False)
out = model.generate(**inputs, max_new_tokens=100, do_sample=True, temperature=1.0, top_p=0.7)
print(tokenizer.batch_decode(out, skip_special_tokens=True)[0])
```
## Citation
```bibtex
@article{wang2026emo,
title = {EMO: Pretraining Mixture of Experts for Emergent Modularity},
author = {Wang, Ryan and Bhagia, Akshita and Min, Sewon},
year = {2026},
url = {https://arxiv.org/abs/2605.06663}
}
```
## Links
- Paper: https://arxiv.org/abs/2605.06663
- Code: https://github.com/allenai/EMO
|