# Layer by Layer: Uncovering Where Multi-Task Learning Happens in Instruction-Tuned Large Language Models

arXiv: https://arxiv.org/abs/2410.20008 · HTML: https://mlsanity.com/paper/2410.20008/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 37 references that are in the map.
- First posted on arXiv 25 Oct 2024.

## Builds on (most cited first, 23 of 23)

- [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)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [The Flan Collection: Designing Data and Methods for Effective Instruction Tuning](https://mlsanity.com/paper/2301.13688/) (2023)
- [Cross-Task Generalization via Natural Language Crowdsourcing Instructions](https://mlsanity.com/paper/2104.08773/) (2021)
- [Improving Text Embeddings with Large Language Models](https://mlsanity.com/paper/2401.00368/) (2023)
- [Muppet: Massive Multi-task Representations with Pre-Finetuning](https://mlsanity.com/paper/2101.11038/) (2021)
- [What do Neural Machine Translation Models Learn about Morphology?](https://mlsanity.com/paper/1704.03471/) (2017)
- [What Happens To BERT Embeddings During Fine-tuning?](https://mlsanity.com/paper/2004.14448/) (2020)
- [Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Information](https://mlsanity.com/paper/1808.08079/) (2018)
- [First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT](https://mlsanity.com/paper/2101.11109/) (2021)
- [Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks](https://mlsanity.com/paper/1801.07772/) (2018)
- [Spectral Editing of Activations for Large Language Model Alignment](https://mlsanity.com/paper/2405.09719/) (2024)
- [A Joint Matrix Factorization Analysis of Multilingual Representations](https://mlsanity.com/paper/2310.15513/) (2023)
- [Understanding Domain Learning in Language Models Through Subpopulation Analysis](https://mlsanity.com/paper/2210.12553/) (2022)

## Built on by (most cited first, 15 of 15)

- [Layer by Layer: Uncovering Hidden Representations in Language Models](https://mlsanity.com/paper/2502.02013/) (2025)
- [Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts](https://mlsanity.com/paper/2510.07358/) (2025)
- [Alignment Quality Index (AQI) : Beyond Refusals: AQI as an Intrinsic Alignment Diagnostic via Latent Geometry, Cluster Divergence, and Layer wise Pooled Representations](https://mlsanity.com/paper/2506.13901/) (2025)
- [Old Habits Die Hard: How Conversational History Geometrically Traps LLMs](https://mlsanity.com/paper/2603.03308/) (2026)
- [FLoRIST: Singular Value Thresholding for Efficient and Accurate Federated Fine-Tuning of Large Language Models](https://mlsanity.com/paper/2506.09199/) (2025)
- [Differentiable Faithfulness Alignment for Cross-Model Circuit Transfer](https://mlsanity.com/paper/2604.24302/) (2026)
- [Parameter Importance-Driven Continual Learning for Foundation Models](https://mlsanity.com/paper/2511.15375/) (2025)
- [Delve into the Applicability of Advanced Optimizers for Multi-Task Learning](https://mlsanity.com/paper/2604.08939/) (2026)
- [Pseudo-Unification: Entropy Probing Reveals Divergent Information Patterns in Unified Multimodal Models](https://mlsanity.com/paper/2604.10949/) (2026)
- [Self Knowledge Re-expression: A Fully Local Method for Adapting LLMs to Tasks Using Intrinsic Knowledge](https://mlsanity.com/paper/2604.22939/) (2026)
- [On the Persistent Effects of Lexicality in Large Language Models](https://mlsanity.com/paper/2606.02750/) (2026)
- [Multi Attribute Bias Mitigation via Representation Learning](https://mlsanity.com/paper/2509.03616/) (2025)
- [The Horizon Threshold in Cooperative Multi-Agent Reward-Free Exploration](https://mlsanity.com/paper/2602.01453/) (2026)
- [The Convergence Gap: Instruction-Tuned Language Models Stabilize Later in the Forward Pass](https://mlsanity.com/paper/2605.07282/) (2026)
- [Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuning](https://mlsanity.com/paper/2606.01967/) (2026)

## Abstract (arXiv)

Fine-tuning pre-trained large language models (LLMs) on a diverse array of tasks has become a common approach for building models that can solve various natural language processing (NLP) tasks. However, where and to what extent these models retain task-specific knowledge remains largely unexplored. This study investigates the task-specific information encoded in pre-trained LLMs and the effects of instruction tuning on their representations across a diverse set of over 60 NLP tasks. We use a set of matrix analysis tools to examine the differences between the way pre-trained and instruction-tuned LLMs store task-specific information. Our findings reveal that while some tasks are already encoded within the pre-trained LLMs, others greatly benefit from instruction tuning. Additionally, we pinpointed the layers in which the model transitions from high-level general representations to more task-oriented representations. This finding extends our understanding of the governing mechanisms of LLMs and facilitates future research in the fields of parameter-efficient transfer learning and multi-task learning.