{"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/towards-tracing-trustworthiness-dynamics","title":"Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models","arxiv_id":"2402.19465","date":"2024-02-29","proceeding":null,"authors":["Chen Qian","Jie Zhang","Wei Yao","Dongrui Liu","Zhenfei Yin","Yu Qiao","Yong liu","Jing Shao"],"abstract":"Ensuring the trustworthiness of large language models (LLMs) is crucial. Most studies concentrate on fully pre-trained LLMs to better understand and improve LLMs' trustworthiness. In this paper, to reveal the untapped potential of pre-training, we pioneer the exploration of LLMs' trustworthiness during this period, focusing on five key dimensions: reliability, privacy, toxicity, fairness, and robustness. To begin with, we apply linear probing to LLMs. The high probing accuracy suggests that \\textit{LLMs in early pre-training can already distinguish concepts in each trustworthiness dimension}. Therefore, to further uncover the hidden possibilities of pre-training, we extract steering vectors from a LLM's pre-training checkpoints to enhance the LLM's trustworthiness. Finally, inspired by~\\citet{choi2023understanding} that mutual information estimation is bounded by linear probing accuracy, we also probe LLMs with mutual information to investigate the dynamics of trustworthiness during pre-training. We are the first to observe a similar two-phase phenomenon: fitting and compression~\\citep{shwartz2017opening}. This research provides an initial exploration of trustworthiness modeling during LLM pre-training, seeking to unveil new insights and spur further developments in the field. We will make our code publicly accessible at \\url{https://github.com/ChnQ/TracingLLM}.","url_abs":"https://arxiv.org/abs/2402.19465v2","url_pdf":"https://arxiv.org/pdf/2402.19465v2.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":"towards-tracing-trustworthiness-dynamics","repo_url":"https://github.com/chnq/tracingllm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":"mutual-information-estimation","task_name":"Mutual Information Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2402.19465","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.19465"}},"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/chnq/tracingllm","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":5,"ran_draft_wrong":1,"unverified":1},"by_repo_kind":{"official":{"samples":7,"ran":6,"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":"b4c2a455a1162ef1","entry":"create_intervene_hook","repo":"chnq/tracingllm","repo_kind":"official","path":"src/eval_trustworthiness.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/eval_trustworthiness.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b4c2a455a1162ef1"}},{"code_sha256_prefix":"2663196c40a3f458","entry":"distmat","repo":"chnq/tracingllm","repo_kind":"official","path":"src/mi_estimators.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/mi_estimators.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2663196c40a3f458"}},{"code_sha256_prefix":"f0744a5e41385253","entry":"get_acts","repo":"chnq/tracingllm","repo_kind":"official","path":"src/generate_activations.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/generate_activations.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f0744a5e41385253"}},{"code_sha256_prefix":"f3fd137e74d1312c","entry":"kernelmat","repo":"chnq/tracingllm","repo_kind":"official","path":"src/mi_estimators.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/mi_estimators.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f3fd137e74d1312c"}},{"code_sha256_prefix":"f0cc63177ce8b7a1","entry":"load_statements","repo":"chnq/tracingllm","repo_kind":"official","path":"src/generate_activations.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/generate_activations.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f0cc63177ce8b7a1"}},{"code_sha256_prefix":"8ede5c4129a729ae","entry":"sigma_estimation","repo":"chnq/tracingllm","repo_kind":"official","path":"src/mi_estimators.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/mi_estimators.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8ede5c4129a729ae"}},{"code_sha256_prefix":"e1c7e2781771e313","entry":"load_model","repo":"chnq/tracingllm","repo_kind":"official","path":"src/generate_activations.py","file_url":"https://github.com/chnq/tracingllm/blob/HEAD/src/generate_activations.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":"e1c7e2781771e313"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}