{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/156794"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/156794","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Interpreting and Editing Memory in Large Transformer Language Models","abstract":"This thesis investigates the mechanisms of factual recall in large language models. We first apply causal interventions to identify neuron activations that are decisive in a model’s factual predictions; surprisingly, we find that factual recall corresponds to a sparse, localizable computation in the MLP weights of the GPT models we study. Harnessing this insight, we then develop methods for efficiently and surgically inserting up to 10,000 new memories into a transformer; these methods perform well in terms of both generalization and specificity. We conclude with some directions for future work.","abstract_html":"This thesis investigates the mechanisms of factual recall in large language models. We first apply causal interventions to identify neuron activations that are decisive in a model’s factual predictions; surprisingly, we find that factual recall corresponds to a sparse, localizable computation in the MLP weights of the GPT models we study. Harnessing this insight, we then develop methods for efficiently and surgically inserting up to 10,000 new memories into a transformer; these methods perform well in terms of both generalization and specificity. We conclude with some directions for future work.","abstract_has_math":false,"creators":["Meng, Kevin"],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Andreas, Jacob D."],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-05","date_published":"2024-05","updated_at":"2026-07-22T22:20:53Z","subjects":[],"languages":[],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"rights_urls":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/156794","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Andreas, Jacob D."]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Meng, Kevin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-09-16T13:49:41Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-09-16T13:49:41Z"]},{"key":"dc:date.issued","label":"Date","values":["2024-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Engineering in Electrical Engineering and Computer Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by-nc-nd/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/156794"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis investigates the mechanisms of factual recall in large language models. We first apply causal interventions to identify neuron activations that are decisive in a model’s factual predictions; surprisingly, we find that factual recall corresponds to a sparse, localizable computation in the MLP weights of the GPT models we study. Harnessing this insight, we then develop methods for efficiently and surgically inserting up to 10,000 new memories into a transformer; these methods perform well in terms of both generalization and specificity. We conclude with some directions for future work."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M.Eng."]},{"key":"dc:title","label":"Title","values":["Interpreting and Editing Memory in Large Transformer Language Models"]}]}],"canonical_facts":{"dc:contributor.advisor":["Andreas, Jacob D."],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Meng, Kevin"],"dc:date.accessioned":["2024-09-16T13:49:41Z"],"dc:date.available":["2024-09-16T13:49:41Z"],"dc:date.issued":["2024-05"],"dc:description.abstract":["This thesis investigates the mechanisms of factual recall in large language models. We first apply causal interventions to identify neuron activations that are decisive in a model’s factual predictions; surprisingly, we find that factual recall corresponds to a sparse, localizable computation in the MLP weights of the GPT models we study. Harnessing this insight, we then develop methods for efficiently and surgically inserting up to 10,000 new memories into a transformer; these methods perform well in terms of both generalization and specificity. We conclude with some directions for future work."],"dc:description.degree":["M.Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/156794"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)","Copyright retained by author(s)"],"dc:rights.uri":["https://creativecommons.org/licenses/by-nc-nd/4.0/"],"dc:title":["Interpreting and Editing Memory in Large Transformer Language Models"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Engineering in Electrical Engineering and Computer Science"]},"updated_at":"2026-07-22T22:20:53Z"}