{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:CVVI6RAOLPMYYTAHB2CQIF7PBL","short_pith_number":"pith:CVVI6RAO","schema_version":"1.0","canonical_sha256":"156a8f440e5bd98c4c070e850417ef0af7a803cf84b5bb118c863cad0a2d236d","source":{"kind":"arxiv","id":"1808.01371","version":2},"attestation_state":"computed","paper":{"title":"Large Scale Language Modeling: Converging on 40GB of Text in Four Hours","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Bryan Catanzaro, Nikolai Yakovenko, Raul Puri, Robert Kirby","submitted_at":"2018-08-03T21:44:29Z","abstract_excerpt":"Recent work has shown how to train Convolutional Neural Networks (CNNs) rapidly on large image datasets, then transfer the knowledge gained from these models to a variety of tasks. Following [Radford 2017], in this work, we demonstrate similar scalability and transfer for Recurrent Neural Networks (RNNs) for Natural Language tasks. By utilizing mixed precision arithmetic and a 32k batch size distributed across 128 NVIDIA Tesla V100 GPUs, we are able to train a character-level 4096-dimension multiplicative LSTM (mLSTM) for unsupervised text reconstruction over 3 epochs of the 40 GB Amazon Revie"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1808.01371","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-08-03T21:44:29Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"6585c45dd0c16b4fc95ab2b7fad93847c86fca3ab916384fb95529f777c7815f","abstract_canon_sha256":"787206eadb43caad5bc3a19ee88e99a69aaa0a2a5575b077e6cfe260485b83a1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:08:20.317905Z","signature_b64":"F/yMQ22RDEqwfdQD9c68Z0okd7N+tngyYKJW5JNhbsXiqKD0JftqtAkr6WYddOPs+MzNeSHvUafQ3/hQmBEyDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"156a8f440e5bd98c4c070e850417ef0af7a803cf84b5bb118c863cad0a2d236d","last_reissued_at":"2026-05-18T00:08:20.317166Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:08:20.317166Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Scale Language Modeling: Converging on 40GB of Text in Four Hours","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Bryan Catanzaro, Nikolai Yakovenko, Raul Puri, Robert Kirby","submitted_at":"2018-08-03T21:44:29Z","abstract_excerpt":"Recent work has shown how to train Convolutional Neural Networks (CNNs) rapidly on large image datasets, then transfer the knowledge gained from these models to a variety of tasks. Following [Radford 2017], in this work, we demonstrate similar scalability and transfer for Recurrent Neural Networks (RNNs) for Natural Language tasks. By utilizing mixed precision arithmetic and a 32k batch size distributed across 128 NVIDIA Tesla V100 GPUs, we are able to train a character-level 4096-dimension multiplicative LSTM (mLSTM) for unsupervised text reconstruction over 3 epochs of the 40 GB Amazon Revie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.01371","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":""},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1808.01371","created_at":"2026-05-18T00:08:20.317291+00:00"},{"alias_kind":"arxiv_version","alias_value":"1808.01371v2","created_at":"2026-05-18T00:08:20.317291+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.01371","created_at":"2026-05-18T00:08:20.317291+00:00"},{"alias_kind":"pith_short_12","alias_value":"CVVI6RAOLPMY","created_at":"2026-05-18T12:32:19.392346+00:00"},{"alias_kind":"pith_short_16","alias_value":"CVVI6RAOLPMYYTAH","created_at":"2026-05-18T12:32:19.392346+00:00"},{"alias_kind":"pith_short_8","alias_value":"CVVI6RAO","created_at":"2026-05-18T12:32:19.392346+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1907.05701","citing_title":"A Highly Efficient Distributed Deep Learning System For Automatic Speech Recognition","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2102.01293","citing_title":"Scaling Laws for Transfer","ref_index":135,"is_internal_anchor":true},{"citing_arxiv_id":"2112.00861","citing_title":"A General Language Assistant as a Laboratory for Alignment","ref_index":177,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05221","citing_title":"Language Models (Mostly) Know What They Know","ref_index":254,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL","json":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL.json","graph_json":"https://pith.science/api/pith-number/CVVI6RAOLPMYYTAHB2CQIF7PBL/graph.json","events_json":"https://pith.science/api/pith-number/CVVI6RAOLPMYYTAHB2CQIF7PBL/events.json","paper":"https://pith.science/paper/CVVI6RAO"},"agent_actions":{"view_html":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL","download_json":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL.json","view_paper":"https://pith.science/paper/CVVI6RAO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1808.01371&json=true","fetch_graph":"https://pith.science/api/pith-number/CVVI6RAOLPMYYTAHB2CQIF7PBL/graph.json","fetch_events":"https://pith.science/api/pith-number/CVVI6RAOLPMYYTAHB2CQIF7PBL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL/action/storage_attestation","attest_author":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL/action/author_attestation","sign_citation":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL/action/citation_signature","submit_replication":"https://pith.science/pith/CVVI6RAOLPMYYTAHB2CQIF7PBL/action/replication_record"}},"created_at":"2026-05-18T00:08:20.317291+00:00","updated_at":"2026-05-18T00:08:20.317291+00:00"}