# DreamReader: An Interpretability Toolkit for Text-to-Image Models

arXiv: https://arxiv.org/abs/2603.13299 · HTML: https://mlsanity.com/paper/2603.13299/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 23 of its 36 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

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

- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [The Platonic Representation Hypothesis](https://mlsanity.com/paper/2405.07987/) (2024)
- [Understanding image representations by measuring their equivariance and equivalence](https://mlsanity.com/paper/1411.5908/) (2014)
- [What the DAAM: Interpreting Stable Diffusion Using Cross Attention](https://mlsanity.com/paper/2210.04885/) (2022)
- [Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models](https://mlsanity.com/paper/2401.06102/) (2024)
- [Design Guidelines for Prompt Engineering Text-to-Image Generative Models](https://mlsanity.com/paper/2109.06977/) (2021)
- [ReFT: Representation Finetuning for Language Models](https://mlsanity.com/paper/2404.03592/) (2024)
- [Improving Dictionary Learning with Gated Sparse Autoencoders](https://mlsanity.com/paper/2404.16014/) (2024)
- [Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Models](https://mlsanity.com/paper/2311.12092/) (2023)
- [Learning Multi-Level Features with Matryoshka Sparse Autoencoders](https://mlsanity.com/paper/2503.17547/) (2025)
- [Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing](https://mlsanity.com/paper/2403.03431/) (2024)
- [Hyperparameters in Reinforcement Learning and How To Tune Them](https://mlsanity.com/paper/2306.01324/) (2023)
- [TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.01261/) (2023)
- [NNsight and NDIF: Democratizing Access to Open-Weight Foundation Model Internals](https://mlsanity.com/paper/2407.14561/) (2024)
- [Shared Global and Local Geometry of Language Model Embeddings](https://mlsanity.com/paper/2503.21073/) (2025)
- [One-Step is Enough: Sparse Autoencoders for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2410.22366/) (2024)
- [On Mechanistic Knowledge Localization in Text-to-Image Generative Models](https://mlsanity.com/paper/2405.01008/) (2024)
- [Beyond Linear Steering: Unified Multi-Attribute Control for Language Models](https://mlsanity.com/paper/2505.24535/) (2025)
- [Rethinking Training for De-biasing Text-to-Image Generation: Unlocking the Potential of Stable Diffusion](https://mlsanity.com/paper/2408.12692/) (2024)
- [Distilling Diversity and Control in Diffusion Models](https://mlsanity.com/paper/2503.10637/) (2025)
- [Prisma: An Open Source Toolkit for Mechanistic Interpretability in Vision and Video](https://mlsanity.com/paper/2504.19475/) (2025)
- [Unveiling Concept Attribution in Diffusion Models](https://mlsanity.com/paper/2412.02542/) (2024)
- [Linear Representation Transferability Hypothesis: Leveraging Small Models to Steer Large Models](https://mlsanity.com/paper/2506.00653/) (2025)

## Abstract (arXiv)

Despite the rapid adoption of text-to-image (T2I) diffusion models, causal and representation-level analysis remains fragmented and largely limited to isolated probing techniques. To address this gap, we introduce DreamReader: a unified framework that formalizes diffusion interpretability as composable representation operators spanning activation extraction, causal patching, structured ablations, and activation steering across modules and timesteps. DreamReader provides a model-agnostic abstraction layer enabling systematic analysis and intervention across diffusion architectures. Beyond consolidating existing methods, DreamReader introduces three novel intervention primitives for diffusion models: (1) representation fine-tuning (LoReFT) for subspace-constrained internal adaptation; (2) classifier-guided gradient steering using MLP probes trained on activations; and (3) component-level cross-model mapping for systematic study of transferability of representations across modalities. These mechanisms allows us to do lightweight white-box interventions on T2I models by drawing inspiration from interpretability techniques on LLMs. We demonstrate DreamReader through controlled experiments that (i) perform activation stitching between two models, and (ii) apply LoReFT to steer multiple activation units, reliably injecting a target concept into the generated images. Experiments are specified declaratively and executed in controlled batched pipelines to enable reproducible large-scale analysis. Across multiple case studies, we show that techniques adapted from language model interpretability yield promising and controllable interventions in diffusion models. DreamReader is released as an open source toolkit for advancing research on T2I interpretability.