Related papers: ParaScopes: What do Language Models Activations En…
The ability of machine learning models to store input information in hidden layer vector embeddings, analogous to the concept of `memory', is widely employed but not well characterized. We find that language model embeddings typically…
A better understanding of the emergent computation and problem-solving capabilities of recent large language models is of paramount importance to further improve them and broaden their applicability. This work investigates how a language…
Large-scale pre-trained vision foundation models, such as CLIP, have become de facto backbones for various vision tasks. However, due to their black-box nature, understanding the underlying rules behind these models' predictions and…
How do language models learn to make predictions during pre-training? To study this, we extract learning curves from five autoregressive English language model pre-training runs, for 1M unseen tokens in context. We observe that the language…
Understanding which information is encoded in deep models of spoken and written language has been the focus of much research in recent years, as it is crucial for debugging and improving these architectures. Most previous work has focused…
As Large Language Models (LLMs) become increasingly integrated into our daily lives, the potential harms from deceptive behavior underlie the need for faithfully interpreting their decision-making. While traditional probing methods have…
Despite the extensive success of pretrained language models as encoders for building NLP systems, they haven't seen prominence as decoders for sequence generation tasks. We explore the question of whether these models can be adapted to be…
Many neural network models nowadays have achieved promising performances in Chit-chat settings. The majority of them rely on an encoder for understanding the post and a decoder for generating the response. Without given assigned semantics,…
Language model intelligence is revolutionizing the way we program materials simulations. However, the diversity of simulation scenarios renders it challenging to precisely transform human language into a tailored simulator. Here, using…
Decoding methods play an indispensable role in converting language models from next-token predictors into practical task solvers. Prior research on decoding methods, primarily focusing on task-specific models, may not extend to the current…
Autoregressive language models trained with next-token prediction generate text by sampling one discrete token at a time. Although very scalable, this objective forces the model to commit at every step, preventing it from exploring or…
Despite the strong performance of large audio language models (LALMs) in various tasks, exactly how and where they integrate acoustic features with textual context remains unclear. We adapt causal tracing to investigate the internal…
Transformer language models are state of the art in a multitude of NLP tasks. Despite these successes, their opaqueness remains problematic. Recent methods aiming to provide interpretability and explainability to black-box models primarily…
Generative large language models (LLMs) excel in natural language processing tasks, yet their inner workings remain underexplored beyond token-level predictions. This study investigates the degree to which these models decide the content of…
The effectiveness of a language model is influenced by its token representations, which must encode contextual information and handle the same word form having a plurality of meanings (polysemy). Currently, none of the common language…
Past research has examined how well these models grasp code syntax, yet their understanding of code semantics still needs to be explored. We extensively analyze seven code models to investigate how code models represent code syntax and…
Language models (LLMs) offer potential as a source of knowledge for agents that need to acquire new task competencies within a performance environment. We describe efforts toward a novel agent capability that can construct cues (or…
Recent progress in large language models (LLMs) has demonstrated the ability to learn and leverage Internet-scale knowledge through pre-training with autoregressive models. Unfortunately, applying such models to settings with embodied…
Chain-of-Thought prompting has significantly enhanced the reasoning capabilities of large language models, with numerous studies exploring factors influencing its performance. However, the underlying mechanisms remain poorly understood. To…
Language models generate responses by producing a series of tokens in immediate succession: the $(K+1)^{th}$ token is an outcome of manipulating $K$ hidden vectors per layer, one vector per preceding token. What if instead we were to let…