# Scaling Laws for Upcycling Mixture-of-Experts Language Models

arXiv: https://arxiv.org/abs/2502.03009 · HTML: https://mlsanity.com/paper/2502.03009/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in MoE large language models, one of 12 directions in Mixture-of-experts.
- Builds on 59 of its 62 references that are in the map.
- First posted on arXiv 5 Feb 2025.

## Placed in

- Mixture-of-experts / [MoE large language models](https://mlsanity.com/moe/moe-large-language-models/)

## Builds on (most cited first, 30 of 59)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models](https://mlsanity.com/paper/2401.06066/) (2024)
- [Crowdsourcing Multiple Choice Science Questions](https://mlsanity.com/paper/1707.06209/) (2017)
- [MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies](https://mlsanity.com/paper/2404.06395/) (2024)
- [DeepSeek LLM: Scaling Open-Source Language Models with Longtermism](https://mlsanity.com/paper/2401.02954/) (2024)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [Deep Learning Scaling is Predictable, Empirically](https://mlsanity.com/paper/1712.00409/) (2017)
- [Scaling Laws for Autoregressive Generative Modeling](https://mlsanity.com/paper/2010.14701/) (2020)
- [Scaling Data-Constrained Language Models](https://mlsanity.com/paper/2305.16264/) (2023)
- [Explaining Neural Scaling Laws](https://mlsanity.com/paper/2102.06701/) (2021)

## Built on by (most cited first, 13 of 13)

- [From Curated Data to Scalable Models: Continual Pre-training of Dense and MoE Large Language Models for Tibetan](https://mlsanity.com/paper/2507.09205/) (2025)
- [Spend Less, Fit Better: Budget-Efficient Scaling Law Fitting via Active Experiment Selection](https://mlsanity.com/paper/2604.22753/) (2026)
- [Evaluating the Robustness of Chinchilla Compute-Optimal Scaling](https://mlsanity.com/paper/2509.23963/) (2025)
- [Reusing Overtrained Language Models Saturates Scaling](https://mlsanity.com/paper/2510.06548/) (2025)
- [Beyond Sunk Costs: Boosting LLM Pre-training Efficiency via Orthogonal Growth of Mixture-of-Experts](https://mlsanity.com/paper/2510.08008/) (2025)
- [Scaling depth capacity via zero/one-layer model expansion](https://mlsanity.com/paper/2511.04981/) (2025)
- [Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better](https://mlsanity.com/paper/2602.05393/) (2026)
- [Expert Upcycling: Shifting the Compute-Efficient Frontier of Mixture-of-Experts](https://mlsanity.com/paper/2604.19835/) (2026)
- [Towards Principled Design of Mixture-of-Experts Language Models under Memory and Inference Constraints](https://mlsanity.com/paper/2601.08215/) (2026)
- [Generalization and Scaling Laws for Mixture-of-Experts Transformers](https://mlsanity.com/paper/2604.09175/) (2026)
- [Enhancing Mixture-of-Experts Specialization via Cluster-Aware Upcycling](https://mlsanity.com/paper/2604.13508/) (2026)
- [Toward LLMs Beyond English-Centric Development](https://mlsanity.com/paper/2605.15613/) (2026)
- [STAPLE: Static Sparse Upcycling of Pretrained Checkpoints via Embedding Modules](https://mlsanity.com/paper/2610.12599/) (2026)

## Abstract (arXiv)

Pretraining large language models (LLMs) is resource-intensive, often requiring months of training time even with high-end GPU clusters. There are two approaches of mitigating such computational demands: reusing smaller models to train larger ones (upcycling), and training computationally efficient models like mixture-of-experts (MoE). In this paper, we study the upcycling of LLMs to MoE models, of which the scaling behavior remains underexplored. Through extensive experiments, we identify empirical scaling laws that describe how performance depends on dataset size and model configuration. Particularly, we show that, while scaling these factors improves performance, there is a novel interaction term between the dense and upcycled training dataset that limits the efficiency of upcycling at large computational budgets. Based on these findings, we provide guidance to scale upcycling, and establish conditions under which upcycling outperforms from-scratch trainings within budget constraints.