{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:OVM55B4OOULKBSVVQ6PUHVEWGS","short_pith_number":"pith:OVM55B4O","canonical_record":{"source":{"id":"2201.11990","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-28T08:59:57Z","cross_cats_sorted":[],"title_canon_sha256":"a8821981d6c4bdeb44cb3a21db1640b9d268b26ece0652e89e6f79292718e7ea","abstract_canon_sha256":"86a01b263314585c601bafccee12ae6cc73f8e9eb8f3276c5c82db667542b812"},"schema_version":"1.0"},"canonical_sha256":"7559de878e7516a0cab5879f43d4963484951c89e0c5e69edc2dc02b28beb2f8","source":{"kind":"arxiv","id":"2201.11990","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.11990","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"arxiv_version","alias_value":"2201.11990v3","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.11990","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"pith_short_12","alias_value":"OVM55B4OOULK","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"OVM55B4OOULKBSVV","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"OVM55B4O","created_at":"2026-05-18T12:33:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:OVM55B4OOULKBSVVQ6PUHVEWGS","target":"record","payload":{"canonical_record":{"source":{"id":"2201.11990","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-28T08:59:57Z","cross_cats_sorted":[],"title_canon_sha256":"a8821981d6c4bdeb44cb3a21db1640b9d268b26ece0652e89e6f79292718e7ea","abstract_canon_sha256":"86a01b263314585c601bafccee12ae6cc73f8e9eb8f3276c5c82db667542b812"},"schema_version":"1.0"},"canonical_sha256":"7559de878e7516a0cab5879f43d4963484951c89e0c5e69edc2dc02b28beb2f8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:22.070115Z","signature_b64":"a5b4qc+Rck0Q+jV2V0gXynHFDdKqpw8JI/u1qJ6KnGuxytQfbB8GOlF9DHa5jhHaanbz4xLGp8Vj76/LJYFFBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7559de878e7516a0cab5879f43d4963484951c89e0c5e69edc2dc02b28beb2f8","last_reissued_at":"2026-05-17T23:39:22.069555Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:22.069555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2201.11990","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MpJkIeXZfIRcoTIQjGl2/bOGMfjJknNpAF4BvtFocq5r+u2C2SeQzPpCN6dQgNFbDL1S/AUwWmJ+jwcWipzUAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T00:37:49.573033Z"},"content_sha256":"3a6d2e6df93da19781467c6e7d3aa38c68515f6d453c68e2cd1f5dba34127649","schema_version":"1.0","event_id":"sha256:3a6d2e6df93da19781467c6e7d3aa38c68515f6d453c68e2cd1f5dba34127649"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:OVM55B4OOULKBSVVQ6PUHVEWGS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Brandon Norick, Bryan Catanzaro, Elton Zhang, George Zerveas, Jared Casper, Julie Bernauer, Michael Houston, Mohammad Shoeybi, Mostofa Patwary, Patrick LeGresley, Rewon Child, Reza Yazdani Aminabadi, Samyam Rajbhandari, Saurabh Tiwary, Shaden Smith, Shrimai Prabhumoye, Vijay Korthikanti, Xia Song, Yuxiong He, Zhun Liu","submitted_at":"2022-01-28T08:59:57Z","abstract_excerpt":"Pretrained general-purpose language models can achieve state-of-the-art accuracies in various natural language processing domains by adapting to downstream tasks via zero-shot, few-shot and fine-tuning techniques. Because of their success, the size of these models has increased rapidly, requiring high-performance hardware, software, and algorithmic techniques to enable training such large models. As the result of a joint effort between Microsoft and NVIDIA, we present details on the training of the largest monolithic transformer based language model, Megatron-Turing NLG 530B (MT-NLG), with 530"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.11990","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-17T23:39:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Id7L/XO2oVFUO8FlPasrZzHLgeZ90fe49LsmZkDt5029tjrDxcQP/3k53e2nds++jNc1yJeA49p2LJjiDMxjCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T00:37:49.573415Z"},"content_sha256":"8f864bdda1983a7b6974bc00320828610202d20b77f7c1f854bcacd1e84a1e01","schema_version":"1.0","event_id":"sha256:8f864bdda1983a7b6974bc00320828610202d20b77f7c1f854bcacd1e84a1e01"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/bundle.json","state_url":"https://pith.science/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