{ "schema": "ddc-token-boundary-ledger/1", "runner_id": "THINK_B_2EP_SIX_ARMS_RUNNER_V3", "bindings": { "config_sha256": "5c380a60cd9fd6e53e02d10c3ab9e2d9ec07b0884ddf3d5878617346d8861fc8", "runner_code_sha256": "fd6d1c6403810acc4460e2ce0fb5f4418881387b11f4bd21f7cd21067faa3a0a", "config_schema_sha256": "d6d13eeee42bf04cf34fc68b2fa7a61b316c4543b77d7d779815b8eddf10c099", "python_binary_sha256": "5700a6a7804feb44da6c963ebd1323982478b7ca57478e4e81981c47818453bd", "environment_lock_sha256": "627d4fe088b8f64841b0520ae5bd9dc5c81e9dcd5969b4bd45c9a9910b2f3c01", "base_composite_sha256": "526963d72561cb6244f60f751312673d2671111d214636ae571de0d62c762760", "base_lock_sha256": "6791648e4030008388f2e9241b23771cb07eb0416fbd7bc3c6777c7c36bda3be", "frozen_rows_sha256": "3a912697f0aa63896d98d5f3a85d511bcfd192937b4f6eec5c3a9daaf4bf8112", "ordered_ids_sha256": "daf31ae3753b428565223c2bc8e3a4115c212977e5e48143bbbe25f32427c818", "update_plan_sha256": "af653aabce045b1194123d1342cc9f29cbd9cd353c9e1fac25d7d5d3998697c6", "data_receipt_sha256": "e66346fb413557929ab53648bc656fb3102557cf3460abb49c7b1c6e7281420d", "decontamination_receipt_sha256": "66c360871b06208ed20e473e57ad0f09c79e4dac1b3124a9461a5b729b487153", "render_receipt_sha256": "e66346fb413557929ab53648bc656fb3102557cf3460abb49c7b1c6e7281420d", "token_bake_sha256": "74234e98afe7498fb5daf1f36ac2d78acc339464f950703b8c019892f982b90b", "continuation_sha256": "74234e98afe7498fb5daf1f36ac2d78acc339464f950703b8c019892f982b90b" }, "completed_optimizer_updates": 175, "next_update_index": 175, "next_ordered_position": 11750, "cumulative_planned_active_tokens": 11441234, "cumulative_realized_active_tokens": 11441234, "optimizer_step_count": 175, "scheduler_step_count": 175, "token_bake": { "enabled": false, "positive_lr_active_tokens": null, "frozen": null, "freeze_digest": null }, "updates": [ { "update_id": 0, "row_count": 63, "row_ids_sha256": "1d0a352c8f286e5f67ced4582b921ad0662ba03de9372347ffde4fdeb1eee079", "input_ids_vector_sha256": "ab189a3e0df1bd3c18963c9a615929dd1f5ddfa4e093b56dc70f3fc9399c3d73", "labels_vector_sha256": "f39f068aa04696c50838a5f47a939a59794ebd11478a5abb87c62e4ebb4b8ea6", "planned_active_tokens": 66184, "realized_active_tokens": 66184, "cumulative_start": 0, "cumulative_end": 66184, "loss_sum": 57623.040100097656, "token_normalized_loss": 0.8706491009926517, "grad_norm_before_clip": 2.130573272705078, "learning_rates_at_step": { "lora": 9.641151740634023e-06, "row_tokens": 9.641151740634023e-06 }, "learning_rates_after_step": { "lora": 1.9201601225975053e-05, "row_tokens": 1.9201601225975053e-05 } }, { "update_id": 1, "row_count": 65, "row_ids_sha256": "83c68fd7e8908efd045166ef0976fc0c377e2d5da28a89c32412a636822abde9", "input_ids_vector_sha256": "44f4cea811dda5f7af8a6adb56f69ab746aed0a35828224aaaed0238e94e6999", "labels_vector_sha256": "52034290fa37c369bc6b1b52026ef3021e2024c0761bf52ebf315d27f35a7208", "planned_active_tokens": 65630, "realized_active_tokens": 65630, "cumulative_start": 66184, "cumulative_end": 131814, "loss_sum": 52948.182525634766, "token_normalized_loss": 0.8067679799731032, "grad_norm_before_clip": 2.1044652462005615, "learning_rates_at_step": { "lora": 1.9201601225975053e-05, "row_tokens": 1.9201601225975053e-05 }, "learning_rates_after_step": { "lora": 2.881216185900704e-05, "row_tokens": 2.881216185900704e-05 } }, { "update_id": 2, "row_count": 62, "row_ids_sha256": "ff28413521384485ebb14cf0836adae175be917e98093e7aae3468babf4f9b4c", "input_ids_vector_sha256": "9bcfea8df80d859cf24a27db93b864eefc7c38cfe7e563155c89fb4c5dbea9d3", "labels_vector_sha256": "0f108d60c74ac068aa33199b4e1fb4edef6cfaadb421bd946eab6bbf626d4479", "planned_active_tokens": 65974, "realized_active_tokens": 65974, "cumulative_start": 131814, "cumulative_end": 197788, "loss_sum": 47840.95251464844, "token_normalized_loss": 0.7251485814813174, "grad_norm_before_clip": 1.944539189338684, "learning_rates_at_step": { "lora": 2.881216185900704e-05, "row_tokens": 2.881216185900704e-05 }, "learning_rates_after_step": { "lora": 3.8331094841173885e-05, "row_tokens": 3.8331094841173885e-05 } }, { "update_id": 3, "row_count": 62, "row_ids_sha256": "8e5c05be13c6ede39f2bdac3c554f4d5588cada49857884abbc0cae1611e2678", "input_ids_vector_sha256": "86abbebc50d6ad32f36df88594f2260c4f3d55eb7d4cd2d4459756e85a2ccb87", "labels_vector_sha256": "5edf39de0efd53f0b1fc74e6e919412595652accdba113e3398230fde66a750f", "planned_active_tokens": 65345, "realized_active_tokens": 65345, "cumulative_start": 197788, "cumulative_end": 263133, "loss_sum": 40284.531829833984, "token_normalized_loss": 0.6164898895069858, "grad_norm_before_clip": 0.9071006774902344, "learning_rates_at_step": { "lora": 3.8331094841173885e-05, "row_tokens": 3.8331094841173885e-05 }, "learning_rates_after_step": { "lora": 4.775373867036479e-05, "row_tokens": 4.775373867036479e-05 } }, { "update_id": 4, "row_count": 66, "row_ids_sha256": "5931c657c5d4db1a5c11681b44864abaad5c37bd2fcba56bafe073839373a919", "input_ids_vector_sha256": "3b7d1dcf2777f07ad9027a5669525cdf6985074c2dc60cf16678ba1d9fe460a8", "labels_vector_sha256": "a01f9873431217db0da4f767c185c4204378eff34edcb98538ba645e9784ddce", "planned_active_tokens": 64684, "realized_active_tokens": 64684, "cumulative_start": 263133, "cumulative_end": 327817, "loss_sum": 35373.712463378906, "token_normalized_loss": 0.5468695885130621, "grad_norm_before_clip": 0.27501359581947327, "learning_rates_at_step": { "lora": 4.775373867036479e-05, "row_tokens": 4.775373867036479e-05 }, "learning_rates_after_step": { "lora": 5.738746114200975e-05, "row_tokens": 5.738746114200975e-05 } }, { "update_id": 5, "row_count": 68, "row_ids_sha256": "5ef2a9ca29eae0f84d11ce72d9af70025b43aa247022365b91c3a827ec0982d7", "input_ids_vector_sha256": "942443854f1b56275de32d55b718f3d92e9b24f4b5766a401b5805178e259c34", "labels_vector_sha256": "6156b0438aa77624c4413aa571f04dbe1f5bb97551bc13f472622cef2be5cb74", "planned_active_tokens": 66133, "realized_active_tokens": 66133, "cumulative_start": 327817, "cumulative_end": 393950, "loss_sum": 35093.10008239746, "token_normalized_loss": 0.5306443089289381, "grad_norm_before_clip": 0.29282692074775696, "learning_rates_at_step": { "lora": 5.738746114200975e-05, "row_tokens": 5.738746114200975e-05 }, "learning_rates_after_step": { "lora": 6.688192123809496e-05, "row_tokens": 6.688192123809496e-05 } }, { "update_id": 6, "row_count": 61, "row_ids_sha256": "3aac1b11101238d2d7b5ed5498fa2ec8ace70c4ac425966ba7999923b81d2477", "input_ids_vector_sha256": "91f123d0c23fac688a29e15025050847b94b648b3e11edddc2b2049e75a375f9", "labels_vector_sha256": "5fbbf0758ce509f5d77a8c30ed919704bc5a30c114e9fec7af5483153f07d576", "planned_active_tokens": 65177, "realized_active_tokens": 65177, "cumulative_start": 393950, "cumulative_end": 459127, "loss_sum": 32181.028198242188, "token_normalized_loss": 0.4937482271083693, "grad_norm_before_clip": 0.23442205786705017, "learning_rates_at_step": { "lora": 6.688192123809496e-05, "row_tokens": 6.688192123809496e-05 }, "learning_rates_after_step": { "lora": 7.647252481521515e-05, "row_tokens": 7.647252481521515e-05 } }, { "update_id": 7, "row_count": 62, "row_ids_sha256": "6cfef37f8a4ca3eaf07fb819d66549b7faab64ce3b67da8070324777f96319ad", "input_ids_vector_sha256": "1488a042caaaf8db33bd55c19a97b563e68b9cbf75c224f2bffb8e5c0730f7ed", "labels_vector_sha256": "e5fc98983646f30dc13fc329f2ce5613b173bab0126f9e9f63f205ff48144403", "planned_active_tokens": 65837, "realized_active_tokens": 65837, "cumulative_start": 459127, "cumulative_end": 524964, "loss_sum": 31847.774826049805, "token_normalized_loss": 0.4837367259451343, "grad_norm_before_clip": 0.2205609679222107, "learning_rates_at_step": { "lora": 7.647252481521515e-05, "row_tokens": 7.647252481521515e-05 }, "learning_rates_after_step": { "lora": 8.608031768136884e-05, "row_tokens": 8.608031768136884e-05 } }, { "update_id": 8, "row_count": 63, "row_ids_sha256": "f7b6a70a721abdb86f0dcae3da25c26aea2e5e022074aba8acb16db311c5738f", "input_ids_vector_sha256": "dbafe7b3e896a5532fcb5614bfb7fb848b168b742b6cc949a2c17efd83948b1b", "labels_vector_sha256": "899499d6ecde6fdccc12c7238d034538752bbd979c97899fb385e1e1e6bcfa7b", "planned_active_tokens": 65955, "realized_active_tokens": 65955, "cumulative_start": 524964, "cumulative_end": 590919, "loss_sum": 28308.705001831055, "token_normalized_loss": 0.4292124175851877, "grad_norm_before_clip": 0.20734237134456635, "learning_rates_at_step": { "lora": 8.608031768136884e-05, "row_tokens": 8.608031768136884e-05 }, "learning_rates_after_step": { "lora": 9.55546750496013e-05, "row_tokens": 9.55546750496013e-05 } }, { "update_id": 9, "row_count": 64, "row_ids_sha256": "9aa08628b8683b149a2987bec38de53ceb0c6fc105e2e5b17b240f94196fbb93", "input_ids_vector_sha256": "23dc31b23af8c73bcd26c3c5437279b310b11086fed81e1a0da46b8d729e5a6b", "labels_vector_sha256": "68163267f563c733eb021ac67e01cdc109bb8a235cc968138ae50938f819d0ec", "planned_active_tokens": 65039, "realized_active_tokens": 65039, "cumulative_start": 590919, "cumulative_end": 655958, "loss_sum": 29991.05549621582, "token_normalized_loss": 0.4611241792803675, "grad_norm_before_clip": 0.21091587841510773, "learning_rates_at_step": { "lora": 9.55546750496013e-05, "row_tokens": 9.55546750496013e-05 }, "learning_rates_after_step": { "lora": 9.999729538681231e-05, "row_tokens": 9.999729538681231e-05 } }, { "update_id": 10, "row_count": 63, "row_ids_sha256": "dce9b8076725d9055632d781f6ec729f713ca227f31d1fb4246ee0705fbadf17", "input_ids_vector_sha256": "9c7b3543ac8d35d92470b110837729f929a9f79323f414eaaf30525d65dafe48", "labels_vector_sha256": "4d83ce9e262aa60525fe3462e0f9e9c4eb48ad8e49024fe662abfe4f3444f066", "planned_active_tokens": 66123, "realized_active_tokens": 66123, "cumulative_start": 655958, "cumulative_end": 722081, "loss_sum": 27184.7479095459, "token_normalized_loss": 0.4111239343276303, "grad_norm_before_clip": 0.20218053460121155, "learning_rates_at_step": { "lora": 9.999729538681231e-05, "row_tokens": 9.999729538681231e-05 }, "learning_rates_after_step": { "lora": 9.997824736095953e-05, "row_tokens": 9.997824736095953e-05 } }, { "update_id": 11, "row_count": 65, "row_ids_sha256": "ba2c14ed44b01e6d0316f8da503c5b553595be20e936050477082b6a62c95c78", "input_ids_vector_sha256": "5731e0a5565690f78cf5444d7d8e47db7fa39c97265e4879d84be18cc5025d52", "labels_vector_sha256": "a9733bc80cce8939c09a6b7806d6b538cf9c984dc49bbd4c53a977b6f9ed1e17", "planned_active_tokens": 65377, "realized_active_tokens": 65377, "cumulative_start": 722081, "cumulative_end": 787458, "loss_sum": 23607.859451293945, "token_normalized_loss": 0.36110343777313036, "grad_norm_before_clip": 0.16408561170101166, "learning_rates_at_step": { "lora": 9.997824736095953e-05, "row_tokens": 9.997824736095953e-05 }, "learning_rates_after_step": { "lora": 9.994133287770612e-05, "row_tokens": 9.994133287770612e-05 } }, { "update_id": 12, "row_count": 65, "row_ids_sha256": "9e064d48cfad0a8050d4c2e5a0df70ee3a090997f40e3b08aa03ee77ad5bb89d", "input_ids_vector_sha256": "71eb30ae94d6209a199eeeb9007f92090c022ecdda7711986d125ff6799ffe8e", "labels_vector_sha256": "437a1c44227428b2f29ed891fb3a393fefa7e4db700b2d62e363b92200c600f1", "planned_active_tokens": 64868, "realized_active_tokens": 64868, "cumulative_start": 787458, "cumulative_end": 852326, "loss_sum": 21392.819427490234, "token_normalized_loss": 0.3297900263225355, "grad_norm_before_clip": 0.13857263326644897, "learning_rates_at_step": { "lora": 9.994133287770612e-05, "row_tokens": 9.994133287770612e-05 }, "learning_rates_after_step": { "lora": 9.988569017766252e-05, "row_tokens": 9.988569017766252e-05 } }, { "update_id": 13, "row_count": 65, "row_ids_sha256": "d0fef968ce527ca29dd0588ca8c0a4a7cd44be3f496ab02a3ed3d8b4180befa5", "input_ids_vector_sha256": "1abb86a702f1321d3e73834ebc1590a1f5ce6c9dcc6cb4fcc768371c0513dc48", "labels_vector_sha256": "6ebc3deba0bf9c943262504066d7a69076fef523e01eec1c10df9c63ebee012b", "planned_active_tokens": 65677, "realized_active_tokens": 65677, "cumulative_start": 852326, "cumulative_end": 918003, "loss_sum": 20565.712936401367, "token_normalized_loss": 0.3131341708117205, "grad_norm_before_clip": 0.1134285256266594, "learning_rates_at_step": { "lora": 9.988569017766252e-05, "row_tokens": 9.988569017766252e-05 }, "learning_rates_after_step": { "lora": 9.981122051475016e-05, "row_tokens": 9.981122051475016e-05 } }, { "update_id": 14, "row_count": 69, "row_ids_sha256": "1f7dbfe9347b4814f341f028f083bad6f3208eee098502ddca459394e0f1f265", "input_ids_vector_sha256": "0203c660c31a1bf6feddc9ea5bf9eec5521021903a7d02b21bde160ddbe5ed59", "labels_vector_sha256": "1ae66a6e06c0cceea6f853255d31ac62a61b85609447399af6ee7c08a1b6fa34", "planned_active_tokens": 66045, "realized_active_tokens": 66045, "cumulative_start": 918003, "cumulative_end": 984048, "loss_sum": 23887.358947753906, "token_normalized_loss": 0.3616830789273057, "grad_norm_before_clip": 0.11442459374666214, "learning_rates_at_step": { "lora": 9.981122051475016e-05, "row_tokens": 9.981122051475016e-05 }, "learning_rates_after_step": { "lora": 9.971957415333032e-05, "row_tokens": 9.971957415333032e-05 } }, { "update_id": 15, "row_count": 69, "row_ids_sha256": "e823ce0ee5cfc57e62c9e719f7159d0ed2d510fc576beee44736014e8c95e2f3", "input_ids_vector_sha256": "396776d0dad3256b160a262f5caf262c6102e924a85ecd22590a15d111527b8b", "labels_vector_sha256": "9e811ae5702fd84be81da52e37279f1192173e9cadb7c7614630cb2c8adb6841", "planned_active_tokens": 65164, "realized_active_tokens": 65164, "cumulative_start": 984048, "cumulative_end": 1049212, "loss_sum": 22872.2361907959, "token_normalized_loss": 0.3509949694738797, "grad_norm_before_clip": 0.11742602288722992, "learning_rates_at_step": { "lora": 9.971957415333032e-05, "row_tokens": 9.971957415333032e-05 }, "learning_rates_after_step": { "lora": 9.960974733200093e-05, "row_tokens": 9.960974733200093e-05 } }, { "update_id": 16, "row_count": 64, "row_ids_sha256": "c9da09cdb9f6ad27b3b693cbae6c19f950eae33b83949f3ce9d88f58927cd604", "input_ids_vector_sha256": "edaa167892548fd3950c0ed4e478c3686449b14930ef3c265a8c42fe6a31f006", "labels_vector_sha256": "dab32bc24483f0926e99e25a87faa2f3fa997ca219f58d4681c04892a4cbf1ca", "planned_active_tokens": 