{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/explaining-how-transformers-use-context-to","title":"Explaining How Transformers Use Context to Build Predictions","arxiv_id":"2305.12535","date":"2023-05-21","proceeding":null,"authors":["Javier Ferrando","Gerard I. Gállego","Ioannis Tsiamas","Marta R. Costa-jussà"],"abstract":"Language Generation Models produce words based on the previous context. Although existing methods offer input attributions as explanations for a model's prediction, it is still unclear how prior words affect the model's decision throughout the layers. In this work, we leverage recent advances in explainability of the Transformer and present a procedure to analyze models for language generation. Using contrastive examples, we compare the alignment of our explanations with evidence of the linguistic phenomena, and show that our method consistently aligns better than gradient-based and perturbation-based baselines. Then, we investigate the role of MLPs inside the Transformer and show that they learn features that help the model predict words that are grammatically acceptable. Lastly, we apply our method to Neural Machine Translation models, and demonstrate that they generate human-like source-target alignments for building predictions.","url_abs":"https://arxiv.org/abs/2305.12535v1","url_pdf":"https://arxiv.org/pdf/2305.12535v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"explaining-how-transformers-use-context-to","repo_url":"https://github.com/mt-upc/logit-explanations","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.12535","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.12535"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mt-upc/logit-explanations","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_draft_wrong":1,"unverified":11},"by_repo_kind":{"official":{"samples":12,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"4f8ea1ef4a3b4d38","entry":"model_preds","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"lm_saliency.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/lm_saliency.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4f8ea1ef4a3b4d38"}},{"code_sha256_prefix":"0b0bac3cd5d8d507","entry":"get_module","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/contributions.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/contributions.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0b0bac3cd5d8d507"}},{"code_sha256_prefix":"0cf59c38d6eded2a","entry":"load_model_tokenizer","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/utils_contributions.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/utils_contributions.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0cf59c38d6eded2a"}},{"code_sha256_prefix":"2bdd8bf55d1dacd1","entry":"normalize_attribution_visualization","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/utils_contributions.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/utils_contributions.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2bdd8bf55d1dacd1"}},{"code_sha256_prefix":"49edb238be0a61d0","entry":"normalize_contributions","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/utils_contributions.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/utils_contributions.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"49edb238be0a61d0"}},{"code_sha256_prefix":"d666903d478de356","entry":"read_attribution_scores","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/utils_evaluation.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/utils_evaluation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d666903d478de356"}},{"code_sha256_prefix":"d84edf920238d8c2","entry":"read_blimp_dataset","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"extract_explanations.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/extract_explanations.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d84edf920238d8c2"}},{"code_sha256_prefix":"4a8acba50f4c94f6","entry":"read_sentence_target_foil","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"src/utils_evaluation.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/src/utils_evaluation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4a8acba50f4c94f6"}},{"code_sha256_prefix":"11312f5a29459cc0","entry":"register_embedding_gradient_hooks","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"lm_saliency.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/lm_saliency.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"11312f5a29459cc0"}},{"code_sha256_prefix":"a5d85e23d4429891","entry":"register_embedding_list_hook","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"lm_saliency.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/lm_saliency.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a5d85e23d4429891"}},{"code_sha256_prefix":"640771894bf085d4","entry":"tokens2words","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"extract_explanations.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/extract_explanations.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"640771894bf085d4"}},{"code_sha256_prefix":"ea07643ad028f852","entry":"track2input_tokens","repo":"mt-upc/logit-explanations","repo_kind":"official","path":"extract_explanations.py","file_url":"https://github.com/mt-upc/logit-explanations/blob/HEAD/extract_explanations.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ea07643ad028f852"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}