Javier Ferrando

I am a final year PhD student at the Universitat Politècnica de Catalunya (UPC), supervised by Marta R. Costa-jussà. Currently, I am also a MATS scholar, mentored by Neel Nanda. During my PhD I have been fortunate enough to do research internships at AI at Meta (FAIR), in the Machine Translation team at Apple, and in the Books Science team at Amazon. Before starting my PhD, I completed a MsC Data Science also at UPC, during which I worked part-time at the Barcelona Supercomputing Center.

My research focuses on interpretability and analysis of models. Currently, I am doing mechanistic interpretability research in Large Language Models, with special interest in improving their safety. Apart from research, I have also been a teacher assistant at several courses in the Computer Science and Telecommunications departments at UPC.

Feel free to give anonymous feedback: here.