65255, "realized_active_tokens": 65255, "cumulative_start": 1049212, "cumulative_end": 1114467, "loss_sum": 17840.16660308838, "token_normalized_loss": 0.2733915654446154, "grad_norm_before_clip": 0.13112713396549225, "learning_rates_at_step": { "lora": 9.960974733200093e-05, "row_tokens": 9.960974733200093e-05 }, "learning_rates_after_step": { "lora": 9.948195404339782e-05, "row_tokens": 9.948195404339782e-05 } }, { "update_id": 17, "row_count": 66, "row_ids_sha256": "0178612463ec3721dee68125bef5cd12db90cff529b24dc98420b14ab8b5dfea", "input_ids_vector_sha256": "23f614129d922ff1b0bcc1a4ddc97e81f431f96047b22da6f1733e49c3a4b472", "labels_vector_sha256": "fe4e49129578252d58501801e9b09ad0d16c3a63f7758435898dd6c7e6077069", "planned_active_tokens": 65227, "realized_active_tokens": 65227, "cumulative_start": 1114467, "cumulative_end": 1179694, "loss_sum": 17641.86093902588, "token_normalized_loss": 0.27046868534542257, "grad_norm_before_clip": 0.13768410682678223, "learning_rates_at_step": { "lora": 9.948195404339782e-05, "row_tokens": 9.948195404339782e-05 }, "learning_rates_after_step": { "lora": 9.933394921487884e-05, "row_tokens": 9.933394921487884e-05 } }, { "update_id": 18, "row_count": 69, "row_ids_sha256": "25e5c2ff876ae27a0784743d35a9131285f30fe860cb9cbbfae93dadc0eaff42", "input_ids_vector_sha256": "1e40ef2891917406986a2fa094df296cc78b9424a300936d7d69cd07489b1f9f", "labels_vector_sha256": "0dea3ff80f61f766f7077a69286c135d37aa2d01e09a8468aa452b55daa0b987", "planned_active_tokens": 66174, "realized_active_tokens": 66174, "cumulative_start": 1179694, "cumulative_end": 1245868, "loss_sum": 22455.481674194336, "token_normalized_loss": 0.3393399473236367, "grad_norm_before_clip": 0.11571381241083145, "learning_rates_at_step": { "lora": 9.933394921487884e-05, "row_tokens": 9.933394921487884e-05 }, "learning_rates_after_step": { "lora": 9.916956403125653e-05, "row_tokens": 9.916956403125653e-05 } }, { "update_id": 19, "row_count": 67, "row_ids_sha256": "2e44a46fe2026ca43b0dd545c0915f9ec95064846921a4c044ba8ec1196842b7", "input_ids_vector_sha256": "9f0c1559524ce4afe5df8d71b529b0440d058ce113f78fe53fdea1717022e9d5", "labels_vector_sha256": "aa0299826dbddcaba8da8a09ee9ac04a96a7683285e4b45f6d9c5f503eab5c66", "planned_active_tokens": 65400, "realized_active_tokens": 65400, "cumulative_start": 1245868, "cumulative_end": 1311268, "loss_sum": 18892.47267150879, "token_normalized_loss": 0.28887572892215274, "grad_norm_before_clip": 0.09118638187646866, "learning_rates_at_step": { "lora": 9.916956403125653e-05, "row_tokens": 9.916956403125653e-05 }, "learning_rates_after_step": { "lora": 9.898598778119266e-05, "row_tokens": 9.898598778119266e-05 } }, { "update_id": 20, "row_count": 64, "row_ids_sha256": "c4698bb80f1dd4fad947efcf9dbafc41abc7db2fa0ecd6637a9c29555a340c4f", "input_ids_vector_sha256": "f9dc70805983ecad7b269c1715822e1e90d8416e439bc0b7c79ff6d06620fede", "labels_vector_sha256": "d78c52c49da8a7224720fa2efc29438f82d22998ebc590ead55f6c29c5756553", "planned_active_tokens": 65826, "realized_active_tokens": 65826, "cumulative_start": 1311268, "cumulative_end": 1377094, "loss_sum": 16558.450790405273, "token_normalized_loss": 0.2515487921247725, "grad_norm_before_clip": 0.08760907500982285, "learning_rates_at_step": { "lora": 9.898598778119266e-05, "row_tokens": 9.898598778119266e-05 }, "learning_rates_after_step": { "lora": 9.878637041785673e-05, "row_tokens": 9.878637041785673e-05 } }, { "update_id": 21, "row_count": 65, "row_ids_sha256": "c78388423f9772f13111802a13e9501fc0a60b16516a26916bfa1b1033fd88ce", "input_ids_vector_sha256": "9c72762b842c8a590c378585fcf02bd5251111aa4996e814ced6878caad7980f", "labels_vector_sha256": "d98b77ec4a340dbc080c7d894ad3003542a0c1d19b4f4ed5b3d19a8310c46173", "planned_active_tokens": 65179, "realized_active_tokens": 65179, "cumulative_start": 1377094, "cumulative_end": 1442273, "loss_sum": 16280.085205078125, "token_normalized_loss": 0.24977500736553376, "grad_norm_before_clip": 0.08939149975776672, "learning_rates_at_step": { "lora": 9.878637041785673e-05, "row_tokens": 9.878637041785673e-05 }, "learning_rates_after_step": { "lora": 9.85690682984278e-05, "row_tokens": 9.85690682984278e-05 } }, { "update_id": 22, "row_count": 69, "row_ids_sha256": "102f40e5eaa1cf3b2ab090382b5103100444c1fd2985f6c8692f0cae11058b27", "input_ids_vector_sha256": "d445ba5d81503e3f62d43e4289c1a067fdee77a15c687f31242a05324bf74875", "labels_vector_sha256": "3742fb944903ad972f364c8dc791e302659a861020e750dccc0a3db1abcd2466", "planned_active_tokens": 65179, "realized_active_tokens": 65179, "cumulative_start": 1442273, "cumulative_end": 1507452, "loss_sum": 19604.333473205566, "token_normalized_loss": 0.30077683722066256, "grad_norm_before_clip": 0.1150469034910202, "learning_rates_at_step": { "lora": 9.85690682984278e-05, "row_tokens": 9.85690682984278e-05 }, "learning_rates_after_step": { "lora": 9.8332555921521e-05, "row_tokens": 9.8332555921521e-05 } }, { "update_id": 23, "row_count": 68, "row_ids_sha256": "640fa4e6f020b3e67e977153f77b10d52eeef5ab4cb7d2713141a303fc1da564", "input_ids_vector_sha256": "8bbbb2102475a1afed8f4dee5d36ac3fa3a09f327869ba6886e0def67cec544d", "labels_vector_sha256": "459a1e18901cb87634f7d02da39ef96947bc96950bca77d8b539133929f99a24", "planned_active_tokens": 65608, "realized_active_tokens": 65608, "cumulative_start": 1507452, "cumulative_end": 1573060, "loss_sum": 18367.376510620117, "token_normalized_loss": 0.27995635456987134, "grad_norm_before_clip": 0.08758990466594696, "learning_rates_at_step": { "lora": 9.8332555921521e-05, "row_tokens": 9.8332555921521e-05 }, "learning_rates_after_step": { "lora": 9.807679176179743e-05, "row_tokens": 9.807679176179743e-05 } }, { "update_id": 24, "row_count": 69, "row_ids_sha256": "c35734ccc8b0c1687097e525c89ea1e304c2ec83a5365ec340f1003ba9774642", "input_ids_vector_sha256": "2339d823464f27e20a529292e34238ab3a01299bbb87b60d332630b370a508f3", "labels_vector_sha256": "3fb51e51b028d5bbece65b0655b67ab700299fb6dfa59c649f55fbd9aa194b63", "planned_active_tokens": 65982, "realized_active_tokens": 65982, "cumulative_start": 1573060, "cumulative_end": 1639042, "loss_sum": 18810.63995361328, "token_normalized_loss": 0.28508744738888303, "grad_norm_before_clip": 0.07312844693660736, "learning_rates_at_step": { "lora": 9.807679176179743e-05, "row_tokens": 9.807679176179743e-05 }, "learning_rates_after_step": { "lora": 9.780527844062164e-05, "row_tokens": 9.780527844062164e-05 } }, { "update_id": 25, "row_count": 64, "row_ids_sha256": "2cbfa68defa91d1ff767ad91cb25728377893a3de9d0812cd5084cb5dd79246c", "input_ids_vector_sha256": "e0151f29a24b3946ae51a7597fc141edfa8d8e78248f7670679e7145c1fa4666", "labels_vector_sha256": "1ef2e9ab16860b95f63376b211d0457c4ba96170a14c65ce687aa26f3693a431", "planned_active_tokens": 65489, "realized_active_tokens": 65489, "cumulative_start": 1639042, "cumulative_end": 1704531, "loss_sum": 25313.56854248047, "token_normalized_loss": 0.3865316090103753, "grad_norm_before_clip": 0.08340176194906235, "learning_rates_at_step": { "lora": 9.780527844062164e-05, "row_tokens": 9.780527844062164e-05 }, "learning_rates_after_step": { "lora": 9.751862511730145e-05, "row_tokens": 9.751862511730145e-05 } }, { "update_id": 26, "row_count": 60, "row_ids_sha256": "7678e04d02e7627a6de8fe64fcd9a7dd3f73f44c8385c2a2a242fa2c3e06d341", "input_ids_vector_sha256": "2c54b02d1b310a6a642a51db7d8e869e1f59ed0e476d25faf7d8bed01da47af9", "labels_vector_sha256": "659b1062ba91437e9ab03792efa593c28b857af2bb38b4a8094c429b9e79f221", "planned_active_tokens": 64971, "realized_active_tokens": 64971, "cumulative_start": 1704531, "cumulative_end": 1769502, "loss_sum": 20852.689331054688, "token_normalized_loss": 0.3209537998653967, "grad_norm_before_clip": 0.07235341519117355, "learning_rates_at_step": { "lora": 9.751862511730145e-05, "row_tokens": 9.751862511730145e-05 }, "learning_rates_after_step": { "lora": 9.721223614445478e-05, "row_tokens": 9.721223614445478e-05 } }, { "update_id": 27, "row_count": 72, "row_ids_sha256": "856aca2f027b29534d04f0cda02b29f8ed45cdc83d403baaada098eb73ea6201", "input_ids_vector_sha256": "3c0fa77f59417c8bd783ef897826abd03f82dedb10c20ac5c0ab1ff7d682393c", "labels_vector_sha256": "47a4aff0b2dafacdeb3750dce97664da2436d4d58ae4cf3ff578104110d640d5", "planned_active_tokens": 65516, "realized_active_tokens": 65516, "cumulative_start": 1769502, "cumulative_end": 1835018, "loss_sum": 17250.570205688477, "token_normalized_loss": 0.26330316572575363, "grad_norm_before_clip": 0.08366767317056656, "learning_rates_at_step": { "lora": 9.721223614445478e-05, "row_tokens": 9.721223614445478e-05 }, "learning_rates_after_step": { "lora": 9.688463386871219e-05, "row_tokens": 9.688463386871219e-05 } }, { "update_id": 28, "row_count": 71, "row_ids_sha256": "972d695ece12f85f9c6ef532680b9295fd3a5b142545967948c6e1c864039af3", "input_ids_vector_sha256": "2920ad95a6d9bd03a7c3c462b01a48352e4a09b15c68d06e206295772fe52f67", "labels_vector_sha256": "e32747214ce9989e592d2cd2d5fe7262a8293446186b1ed451749be1ca2a7e25", "planned_active_tokens": 66289, "realized_active_tokens": 66289, "cumulative_start": 1835018, "cumulative_end": 1901307, "loss_sum": 15775.622619628906, "token_normalized_loss": 0.23798251021480044, "grad_norm_before_clip": 0.06702205538749695, "learning_rates_at_step": { "lora": 9.688463386871219e-05, "row_tokens": 9.688463386871219e-05 }, "learning_rates_after_step": { "lora": 9.654569120943919e-05, "row_tokens": 9.654569120943919e-05 } }, { "update_id": 29, "row_count": 72, "row_ids_sha256": "b99b9e21902f37b799218894f51850d7af2070b8030e93d0f96a1485d2357d2d", "input_ids_vector_sha256": "fabc20044b8658525efda7174176bd43055f52b76eab581281ccace4bad9a7e2", "labels_vector_sha256": "998074668dd380ab22572cddb0424bab6c8584a6f8373d71dd96431549b211ac", "planned_active_tokens": 65120, "realized_active_tokens": 65120, "cumulative_start": 1901307, "cumulative_end": 1966427, "loss_sum": 15416.36563873291, "token_normalized_loss": 0.23673780157759383, "grad_norm_before_clip": 0.06540918350219727, "learning_rates_at_step": { "lora": 9.654569120943919e-05, "row_tokens": 9.654569120943919e-05 }, "learning_rates_after_step": { "lora": 9.618653934665645e-05, "row_tokens": 9.618653934665645e-05 } }, { "update_id": 30, "row_count": 72, "row_ids_sha256": "ce0d4a57e406e31bf79a7e5fda8068661d8168d305a932c04143f73da8b77c5c", "input_ids_vector_sha256": "0d446dcf176dfdd4f171e3c4106be8f062594fb4be8569710d3103741bacb864", "labels_vector_sha256": "a5cba27676f7367ec06e2ddc816ad8cdf5f8053bcc1bc4f49e429db085982f40", "planned_active_tokens": 65722, "realized_active_tokens": 65722, "cumulative_start": 1966427, "cumulative_end": 2032149, "loss_sum": 14481.376586914062, "token_normalized_loss": 0.22034290780734095, "grad_norm_before_clip": 0.06471860408782959, "learning_rates_at_step": { "lora": 9.618653934665645e-05, "row_tokens": 9.618653934665645e-05 }, "learning_rates_after_step": { "lora": 9.581262382399307e-05, "row_tokens": 9.581262382399307e-05 } }, { "update_id": 31, "row_count": 67, "row_ids_sha256": "93a16b88c9af68fc323be0e0397e954b34c6a5cc3ae0012ffbc13847b8c25edd", "input_ids_vector_sha256": "65d4423c5dc4530f83a2d04854dfe585c65b72866ad3447d39278832788b73b9", "labels_vector_sha256": "3c2ae6777fc9ea2b8bdff9c9075cec77f4df9dda9d2c26e2ed0af59bfb0996d8", "planned_active_tokens": 65336, "realized_active_tokens": 65336, "cumulative_start": 2032149, "cumulative_end": 2097485, "loss_sum": 12970.317554473877, "token_normalized_loss": 0.19851716594945937, "grad_norm_before_clip": 0.05869753658771515, "learning_rates_at_step": { "lora": 9.581262382399307e-05, "row_tokens": 9.581262382399307e-05 }, "learning_rates_after_step": { "lora": 9.542084300860709e-05, "row_tokens": 9.542084300860709e-05 } }, { "update_id": 32, "row_count": 71, "row_ids_sha256": "9e99e39d47d9076d973e4c697efe5a200c118edaf26134919d26348d81bb8152", "input_ids_vector_sha256": "5913ce4f4876ffc2a24777145745eab9636974d8c78b43ee7d636749b3ccd45f", "labels_vector_sha256": "8aaa331bd5bce9525bb2448427bb2725bb5ec0b66c3dfa890d1de4fada27e822", "planned_active_tokens": 65529, "realized_active_tokens": 65529, "cumulative_start": 2097485, "cumulative_end": 2163014, "loss_sum": 14857.14852142334, "token_normalized_loss": 0.22672631234145707, "grad_norm_before_clip": 0.06140521168708801, "learning_rates_at_step": { "lora": 9.542084300860709e-05, "row_tokens": 9.542084300860709e-05 }, "learning_rates_after_step": { "lora": 9.50108302711997e-05, "row_tokens": 9.50108302711997e-05 } }, { "update_id": 33, "row_count": 73, "row_ids_sha256": "3db123285a1e354088d5ef41b25c584df8d9335411e613bd6d8b96ee9ecd906a", "input_ids_vector_sha256": "c6671b76f69784c51aee7fb29ae6689cb1911c240786acba116b7dcd1c2c2146", "labels_vector_sha256": "3384e4363747dc64837048d48494f944c33721bb8e9fbcb35b425fc6179b859c", "planned_active_tokens": 65779, "realized_active_tokens": 65779, "cumulative_start": 2163014, "cumulative_end": 2228793, "loss_sum": 12555.756797790527, "token_normalized_loss": 0.19087789108667702, "grad_norm_before_clip": 0.06371292471885681, "learning_rates_at_step": { "lora": 9.50108302711997e-05, "row_tokens": 9.50108302711997e-05 }, "learning_rates_after_step": { "lora": 9.458339300796717e-05, "row_tokens": 9.458339300796717e-05 } }, { "update_id": 34, "row_count": 77, "row_ids_sha256": "b2cd4ac8cd162d1298d6578c3b341e469c62da7a223329c2a8d83f936a9f2fc9", "input_ids_vector_sha256": "6cc7f7ba437bf731c48c911e1182384bab2d75a9b7b155ee261ead8fe090222c", "labels_vector_sha256": "c49abf4c2427dfa791b87b6f2c8e0bf40096ff4884f499844f7e69ca1e049438", "planned_active_tokens": 65901, "realized_active_tokens": 65901, "cumulative_start": 2228793, "cumulative_end": 2294694, "loss_sum": 14059.703887939453, "token_normalized_loss": 0.2133458352367863, "grad_norm_before_clip": 0.05596740543842316, "learning_rates_at_step": { "lora": 9.458339300796717e-05, "row_tokens": 9.458339300796717e-05 }, "learning_rates_after_step": { "lora": 9.414667793277241e-05, "row_tokens": 9.414667793277241e-05 } }, { "update_id": 35, "row_count": 69, "row_ids_sha256": "ee7819c0c2f6020843e6178f1874707e0dce1ca6d01c98aa00de7ee1c4811156", "input_ids_vector_sha256": "fd9c57d373717b7d86f14a16b385fea9dd8158142b9a750bccf7b19f115bb413", "labels_vector_sha256": "700e63b76f3f9a982432d7d0ad2440fbb5961dae4273a791b0a4d7d22c550ea7", "planned_active_tokens": 