{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:7TN4LDUIK53ANGHLCDV353HZCU","short_pith_number":"pith:7TN4LDUI","canonical_record":{"source":{"id":"2402.07871","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:33:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"93c97d780071f155cc084a5326ecff132cab7746b1aea2666a386872819ebbe8","abstract_canon_sha256":"e5c9cc1192d45b36808cc718fd14d24e24ded363db4246d3ec024a2beadc1432"},"schema_version":"1.0"},"canonical_sha256":"fcdbc58e8857760698eb10ebbeecf9153af746bcd916fdc7e1b4929ce4129182","source":{"kind":"arxiv","id":"2402.07871","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.07871","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"arxiv_version","alias_value":"2402.07871v1","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.07871","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_12","alias_value":"7TN4LDUIK53A","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_16","alias_value":"7TN4LDUIK53ANGHL","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_8","alias_value":"7TN4LDUI","created_at":"2026-07-05T07:44:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:7TN4LDUIK53ANGHLCDV353HZCU","target":"record","payload":{"canonical_record":{"source":{"id":"2402.07871","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:33:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"93c97d780071f155cc084a5326ecff132cab7746b1aea2666a386872819ebbe8","abstract_canon_sha256":"e5c9cc1192d45b36808cc718fd14d24e24ded363db4246d3ec024a2beadc1432"},"schema_version":"1.0"},"canonical_sha256":"fcdbc58e8857760698eb10ebbeecf9153af746bcd916fdc7e1b4929ce4129182","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:44:13.520447Z","signature_b64":"rVBDJry3g4FCQHHoT+3/Nd9FiAFpgpTkFCz7BUvBkgxxIrRRUIFRji+o+YvaV9YoeGlUl1VJyoM9AFWscUVbBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcdbc58e8857760698eb10ebbeecf9153af746bcd916fdc7e1b4929ce4129182","last_reissued_at":"2026-07-05T07:44:13.519928Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:44:13.519928Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.07871","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:44:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"23n54B2Oyfk4j7GXcH7VyN6dRn14qm8Q+ed3u4RiyGmelMGrFjBG5+aDxscrO/Url8kJXzYPyz+5L5gglAMACw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:37:46.752262Z"},"content_sha256":"c6cc456051c019be246eff753a123b82ec059b4d2654524c8ed5f6e668e6fbd7","schema_version":"1.0","event_id":"sha256:c6cc456051c019be246eff753a123b82ec059b4d2654524c8ed5f6e668e6fbd7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:7TN4LDUIK53ANGHLCDV353HZCU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Scaling Laws for Fine-Grained Mixture of Experts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jakub Krajewski, Jan Ludziejewski, Kamil Adamczewski, Kamil Ciebiera, Krystian Kr\\'ol, Maciej Pi\\'oro, Marek Cygan, Micha{\\l} Krutul, Piotr Sankowski, Sebastian Jaszczur, Szymon Antoniak, Tomasz Odrzyg\\'o\\'zd\\'z","submitted_at":"2024-02-12T18:33:47Z","abstract_excerpt":"Mixture of Experts (MoE) models have emerged as a primary solution for reducing the computational cost of Large Language Models. In this work, we analyze their scaling properties, incorporating an expanded range of variables. Specifically, we introduce a new hyperparameter, granularity, whose adjustment enables precise control over the size of the experts. Building on this, we establish scaling laws for fine-grained MoE, taking into account the number of training tokens, model size, and granularity. Leveraging these laws, we derive the optimal training configuration for a given computational b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.07871","kind":"arxiv","version":1},"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/2402.07871/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":"2026-07-05T07:44:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jEKqRyAmZuZ6rIuXvThxEawmlKoUkJS8EuRdQR0qDRppKKSwZ/luE6vIn8+eaAwGYAvEfiqChKVa1lUVitipDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T08:37:46.752648Z"},"content_sha256":"f1a07cf925017700b6ebcb5e40df2085d4c65086c78b8e99cf7f751ae4f96468","schema_version":"1.0","event_id":"sha256:f1a07cf925017700b6ebcb5e40df2085d4c65086c78b8e99cf7f751ae4f96468"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7TN4LDUIK53ANGHLCDV353HZCU/bundle.json","state_url":"https://pith.science