# Understanding the Interplay of Scale, Data, and Bias in Language Models: A Case Study with BERT

arXiv: https://arxiv.org/abs/2407.21058 · HTML: https://mlsanity.com/paper/2407.21058/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 53 references that are in the map.
- First posted on arXiv 25 Jul 2024.

## Builds on (most cited first, 30 of 32)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Deep Learning Scaling is Predictable, Empirically](https://mlsanity.com/paper/1712.00409/) (2017)
- [CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data](https://mlsanity.com/paper/1911.00359/) (2019)
- [Scaling Laws for Autoregressive Generative Modeling](https://mlsanity.com/paper/2010.14701/) (2020)
- [TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification](https://mlsanity.com/paper/2010.12421/) (2020)
- [On Measuring Social Biases in Sentence Encoders](https://mlsanity.com/paper/1903.10561/) (2019)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [Measuring Bias in Contextualized Word Representations](https://mlsanity.com/paper/1906.07337/) (2019)
- [Scaling Laws for Transfer](https://mlsanity.com/paper/2102.01293/) (2021)
- [Unified Scaling Laws for Routed Language Models](https://mlsanity.com/paper/2202.01169/) (2022)
- [Social Biases in NLP Models as Barriers for Persons with Disabilities](https://mlsanity.com/paper/2005.00813/) (2020)
- [UnQovering Stereotyping Biases via Underspecified Questions](https://mlsanity.com/paper/2010.02428/) (2020)
- [Scaling Laws vs Model Architectures: How does Inductive Bias Influence Scaling?](https://mlsanity.com/paper/2207.10551/) (2022)
- [Automated Experiments on Ad Privacy Settings: A Tale of Opacity, Choice, and Discrimination](https://mlsanity.com/paper/1408.6491/) (2014)
- [Unmasking Contextual Stereotypes: Measuring and Mitigating BERT's Gender Bias](https://mlsanity.com/paper/2010.14534/) (2020)
- [Hurtful Words: Quantifying Biases in Clinical Contextual Word Embeddings](https://mlsanity.com/paper/2003.11515/) (2020)
- [Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics](https://mlsanity.com/paper/2110.01518/) (2021)

## Built on by (most cited first, 5 of 5)

- [Explicit vs. Implicit: Investigating Social Bias in Large Language Models through Self-Reflection](https://mlsanity.com/paper/2501.02295/) (2025)
- [Unequal Opportunities: Examining the Bias in Geographical Recommendations by Large Language Models](https://mlsanity.com/paper/2504.05325/) (2025)
- [A dataset of questions on decision-theoretic reasoning in Newcomb-like problems](https://mlsanity.com/paper/2411.10588/) (2024)
- [Fairness Metric Design Exploration in Multi-Domain Moral Sentiment Classification using Transformer-Based Models](https://mlsanity.com/paper/2510.11222/) (2025)
- [Critical Acclaim Orientation in Large Language Models: Evidence from Film Preference Elicitation](https://mlsanity.com/paper/2608.06955/) (2026)

## Abstract (arXiv)

In the current landscape of language model research, larger models, larger datasets and more compute seems to be the only way to advance towards intelligence. While there have been extensive studies of scaling laws and models' scaling behaviors, the effect of scale on a model's social biases and stereotyping tendencies has received less attention. In this study, we explore the influence of model scale and pre-training data on its learnt social biases. We focus on BERT -- an extremely popular language model -- and investigate biases as they show up during language modeling (upstream), as well as during classification applications after fine-tuning (downstream). Our experiments on four architecture sizes of BERT demonstrate that pre-training data substantially influences how upstream biases evolve with model scale. With increasing scale, models pre-trained on large internet scrapes like Common Crawl exhibit higher toxicity, whereas models pre-trained on moderated data sources like Wikipedia show greater gender stereotypes. However, downstream biases generally decrease with increasing model scale, irrespective of the pre-training data. Our results highlight the qualitative role of pre-training data in the biased behavior of language models, an often overlooked aspect in the study of scale. Through a detailed case study of BERT, we shed light on the complex interplay of data and model scale, and investigate how it translates to concrete biases.