64845, "realized_active_tokens": 64845, "cumulative_start": 2294694, "cumulative_end": 2359539, "loss_sum": 12368.137687683105, "token_normalized_loss": 0.1907338682656042, "grad_norm_before_clip": 0.05099748447537422, "learning_rates_at_step": { "lora": 9.414667793277241e-05, "row_tokens": 9.414667793277241e-05 }, "learning_rates_after_step": { "lora": 9.368313488551548e-05, "row_tokens": 9.368313488551548e-05 } }, { "update_id": 36, "row_count": 69, "row_ids_sha256": "12f26722aaf0eaf97160e9f9044b3c5eca25a89de515b75c434476682535f83a", "input_ids_vector_sha256": "241235f9bdd7213600dcc980c27e415836f4f58a24214daf5510f26ae9dce8f1", "labels_vector_sha256": "f2025dbabcf0b8467404e61501cc71b95114a2c38d433b7afd2f800c4bbffc08", "planned_active_tokens": 66392, "realized_active_tokens": 66392, "cumulative_start": 2359539, "cumulative_end": 2425931, "loss_sum": 12313.798965454102, "token_normalized_loss": 0.1854711255189496, "grad_norm_before_clip": 0.051004402339458466, "learning_rates_at_step": { "lora": 9.368313488551548e-05, "row_tokens": 9.368313488551548e-05 }, "learning_rates_after_step": { "lora": 9.321035817213665e-05, "row_tokens": 9.321035817213665e-05 } }, { "update_id": 37, "row_count": 71, "row_ids_sha256": "3c1deee8b4f00aa84e1850a830a89d48afcf4b1df51d71b251c76312eb4de007", "input_ids_vector_sha256": "12e996a0882847f8b323ec4fe9cadae6b1a501558c971316b6120eaf8b73148f", "labels_vector_sha256": "7872b7aa490f33ab5bdd9702a1df52d21b80f14e4f2d256760adf995c550258a", "planned_active_tokens": 65413, "realized_active_tokens": 65413, "cumulative_start": 2425931, "cumulative_end": 2491344, "loss_sum": 11421.17261505127, "token_normalized_loss": 0.17460096028390792, "grad_norm_before_clip": 0.062253303825855255, "learning_rates_at_step": { "lora": 9.321035817213665e-05, "row_tokens": 9.321035817213665e-05 }, "learning_rates_after_step": { "lora": 9.272820864295033e-05, "row_tokens": 9.272820864295033e-05 } }, { "update_id": 38, "row_count": 70, "row_ids_sha256": "71d635012e30c8e3ab85fc726c55a33c9ffd42d82d20f212d1a92983989c0864", "input_ids_vector_sha256": "b6ca233a6aacd22523b6ae2091f8f83533592d5030f20034b56e1a401117d728", "labels_vector_sha256": "9b885fc413643ec6929871200529e3ff48498b0c3d3224a53caea2cd200f248d", "planned_active_tokens": 64569, "realized_active_tokens": 64569, "cumulative_start": 2491344, "cumulative_end": 2555913, "loss_sum": 11176.435283660889, "token_normalized_loss": 0.17309289726743313, "grad_norm_before_clip": 0.05578511208295822, "learning_rates_at_step": { "lora": 9.272820864295033e-05, "row_tokens": 9.272820864295033e-05 }, "learning_rates_after_step": { "lora": 9.222275611266875e-05, "row_tokens": 9.222275611266875e-05 } }, { "update_id": 39, "row_count": 73, "row_ids_sha256": "22768f4215de3dc54a1fe6d4bb46a8f36658a1f406db4ac4335eaa9e11821dd7", "input_ids_vector_sha256": "1c24a717ac3da1c9ed378058f873a8f04d3aecf80ba2e104126efc1ce3a7b770", "labels_vector_sha256": "7530f8e54d07887796722249bba6fc687c8c8bb8a89efee95f4daf00dfb1cc87", "planned_active_tokens": 65605, "realized_active_tokens": 65605, "cumulative_start": 2555913, "cumulative_end": 2621518, "loss_sum": 12201.446716308594, "token_normalized_loss": 0.1859834877876472, "grad_norm_before_clip": 0.05482906848192215, "learning_rates_at_step": { "lora": 9.222275611266875e-05, "row_tokens": 9.222275611266875e-05 }, "learning_rates_after_step": { "lora": 9.170016958017611e-05, "row_tokens": 9.170016958017611e-05 } }, { "update_id": 40, "row_count": 71, "row_ids_sha256": "780b88fc570393322678eead0b3372bb1cb5d1db5df632d3defef6be6039a731", "input_ids_vector_sha256": "fc832d7190a21f4c2d7a73bd75bc6a3c0b791c036e9a90eb99a672c1259138ef", "labels_vector_sha256": "0f84ee857c8692c9aa19ca1089ff98b92fc1527533d49c07228088c29b97b207", "planned_active_tokens": 65807, "realized_active_tokens": 65807, "cumulative_start": 2621518, "cumulative_end": 2687325, "loss_sum": 11386.783065795898, "token_normalized_loss": 0.17303300660713752, "grad_norm_before_clip": 0.052972033619880676, "learning_rates_at_step": { "lora": 9.170016958017611e-05, "row_tokens": 9.170016958017611e-05 }, "learning_rates_after_step": { "lora": 9.115922234457153e-05, "row_tokens": 9.115922234457153e-05 } }, { "update_id": 41, "row_count": 69, "row_ids_sha256": "1d63f8d33871fb63d97c11f3fde9e435023b0482bc256a4c52555ed6f94038ae", "input_ids_vector_sha256": "8d9f3d561fa9658eb18ee27dac831f491562fa6482cb4ef0e9bc9781e0d3689b", "labels_vector_sha256": "cc240adbe87f5454124a280f1900f466d5ad045d2b4339a6209ef051f3f8fed9", "planned_active_tokens": 66163, "realized_active_tokens": 66163, "cumulative_start": 2687325, "cumulative_end": 2753488, "loss_sum": 10402.826175689697, "token_normalized_loss": 0.15723026730483347, "grad_norm_before_clip": 0.05018887668848038, "learning_rates_at_step": { "lora": 9.115922234457153e-05, "row_tokens": 9.115922234457153e-05 }, "learning_rates_after_step": { "lora": 9.061630442221628e-05, "row_tokens": 9.061630442221628e-05 } }, { "update_id": 42, "row_count": 66, "row_ids_sha256": "bdb9bdadb82a90439fde49b33f160b040505e66cd7cd207162aac3313961b98e", "input_ids_vector_sha256": "4aca6a6ba29d0102d70a44fc77ffcd45bff9e295d02a6fcc92b77cbe0808e99e", "labels_vector_sha256": "ac37b7c5fe8216d7ece5891f9dc4de351ce426f5b79b17b1f3aa1d7365438a87", "planned_active_tokens": 64590, "realized_active_tokens": 64590, "cumulative_start": 2753488, "cumulative_end": 2818078, "loss_sum": 14664.625457763672, "token_normalized_loss": 0.2270417318124117, "grad_norm_before_clip": 0.0597090870141983, "learning_rates_at_step": { "lora": 9.061630442221628e-05, "row_tokens": 9.061630442221628e-05 }, "learning_rates_after_step": { "lora": 9.005057962890521e-05, "row_tokens": 9.005057962890521e-05 } }, { "update_id": 43, "row_count": 75, "row_ids_sha256": "ce42f80586aa05bc2f179fe94d927a5a547bd7ea758339cbde5baf884e1b8bc5", "input_ids_vector_sha256": "2a7fc7c4ed136742344012ec43c2cc6beecc9b6e525b1878f85f8e138ef84ef2", "labels_vector_sha256": "f82e8225320f9a83ea490beb8674b2cb625a74ec73e5aa3cab6a3a4008badaac", "planned_active_tokens": 65545, "realized_active_tokens": 65545, "cumulative_start": 2818078, "cumulative_end": 2883623, "loss_sum": 14651.431816101074, "token_normalized_loss": 0.22353241004044663, "grad_norm_before_clip": 0.055438678711652756, "learning_rates_at_step": { "lora": 9.005057962890521e-05, "row_tokens": 9.005057962890521e-05 }, "learning_rates_after_step": { "lora": 8.946809304617247e-05, "row_tokens": 8.946809304617247e-05 } }, { "update_id": 44, "row_count": 72, "row_ids_sha256": "b053d25137c30ca8ef5b6b8ffca884043a3800f45e47c2fb9712759292d26c3a", "input_ids_vector_sha256": "e5d17d02a51a2d6ec45de3caca12c270542ecaaac641bc7eaf70ee9f9f03d323", "labels_vector_sha256": "6d1a26ec0a234a61b3669ce764f16a806844e5337ab5674c870247ed67270e3c", "planned_active_tokens": 65777, "realized_active_tokens": 65777, "cumulative_start": 2883623, "cumulative_end": 2949400, "loss_sum": 13961.837329864502, "token_normalized_loss": 0.21226017194254074, "grad_norm_before_clip": 0.05710973963141441, "learning_rates_at_step": { "lora": 8.946809304617247e-05, "row_tokens": 8.946809304617247e-05 }, "learning_rates_after_step": { "lora": 8.88659091432649e-05, "row_tokens": 8.88659091432649e-05 } }, { "update_id": 45, "row_count": 72, "row_ids_sha256": "6e86ea43727e76ed0e124108a72d790ac9bd002ecdde4afad97d3916adf4110e", "input_ids_vector_sha256": "9864f0fbf229e88625e9fb4875c74793b1bb20511c4848c8a76b61b90de00a09", "labels_vector_sha256": "c2839fd06a85d3d64f5598815a70473855c43dcc17e64e8c67906da00a92240e", "planned_active_tokens": 66335, "realized_active_tokens": 66335, "cumulative_start": 2949400, "cumulative_end": 3015735, "loss_sum": 12983.625923156738, "token_normalized_loss": 0.1957281363255708, "grad_norm_before_clip": 0.050991255789995193, "learning_rates_at_step": { "lora": 8.88659091432649e-05, "row_tokens": 8.88659091432649e-05 }, "learning_rates_after_step": { "lora": 8.826028653075119e-05, "row_tokens": 8.826028653075119e-05 } }, { "update_id": 46, "row_count": 73, "row_ids_sha256": "e110e5d85d158d71834ce404ecaa0dfd497d30a13b9ae80f7f8bd036aa11d3fe", "input_ids_vector_sha256": "d65acf7bd54f2f3f590286d98c3db54a0c05b2c62a7915b5e256a87a68ea019d", "labels_vector_sha256": "f1b1a4ee65a219f1d6ba4492d88a0bd9764a99e2527aa1f9a3d0b8703f34a0bf", "planned_active_tokens": 65149, "realized_active_tokens": 65149, "cumulative_start": 3015735, "cumulative_end": 3080884, "loss_sum": 12658.143417358398, "token_normalized_loss": 0.1942952833866736, "grad_norm_before_clip": 0.05048849806189537, "learning_rates_at_step": { "lora": 8.826028653075119e-05, "row_tokens": 8.826028653075119e-05 }, "learning_rates_after_step": { "lora": 8.764212463334784e-05, "row_tokens": 8.764212463334784e-05 } }, { "update_id": 47, "row_count": 72, "row_ids_sha256": "95ef11a341c49a26208f375beacad400cedc3e7ff446dc38ce6a1c1fc41af369", "input_ids_vector_sha256": "57c9758fd92556851fea90d4e6026cacac7da845675681a24c6d8b988e1703ba", "labels_vector_sha256": "e172a48b72c6197808e17243db252e8596de98d31d6851898dd604ae98d75b24", "planned_active_tokens": 65012, "realized_active_tokens": 65012, "cumulative_start": 3080884, "cumulative_end": 3145896, "loss_sum": 11654.413841247559, "token_normalized_loss": 0.1792655792968615, "grad_norm_before_clip": 0.05710308253765106, "learning_rates_at_step": { "lora": 8.764212463334784e-05, "row_tokens": 8.764212463334784e-05 }, "learning_rates_after_step": { "lora": 8.700525098604702e-05, "row_tokens": 8.700525098604702e-05 } }, { "update_id": 48, "row_count": 76, "row_ids_sha256": "ab227911cfc013a419589a24ab74f0cb444069d585a847001921e976ebf4f43c", "input_ids_vector_sha256": "ad48187f54fee47e1e5d2ac2b6d6f03e04ea6f96fc5a76df05f0d39c366f0d86", "labels_vector_sha256": "d0e86571d63fcfd2723e0fbc55870fc927e90d3c0b2fb92afa1c1569d6ec19d9", "planned_active_tokens": 65535, "realized_active_tokens": 65535, "cumulative_start": 3145896, "cumulative_end": 3211431, "loss_sum": 12928.416206359863, "token_normalized_loss": 0.19727498598245005, "grad_norm_before_clip": 0.052575401961803436, "learning_rates_at_step": { "lora": 8.700525098604702e-05, "row_tokens": 8.700525098604702e-05 }, "learning_rates_after_step": { "lora": 8.635147698786666e-05, "row_tokens": 8.635147698786666e-05 } }, { "update_id": 49, "row_count": 76, "row_ids_sha256": "9daa9aab50ec17ec157442672130b52fc2cd193bebd0f094c19965e4233a9097", "input_ids_vector_sha256": "4ecaabbca6ff9f6fd9d051c3f7fa120d21ca4ec4abf190a3fd339bd41888c690", "labels_vector_sha256": "e27cf8d5ed38fc6f891f607cf7324ee441ac7a7c2b019355610315ae56e251e5", "planned_active_tokens": 65868, "realized_active_tokens": 65868, "cumulative_start": 3211431, "cumulative_end": 3277299, "loss_sum": 12219.417446136475, "token_normalized_loss": 0.18551371600984506, "grad_norm_before_clip": 0.05162892863154411, "learning_rates_at_step": { "lora": 8.635147698786666e-05, "row_tokens": 8.635147698786666e-05 }, "learning_rates_after_step": { "lora": 8.568669081048716e-05, "row_tokens": 8.568669081048716e-05 } }, { "update_id": 50, "row_count": 70, "row_ids_sha256": "73cd7752d9419fa937b911b804ca26c5b534f0c05e94a8a314dc7f9422366c8f", "input_ids_vector_sha256": "b595785e84891fd09d7b00a146586066fee4b597c30ac78897abf197897a63fc", "labels_vector_sha256": "fef0085b7f19029b4f5a1037f83be7fd26ca87d3f055780265776aa96b7d285f", "planned_active_tokens": 65629, "realized_active_tokens": 65629, "cumulative_start": 3277299, "cumulative_end": 3342928, "loss_sum": 12200.997806549072, "token_normalized_loss": 0.18590863500204288, "grad_norm_before_clip": 0.04888525977730751, "learning_rates_at_step": { "lora": 8.568669081048716e-05, "row_tokens": 8.568669081048716e-05 }, "learning_rates_after_step": { "lora": 8.5008570205724e-05, "row_tokens": 8.5008570205724e-05 } }, { "update_id": 51, "row_count": 72, "row_ids_sha256": "622a0f8d08fa81e8c679dde13816143f08f1c948ee878c5bca5ac3b056a19b27", "input_ids_vector_sha256": "6cf790e82e310330483ffc572ff53296e53d0e1ece91585ba9bd7b8d45c160a3", "labels_vector_sha256": "30477903ba701d357fa24111b5a85ccf6a60be13937afc79ab844fd7f7b03fd4", "planned_active_tokens": 65650, "realized_active_tokens": 65650, "cumulative_start": 3342928, "cumulative_end": 3408578, "loss_sum": 11959.353622436523, "token_normalized_loss": 0.1821683720096957, "grad_norm_before_clip": 0.04693548381328583, "learning_rates_at_step": { "lora": 8.5008570205724e-05, "row_tokens": 8.5008570205724e-05 }, "learning_rates_after_step": { "lora": 8.432246105280315e-05, "row_tokens": 8.432246105280315e-05 } }, { "update_id": 52, "row_count": 69, "row_ids_sha256": "5a48cbcf651c065162ad2e9669334cd1ab307286968fecbaaade554f71b2ab71", "input_ids_vector_sha256": "c9e2005191dd6553e58941a5dba12d8fe164e18976a7ce780cc72ab3cb788731", "labels_vector_sha256": "d40a057f4ece667098b0efc823231c9147459b9fa9cfa23d427d6827c554612e", "planned_active_tokens": 65190, "realized_active_tokens": 65190, "cumulative_start": 3408578, "cumulative_end": 3473768, "loss_sum": 10253.744621276855, "token_normalized_loss": 0.15729014605425456, "grad_norm_before_clip": 0.0450500063598156, "learning_rates_at_step": { "lora": 8.432246105280315e-05, "row_tokens": 8.432246105280315e-05 }, "learning_rates_after_step": { "lora": 8.362191696099518e-05, "row_tokens": 8.362191696099518e-05 } }, { "update_id": 53, "row_count": 75, "row_ids_sha256": "8025d069faed7d53137e8f01c8076caf9f4e22f0f330e1b0ac480811aa90d945", "input_ids_vector_sha256": "f686f8e4ff9212e9b5975e66218b2e3766468b18e2104653237e88a515a93666", "labels_vector_sha256": "b7951f33d31300a3a6eef801dab56f3dead7e6df1d0ae3cc2d4a6444c0230ffe", "planned_active_tokens": 65374, "realized_active_tokens": 65374, "cumulative_start": 3473768, "cumulative_end": 3539142, "loss_sum": 11676.723430633545, "token_normalized_loss": 0.1786141804178044, "grad_norm_before_clip": 0.04766929894685745, "learning_rates_at_step": { "lora": 8.362191696099518e-05, "row_tokens": 8.362191696099518e-05 }, "learning_rates_after_step": { "lora": 8.290518636915056e-05, "row_tokens": 8.290518636915056e-05 } }, { "update_id": 54, "row_count": 