/pith/7TN4LDUIK53ANGHLCDV353HZCU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7TN4LDUIK53ANGHLCDV353HZCU/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":"2026-07-07T08:37:46Z","links":{"resolver":"https://pith.science/pith/7TN4LDUIK53ANGHLCDV353HZCU","bundle":"https://pith.science/pith/7TN4LDUIK53ANGHLCDV353HZCU/bundle.json","state":"https://pith.science/pith/7TN4LDUIK53ANGHLCDV353HZCU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7TN4LDUIK53ANGHLCDV353HZCU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7TN4LDUIK53ANGHLCDV353HZCU","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":"e5c9cc1192d45b36808cc718fd14d24e24ded363db4246d3ec024a2beadc1432","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:33:47Z","title_canon_sha256":"93c97d780071f155cc084a5326ecff132cab7746b1aea2666a386872819ebbe8"},"schema_version":"1.0","source":{"id":"2402.07871","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.07871","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"arxiv_version","alias_value":"2402.07871v1","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.07871","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_12","alias_value":"7TN4LDUIK53A","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_16","alias_value":"7TN4LDUIK53ANGHL","created_at":"2026-07-05T07:44:13Z"},{"alias_kind":"pith_short_8","alias_value":"7TN4LDUI","created_at":"2026-07-05T07:44:13Z"}],"graph_snapshots":[{"event_id":"sha256:f1a07cf925017700b6ebcb5e40df2085d4c65086c78b8e99cf7f751ae4f96468","target":"graph","created_at":"2026-07-05T07:44:13Z","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/2402.07871/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Mixture of Experts (MoE) models have emerged as a primary solution for reducing the computational cost of Large Language Models. In this work, we analyze their scaling properties, incorporating an expanded range of variables. Specifically, we introduce a new hyperparameter, granularity, whose adjustment enables precise control over the size of the experts. Building on this, we establish scaling laws for fine-grained MoE, taking into account the number of training tokens, model size, and granularity. Leveraging these laws, we derive the optimal training configuration for a given computational b","authors_text":"Jakub Krajewski, Jan Ludziejewski, Kamil Adamczewski, Kamil Ciebiera, Krystian Kr\\'ol, Maciej Pi\\'oro, Marek Cygan, Micha{\\l} Krutul, Piotr Sankowski, Sebastian Jaszczur, Szymon Antoniak, Tomasz Odrzyg\\'o\\'zd\\'z","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:33:47Z","title":"Scaling Laws for Fine-Grained Mixture of Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.07871","kind":"arxiv","version":1},"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:c6cc456051c019be246eff753a123b82ec059b4d2654524c8ed5f6e668e6fbd7","target":"record","created_at":"2026-07-05T07:44:13Z","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":"e5c9cc1192d45b36808cc718fd14d24e24ded363db4246d3ec024a2beadc1432","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-12T18:33:47Z","title_canon_sha256":"93c97d780071f155cc084a5326ecff132cab7746b1aea2666a386872819ebbe8"},"schema_version":"1.0","source":{"id":"2402.07871","kind":"arxiv","version":1}},"canonical_sha256":"fcdbc58e8857760698eb10ebbeecf9153af746bcd916fdc7e1b4929ce4129182","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fcdbc58e8857760698eb10ebbeecf9153af746bcd916fdc7e1b4929ce4129182","first_computed_at":"2026-07-05T07:44:13.519928Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:44:13.519928Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rVBDJry3g4FCQHHoT+3/Nd9FiAFpgpTkFCz7BUvBkgxxIrRRUIFRji+o+YvaV9YoeGlUl1VJyoM9AFWscUVbBA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:44:13.520447Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.07871","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c6cc456051c019be246eff753a123b82ec059b4d2654524c8ed5f6e668e6fbd7","sha256:f1a07cf925017700b6ebcb5e40df2085d4c65086c78b8e99cf7f751ae4f96468"],"state_sha256":"da73300da8b6371a0b943efd92fec4f4be4921772e172437f1bb4dc460730a00"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AG6MVV7deYzvVWJ9wfGHp8DdsyprNYtI2FhNn1GrVWaHGYcv9p+JmJv7/+2mGWfnA6dKqYWsVDEkXw+Dl6FsBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T08:37:46.754618Z","bundle_sha256":"9b63d4d3e75dc305b8e9e454ebf19b4c60908b8582cd9d3f5162ff045e817b75"}}