# Character-level Chinese Backpack Language Models

arXiv: https://arxiv.org/abs/2310.12751 · HTML: https://mlsanity.com/paper/2310.12751/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge and representations inside LLMs, one of 13 directions in Mechanistic interpretability.
- Builds on 34 of its 62 references that are in the map.
- First posted on arXiv 19 Oct 2023.

## Placed in

- Mechanistic interpretability / [Knowledge and representations inside LLMs](https://mlsanity.com/mechinterp/knowledge-and-representations-inside-llms/)

## Builds on (most cited first, 30 of 34)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Distributed Representations of Sentences and Documents](https://mlsanity.com/paper/1405.4053/) (2014)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [Recipes for building an open-domain chatbot](https://mlsanity.com/paper/2004.13637/) (2020)
- [CLUE: A Chinese Language Understanding Evaluation Benchmark](https://mlsanity.com/paper/2004.05986/) (2020)
- [Revisiting Pre-Trained Models for Chinese Natural Language Processing](https://mlsanity.com/paper/2004.13922/) (2020)
- [A Convolutional Neural Network for Modelling Sentences](https://mlsanity.com/paper/1404.2188/) (2014)
- [ParlAI: A Dialog Research Software Platform](https://mlsanity.com/paper/1705.06476/) (2017)
- [A Span-Extraction Dataset for Chinese Machine Reading Comprehension](https://mlsanity.com/paper/1810.07366/) (2018)
- [MEGATRON-CNTRL: Controllable Story Generation with External Knowledge Using Large-Scale Language Models](https://mlsanity.com/paper/2010.00840/) (2020)
- [Few-Shot Question Answering by Pretraining Span Selection](https://mlsanity.com/paper/2101.00438/) (2021)
- [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://mlsanity.com/paper/2012.00413/) (2020)
- [SPINE: SParse Interpretable Neural Embeddings](https://mlsanity.com/paper/1711.08792/) (2017)
- [What are the biases in my word embedding?](https://mlsanity.com/paper/1812.08769/) (2018)
- [Semantic Structure and Interpretability of Word Embeddings](https://mlsanity.com/paper/1711.00331/) (2017)
- [Multilingual Models for Compositional Distributed Semantics](https://mlsanity.com/paper/1404.4641/) (2014)
- [Is Word Segmentation Necessary for Deep Learning of Chinese Representations?](https://mlsanity.com/paper/1905.05526/) (2019)
- [GraphPlan: Story Generation by Planning with Event Graph](https://mlsanity.com/paper/2102.02977/) (2021)

## Built on by (most cited first, 1 of 1)

- [Towards Intrinsic Interpretability of Large Language Models:A Survey of Design Principles and Architectures](https://mlsanity.com/paper/2604.16042/) (2026)

## Abstract (arXiv)

The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. However, Backpacks' reliance on token-defined meaning raises questions as to their potential for languages other than English, a language for which subword tokenization provides a reasonable approximation for lexical items. In this work, we train, evaluate, interpret, and control Backpack language models in character-tokenized Chinese, in which words are often composed of many characters. We find that our (134M parameter) Chinese Backpack language model performs comparably to a (104M parameter) Transformer, and learns rich character-level meanings that log-additively compose to form word meanings. In SimLex-style lexical semantic evaluations, simple averages of Backpack character senses outperform input embeddings from a Transformer. We find that complex multi-character meanings are often formed by using the same per-character sense weights consistently across context. Exploring interpretability-through control, we show that we can localize a source of gender bias in our Backpacks to specific character senses and intervene to reduce the bias.