71, "row_ids_sha256": "3f8e2537031ee7ec5c0ebdd0f124634e11c730efd9ea622a603ea9f66ebbe420", "input_ids_vector_sha256": "c4860e106861a8f7c04ac9b2cf73705436d5443b34af1c94021eb9ec31ca49d6", "labels_vector_sha256": "25f4b86baf3a08da5a716940befb27a75bc5db30209797f53df56e1c06f3cf74", "planned_active_tokens": 65731, "realized_active_tokens": 65731, "cumulative_start": 3539142, "cumulative_end": 3604873, "loss_sum": 11334.259449005127, "token_normalized_loss": 0.17243400296671474, "grad_norm_before_clip": 0.05148300901055336, "learning_rates_at_step": { "lora": 8.290518636915056e-05, "row_tokens": 8.290518636915056e-05 }, "learning_rates_after_step": { "lora": 8.21821714538094e-05, "row_tokens": 8.21821714538094e-05 } }, { "update_id": 55, "row_count": 69, "row_ids_sha256": "9aca459aa271fd9c7f1b04e8bd4b048ccaca1bc4bc9f4b2b76fe1e0bb7e499e2", "input_ids_vector_sha256": "80c3c4ef69a8d2a8d95cd4a1d7d09a1aacb712b195b020cfd771f0666e8b1f31", "labels_vector_sha256": "04d665ab56b8160e374f7afdd131da1b6fa153d0c7f62ad82237814189d5ff69", "planned_active_tokens": 65208, "realized_active_tokens": 65208, "cumulative_start": 3604873, "cumulative_end": 3670081, "loss_sum": 12261.371433258057, "token_normalized_loss": 0.18803477231717053, "grad_norm_before_clip": 0.05183539167046547, "learning_rates_at_step": { "lora": 8.21821714538094e-05, "row_tokens": 8.21821714538094e-05 }, "learning_rates_after_step": { "lora": 8.14432787625836e-05, "row_tokens": 8.14432787625836e-05 } }, { "update_id": 56, "row_count": 73, "row_ids_sha256": "dcda659b74ccbb373e84c158be3050658d801d1c09df840396cb63b9e23b0e32", "input_ids_vector_sha256": "a78bbc22064ce36e3aba796ee420db4649d4335211e513502a8efb5450177c07", "labels_vector_sha256": "e645568cbe5a700c2d81b700276e487bf6c94421657430c32faa26168301dfcf", "planned_active_tokens": 65578, "realized_active_tokens": 65578, "cumulative_start": 3670081, "cumulative_end": 3735659, "loss_sum": 13299.289775848389, "token_normalized_loss": 0.2028010884114854, "grad_norm_before_clip": 0.05956578627228737, "learning_rates_at_step": { "lora": 8.14432787625836e-05, "row_tokens": 8.14432787625836e-05 }, "learning_rates_after_step": { "lora": 8.068806297073575e-05, "row_tokens": 8.068806297073575e-05 } }, { "update_id": 57, "row_count": 70, "row_ids_sha256": "5024ac47b2eee2133ab9131268be7f221fdd1e6b8e6d93428e7e9d5edcf42738", "input_ids_vector_sha256": "571ca8a6966fd719a059be4b99771441f2444a5ab34de112ab7f4bd68e264aa0", "labels_vector_sha256": "01aa40a2b808a0c1473ccbb40ca1fc9daf72490c1e9c7049c15e318b8461ebf9", "planned_active_tokens": 65993, "realized_active_tokens": 65993, "cumulative_start": 3735659, "cumulative_end": 3801652, "loss_sum": 13658.010833740234, "token_normalized_loss": 0.20696150855000128, "grad_norm_before_clip": 0.05424508824944496, "learning_rates_at_step": { "lora": 8.068806297073575e-05, "row_tokens": 8.068806297073575e-05 }, "learning_rates_after_step": { "lora": 7.99311899537653e-05, "row_tokens": 7.99311899537653e-05 } }, { "update_id": 58, "row_count": 72, "row_ids_sha256": "ca39012216064f366390a79fab5c218c9aad3856e424f34bd47adeeda83c0f21", "input_ids_vector_sha256": "9e6b1e0aca498549cb8dcc43a846227b570984c741bcf02429ad668561f3de7b", "labels_vector_sha256": "b84a9a99ae99bb26c94031e9737df041f814233c726c691968f28f0326825a5d", "planned_active_tokens": 65160, "realized_active_tokens": 65160, "cumulative_start": 3801652, "cumulative_end": 3866812, "loss_sum": 12616.000995635986, "token_normalized_loss": 0.19361573044254124, "grad_norm_before_clip": 0.057057298719882965, "learning_rates_at_step": { "lora": 7.99311899537653e-05, "row_tokens": 7.99311899537653e-05 }, "learning_rates_after_step": { "lora": 7.915078955489767e-05, "row_tokens": 7.915078955489767e-05 } }, { "update_id": 59, "row_count": 74, "row_ids_sha256": "b52a2e192667925a74f7eefd591397322121567161c41367dfbe6df72c34180b", "input_ids_vector_sha256": "0cec5b64c19ba959c98b45811cf8aa1d3480cb4ae9851f8d851f4f5cc7275049", "labels_vector_sha256": "ee5484d1e760e0bdba7b0062e68df16e2f29490f90a365879fdde6895486e8f9", "planned_active_tokens": 66229, "realized_active_tokens": 66229, "cumulative_start": 3866812, "cumulative_end": 3933041, "loss_sum": 11953.22282409668, "token_normalized_loss": 0.18048321466573072, "grad_norm_before_clip": 0.055502913892269135, "learning_rates_at_step": { "lora": 7.915078955489767e-05, "row_tokens": 7.915078955489767e-05 }, "learning_rates_after_step": { "lora": 7.83717713400463e-05, "row_tokens": 7.83717713400463e-05 } }, { "update_id": 60, "row_count": 73, "row_ids_sha256": "32a5c07f4e1d51b9e2793d4eedf7b7b9088d7c8fdc5bb8de85e0d8d60c02e32f", "input_ids_vector_sha256": "8b0df25ed59b3a15a000bfb911b919ceb9c3735d1f13942222a206b653dbced3", "labels_vector_sha256": "3992cccf2533b53ef39cf4c5dd50c5a9f61c09abe1dec5e461ddadfea41d045a", "planned_active_tokens": 65207, "realized_active_tokens": 65207, "cumulative_start": 3933041, "cumulative_end": 3998248, "loss_sum": 12276.744953155518, "token_normalized_loss": 0.18827342084677284, "grad_norm_before_clip": 0.05334367975592613, "learning_rates_at_step": { "lora": 7.83717713400463e-05, "row_tokens": 7.83717713400463e-05 }, "learning_rates_after_step": { "lora": 7.757073935346118e-05, "row_tokens": 7.757073935346118e-05 } }, { "update_id": 61, "row_count": 71, "row_ids_sha256": "c4011b1be6b5a06a7837bd99a00097967e4167daf2093f03c05d06c1b9e9c265", "input_ids_vector_sha256": "29c9cc37b73e319b10ccab5075e284314b344d4bc60a5a230b8c39edf3a475c1", "labels_vector_sha256": "68fccad3960333164e45dcc8048db1664b651e47bd732871eab4b3889ab0dcc9", "planned_active_tokens": 66169, "realized_active_tokens": 66169, "cumulative_start": 3998248, "cumulative_end": 4064417, "loss_sum": 12081.865547180176, "token_normalized_loss": 0.182591025210902, "grad_norm_before_clip": 0.050717491656541824, "learning_rates_at_step": { "lora": 7.757073935346118e-05, "row_tokens": 7.757073935346118e-05 }, "learning_rates_after_step": { "lora": 7.67745317010779e-05, "row_tokens": 7.67745317010779e-05 } }, { "update_id": 62, "row_count": 70, "row_ids_sha256": "f748a1d9a811dc83e902f25dbf051a0d548c3eb9a31b0a79979839982630d124", "input_ids_vector_sha256": "71659187a08c4e0f70a50184bcb79742c20035f82988acf0f0125314f1f2f8e3", "labels_vector_sha256": "ee36f56db6a6fb7384e20804faf7bfd4a0132211287cb31658ca9472c11b2db1", "planned_active_tokens": 64943, "realized_active_tokens": 64943, "cumulative_start": 4064417, "cumulative_end": 4129360, "loss_sum": 11723.530261993408, "token_normalized_loss": 0.1805203064532499, "grad_norm_before_clip": 0.0468483529984951, "learning_rates_at_step": { "lora": 7.67745317010779e-05, "row_tokens": 7.67745317010779e-05 }, "learning_rates_after_step": { "lora": 7.595786659740063e-05, "row_tokens": 7.595786659740063e-05 } }, { "update_id": 63, "row_count": 70, "row_ids_sha256": "59383979ad9506bde263cd323f77c55442edda14cff9db558049230d9789caf4", "input_ids_vector_sha256": "0d2101ac81beff7ad81928b3f995db062d06dadf892ee152971d46bcbcdddf78", "labels_vector_sha256": "30918f5d15c40f53b9ba824690a6e73ffb63ebdca0cef1dd2bf553cca4a80bc5", "planned_active_tokens": 65810, "realized_active_tokens": 65810, "cumulative_start": 4129360, "cumulative_end": 4195170, "loss_sum": 10324.974689483643, "token_normalized_loss": 0.1568906653925489, "grad_norm_before_clip": 0.047296956181526184, "learning_rates_at_step": { "lora": 7.595786659740063e-05, "row_tokens": 7.595786659740063e-05 }, "learning_rates_after_step": { "lora": 7.514433532562401e-05, "row_tokens": 7.514433532562401e-05 } }, { "update_id": 64, "row_count": 75, "row_ids_sha256": "36fc9bce54eee0ac110453f21b79ce78b28d37cef50daca1498c65ada63d8e2a", "input_ids_vector_sha256": "3493fc0eae1f3bb8389cfd88f34f22b8aaeac66c9dda0937732711e05de1a1b4", "labels_vector_sha256": "344392a47ccd3114bfcbab0b00af98d5fe670a623404d8b22d7dfc8595b438da", "planned_active_tokens": 64802, "realized_active_tokens": 64802, "cumulative_start": 4195170, "cumulative_end": 4259972, "loss_sum": 11055.964275360107, "token_normalized_loss": 0.17061146685843195, "grad_norm_before_clip": 0.05265768617391586, "learning_rates_at_step": { "lora": 7.514433532562401e-05, "row_tokens": 7.514433532562401e-05 }, "learning_rates_after_step": { "lora": 7.431177638355628e-05, "row_tokens": 7.431177638355628e-05 } }, { "update_id": 65, "row_count": 71, "row_ids_sha256": "fdd793a7740509ae1e6e002a1c0a455f6bcdf98e3fb17213b725f6c04df81f5c", "input_ids_vector_sha256": "e573d6e6270b51ec4d358fdea4797839038a14f3a3dac054cd0d12381117e609", "labels_vector_sha256": "c7a9984730ca5be6fdd08adb709170fe8d6ef9b9cf40b0aba62ae4129cfa87c4", "planned_active_tokens": 65587, "realized_active_tokens": 65587, "cumulative_start": 4259972, "cumulative_end": 4325559, "loss_sum": 10636.319984436035, "token_normalized_loss": 0.16217116173077034, "grad_norm_before_clip": 0.050109073519706726, "learning_rates_at_step": { "lora": 7.431177638355628e-05, "row_tokens": 7.431177638355628e-05 }, "learning_rates_after_step": { "lora": 7.347305370144464e-05, "row_tokens": 7.347305370144464e-05 } }, { "update_id": 66, "row_count": 75, "row_ids_sha256": "3267a3669f237caa9c69147384cd29b550ec2e04ae843fd8627f3756680f7208", "input_ids_vector_sha256": "e61dfb4d5ea61bb4427210ebe407a42ad62f21ae1b83723ec9566f3f1848da68", "labels_vector_sha256": "3604d9afe6c2bd3188bb251b4f3c3ae0714652dc49621486ebd24cd3a86d930f", "planned_active_tokens": 65373, "realized_active_tokens": 65373, "cumulative_start": 4325559, "cumulative_end": 4390932, "loss_sum": 10591.345710754395, "token_normalized_loss": 0.16201406866373572, "grad_norm_before_clip": 0.04960019886493683, "learning_rates_at_step": { "lora": 7.347305370144464e-05, "row_tokens": 7.347305370144464e-05 }, "learning_rates_after_step": { "lora": 7.261223782279453e-05, "row_tokens": 7.261223782279453e-05 } }, { "update_id": 67, "row_count": 76, "row_ids_sha256": "a99d90edec86c187b74428ffc3d362e12af15d4c35d394c89bcb4b8989985234", "input_ids_vector_sha256": "15df994c861fb3425ff4be742a1d1b0a2d2a1e3e437f7c43d645f958c12442f5", "labels_vector_sha256": "af2f31d9b6f7e04c2633409dcc5dd5c7f30fae040a296d86582fa7ebcdeff08f", "planned_active_tokens": 66412, "realized_active_tokens": 66412, "cumulative_start": 4390932, "cumulative_end": 4457344, "loss_sum": 10059.00280380249, "token_normalized_loss": 0.1514636331356154, "grad_norm_before_clip": 0.047975797206163406, "learning_rates_at_step": { "lora": 7.261223782279453e-05, "row_tokens": 7.261223782279453e-05 }, "learning_rates_after_step": { "lora": 7.176190211974081e-05, "row_tokens": 7.176190211974081e-05 } }, { "update_id": 68, "row_count": 73, "row_ids_sha256": "f9d5933bce1b0af8f1473c3fa76637f9d95dff70ff0dd8e79a3951eeb1d33405", "input_ids_vector_sha256": "58785ee1dacec7bea1f31a6140dcd28c4c217e226c96f49c9bc567a4a9879e9a", "labels_vector_sha256": "c0275ae7b129a1e5fbfd54e378524adbfe85843d904fc7368f2a12f7e4fd862e", "planned_active_tokens": 64968, "realized_active_tokens": 64968, "cumulative_start": 4457344, "cumulative_end": 4522312, "loss_sum": 10534.569999694824, "token_normalized_loss": 0.16215013544660178, "grad_norm_before_clip": 0.0528222881257534, "learning_rates_at_step": { "lora": 7.176190211974081e-05, "row_tokens": 7.176190211974081e-05 }, "learning_rates_after_step": { "lora": 7.089231779464528e-05, "row_tokens": 7.089231779464528e-05 } }, { "update_id": 69, "row_count": 77, "row_ids_sha256": "ede80b759cb2e695531602ae7a6bf2d75f7ab74c2ca8ff511dc4462586499416", "input_ids_vector_sha256": "1adc5e145a846fdc4ad7012552133f410105f245b4bc9344587c40df128914bd", "labels_vector_sha256": "5fae954111705f9fbad85047c63fed76805050226731f45c202f2247a06a5f4a", "planned_active_tokens": 65828, "realized_active_tokens": 65828, "cumulative_start": 4522312, "cumulative_end": 4588140, "loss_sum": 13539.852436065674, "token_normalized_loss": 0.20568530771200208, "grad_norm_before_clip": 0.05885227024555206, "learning_rates_at_step": { "lora": 7.089231779464528e-05, "row_tokens": 7.089231779464528e-05 }, "learning_rates_after_step": { "lora": 7.002646493164387e-05, "row_tokens": 7.002646493164387e-05 } }, { "update_id": 70, "row_count": 61, "row_ids_sha256": "0a1edb7c92e551be6ee2a3cafac6156f84934914e81a55f49aaf0103f2852146", "input_ids_vector_sha256": "d3d892dc3c4432593a271605758cb7b6dcc0237f7c2140fb87791280cef0feca", "labels_vector_sha256": "a186984b2529984b5fd733ecc96000f1f0e292aef26806f73fd12332c621fc4b", "planned_active_tokens": 64972, "realized_active_tokens": 64972, "cumulative_start": 4588140, "cumulative_end": 4653112, "loss_sum": 12843.526168823242, "token_normalized_loss": 0.1976778638309309, "grad_norm_before_clip": 0.054899465292692184, "learning_rates_at_step": { "lora": 7.002646493164387e-05, "row_tokens": 7.002646493164387e-05 }, "learning_rates_after_step": { "lora": 6.912820719601565e-05, "row_tokens": 6.912820719601565e-05 } }, { "update_id": 71, "row_count": 53, "row_ids_sha256": "cad1a4fdc3380926940ae8d325c7c7896bd5d39db57d9a788f705eb81b84315a", "input_ids_vector_sha256": "5e9f72b33af70b7e60154052a80e6ab261cbe144782f13282ed3f9b99189c278", "labels_vector_sha256": "af8bc3bc375f7539c0a487b534e2d9878e9b73412d39ebeb353733f7175500cb", "planned_active_tokens": 66839, "realized_active_tokens": 66839, "cumulative_start": 4653112, "cumulative_end": 4719951, "loss_sum": 15320.799278259277, "token_normalized_loss": 0.22921945687785988, "grad_norm_before_clip": 0.05518115311861038, "learning_rates_at_step": { "lora": 6.912820719601565e-05, "row_tokens": 6.912820719601565e-05 }, "learning_rates_after_step": { "lora": 6.824693331615431e-05, "row_tokens": 6.824693331615431e-05 } }, { "update_id": 72, "row_count": 51, "row_ids_sha256": "1e2e6b64b03504df26a7fe6e3f24e68517362a53674e29c906a10bb429db0204", "input_ids_vector_sha256": "30bce7e93d9236c35a480728656f3fe9c96b20045432ee7bf9085ff9d9d68e12", "labels_vector_sha256": "32fff4e3f6ff8a79f3f5111c7ca23280069553f261f99a3ea715147491692fb6", "planned_active_tokens": 65054, "realized_active_tokens": 65054, "cumulative_start": 4719951, "cumulative_end": 4785005, "loss_sum": 13723.007751464844, "token_normalized_loss": 0.2109479471126271, "grad_norm_before_clip": 0.05254993215203285, "learning_rates_at_step": { "lora": 6.824693331615431e-05, "row_tokens": 6.824693331615431e-05 }, "learning_rates_after_step": { "lora": 6.734608487040944e-05, "row_tokens": 6.734608487040944e-05 } }, { "update_id": 73, "row_count": 47, "row_ids_sha256": "390b7dd24d0ffbf25d49634f096718e68a234863d36a1710af2e93da9745bf5e", "input_ids_vector_sha256": "94acb64853e653891d528649fbf99ad959b3749103f9c074980a3ea78a73a132", "labels_vector_sha256": "2b639a8e1cfce3e2f3ecea441e5c3d05471b919461218af2cc191eeee5b4b829", "planned_active_tokens": 66002, "realized_active_tokens": 66002, "cumulative_start": 4785005, "cumulative_end": 4851007, "loss_sum": 14385.636413574219, "token_normalized_loss": 0.21795758330920606, "grad_norm_before_clip": 0.051704391837120056, "learning_rates_at_step": { "lora": 6.734608487040944e-05, "row_tokens": 6.734608487040944e-05 }, "learning_rates_after_step": { "lora": 6.644636128445993e-05, "row_tokens": 6.644636128445993e-05 } }, { "update_id": 74, "row_count": 56, "row_ids_sha256": "2f112a3cc56b5492ae5ac57f543030cc04871f3bd855b48120b2bc5f18981be2", "input_ids_vector_sha256": "9a138727f3d5e8c6d0d88f2f7a9d62a34b9af10ed4d9772486c48b6961678737", "labels_vector_sha256": "6a2376d9f2cc8e7127c81d3efb8c4beef511dcc48b63d0d05d3e0102ef5d2af9", "planned_active_tokens": 65453, "realized_active_tokens": 65453, "cumulative_start": 4851007, "cumulative_end": 4916460, "loss_sum": 15793.368549346924, "token_normalized_loss": 0.24129327226172861, "grad_norm_before_clip": 0.06016969308257103, "learning_rates_at_step": { "lora": 6.644636128445993e-05, "row_tokens": 6.644636128445993e-05 }, "learning_rates_after_step": { "lora": 6.555234186720564e-05, "row_tokens": 6.555234186720564e-05 } }, { "update_id": 75, "row_count": 47, "row_ids_sha256": "b9b26ce455517fa711debb9bb4ac204744593c4e5c060361af5c300ffbbf0125", "input_ids_vector_sha256": "12e90b9875464dfe0239e459d210f05a2f33606d7a41e003d6705d0cb3612801", "labels_vector_sha256": "6574e0666da9ad93c8bc478c3c980c99cd14902599c90a822ede23bbf4e57349", "planned_active_tokens": 64609, "realized_active_tokens": 64609, "cumulative_start": 4916460, "cumulative_end": 4981069, "loss_sum": 13430.574440002441, "token_normalized_loss": 0.20787466823511339, "grad_norm_before_clip": 0.0565933957695961, "learning_rates_at_step": { "lora": 6.555234186720564e-05, "row_tokens": 6.555234186720564e-05 }, "learning_rates_after_step": { "lora": 6.463099710335184e-05, "row_tokens": 6.463099710335184e-05 } }, { "update_id": 76, "row_count": 43, "row_ids_sha256": "611fea21756b647c3ff635009270da3bc0a496075cdfbb71d1bc7750da77d9ce", "input_ids_vector_sha256": "74cceb959bef59dc8f00887a986160bff20eaf8c7718446421e7b4ae12fee7e5", "labels_vector_sha256": "b34eec1d12ab66e25741d3d02345c05d7d61ce0acaf8f0ed4c6a129f4379c7ab", "planned_active_tokens": 66169, "realized_active_tokens": 66169, "cumulative_start": 4981069, "cumulative_end": 5047238, "loss_sum": 14664.778610229492, "token_normalized_loss": 0.22162611812524735, "grad_norm_before_clip": 0.05549653246998787, "learning_rates_at_step": { "lora": 6.463099710335184e-05, "row_tokens": 6.463099710335184e-05 }, "learning_rates_after_step": { "lora": 6.370209345283557e-05, "row_tokens": 6.370209345283557e-05 } }, { "update_id": 77, "row_count": 49, "row_ids_sha256": "4d1fc5f469e31104b1b442e4b3ecbfe702627278a206547619dce63b87e9ad60", "input_ids_vector_sha256": "57c7d87b19347b54b74e375d1f52043cd4aa6fdb607ca1ef1dadac86b94e68be", "labels_vector_sha256": "a30ef100073e3992b9640dc2a5d291e05a2159a273a9f908ba777d4cade8859f", "planned_active_tokens": 66318, "realized_active_tokens": 66318, "cumulative_start": 5047238, "cumulative_end": 5113556, "loss_sum": 15501.397819519043, "token_normalized_loss": 0.23374344551281767, "grad_norm_before_clip": 0.05308637022972107, "learning_rates_at_step": { "lora": 6.370209345283557e-05, "row_tokens": 6.370209345283557e-05 }, "learning_rates_after_step": { "lora": 6.278167437368811e-05, "row_tokens": 6.278167437368811e-05 } }, { "update_id": 78, "row_count": 55, "row_ids_sha256": "d6fb9dcd607b152d73b99d82d86cce607177c07cb71d7c61f52cf4a97ae01485", "input_ids_vector_sha256": "095d53a193dd10268f39a8cca3f18bf940c64e4f1bf4810dc8fad2fe3a0ef912", "labels_vector_sha256": "87a00ea3422c7a8441b8604cd2c6b1ae2df88296e364fc3757871eaee39d114c", "planned_active_tokens": 65353, "realized_active_tokens": 65353, "cumulative_start": 5113556, "cumulative_end": 5178909, "loss_sum": 14439.421787261963, "token_normalized_loss": 0.22094504899946388, "grad_norm_before_clip": 0.05024215206503868, "learning_rates_at_step": { "lora": 6.278167437368811e-05, "row_tokens": 6.278167437368811e-05 }, "learning_rates_after_step": { "lora": 6.187375101182396e-05, "row_tokens": 6.187375101182396e-05 } }, { "update_id": 79, "row_count": 72, "row_ids_sha256": "de043b1a342579573d9cd5655efdce68b210a320f3d41db6c805bc00c4c19d38", "input_ids_vector_sha256": "f60ebe84a7ac051e6c74d558dd08736bb5ad74987e5e99b5ab33f781003d9a51", "labels_vector_sha256": "f33b7d42dcc372b69b4fbe444d87f68fce18c5720ef8f7afd681fb0c7cd9df17", "planned_active_tokens": 64144, "realized_active_tokens": 64144, "cumulative_start": 5178909, "cumulative_end": 5243053, "loss_sum": 14542.30244064331, "token_normalized_loss": 0.22671337055131127, "grad_norm_before_clip": 0.05507722869515419, "learning_rates_at_step": { "lora": 6.187375101182396e-05, "row_tokens": 6.187375101182396e-05 }, "learning_rates_after_step": { "lora": 6.094336097185753e-05, "row_tokens": 6.094336097185753e-05 } }, { "update_id": 80, "row_count": 71, "row_ids_sha256": "fa5f81f7f95aca5a08c4772eaad27c485757d9514dfbb7cf6be86febe25fe2e9", "input_ids_vector_sha256": "a006087e8ba184738c766b0b1e919a5efef26b1db6fb19e684d42a0db01bc1d8", "labels_vector_sha256": "593853d0bef61f142216ea44f500ed8836390e9cf7e4012f5a088d4f896a31b6", "planned_active_tokens": 65428, "realized_active_tokens": 65428, "cumulative_start": 5243053, "cumulative_end": 5308481, "loss_sum": 14436.13883972168, "token_normalized_loss": 0.2206416035905374, "grad_norm_before_clip": 0.05398823693394661, "learning_rates_at_step": { "lora": 6.094336097185753e-05, "row_tokens": 6.094336097185753e-05 }, "learning_rates_after_step": { "lora": 6.0007785935576964e-05, "row_tokens": 6.0007785935576964e-05 } }, { "update_id": 81, "row_count": 70, "row_ids_sha256": "91c8e6eb328170a3718569ad777397164855c6c1f5d1544cf4b9979b3a8a0da9", "input_ids_vector_sha256": "e10ed3420a274a7bc75ab599f0a8ccf49f91a85324b992a4442bc5b64e29adab", "labels_vector_sha256": "883f757081b0d237835e474c7c2f6ad2d44b084c4f1ccc6ce7e73d5a8704ee04", "planned_active_tokens": 65511, "realized_active_tokens": 65511, "cumulative_start": 5308481, "cumulative_end": 5373992, "loss_sum": 15145.5953540802, "token_normalized_loss": 0.23119163734457115, "grad_norm_before_clip": 0.05249832198023796, "learning_rates_at_step": { "lora": 6.0007785935576964e-05, "row_tokens": 6.0007785935576964e-05 }, "learning_rates_after_step": { "lora": 5.9066621392470025e-05, "row_tokens": 5.9066621392470025e-05 } }, { "update_id": 82, "row_count": 69, "row_ids_sha256": "2cb2be5f2244cdf359ba8ea3aa05bb950409cdaa41d70a85a885ed8ea828805a", "input_ids_vector_sha256": "2ca2b95c85539020e8ceb59b9c2961a1e51ba92ebba3b9aeb185f541a0566045", "labels_vector_sha256": "79cea892a1b397a9aaf403239e305448e6929c0ca7c6aa9681a1dcafd1396c66", "planned_active_tokens": 65645, "realized_active_tokens": 65645, "cumulative_start": 5373992, "cumulative_end": 5439637, "loss_sum": 17087.394828796387, "token_normalized_loss": 0.26030002024215687, "grad_norm_before_clip": 0.05664542689919472, "learning_rates_at_step": { "lora": 5.9066621392470025e-05, "row_tokens": 5.9066621392470025e-05 }, "learning_rates_after_step": { "lora": 5.8115738686641926e-05, "row_tokens": 5.8115738686641926e-05 } }, { "update_id": 83, "row_count": 66, "row_ids_sha256": "f63490ccb9bee55a6f9c0b86b0020f6b328d4058de113d1f580d18ae3a435cc6", "input_ids_vector_sha256": "b70aa04cb28eb8c684cf9aa279a647124469dcbed5e9559ae090e8a485e093b9", "labels_vector_sha256": "fc0358736bf94cbdf2d820feab04f5d54da8a9a32096afa49ae8439f59f6b7fc", "planned_active_tokens": 66088, "realized_active_tokens": 66088, "cumulative_start": 5439637, "cumulative_end": 5505725, "loss_sum": 15596.831939697266, "token_normalized_loss": 0.23600096749330082, "grad_norm_before_clip": 0.053290873765945435, "learning_rates_at_step": { "lora": 5.8115738686641926e-05, "row_tokens": 5.8115738686641926e-05 }, "learning_rates_after_step": { "lora": 5.717523095845843e-05, "row_tokens": 5.717523095845843e-05 } }, { "update_id": 84, "row_count": 69, "row_ids_sha256": "b6cdc0731f25eff598dc5d0e9a1bed9b6de264af4cdfdda625f8445e3940236d", "input_ids_vector_sha256": "c84d0aa861521ae117e706cc936debe76753c1f8233c1930bf6700b61bfafc23", "labels_vector_sha256": "e8634397d522caea276f31139115f15a309bb14e698468c24336b7902692a726", "planned_active_tokens": 65161, "realized_active_tokens": 65161, "cumulative_start": 5505725, "cumulative_end": 5570886, "loss_sum": 13740.693927764893, "token_normalized_loss": 0.21087297505816197, "grad_norm_before_clip": 0.05295582115650177, "learning_rates_at_step": { "lora": 5.717523095845843e-05, "row_tokens": 5.717523095845843e-05 }, "learning_rates_after_step": { "lora": 5.6222240231654236e-05, "row_tokens": 5.6222240231654236e-05 } }, { "update_id": 85, "row_count": 65, "row_ids_sha256": "ed5a28796affb7d848bd1083597fe21c308ad84722c8ea7320e9559bca64b10d", "input_ids_vector_sha256": "d643874744c9499238358fd40f9925a87059be011a353c30793f6be67565a7b5", "labels_vector_sha256": "26af0ea0b628a5b9aebb5c7be5a4628954d74a1feb00738bafce174f12e38859", "planned_active_tokens": 65843, "realized_active_tokens": 65843, "cumulative_start": 5570886, "cumulative_end": 5636729, "loss_sum": 15370.643424987793, "token_normalized_loss": 0.23344385014333782, "grad_norm_before_clip": 0.053080227226018906, "learning_rates_at_step": { "lora": 5.6222240231654236e-05, "row_tokens": 5.6222240231654236e-05 }, "learning_rates_after_step": { "lora": 5.5256446584117946e-05, "row_tokens": 5.5256446584117946e-05 } }, { "update_id": 86, "row_count": 62, "row_ids_sha256": "9baa90a3224bdb5ac6acec6206fea7223f9a74d8879293478f40515c805963f5", "input_ids_vector_sha256": "73547b2eca4df955157f0371def355b8df6ac815201451c58ab65d72f9d6843e", "labels_vector_sha256": "3aac2ccd5acd41f4a3f4d20d286b16abd75b40e697e156e6ef641ce7111ecd2b", "planned_active_tokens": 66566, "realized_active_tokens": 66566, "cumulative_start": 5636729, "cumulative_end": 5703295, "loss_sum": 15591.915410995483, "token_normalized_loss": 0.23423242212233697, "grad_norm_before_clip": 0.05596103146672249, "learning_rates_at_step": { "lora": 5.5256446584117946e-05, "row_tokens": 5.5256446584117946e-05 }, "learning_rates_after_step": { "lora": 5.432783757482711e-05, "row_tokens": 5.432783757482711e-05 } }, { "update_id": 87, "row_count": 60, "row_ids_sha256": "5557c61c07cba2da2a62581ace3956f1124543a32db6d1c63282655dc706c4b8", "input_ids_vector_sha256": "16e33e53350dda65769b62cf1ce42b93af88b5a8afedf4d945615a162788fdb6", "labels_vector_sha256": "fe21d620bfd6195292c846a38c0bed9a210fab6f3be5b211425584a5295f3db5", "planned_active_tokens": 63874, "realized_active_tokens": 63874, "cumulative_start": 5703295, "cumulative_end": 5767169, "loss_sum": 18388.72794342041, "token_normalized_loss": 0.2878906588505559, "grad_norm_before_clip": 0.06538720428943634, "learning_rates_at_step": { "lora": 5.432783757482711e-05, "row_tokens": 5.432783757482711e-05 }, "learning_rates_after_step": { "lora": 5.337292026844935e-05, "row_tokens": 5.337292026844935e-05 } }, { "update_id": 88, "row_count": 66, "row_ids_sha256": "606880848430a6e83226c3a3a7052e8f9fb1c4e47433fd6d7b593c2eedb3c9b4", "input_ids_vector_sha256": "15ef7e2dcddd6ac7aff154c8ddf6e48470ecdbb3d80db2f922e9058f01ba37c6", "labels_vector_sha256": "a5a0763f449b0edecb06f6f50154a705795b80813775f4e713745ccdba0a407c", "planned_active_tokens": 65576, "realized_active_tokens": 65576, "cumulative_start": 5767169, "cumulative_end": 5832745, "loss_sum": 15537.892261505127, "token_normalized_loss": 0.23694480086472378, "grad_norm_before_clip": 0.05881265178322792, "learning_rates_at_step": { "lora": 5.337292026844935e-05, "row_tokens": 5.337292026844935e-05 }, "learning_rates_after_step": { "lora": 5.2411557245439344e-05, "row_tokens": 5.2411557245439344e-05 } }, { "update_id": 89, "row_count": 62, "row_ids_sha256": "385d39e7f5112420b3ce527d2c8230450903a2b4a5de70e164d595758740fa61", "input_ids_vector_sha256": "e9ba1b38784bba7fa27f20f222dde28d94406e77d7cf7bf3a10ca85ce0e134ad", "labels_vector_sha256": "edffe80d40e929aa479796c9fb27c63bb28e25a9753ad025a820216e0bcb4cdf", "planned_active_tokens": 65933, "realized_active_tokens": 65933, "cumulative_start": 5832745, "cumulative_end": 5898678, "loss_sum": 14869.551536560059, "token_normalized_loss": 0.2255251776281992, "grad_norm_before_clip": 0.06194888800382614, "learning_rates_at_step": { "lora": 5.2411557245439344e-05, "row_tokens": 5.2411557245439344e-05 }, "learning_rates_after_step": { "lora": 5.1460585047788825e-05, "row_tokens": 5.1460585047788825e-05 } }, { "update_id": 90, "row_count": 62, "row_ids_sha256": "0fb8b1ff5399d9763449d3476262c30c2266101c095325089b5e3f79369bdf71", "input_ids_vector_sha256": "20e1eaf41a964201e7bcf35753bcfd10c798f87e075276c5a6e6c449959d2c50", "labels_vector_sha256": "f4f176aa255cd641c71aaf1a62ad47865057b3f89479b0baf485fa454ae49d17", "planned_active_tokens": 65160, "realized_active_tokens": 65160, "cumulative_start": 5898678, "cumulative_end": 5963838, "loss_sum": 16552.224716186523, "token_normalized_loss": 0.2540243203834641, "grad_norm_before_clip": 0.06755372881889343, "learning_rates_at_step": { "lora": 5.1460585047788825e-05, "row_tokens": 5.1460585047788825e-05 }, "learning_rates_after_step": { "lora": 5.0502409291847497e-05, "row_tokens": 5.0502409291847497e-05 } }, { "update_id": 91, "row_count": 64, "row_ids_sha256": "d3b7e949cf6919a0f07db041c7f838a3b4b9ee0ca39ba49c000468074ad824cb", "input_ids_vector_sha256": "07c63c9874bc7e5e2faea30a1ed9bba76284d811b1223c9538d586e7a80590a4", "labels_vector_sha256": "66b0640a7db0a48b792e837ed7a7dbf83a82266a0c0cfc1b3cee44f295a4a606", "planned_active_tokens": 65617, "realized_active_tokens": 65617, "cumulative_start": 5963838, "cumulative_end": 6029455, "loss_sum": 16769.563850402832, "token_normalized_loss": 