{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:ZFAOUUES2M4IRD4GAKKI3BDCYQ","short_pith_number":"pith:ZFAOUUES","canonical_record":{"source":{"id":"2412.13795","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-18T12:39:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"acbd45157995c6726c310b0816e28cbc2b76ba392de060f11d94fd175e6009ca","abstract_canon_sha256":"c8e1842583ff6260b22a3a9aa809cb2874ab02e5b02665bd485af6364d1e6cb1"},"schema_version":"1.0"},"canonical_sha256":"c940ea5092d338888f8602948d8462c40d30c5c3a72c490d6ea06a6205869ec8","source":{"kind":"arxiv","id":"2412.13795","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.13795","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"arxiv_version","alias_value":"2412.13795v2","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13795","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_12","alias_value":"ZFAOUUES2M4I","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_16","alias_value":"ZFAOUUES2M4IRD4G","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_8","alias_value":"ZFAOUUES","created_at":"2026-07-05T11:47:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:ZFAOUUES2M4IRD4GAKKI3BDCYQ","target":"record","payload":{"canonical_record":{"source":{"id":"2412.13795","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-18T12:39:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"acbd45157995c6726c310b0816e28cbc2b76ba392de060f11d94fd175e6009ca","abstract_canon_sha256":"c8e1842583ff6260b22a3a9aa809cb2874ab02e5b02665bd485af6364d1e6cb1"},"schema_version":"1.0"},"canonical_sha256":"c940ea5092d338888f8602948d8462c40d30c5c3a72c490d6ea06a6205869ec8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:47:59.091493Z","signature_b64":"csPin+QtoYK+c6WesNU65ZFCA4gBgIBrVHyXFn0z/tLjZLAU6LyNQsTjNhK1fOaxao6ZQmvrL5AtRoUN4cljCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c940ea5092d338888f8602948d8462c40d30c5c3a72c490d6ea06a6205869ec8","last_reissued_at":"2026-07-05T11:47:59.091050Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:47:59.091050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.13795","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:47:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m0zg2VaxgCInTDna+phPvkswnuPrD4XwaNpqr8iLWlfyt35lgUtKFjFaHEXtthnmie0Vx7DQUOLa03V/KQ7LDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T00:18:24.295296Z"},"content_sha256":"cb5f1c74408dbba1b908fe8c898c8403336405d765d29d98674a1a5004488c41","schema_version":"1.0","event_id":"sha256:cb5f1c74408dbba1b908fe8c898c8403336405d765d29d98674a1a5004488c41"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:ZFAOUUES2M4IRD4GAKKI3BDCYQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Lu Yin, Pengxiang Li, Shiwei Liu","submitted_at":"2024-12-18T12:39:53Z","abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable success, yet recent findings reveal that their deeper layers often contribute minimally and can be pruned without affecting overall performance. While some view this as an opportunity for model compression, we identify it as a training shortfall rooted in the widespread use of Pre-Layer Normalization (Pre-LN). We demonstrate that Pre-LN, commonly employed in models like GPT and LLaMA, leads to diminished gradient norms in its deeper layers, reducing their effectiveness. In contrast, Post-Layer Normalization (Post-LN) preserves larger gradie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13795","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2412.13795/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:47:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UtegdsqmKEFqEmIkygdpcCbejyoC4gmXRHvQrZxaARll1ETTHLM24LAP89QT17WST3UDsEBykE6vbyAen0dQDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T00:18:24.296162Z"},"content_sha256":"1a9340fe1c30f60e57221b16e623fbef2e4a15e4e73c10eed5fc150a81f679dd","schema_version":"1.0","event_id":"sha256:1a9340fe1c30f60e57221b16e623fbef2e4a15e4e73c10eed5fc150a81f679dd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/bundle.json","state_url":"https://pith.science/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T00:18:24Z","links":{"resolver":"https://pith.science/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ","bundle":"https://pith.science/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/bundle.json","state":"https://pith.science/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZFAOUUES2M4IRD4GAKKI3BDCYQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZFAOUUES2M4IRD4GAKKI3BDCYQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c8e1842583ff6260b22a3a9aa809cb2874ab02e5b02665bd485af6364d1e6cb1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-18T12:39:53Z","title_canon_sha256":"acbd45157995c6726c310b0816e28cbc2b76ba392de060f11d94fd175e6009ca"},"schema_version":"1.0","source":{"id":"2412.13795","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.13795","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"arxiv_version","alias_value":"2412.13795v2","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.13795","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_12","alias_value":"ZFAOUUES2M4I","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_16","alias_value":"ZFAOUUES2M4IRD4G","created_at":"2026-07-05T11:47:59Z"},{"alias_kind":"pith_short_8","alias_value":"ZFAOUUES","created_at":"2026-07-05T11:47:59Z"}],"graph_snapshots":[{"event_id":"sha256:1a9340fe1c30f60e57221b16e623fbef2e4a15e4e73c10eed5fc150a81f679dd","target":"graph","created_at":"2026-07-05T11:47:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.13795/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable success, yet recent findings reveal that their deeper layers often contribute minimally and can be pruned without affecting overall performance. While some view this as an opportunity for model compression, we identify it as a training shortfall rooted in the widespread use of Pre-Layer Normalization (Pre-LN). We demonstrate that Pre-LN, commonly employed in models like GPT and LLaMA, leads to diminished gradient norms in its deeper layers, reducing their effectiveness. In contrast, Post-Layer Normalization (Post-LN) preserves larger gradie","authors_text":"Lu Yin, Pengxiang Li, Shiwei Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-18T12:39:53Z","title":"Mix-LN: Unleashing the Power of Deeper Layers by Combining Pre-LN and Post-LN"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.13795","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cb5f1c74408dbba1b908fe8c898c8403336405d765d29d98674a1a5004488c41","target":"record","created_at":"2026-07-05T11:47:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c8e1842583ff6260b22a3a9aa809cb2874ab02e5b02665bd485af6364d1e6cb1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-18T12:39:53Z","title_canon_sha256":"acbd45157995c6726c310b0816e28cbc2b76ba392de060f11d94fd175e6009ca"},"schema_version":"1.0","source":{"id":"2412.13795","kind":"arxiv","version":2}},"canonical_sha256":"c940ea5092d338888f8602948d8462c40d30c5c3a72c490d6ea06a6205869ec8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c940ea5092d338888f8602948d8462c40d30c5c3a72c490d6ea06a6205869ec8","first_computed_at":"2026-07-05T11:47:59.091050Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:47:59.091050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"csPin+QtoYK+c6WesNU65ZFCA4gBgIBrVHyXFn0z/tLjZLAU6LyNQsTjNhK1fOaxao6ZQmvrL5AtRoUN4cljCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:47:59.091493Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.13795","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cb5f1c74408dbba1b908fe8c898c8403336405d765d29d98674a1a5004488c41","sha256:1a9340fe1c30f60e57221b16e623fbef2e4a15e4e73c10eed5fc150a81f679dd"],"state_sha256":"263ec1840eb598369751912429bef4987f602762292ed485a7030bd48cb4df10"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hzwqR9x/9LhpNUMwbP0a76I4ry38nQqhm2G0VuS28A9ruu5echOGzshb2+wvW39SQwTbC065nRh5txFtSNmCCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T00:18:24.303342Z","bundle_sha256":"c4d811fbc0c9e9ff181f29bfcf218ca183829f828c02f20fb422557e4293f937"}}