/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":"2026-05-26T00:37:49Z","links":{"resolver":"https://pith.science/pith/OVM55B4OOULKBSVVQ6PUHVEWGS","bundle":"https://pith.science/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/bundle.json","state":"https://pith.science/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OVM55B4OOULKBSVVQ6PUHVEWGS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:OVM55B4OOULKBSVVQ6PUHVEWGS","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":"86a01b263314585c601bafccee12ae6cc73f8e9eb8f3276c5c82db667542b812","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-28T08:59:57Z","title_canon_sha256":"a8821981d6c4bdeb44cb3a21db1640b9d268b26ece0652e89e6f79292718e7ea"},"schema_version":"1.0","source":{"id":"2201.11990","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.11990","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"arxiv_version","alias_value":"2201.11990v3","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.11990","created_at":"2026-05-17T23:39:22Z"},{"alias_kind":"pith_short_12","alias_value":"OVM55B4OOULK","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"OVM55B4OOULKBSVV","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"OVM55B4O","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:8f864bdda1983a7b6974bc00320828610202d20b77f7c1f854bcacd1e84a1e01","target":"graph","created_at":"2026-05-17T23:39:22Z","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"},"paper":{"abstract_excerpt":"Pretrained general-purpose language models can achieve state-of-the-art accuracies in various natural language processing domains by adapting to downstream tasks via zero-shot, few-shot and fine-tuning techniques. Because of their success, the size of these models has increased rapidly, requiring high-performance hardware, software, and algorithmic techniques to enable training such large models. As the result of a joint effort between Microsoft and NVIDIA, we present details on the training of the largest monolithic transformer based language model, Megatron-Turing NLG 530B (MT-NLG), with 530","authors_text":"Brandon Norick, Bryan Catanzaro, Elton Zhang, George Zerveas, Jared Casper, Julie Bernauer, Michael Houston, Mohammad Shoeybi, Mostofa Patwary, Patrick LeGresley, Rewon Child, Reza Yazdani Aminabadi, Samyam Rajbhandari, Saurabh Tiwary, Shaden Smith, Shrimai Prabhumoye, Vijay Korthikanti, Xia Song, Yuxiong He, Zhun Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-28T08:59:57Z","title":"Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.11990","kind":"arxiv","version":3},"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":"sha256:3a6d2e6df93da19781467c6e7d3aa38c68515f6d453c68e2cd1f5dba34127649","target":"record","created_at":"2026-05-17T23:39:22Z","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":"86a01b263314585c601bafccee12ae6cc73f8e9eb8f3276c5c82db667542b812","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-28T08:59:57Z","title_canon_sha256":"a8821981d6c4bdeb44cb3a21db1640b9d268b26ece0652e89e6f79292718e7ea"},"schema_version":"1.0","source":{"id":"2201.11990","kind":"arxiv","version":3}},"canonical_sha256":"7559de878e7516a0cab5879f43d4963484951c89e0c5e69edc2dc02b28beb2f8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7559de878e7516a0cab5879f43d4963484951c89e0c5e69edc2dc02b28beb2f8","first_computed_at":"2026-05-17T23:39:22.069555Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:22.069555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"a5b4qc+Rck0Q+jV2V0gXynHFDdKqpw8JI/u1qJ6KnGuxytQfbB8GOlF9DHa5jhHaanbz4xLGp8Vj76/LJYFFBQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:22.070115Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.11990","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a6d2e6df93da19781467c6e7d3aa38c68515f6d453c68e2cd1f5dba34127649","sha256:8f864bdda1983a7b6974bc00320828610202d20b77f7c1f854bcacd1e84a1e01"],"state_sha256":"7d4e6cf7e57a38d00a346804798ef200f0dbdf03b46ce5e235f8025d855e95dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/z9SRjSADnRDuzjIJCSOObN1I31rwKsxHaeyfK35FEECsgZsPdvzfMRrqWZPW4TPUdypwEwZ4yLY/HXSeVQ9Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T00:37:49.575637Z","bundle_sha256":"5b64239aaaa831a491832c0712ec5cd603bd7fbdc34f1fe9da7b2b80a6149e74"}}