0.2555673659326521, "grad_norm_before_clip": 0.06476084142923355, "learning_rates_at_step": { "lora": 5.0502409291847497e-05, "row_tokens": 5.0502409291847497e-05 }, "learning_rates_after_step": { "lora": 4.954413658973792e-05, "row_tokens": 4.954413658973792e-05 } }, { "update_id": 92, "row_count": 70, "row_ids_sha256": "e178b531ef74750a7569ec61b51e1798b025a2c8e5815ff36e81e6264af626ec", "input_ids_vector_sha256": "85c8f1e9c3682329e48415f38b025f64f03fac6ce085251ef615b98b7063c709", "labels_vector_sha256": "a526624c2cee3c3d8f59020403788ee7608eadd3721382e720ff06ef893366d7", "planned_active_tokens": 65611, "realized_active_tokens": 65611, "cumulative_start": 6029455, "cumulative_end": 6095066, "loss_sum": 16073.283752441406, "token_normalized_loss": 0.2449784906866441, "grad_norm_before_clip": 0.060915134847164154, "learning_rates_at_step": { "lora": 4.954413658973792e-05, "row_tokens": 4.954413658973792e-05 }, "learning_rates_after_step": { "lora": 4.857995784731376e-05, "row_tokens": 4.857995784731376e-05 } }, { "update_id": 93, "row_count": 64, "row_ids_sha256": "b03887b7ba818319abcc634889679205fcf126f9e01b9823e150a36c14d650ca", "input_ids_vector_sha256": "dcebbcbce0c237bdd2efdd5ad9bdfe1dd26ebae5f811b90d7d76ab5836a450c3", "labels_vector_sha256": "99281bfc2d7b2519e1acb3b5f1e6e8c0055fabb9679ccdc226899db0281671a1", "planned_active_tokens": 66027, "realized_active_tokens": 66027, "cumulative_start": 6095066, "cumulative_end": 6161093, "loss_sum": 15217.403266906738, "token_normalized_loss": 0.23047243198853104, "grad_norm_before_clip": 0.05744441971182823, "learning_rates_at_step": { "lora": 4.857995784731376e-05, "row_tokens": 4.857995784731376e-05 }, "learning_rates_after_step": { "lora": 4.763289514881747e-05, "row_tokens": 4.763289514881747e-05 } }, { "update_id": 94, "row_count": 59, "row_ids_sha256": "69da673b7e81059d8374e069bdd07ace3354c70274293764e2b2c67800a0d661", "input_ids_vector_sha256": "f7fa302df05fc0d7429c00051cde3c0cc03db921c75e5cdbfc0d689cf2349a4c", "labels_vector_sha256": "c62c60f09c2a7c6ad87f19d6f76517e6d7c7da45647fc01cb28fe77c0a6704de", "planned_active_tokens": 64890, "realized_active_tokens": 64890, "cumulative_start": 6161093, "cumulative_end": 6225983, "loss_sum": 14584.092697143555, "token_normalized_loss": 0.22475100473329565, "grad_norm_before_clip": 0.05633697658777237, "learning_rates_at_step": { "lora": 4.763289514881747e-05, "row_tokens": 4.763289514881747e-05 }, "learning_rates_after_step": { "lora": 4.667636496038114e-05, "row_tokens": 4.667636496038114e-05 } }, { "update_id": 95, "row_count": 64, "row_ids_sha256": "0ebe39094e75f5140319443fc82ae1a386a118a7f70434e835d9c074e714f0da", "input_ids_vector_sha256": "f2f7040b12e6c73d883d465feb74d10c19641c0048f32b2c7f3f7687860916f7", "labels_vector_sha256": "d0694d8bc7ba9d76a8fd59ebd86c19cc1de620749a2e08b11557b6a4ac0b262a", "planned_active_tokens": 65598, "realized_active_tokens": 65598, "cumulative_start": 6225983, "cumulative_end": 6291581, "loss_sum": 13136.748161315918, "token_normalized_loss": 0.20026141286801302, "grad_norm_before_clip": 0.05697782710194588, "learning_rates_at_step": { "lora": 4.667636496038114e-05, "row_tokens": 4.667636496038114e-05 }, "learning_rates_after_step": { "lora": 4.571159637926643e-05, "row_tokens": 4.571159637926643e-05 } }, { "update_id": 96, "row_count": 63, "row_ids_sha256": "a8d024847fd7ad74b104fe146bd9647cd51f2b8f42a0ef812d129d1238f356cb", "input_ids_vector_sha256": "b083913fa8666f066c2e3b8ada090f27f328d7aa3b0022d234a2c8a8e64e67fc", "labels_vector_sha256": "59a5aa497c06dd950b2c8a443164fd5887bf28b50e37d0e726845e4d8fa0613f", "planned_active_tokens": 66248, "realized_active_tokens": 66248, "cumulative_start": 6291581, "cumulative_end": 6357829, "loss_sum": 15527.492401123047, "token_normalized_loss": 0.23438431954357938, "grad_norm_before_clip": 0.06094866245985031, "learning_rates_at_step": { "lora": 4.571159637926643e-05, "row_tokens": 4.571159637926643e-05 }, "learning_rates_after_step": { "lora": 4.477032311773695e-05, "row_tokens": 4.477032311773695e-05 } }, { "update_id": 97, "row_count": 63, "row_ids_sha256": "c26b0086921c32e38f0c91006645c5af6dc288e2c667806dbcb3126272a46607", "input_ids_vector_sha256": "1902d6005413140ea92d46a2724308ee0df2def83824e96133439f4260fcfcac", "labels_vector_sha256": "bd117aed6fff8e046d0a4ae1c5d44d8d768b296836166bb48c6f9f719c23a1c0", "planned_active_tokens": 64741, "realized_active_tokens": 64741, "cumulative_start": 6357829, "cumulative_end": 6422570, "loss_sum": 14971.440120697021, "token_normalized_loss": 0.23125129548040688, "grad_norm_before_clip": 0.06062989681959152, "learning_rates_at_step": { "lora": 4.477032311773695e-05, "row_tokens": 4.477032311773695e-05 }, "learning_rates_after_step": { "lora": 4.380986097156624e-05, "row_tokens": 4.380986097156624e-05 } }, { "update_id": 98, "row_count": 64, "row_ids_sha256": "d9bd06be6c1706f0e8e090a0eeea6adbef2c31b7f49df8392ed2d148df50f599", "input_ids_vector_sha256": "81e06ce007d3cc15fdb2c199e49491a50e5161ba26588d2efcf8eb158c021f9a", "labels_vector_sha256": "e46994cae5cdeb17de0bea05808c34e4b47d11fc617acf83f58a9fa4922235e3", "planned_active_tokens": 66194, "realized_active_tokens": 66194, "cumulative_start": 6422570, "cumulative_end": 6488764, "loss_sum": 13385.710548400879, "token_normalized_loss": 0.20221939372754144, "grad_norm_before_clip": 0.05466971918940544, "learning_rates_at_step": { "lora": 4.380986097156624e-05, "row_tokens": 4.380986097156624e-05 }, "learning_rates_after_step": { "lora": 4.28577556165354e-05, "row_tokens": 4.28577556165354e-05 } }, { "update_id": 99, "row_count": 66, "row_ids_sha256": "075ea35440abcf86cce5f913293d3a4cac50626059267291eae528d932e29a4c", "input_ids_vector_sha256": "f5ea2f243e20cf6d7942637e99086a2671a320d6516f5d13f27399c7637ca358", "labels_vector_sha256": "2df0903c1cb5ebffd60e446eb73c7a0b56f6c70b3dd9e46cfa003d0695c4017c", "planned_active_tokens": 65776, "realized_active_tokens": 65776, "cumulative_start": 6488764, "cumulative_end": 6554540, "loss_sum": 12451.814453125, "token_normalized_loss": 0.1893063496279038, "grad_norm_before_clip": 0.0533260777592659, "learning_rates_at_step": { "lora": 4.28577556165354e-05, "row_tokens": 4.28577556165354e-05 }, "learning_rates_after_step": { "lora": 4.1906787969619934e-05, "row_tokens": 4.1906787969619934e-05 } }, { "update_id": 100, "row_count": 64, "row_ids_sha256": "e700a5cf4712305eaea42947a86ec316a0e79a5f8ca1ef8cd24d0dfbcc17bca9", "input_ids_vector_sha256": "2b012653187f24b20976622ed06fbafa5a07a1540b0d08e02d4f4b188a90a7b4", "labels_vector_sha256": "bb673b93faddd5d484e232d4fa15702817164e200870d21a6b9b3937a7dc1437", "planned_active_tokens": 65880, "realized_active_tokens": 65880, "cumulative_start": 6554540, "cumulative_end": 6620420, "loss_sum": 12411.509559631348, "token_normalized_loss": 0.18839571280557602, "grad_norm_before_clip": 0.05825131759047508, "learning_rates_at_step": { "lora": 4.1906787969619934e-05, "row_tokens": 4.1906787969619934e-05 }, "learning_rates_after_step": { "lora": 4.097835438054806e-05, "row_tokens": 4.097835438054806e-05 } }, { "update_id": 101, "row_count": 69, "row_ids_sha256": "0de81386c60ac20fa76ea429b227b0eb4527ca9af2195d292f5e60c22d41e145", "input_ids_vector_sha256": "2278c0e6c92d4e969831f52c1d8b7685c65a8b5c2ae4b1ed8ed509fe31f89f32", "labels_vector_sha256": "f3b89e11181d56aed3da9a228908346fef916b555d23cfa696fcf044225444b4", "planned_active_tokens": 64520, "realized_active_tokens": 64520, "cumulative_start": 6620420, "cumulative_end": 6684940, "loss_sum": 14939.68977355957, "token_normalized_loss": 0.2315512984122686, "grad_norm_before_clip": 0.06096624583005905, "learning_rates_at_step": { "lora": 4.097835438054806e-05, "row_tokens": 4.097835438054806e-05 }, "learning_rates_after_step": { "lora": 4.0031627133108116e-05, "row_tokens": 4.0031627133108116e-05 } }, { "update_id": 102, "row_count": 70, "row_ids_sha256": "54d8382aedc1a8f3cba5ac1e62ed4872dff330b29dc67888391fb9967fb3e2c7", "input_ids_vector_sha256": "ab32155f1ef64bd9bc7857a204ce1e1577f4ec54682b9647484eb94c0de21384", "labels_vector_sha256": "692479189944c0270f08b267ce926660a4c90cbf7a1d4106c8aff367b6499b85", "planned_active_tokens": 66022, "realized_active_tokens": 66022, "cumulative_start": 6684940, "cumulative_end": 6750962, "loss_sum": 15076.444633483887, "token_normalized_loss": 0.22835486100820768, "grad_norm_before_clip": 0.05613531917333603, "learning_rates_at_step": { "lora": 4.0031627133108116e-05, "row_tokens": 4.0031627133108116e-05 }, "learning_rates_after_step": { "lora": 3.9097259516638026e-05, "row_tokens": 3.9097259516638026e-05 } }, { "update_id": 103, "row_count": 64, "row_ids_sha256": "589c9de910480092e2c8c03bab757c0816084941a9b424a9ba5562143d9e8fbc", "input_ids_vector_sha256": "49ed642cfb29fcbcab23f2449a24a2a0642d0ede107a6c6b601d763470e8c6ad", "labels_vector_sha256": "79bb2a1283b5140ccc3cba48671953381a23c10f0aeeea056d02923f0be16103", "planned_active_tokens": 65415, "realized_active_tokens": 65415, "cumulative_start": 6750962, "cumulative_end": 6816377, "loss_sum": 13168.647598266602, "token_normalized_loss": 0.20130929600652148, "grad_norm_before_clip": 0.056113243103027344, "learning_rates_at_step": { "lora": 3.9097259516638026e-05, "row_tokens": 3.9097259516638026e-05 }, "learning_rates_after_step": { "lora": 3.816677341214169e-05, "row_tokens": 3.816677341214169e-05 } }, { "update_id": 104, "row_count": 66, "row_ids_sha256": "a50f1c1b2a92a4fa17ec693eef35239773ff389c87b71f0f6d20166efae6a241", "input_ids_vector_sha256": "30d69d3e0f363d4231c4b2a452bc6c27c2dc56b1d5bac87ce4093830b6c77358", "labels_vector_sha256": "7c6e13adcec715da03b8ff6f7230289ce16d0fdb84afb555319a5dcbf5ac1ead", "planned_active_tokens": 65422, "realized_active_tokens": 65422, "cumulative_start": 6816377, "cumulative_end": 6881799, "loss_sum": 11327.32543182373, "token_normalized_loss": 0.1731424510382399, "grad_norm_before_clip": 0.056043967604637146, "learning_rates_at_step": { "lora": 3.816677341214169e-05, "row_tokens": 3.816677341214169e-05 }, "learning_rates_after_step": { "lora": 3.723970501415329e-05, "row_tokens": 3.723970501415329e-05 } }, { "update_id": 105, "row_count": 66, "row_ids_sha256": "e5bb36ada9195f1eafdff1dfac8ac75d41f592f5d255e25e7b4690119e307901", "input_ids_vector_sha256": "fc24f1cfb3f6ce55bac9e86b7fc9fd4b42acce833aae299a27fa34385363f11e", "labels_vector_sha256": "9dcc3fb2b0a376a4390d7d525c51c8b8638bc71ab40d78f29589f07986ade370", "planned_active_tokens": 65486, "realized_active_tokens": 65486, "cumulative_start": 6881799, "cumulative_end": 6947285, "loss_sum": 14285.210441589355, "token_normalized_loss": 0.21814144155375736, "grad_norm_before_clip": 0.05943690612912178, "learning_rates_at_step": { "lora": 3.723970501415329e-05, "row_tokens": 3.723970501415329e-05 }, "learning_rates_after_step": { "lora": 3.632288299448241e-05, "row_tokens": 3.632288299448241e-05 } }, { "update_id": 106, "row_count": 68, "row_ids_sha256": "f77c2ac0e2252f7e6a668fda7d3eb5f2807614e0b48d6fd5b5a8f0b4c11818ba", "input_ids_vector_sha256": "d06d558ac658f594e16a6319a70b60f4519c4a1e4b08dffee57e661f02571525", "labels_vector_sha256": "7a9518a927afb17769a8af423573a2324a65ba4054d43dea4dd2e09a83775e25", "planned_active_tokens": 65089, "realized_active_tokens": 65089, "cumulative_start": 6947285, "cumulative_end": 7012374, "loss_sum": 14929.785842895508, "token_normalized_loss": 0.22937494573423325, "grad_norm_before_clip": 0.060550641268491745, "learning_rates_at_step": { "lora": 3.632288299448241e-05, "row_tokens": 3.632288299448241e-05 }, "learning_rates_after_step": { "lora": 3.540332183107734e-05, "row_tokens": 3.540332183107734e-05 } }, { "update_id": 107, "row_count": 67, "row_ids_sha256": "bec6ef197c5441abca8b4172e84d3e906de2c118271c2cb6b88eeb5a436870dc", "input_ids_vector_sha256": "492dc96b9894e4af9ccbf4fc841cea6fe9b9dd64b974eb6fd267f43945f33fda", "labels_vector_sha256": "9b919bb3897f7ea7eeed4781c7745b71e41323e093bcb36f6040c23f33f73cab", "planned_active_tokens": 65639, "realized_active_tokens": 65639, "cumulative_start": 7012374, "cumulative_end": 7078013, "loss_sum": 12683.062160491943, "token_normalized_loss": 0.19322448788817537, "grad_norm_before_clip": 0.053244173526763916, "learning_rates_at_step": { "lora": 3.540332183107734e-05, "row_tokens": 3.540332183107734e-05 }, "learning_rates_after_step": { "lora": 3.448812710027065e-05, "row_tokens": 3.448812710027065e-05 } }, { "update_id": 108, "row_count": 64, "row_ids_sha256": "b1feec97d6f011fb886aeb4c3f5eea1d2e8e61813781bc5acd651c0d66d6d763", "input_ids_vector_sha256": "e271ad6e3db9a8aa58c5354f868df4e757fe313ce428eca7ad8943ad6401b025", "labels_vector_sha256": "f48c5d81c6b5dd8b2bacb6f7740db2bfabcf4310e649deac07c508931fe148cc", "planned_active_tokens": 65711, "realized_active_tokens": 65711, "cumulative_start": 7078013, "cumulative_end": 7143724, "loss_sum": 11536.959907531738, "token_normalized_loss": 0.17557121193607977, "grad_norm_before_clip": 0.056289881467819214, "learning_rates_at_step": { "lora": 3.448812710027065e-05, "row_tokens": 3.448812710027065e-05 }, "learning_rates_after_step": { "lora": 3.357666096499278e-05, "row_tokens": 3.357666096499278e-05 } }, { "update_id": 109, "row_count": 67, "row_ids_sha256": "fcecde696d4769a35402c68fa987a23de7d90ac4524ba418a2ba4ff340ea1d70", "input_ids_vector_sha256": "b165bf002c218857661032bf533970823be06d0ccfd7d18503fb8bc90b098c6a", "labels_vector_sha256": "f14e598e972aa6821327a6125b3560471dff05870f537f7754fb7530bb08aebc", "planned_active_tokens": 65855, "realized_active_tokens": 65855, "cumulative_start": 7143724, "cumulative_end": 7209579, "loss_sum": 14507.54305267334, "token_normalized_loss": 0.2202952403412549, "grad_norm_before_clip": 0.06281700730323792, "learning_rates_at_step": { "lora": 3.357666096499278e-05, "row_tokens": 3.357666096499278e-05 }, "learning_rates_after_step": { "lora": 3.2668902172065616e-05, "row_tokens": 3.2668902172065616e-05 } }, { "update_id": 110, "row_count": 70, "row_ids_sha256": "65d1d91cc2d41f7975f02fc4973dcb2b03b90273a224b52eebc287bf5dffc084", "input_ids_vector_sha256": "38dd3d192142721133e97931ae2f75c2a6a367118a149f6584d6431359954576", "labels_vector_sha256": "179265b7405ad6cac423245ed383bbc3b828eb2276ce9faf8db23ed2e7778552", "planned_active_tokens": 