Search for: All records

Transformers have revolutionized machine learning, yet their inner workings remain opaque to many. We present TRANSFORMER EXPLAINER, an interactive visualization tool designed for non-experts to learn about Transformers through the GPT-2 model. Our tool helps users understand complex Transformer concepts by integrating a model overview and smooth transitions across abstraction levels of math operations and model structures. It runs a live GPT-2 model locally in the user’s browser, empowering users to experiment with their own input and observe in real-time how the internal components and parameters of the Transformer work together to predict the next tokens. 125,000 users have used our open-source tool at https://poloclub.github.io/ transformer-explainer/. 

Animal behavior spans many timescales, from short, seconds-scale actions to daily rhythms over many hours to life-long changes during aging. To access longer timescales of behavior, we continuously recorded individualDrosophila melanogasterat 100 frames per second for up to 7 days at a time in featureless arenas on sucrose-agarose media. We use the deep learning framework SLEAP to produce a full-body postural dataset for 47 individuals resulting in nearly 2 billion pose instances. We identify stereotyped behaviors such as grooming, proboscis extension, and locomotion and use the resulting ethograms to explore how the flies’ behavior varies across time of day and days in the experiment. We find distinct daily patterns in all stereotyped behaviors, adding specific information about trends in different grooming modalities, proboscis extension duration, and locomotion speed to what is known about theD. melanogastercircadian cycle. Using our holistic measurements of behavior, we find that the hour after dawn is a unique time point in the flies’ daily pattern of behavior, and that the behavioral composition of this hour tracks well with other indicators of health such as locomotion speed and the fraction of time spend moving vs. resting. The method, data, and analysis presented here give us a new and clearer picture ofD. melanogasterbehavior across timescales, revealing novel features that hint at unexplored underlying biological mechanisms. 

Autistic and neurotypical children do not handle audiovisual speech in the same manner. Current evidence suggests that this difference occurs at the level of cue combination. Here, we test whether differences in autistic and neurotypical audiovisual speech perception can be explained by a neural theory of sensory perception in autism, which proposes that heightened levels of neural excitation can account for sensory differences in autism. Through a linking hypothesis that integrates a standard probabilistic cognitive model of cue integration with representations of neural activity, we derive a model that can simulate audio-visual speech perception at a neural population level. Simulations of an audiovisual lexical identification task demonstrate that heightened levels of neural excitation at the level of cue combination cannot account for the observed differences in autistic and neurotypical children's audiovisual speech perception. 

Photochromic radionuclide-based “claw machines” characterizedviaa combination of isothermal titration calorimetry and spectroscopic analysis unlock a pathway for on demand radionuclide capture and release. 

Regulation of ion channel expression on the plasma membrane is a major determinant of neuronal excitability, and identifying the underlying mechanisms of this expression is critical to our understanding of neurons. Here, we present two orthogonal strategies to label extracellular sites of the ion channel TRPV1 that minimally perturb its function. We use the amber codon suppression technique to introduce a non-canonical amino acid (ncAA) with tetrazine click chemistry, compatible with a trans-cyclooctene coupled fluorescent dye. Additionally, by inserting the circularly permutated HaloTag (cpHaloTag) in an extracellular loop of TRPV1, we can incorporate a fluorescent dye of our choosing. Optimization of ncAA insertion sites was accomplished by screening residue positions between the S1 and S2 transmembrane domains with elevated missense variants in the human population. We identified T468 as a rapid labeling site (∼5 min) based on functional and biochemical assays in HEK293T/17 cells. Through adapting linker lengths and backbone placement of cpHaloTag on the extracellular side of TRPV1, we generated a fully functional channel construct, TRPV1exCellHalo, with intact wild-type gating properties. We used TRPV1exCellHalo in a single molecule experiment to track TRPV1 on the cell surface and validate studies that show decreased mobility of the channel upon activation. The application of these extracellular label TRPV1 (exCellTRPV1) constructs to track surface localization of the channel will shed significant light on the mechanisms regulating its expression and provide a general scheme to introduce similar modifications to other cell surface receptors.