{"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/beyond-surface-probing-llama-across-scales","title":"Is Bigger and Deeper Always Better? Probing LLaMA Across Scales and Layers","arxiv_id":"2312.04333","date":"2023-12-07","proceeding":null,"authors":["Nuo Chen","Ning Wu","Shining Liang","Ming Gong","Linjun Shou","Dongmei Zhang","Jia Li"],"abstract":"This paper presents an in-depth analysis of Large Language Models (LLMs), focusing on LLaMA, a prominent open-source foundational model in natural language processing. Instead of assessing LLaMA through its generative output, we design multiple-choice tasks to probe its intrinsic understanding in high-order tasks such as reasoning and computation. We examine the model horizontally, comparing different sizes, and vertically, assessing different layers. We unveil several key and uncommon findings based on the designed probing tasks: (1) Horizontally, enlarging model sizes almost could not automatically impart additional knowledge or computational prowess. Instead, it can enhance reasoning abilities, especially in math problem solving, and helps reduce hallucinations, but only beyond certain size thresholds; (2) In vertical analysis, the lower layers of LLaMA lack substantial arithmetic and factual knowledge, showcasing logical thinking, multilingual and recognitive abilities, with top layers housing most computational power and real-world knowledge.","url_abs":"https://arxiv.org/abs/2312.04333v4","url_pdf":"https://arxiv.org/pdf/2312.04333v4.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":"beyond-surface-probing-llama-across-scales","repo_url":"https://github.com/nuochenpku/llama_analysis","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"math","task_name":"Math"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"world-knowledge","task_name":"World Knowledge"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2312.04333","atlas_url":"https://app.syntology.ai/?focus=2312.04333","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2312.04333"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/nuochenpku/llama_analysis","reach":{"status":"ok"}}],"summary":{"ran":2,"unverified":4},"by_repo_kind":{"official":{"samples":6,"ran":2,"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":6,"samples":[{"code_sha256_prefix":"c32369483bc86092","entry":"format_math","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"gsm8k_mc_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/gsm8k_mc_eval.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c32369483bc86092"}},{"code_sha256_prefix":"a1d76be88810d02c","entry":"split_multi_answer","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"factural_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/factural_eval.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a1d76be88810d02c"}},{"code_sha256_prefix":"158b8847881a64b3","entry":"format_best","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"factural_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/factural_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"158b8847881a64b3"}},{"code_sha256_prefix":"2a4c0ecd291e1795","entry":"load_csv","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"factural_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/factural_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2a4c0ecd291e1795"}},{"code_sha256_prefix":"c852bd4d2face0a1","entry":"split_multi_answer","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"tfqa_mc_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/tfqa_mc_eval.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c852bd4d2face0a1"}},{"code_sha256_prefix":"559b132bd75ccec7","entry":"split_multi_answer","repo":"nuochenpku/llama_analysis","repo_kind":"official","path":"gsm8k_mc_eval.py","file_url":"https://github.com/nuochenpku/llama_analysis/blob/HEAD/gsm8k_mc_eval.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"559b132bd75ccec7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}