66028, "realized_active_tokens": 66028, "cumulative_start": 7209579, "cumulative_end": 7275607, "loss_sum": 14852.402214050293, "token_normalized_loss": 0.22494096768113972, "grad_norm_before_clip": 0.059342991560697556, "learning_rates_at_step": { "lora": 3.2668902172065616e-05, "row_tokens": 3.2668902172065616e-05 }, "learning_rates_after_step": { "lora": 3.177613158267546e-05, "row_tokens": 3.177613158267546e-05 } }, { "update_id": 111, "row_count": 68, "row_ids_sha256": "3bedd80329d8d6e95053ad0015acb025e16883202187e70aa25366ee45de3ad6", "input_ids_vector_sha256": "e273880315d3264657b7493f2d6f8c6aa0b2ef5ee014517d42caf60cb45aa79f", "labels_vector_sha256": "c39c1c08ddcaaf67a23c24533079740767e8ca1bbf1a84ff7d7f1bb473944af3", "planned_active_tokens": 65400, "realized_active_tokens": 65400, "cumulative_start": 7275607, "cumulative_end": 7341007, "loss_sum": 11892.257598876953, "token_normalized_loss": 0.18183880120606963, "grad_norm_before_clip": 0.05480519309639931, "learning_rates_at_step": { "lora": 3.177613158267546e-05, "row_tokens": 3.177613158267546e-05 }, "learning_rates_after_step": { "lora": 3.089499230940841e-05, "row_tokens": 3.089499230940841e-05 } }, { "update_id": 112, "row_count": 66, "row_ids_sha256": "8f2e0b5b83784cda94261e3176159c251746b817a95273ec2a3f198dc4acea95", "input_ids_vector_sha256": "e7591af079a9a0fbbce66b61ed3e1386623e42125c7486a81d2cb4aaab118a66", "labels_vector_sha256": "5a76cc6a6f1acad8f0995861a7a162e2cc8d0c341862dfa7adb559796e01615a", "planned_active_tokens": 65031, "realized_active_tokens": 65031, "cumulative_start": 7341007, "cumulative_end": 7406038, "loss_sum": 21395.05453491211, "token_normalized_loss": 0.32899777851966155, "grad_norm_before_clip": 0.06811846047639847, "learning_rates_at_step": { "lora": 3.089499230940841e-05, "row_tokens": 3.089499230940841e-05 }, "learning_rates_after_step": { "lora": 3.0005029687766427e-05, "row_tokens": 3.0005029687766427e-05 } }, { "update_id": 113, "row_count": 61, "row_ids_sha256": "685618728dbcb2d6eafeab1b99d8c42bb7d30513bb92173c6487d6704ebd0ba9", "input_ids_vector_sha256": "fd390f02f9a6791cde798e1139bdcfabac0b7df0fb57a83118a27e455e07671a", "labels_vector_sha256": "00b1696cd4fa7e1136d459b2955f97e21cb950ff030a5db9326fb445906a17d8", "planned_active_tokens": 66205, "realized_active_tokens": 66205, "cumulative_start": 7406038, "cumulative_end": 7472243, "loss_sum": 20333.01313018799, "token_normalized_loss": 0.30712201691999075, "grad_norm_before_clip": 0.06504218280315399, "learning_rates_at_step": { "lora": 3.0005029687766427e-05, "row_tokens": 3.0005029687766427e-05 }, "learning_rates_after_step": { "lora": 2.9140343689528482e-05, "row_tokens": 2.9140343689528482e-05 } }, { "update_id": 114, "row_count": 67, "row_ids_sha256": "e7826d8c42b4e406e923b010d5192a2a99c59eca3fca7a9229f8b551dfe75b0a", "input_ids_vector_sha256": "239b2375076006d530654f19bb3eadb791c4e6cf1df48aa0bcccff9316ba6ea8", "labels_vector_sha256": "8dbf20fa23c00d72d46873eb58440ac41353671c977d4690833f26086f9e644e", "planned_active_tokens": 64864, "realized_active_tokens": 64864, "cumulative_start": 7472243, "cumulative_end": 7537107, "loss_sum": 14075.810207366943, "token_normalized_loss": 0.2170049674298061, "grad_norm_before_clip": 0.06465376913547516, "learning_rates_at_step": { "lora": 2.9140343689528482e-05, "row_tokens": 2.9140343689528482e-05 }, "learning_rates_after_step": { "lora": 2.8279488984430903e-05, "row_tokens": 2.8279488984430903e-05 } }, { "update_id": 115, "row_count": 73, "row_ids_sha256": "b8c9f4e480d145de6b714fcbf027a019f4d0c5a5a7e174377d268b17e2d581f0", "input_ids_vector_sha256": "8fea6b2368e37340daf15d26824c942e089a736acb2fb9d2a08bd8cbc0280aec", "labels_vector_sha256": "1129b7363c84a441d574132fb379677d6ecfb84ed97ec79f17bb1d5e078402b6", "planned_active_tokens": 65142, "realized_active_tokens": 65142, "cumulative_start": 7537107, "cumulative_end": 7602249, "loss_sum": 12664.333644866943, "token_normalized_loss": 0.19441118855526301, "grad_norm_before_clip": 0.06473589688539505, "learning_rates_at_step": { "lora": 2.8279488984430903e-05, "row_tokens": 2.8279488984430903e-05 }, "learning_rates_after_step": { "lora": 2.742157279537002e-05, "row_tokens": 2.742157279537002e-05 } }, { "update_id": 116, "row_count": 71, "row_ids_sha256": "cc62116ff5c8c7813e8c66ac57b629f9927e6842a5cf6ec0cb6e31b813f7d128", "input_ids_vector_sha256": "4a1e6aca0444d6ad60fc50ed479b4bcecfb49e2879dfda2279730b8302fd674a", "labels_vector_sha256": "8be949c2a338bb812de7bfbbd77372ca53f5293f95284044de5695df4bde72af", "planned_active_tokens": 65520, "realized_active_tokens": 65520, "cumulative_start": 7602249, "cumulative_end": 7667769, "loss_sum": 12016.830284118652, "token_normalized_loss": 0.18340705561841655, "grad_norm_before_clip": 0.05898382142186165, "learning_rates_at_step": { "lora": 2.742157279537002e-05, "row_tokens": 2.742157279537002e-05 }, "learning_rates_after_step": { "lora": 2.6570030282171764e-05, "row_tokens": 2.6570030282171764e-05 } }, { "update_id": 117, "row_count": 73, "row_ids_sha256": "350164cbb83643412f4dda93d1bd45a317b18fb0210b98c9d2492dcb1bc5002b", "input_ids_vector_sha256": "355b3c11af40e0948f1288b453df12ebb7846f471f2a482d69c19041ed1a7711", "labels_vector_sha256": "a31112c0f0f84f5526f5931406b5ae92c9d713dcae5751b61977031f42bfedef", "planned_active_tokens": 65667, "realized_active_tokens": 65667, "cumulative_start": 7667769, "cumulative_end": 7733436, "loss_sum": 11898.190662384033, "token_normalized_loss": 0.18118980100178222, "grad_norm_before_clip": 0.057161033153533936, "learning_rates_at_step": { "lora": 2.6570030282171764e-05, "row_tokens": 2.6570030282171764e-05 }, "learning_rates_after_step": { "lora": 2.5727159714644677e-05, "row_tokens": 2.5727159714644677e-05 } }, { "update_id": 118, "row_count": 67, "row_ids_sha256": "cb401275af207714fd2c0580437b6a9ea2c95868576b98daaea7739b27e73d6c", "input_ids_vector_sha256": "81563fcde8070d0f391386f56f7d3759aa757a4715620323b78cfd27802ad225", "labels_vector_sha256": "48d156d9abcab07bff5c3b4042ecd22562edfdda39ddd67c409e4ba87eed8c79", "planned_active_tokens": 65663, "realized_active_tokens": 65663, "cumulative_start": 7733436, "cumulative_end": 7799099, "loss_sum": 10404.778907775879, "token_normalized_loss": 0.15845725763026178, "grad_norm_before_clip": 0.05861957371234894, "learning_rates_at_step": { "lora": 2.5727159714644677e-05, "row_tokens": 2.5727159714644677e-05 }, "learning_rates_after_step": { "lora": 2.4891867577577693e-05, "row_tokens": 2.4891867577577693e-05 } }, { "update_id": 119, "row_count": 70, "row_ids_sha256": "3544eb1ee38ee9c6af908cb719cb27fb04a99185b4d73a8a65a7d4db41f21596", "input_ids_vector_sha256": "2ed5ae94f53a64fce71320f135d0082d1ab8a25d39c7da82417d7aae9f51c3b8", "labels_vector_sha256": "deb2a499d610c482afcda3d7a5fe050aff42d6107f2898f191b92f4839cf8417", "planned_active_tokens": 65770, "realized_active_tokens": 65770, "cumulative_start": 7799099, "cumulative_end": 7864869, "loss_sum": 12028.227264404297, "token_normalized_loss": 0.1828831878425467, "grad_norm_before_clip": 0.05768251419067383, "learning_rates_at_step": { "lora": 2.4891867577577693e-05, "row_tokens": 2.4891867577577693e-05 }, "learning_rates_after_step": { "lora": 2.406373245839954e-05, "row_tokens": 2.406373245839954e-05 } }, { "update_id": 120, "row_count": 73, "row_ids_sha256": "a99ef1908bc52ad2145b7890eda024133885adecfd1c60c8b5003804b25c8747", "input_ids_vector_sha256": "aba34491f3b5238b0b0249fafffaeba87465698548b53cc6f940a91cd3726652", "labels_vector_sha256": "6a36deca0b05cee5b8a34af0cb50136a9bdceb8f790f05cd8d49e4025f904d84", "planned_active_tokens": 65939, "realized_active_tokens": 65939, "cumulative_start": 7864869, "cumulative_end": 7930808, "loss_sum": 10169.787521362305, "token_normalized_loss": 0.15423023584467924, "grad_norm_before_clip": 0.058216195553541183, "learning_rates_at_step": { "lora": 2.406373245839954e-05, "row_tokens": 2.406373245839954e-05 }, "learning_rates_after_step": { "lora": 2.3256041498641724e-05, "row_tokens": 2.3256041498641724e-05 } }, { "update_id": 121, "row_count": 76, "row_ids_sha256": "962e335a12d5a001f58e2cb078cc8ced6c765dd4f7f15ac13f28dc55c511e2f4", "input_ids_vector_sha256": "03e84800bc721ce6f73a1f89d9de7945a9b611afbc132dd87b474b6fea80957b", "labels_vector_sha256": "009835ffc981b20f4b24f638f7a785bbd5974776aec89e4038699cd03aeaa5ab", "planned_active_tokens": 65062, "realized_active_tokens": 65062, "cumulative_start": 7930808, "cumulative_end": 7995870, "loss_sum": 10617.065353393555, "token_normalized_loss": 0.16318381472124366, "grad_norm_before_clip": 0.05630940943956375, "learning_rates_at_step": { "lora": 2.3256041498641724e-05, "row_tokens": 2.3256041498641724e-05 }, "learning_rates_after_step": { "lora": 2.244861250377864e-05, "row_tokens": 2.244861250377864e-05 } }, { "update_id": 122, "row_count": 72, "row_ids_sha256": "bd4dcb4979a97d151c4fc8d78085d33edbe96cad76b4e4687df93e13ac6e0f30", "input_ids_vector_sha256": "4b2c3b07be0b8811895824bf4eebfea3f2f2fd851e4e752746f86459a389acf4", "labels_vector_sha256": "ada0c135e38bec43b6eb9fcece23f47f130bc1852a45a0088e0406cea596f8be", "planned_active_tokens": 65833, "realized_active_tokens": 65833, "cumulative_start": 7995870, "cumulative_end": 8061703, "loss_sum": 10186.93236541748, "token_normalized_loss": 0.15473899663417254, "grad_norm_before_clip": 0.058929480612277985, "learning_rates_at_step": { "lora": 2.244861250377864e-05, "row_tokens": 2.244861250377864e-05 }, "learning_rates_after_step": { "lora": 2.166103393890323e-05, "row_tokens": 2.166103393890323e-05 } }, { "update_id": 123, "row_count": 68, "row_ids_sha256": "7779332fba5cfef5db6cd5fc46ee3b1e2a0715478f4f68111373ff217a0f8acf", "input_ids_vector_sha256": "5840807539254c6490288fd336540eeee4f8ecff94146a826009292628f06916", "labels_vector_sha256": "52153503504be3b3fd51889184bb1d99a2d9b31379e113a35d2e99d5b26e7ac1", "planned_active_tokens": 65030, "realized_active_tokens": 65030, "cumulative_start": 8061703, "cumulative_end": 8126733, "loss_sum": 9978.282817840576, "token_normalized_loss": 0.15344122432478205, "grad_norm_before_clip": 0.055067963898181915, "learning_rates_at_step": { "lora": 2.166103393890323e-05, "row_tokens": 2.166103393890323e-05 }, "learning_rates_after_step": { "lora": 2.0871998653006207e-05, "row_tokens": 2.0871998653006207e-05 } }, { "update_id": 124, "row_count": 71, "row_ids_sha256": "a197b292536012224dfd9250dd284cd4c25606ba38e34fcc6f0ba26346f5cb89", "input_ids_vector_sha256": "5ebf3bf3a99c533e09eaa2cbf6dc998189aa15c217bae1c1f2f36c8b6367cc96", "labels_vector_sha256": "99a2af4b66b4a033cb5496110dd376abfc40fad514f1ff2911e6fa4ba4804010", "planned_active_tokens": 66014, "realized_active_tokens": 66014, "cumulative_start": 8126733, "cumulative_end": 8192747, "loss_sum": 9733.11689376831, "token_normalized_loss": 0.14744019289496638, "grad_norm_before_clip": 0.055539824068546295, "learning_rates_at_step": { "lora": 2.0871998653006207e-05, "row_tokens": 2.0871998653006207e-05 }, "learning_rates_after_step": { "lora": 2.0107771899234783e-05, "row_tokens": 2.0107771899234783e-05 } }, { "update_id": 125, "row_count": 71, "row_ids_sha256": "029fd38b4efc9b8e1313eae33dfbae3c6ffdca0d6a111fe30a78642bf4757922", "input_ids_vector_sha256": "772b3adcced82721f63d02f3d616c3f851d2b79920b8989d11e465e674497991", "labels_vector_sha256": "049d4ddce991ddb9c17311b39f184169510d6e7cf383026f08def46133aec486", "planned_active_tokens": 64820, "realized_active_tokens": 64820, "cumulative_start": 8192747, "cumulative_end": 8257567, "loss_sum": 8987.342416763306, "token_normalized_loss": 0.1386507623690729, "grad_norm_before_clip": 0.0574454627931118, "learning_rates_at_step": { "lora": 2.0107771899234783e-05, "row_tokens": 2.0107771899234783e-05 }, "learning_rates_after_step": { "lora": 1.9342584118050345e-05, "row_tokens": 1.9342584118050345e-05 } }, { "update_id": 126, "row_count": 72, "row_ids_sha256": "111f569f4f01cb7919ecb3e616eb555c132df5c40f275b9f5c0e476a3c6fa73d", "input_ids_vector_sha256": "60412e4604abe702c7655218010ccb940563070f05b4706a2318fa0671f5f913", "labels_vector_sha256": "8614b22d1590f4bac3c889535d6d999cb5a5dfd9ebd4f14af59d774b0e6dc700", "planned_active_tokens": 65832, "realized_active_tokens": 65832, "cumulative_start": 8257567, "cumulative_end": 8323399, "loss_sum": 8977.555282592773, "token_normalized_loss": 0.13637069028121238, "grad_norm_before_clip": 0.05559604987502098, "learning_rates_at_step": { "lora": 1.9342584118050345e-05, "row_tokens": 1.9342584118050345e-05 }, "learning_rates_after_step": { "lora": 1.8594313098573013e-05, "row_tokens": 1.8594313098573013e-05 } }, { "update_id": 127, "row_count": 72, "row_ids_sha256": "63f6455d69f9d402fb8328ff3f2ab1d9c25f298ddfa20528db169ce7517b857b", "input_ids_vector_sha256": "217c12edd9bb2c5c8ce00ed360b0c1284e1207eb34909f1a61760b417902006e", "labels_vector_sha256": "bf42bd94beb978effbe405c7f1a2bd0cccd3309d7c1435f71deb0d98cdcd9e96", "planned_active_tokens": 65341, "realized_active_tokens": 65341, "cumulative_start": 8323399, "cumulative_end": 8388740, "loss_sum": 9994.487823486328, "token_normalized_loss": 0.15295890518183572, "grad_norm_before_clip": 0.06045476347208023, "learning_rates_at_step": { "lora": 1.8594313098573013e-05, "row_tokens": 1.8594313098573013e-05 }, "learning_rates_after_step": { "lora": 1.7854395386626138e-05, "row_tokens": 1.7854395386626138e-05 } }, { "update_id": 128, "row_count": 67, "row_ids_sha256": "be0436320e01c23787ff6f31509a2a14203e512f6b448eb646629c1956ae1627", "input_ids_vector_sha256": "b50937580fec694b45a6d1b0f5cff99071ee2ed76f7ad9f43bae3f2e72944aed", "labels_vector_sha256": "60d5076ee989a3a82bdc64093c30215b87463bd068e8c1795a22c3c5285ecc42", "planned_active_tokens": 65617, "realized_active_tokens": 65617, "cumulative_start": 8388740, "cumulative_end": 8454357, "loss_sum": 8424.772136688232, "token_normalized_loss": 0.12839313191228238, "grad_norm_before_clip": 0.05375078693032265, "learning_rates_at_step": { "lora": 1.7854395386626138e-05, "row_tokens": 1.7854395386626138e-05 }, "learning_rates_after_step": { "lora": 1.7116912164991384e-05, "row_tokens": 1.7116912164991384e-05 } }, { "update_id": 129, "row_count": 67, "row_ids_sha256": "ac7a21c1802d2891f434eea15ce00ee2f62ee17ee8a556e5bae302926aa348e1", "input_ids_vector_sha256": "c42c64891e055ef4f62b2a6b7d93a50a35d17bda6a48f707ee486cbf3a3903de", "labels_vector_sha256": "1c8e74c7e7fffe883cf1375664ec52b31653a0ef138355a6e8e9bc1dfae85fc5", "planned_active_tokens": 66469, "realized_active_tokens": 66469, "cumulative_start": 8454357, "cumulative_end": 8520826, "loss_sum": 11374.501007080078, "token_normalized_loss": 0.1711249004359939, "grad_norm_before_clip": 0.05693138763308525, "learning_rates_at_step": { "lora": 1.7116912164991384e-05, "row_tokens": 1.7116912164991384e-05 }, "learning_rates_after_step": { "lora": 1.6410321508814923e-05, "row_tokens": 1.6410321508814923e-05 } }, { "update_id": 130, "row_count": 74, "row_ids_sha256": "3c1b35b897a2c0680af8bd5801ae3678beb7de3428ca2556570e8e60494eeef3", "input_ids_vector_sha256": "45c103ead98a0f77ce4ed786a4901bc2a20dde323c9b5d619ba648cf90d3f5e3", "labels_vector_sha256": "a89d73ef7324aced01884d65a54a112bdecd1c03825852918a1716602448b600", "planned_active_tokens": 64759, "realized_active_tokens": 64759, "cumulative_start": 8520826, "cumulative_end": 8585585, "loss_sum": 12157.519046783447, "token_normalized_loss": 0.18773481750464718, "grad_norm_before_clip": 0.06209832802414894, "learning_rates_at_step": { "lora": 1.6410321508814923e-05, "row_tokens": 1.6410321508814923e-05 }, "learning_rates_after_step": { "lora": 1.570909581773237e-05, "row_tokens": 1.570909581773237e-05 } }, { "update_id": 131, "row_count": 71, "row_ids_sha256": "373ad9c269656c47aa16fc6a139bd54680d850d16d10da3a328f2e946cfec0c3", "input_ids_vector_sha256": "042c35599f7aa4e6fd18d2f6f58356de9ab81259376cb3deee116ff37fb15313", "labels_vector_sha256": "b1750e8fe1ffcc2e87daca78180ce6fd32af418db651edfcb6262c277d01d334", "planned_active_tokens": 65385, "realized_active_tokens": 65385, "cumulative_start": 8585585, "cumulative_end": 8650970, "loss_sum": 11561.814865112305, "token_normalized_loss": 0.17682671660338464, "grad_norm_before_clip": 0.05990307778120041, "learning_rates_at_step": { "lora": 1.570909581773237e-05, "row_tokens": 1.570909581773237e-05 }, "learning_rates_after_step": { "lora": 1.5016632617195059e-05, "row_tokens": 1.5016632617195059e-05 } }, { "update_id": 132, "row_count": 74, "row_ids_sha256": "9971e5843b959718ab119044abfc2ff02668f776317277151fbd47f8500d0089", "input_ids_vector_sha256": "6341a3ad40cd40996bea5659e96dfddf867867c0d63a712b3da3fb5a3ac946b7", "labels_vector_sha256": "fc1fa756283fd30e758fb349053b54c40485cf2cbe5bbadf9a0e91c5722d886d", "planned_active_tokens": 65744, "realized_active_tokens": 65744, "cumulative_start": 8650970, "cumulative_end": 8716714, "loss_sum": 11537.68090057373, "token_normalized_loss": 0.17549405117689418, "grad_norm_before_clip": 0.06132487952709198, "learning_rates_at_step": { "lora": 1.5016632617195059e-05, "row_tokens": 1.5016632617195059e-05 }, "learning_rates_after_step": { "lora": 1.4343419039893918e-05, "row_tokens": 1.4343419039893918e-05 } }, { "update_id": 133, "row_count": 69, "row_ids_sha256": "0b82a6f4ecf980505deeaa662f3a9d907777b6b6ed565a6dc4ca9d3f7a918423", "input_ids_vector_sha256": "46b48aa66b991386824834ee738578c6e4e69c84bb8ce1db0567206adb2dc30a", "labels_vector_sha256": "afdbdce5c4969a592436bf36383ea9e4ec1e6ec2abbba64dc8a210c88ca2852d", "planned_active_tokens": 65124, "realized_active_tokens": 65124, "cumulative_start": 8716714, "cumulative_end": 8781838, "loss_sum": 10876.983951568604, "token_normalized_loss": 0.16701959264738964, "grad_norm_before_clip": 0.06020372733473778, "learning_rates_at_step": { "lora": 1.4343419039893918e-05, "row_tokens": 1.4343419039893918e-05 }, "learning_rates_after_step": { "lora": 1.3677306937236784e-05, "row_tokens": 1.3677306937236784e-05 } }, { "update_id": 134, "row_count": 74, "row_ids_sha256": "7249bdd188770ceb340e50089a557eeb308ca4dd063e5b1da27735cd272435cd", "input_ids_vector_sha256": "72a92c2914a5af333361eb0a705966015a95d469ffdfab24419fcf8a206e0537", "labels_vector_sha256": "b0ef283c50c4aba47be23a8f63c5df5dc1003bd046ced14a2c9bccd8d190d286", "planned_active_tokens": 65702, "realized_active_tokens": 65702, "cumulative_start": 8781838, "cumulative_end": 8847540, "loss_sum": 9814.46354675293, "token_normalized_loss": 0.14937845951040957, "grad_norm_before_clip": 0.05575224384665489, "learning_rates_at_step": { "lora": 1.3677306937236784e-05, "row_tokens": 1.3677306937236784e-05 }, "learning_rates_after_step": { "lora": 1.3027159577460108e-05, "row_tokens": 1.3027159577460108e-05 } }, { "update_id": 135, "row_count": 76, "row_ids_sha256": "9a201b9bae5fa15a993fc8dcf5dfe8a85bd34981a7d4d9754c0a4d62700a8da0", "input_ids_vector_sha256": "7a94b366c9885e2a18395613355e5a4e597353b78c94d433b85141625a03fbe8", "labels_vector_sha256": "5f65dc1bf038298ff24d01c75c8a73072aa2ae6a4599906f4c243dce67c0c586", "planned_active_tokens": 65439, "realized_active_tokens": 65439, "cumulative_start": 8847540, "cumulative_end": 8912979, "loss_sum": 10541.658748626709, "token_normalized_loss": 0.16109137897319195, "grad_norm_before_clip": 0.05684371292591095, "learning_rates_at_step": { "lora": 1.3027159577460108e-05, "row_tokens": 1.3027159577460108e-05 }, "learning_rates_after_step": { "lora": 1.2387750110024704e-05, "row_tokens": 1.2387750110024704e-05 } }, { "update_id": 136, "row_count": 76, "row_ids_sha256": "5a7e0f2442ad702710857670e3fbf847336d86f174eaf972fedebf019ce93f38", "input_ids_vector_sha256": "af3428de21269ae55268993602503ac9a90e243b4f3832f259aa2f1a2dbf9a86", "labels_vector_sha256": "81c9f26da20468651196ecab0e7f341f40c270bb05a29b19f94b73617dbe0676", "planned_active_tokens": 65727, "realized_active_tokens": 65727, "cumulative_start": 8912979, "cumulative_end": 8978706, "loss_sum": 10295.742664337158, "token_normalized_loss": 0.15664403767610202, "grad_norm_before_clip": 0.05563507229089737, "learning_rates_at_step": { "lora": 1.2387750110024704e-05, "row_tokens": 1.2387750110024704e-05 }, "learning_rates_after_step": { "lora": 1.1757067325464932e-05, "row_tokens": 1.1757067325464932e-05 } }, { "update_id": 137, "row_count": 71, "row_ids_sha256": "0941e35572934b989e7c2a23b050ccc514185f48b729fdedfcca60c07b86663a", "input_ids_vector_sha256": "297c81b8fa833296c465b9762d52189e88059c246e810ad26489a3726f09bff3", "labels_vector_sha256": "0903f3af9df1b834f8b1ef9654f3af5bea86661a3d4c9ef62273b2a13401d61a", "planned_active_tokens": 66273, "realized_active_tokens": 66273, "cumulative_start": 8978706, "cumulative_end": 9044979, "loss_sum": 10267.769477844238, "token_normalized_loss": 0.1549314121564474, "grad_norm_before_clip": 0.05415307730436325, "learning_rates_at_step": { "lora": 1.1757067325464932e-05, "row_tokens": 1.1757067325464932e-05 }, "learning_rates_after_step": { "lora": 1.1153062962038264e-05, "row_tokens": 1.1153062962038264e-05 } }, { "update_id": 138, "row_count": 73, "row_ids_sha256": "7548f2745f8185814cd744a15d4514ccc8189e68baa9529765f83377d9188cca", "input_ids_vector_sha256": "551f3ff8c62f8f58621b3225a8fe7058f2d3e3ec0bf8acf7923c300b3a439912", "labels_vector_sha256": "da76a6948d8c923978df283664e780ea796e11f7a7b8e136010425697f3f7872", "planned_active_tokens": 64930, "realized_active_tokens": 64930, "cumulative_start": 9044979, "cumulative_end": 9109909, "loss_sum": 10559.947528839111, "token_normalized_loss": 0.16263587754257064, "grad_norm_before_clip": 0.057802073657512665, "learning_rates_at_step": { "lora": 1.1153062962038264e-05, "row_tokens": 1.1153062962038264e-05 }, "learning_rates_after_step": { "lora": 1.055387670722936e-05, "row_tokens": 1.055387670722936e-05 } }, { "update_id": 139, "row_count": 71, "row_ids_sha256": "b20e3f23068c425a4e5b95ce025d68c883eb1f47391f523710ff987e2204377b", "input_ids_vector_sha256": "1e631e576763eec04cc24486e4de798ba08b5180daf5ac70a85a035612688290", "labels_vector_sha256": "cef051475d7ae0741b0b46fbfa29359719bb15ad9158d20640b7b91c6c0215c3", "planned_active_tokens": 65950, "realized_active_tokens": 65950, "cumulative_start": 9109909, "cumulative_end": 9175859, "loss_sum": 8825.369197845459, "token_normalized_loss": 0.13381909321979468, "grad_norm_before_clip": 0.052064698189496994, "learning_rates_at_step": { "lora": 1.055387670722936e-05, "row_tokens": 1.055387670722936e-05 }, "learning_rates_after_step": { "lora": 9.978818222052544e-06, "row_tokens": 9.978818222052544e-06 } }, { "update_id": 140, "row_count": 71, "row_ids_sha256": "8b0d5c39bc24e9cf25e75bbf1ec7cb2b19354352012f8d8e4869b2f4e5630828", "input_ids_vector_sha256": "f47b6a700fb5a273e0c2ef2c647a75b0b46e84937c8c42f7a624d719bc436941", "labels_vector_sha256": "dc4eba053f072d15cf1512f0932304f0ea2782509341d202378f96d3ebfb9f5b", "planned_active_tokens": 64866, "realized_active_tokens": 64866, "cumulative_start": 9175859, "cumulative_end": 9240725, "loss_sum": 9156.802730560303, "token_normalized_loss": 0.14116490504363308, "grad_norm_before_clip": 0.0563594289124012, "learning_rates_at_step": { "lora": 9.978818222052544e-06, "row_tokens": 9.978818222052544e-06 }, "learning_rates_after_step": { "lora": 9.405385191662685e-06, "row_tokens": 9.405385191662685e-06 } }, { "update_id": 141, "row_count": 72, "row_ids_sha256": "d00f689454e0301eae7770c4a4b9cb5605439519dd6c7db318e76d27bdfd5e34", "input_ids_vector_sha256": "a933ab2070645bc7f287d7db9042f3ab3cd7b1585b606297b0bd26b2259ba915", "labels_vector_sha256": "eec77880b0edeb1bec83cacb748303ea0af13737fa4169ebceceb14b788f1aa5", "planned_active_tokens": 66360, "realized_active_tokens": 66360, "cumulative_start": 9240725, "cumulative_end": 9307085, "loss_sum": 9758.773998260498, "token_normalized_loss": 0.14705807712869948, "grad_norm_before_clip": 0.05583694949746132, "learning_rates_at_step": { "lora": 9.405385191662685e-06, "row_tokens": 9.405385191662685e-06 }, "learning_rates_after_step": { "lora": 8.854647918058224e-06, "row_tokens": 8.854647918058224e-06 } }, { "update_id": 142, "row_count": 73, "row_ids_sha256": "8745766bba400fed8848ee189aea60640b8d8257f6bfea75173087c77aaa2200", "input_ids_vector_sha256": "c2522a4ffb2cc1199383cd1d4a7b78863f2f9ce236e35582481cd6a40d739d62", "labels_vector_sha256": "658bd1dc9139c1ddb900f090b056077144758311d77c5efe3f8ceef1efb281dd", "planned_active_tokens": 65463, "realized_active_tokens": 65463, "cumulative_start": 9307085, "cumulative_end": 9372548, "loss_sum": 9384.759449005127, "token_normalized_loss": 0.143359752058493, "grad_norm_before_clip": 0.05544775724411011, "learning_rates_at_step": { "lora": 8.854647918058224e-06, "row_tokens": 8.854647918058224e-06 }, "learning_rates_after_step": { "lora": 8.31573730026513e-06, "row_tokens": 8.31573730026513e-06 } }, { "update_id": 143, "row_count": 70, "row_ids_sha256": "657b7c3f6db9550b3c064c249aeb098b6a6da5db0916a0ac61343c21f7a15c1a", "input_ids_vector_sha256": "f913e91f696cf52254588fd09abce99469f568794b738aa8ca225ebf473b00f6", "labels_vector_sha256": "f5058498bf45fab4614a253154de53dc66116700ade6b6bfc84234a03edaf314", "planned_active_tokens": 65862, "realized_active_tokens": 65862, "cumulative_start": 9372548, "cumulative_end": 9438410, "loss_sum": 12264.032722473145, "token_normalized_loss": 0.18620802165851544, "grad_norm_before_clip": 0.06367131322622299, "learning_rates_at_step": { "lora": 8.31573730026513e-06, "row_tokens": 8.31573730026513e-06 }, "learning_rates_after_step": { "lora": 7.803863380602127e-06, "row_tokens": 7.803863380602127e-06 } }, { "update_id": 144, "row_count": 69, "row_ids_sha256": "cc415e7bc8ad6cfbae0097db44c298321dcb1e96b641d1bd74b21fa8b9d193c0", "input_ids_vector_sha256": "8c3f4a927f2d1b011ca9c2fe8aa07b776772769ef109833ef64fd3fdb53047c1", "labels_vector_sha256": "5f581553d7542cc8861904d8bdf19aed0d51ed38b64426c8fdcf94c835ba8ee1", "planned_active_tokens": 64380, "realized_active_tokens": 64380, "cumulative_start": 9438410, "cumulative_end": 9502790, "loss_sum": 10883.519382476807, "token_normalized_loss": 0.16905124856285814, "grad_norm_before_clip": 0.06149400398135185, "learning_rates_at_step": { "lora": 7.803863380602127e-06, "row_tokens": 7.803863380602127e-06 }, "learning_rates_after_step": { "lora": 7.291593949623754e-06, "row_tokens": 7.291593949623754e-06 } }, { "update_id": 145, "row_count": 73, "row_ids_sha256": "31a904b0aca6d42a8678228ac45bc0e0a2a36d4be16d8d30343d2035328849b2", "input_ids_vector_sha256": "6e90420956cc31afa6e5649887e66b08194b9ad37f96f793f2aac09bfa17a085", "labels_vector_sha256": "c4985904bb5d990a2ec43417353169cef394f5687a06173c3a37da887d01f6b7", "planned_active_tokens": 66396, "realized_active_tokens": 66396, "cumulative_start": 9502790, "cumulative_end": 9569186, "loss_sum": 11332.463512420654, "token_normalized_loss": 0.17067991313363237, "grad_norm_before_clip": 0.05694013461470604, "learning_rates_at_step": { "lora": 7.291593949623754e-06, "row_tokens": 7.291593949623754e-06 }, "learning_rates_after_step": { "lora": 6.802904703230067e-06, "row_tokens": 6.802904703230067e-06 } }, { "update_id": 146, "row_count": 72, "row_ids_sha256": "0b0c743530b9c9b8bd2b8946c0ba8cbee6ee941445111d7b9a08c87c2739753e", "input_ids_vector_sha256": "61b02b284ddba217c8018f1cba50503cb6437c4b9590eae8dad53524072f1106", "labels_vector_sha256": "7e62403d8fbfdd83236c190047c08bcee0e88f891f48338bdb63138d0500b736", "planned_active_tokens": 65374, "realized_active_tokens": 65374, "cumulative_start": 9569186, "cumulative_end": 9634560, "loss_sum": 10577.7036819458, "token_normalized_loss": 0.16180291372634076, "grad_norm_before_clip": 0.05710397660732269, "learning_rates_at_step": { "lora": 6.802904703230067e-06, "row_tokens": 6.802904703230067e-06 }, "learning_rates_after_step": { "lora": 6.3342150666999115e-06, "row_tokens": 6.3342150666999115e-06 } }, { "update_id": 147, "row_count": 72, "row_ids_sha256": "528bb8ae6a9a421ed92fe32bc8e1bcae70e40aa724accc694dbc691d79780849", "input_ids_vector_sha256": "f1d1e541a485892839e5fe8aa3d71bcc6438991fb7cd04f8fadf9845ebbd7c86", "labels_vector_sha256": "2107e263154cad9ae37d2a86e037cdd22651528287c2f06a006548decbceec75", "planned_active_tokens": 64777, "realized_active_tokens": 64777, "cumulative_start": 9634560, "cumulative_end": 9699337, "loss_sum": 9908.456253051758, "token_normalized_loss": 0.1529625677794859, "grad_norm_before_clip": 0.05595850571990013, "learning_rates_at_step": { "lora": 6.3342150666999115e-06, "row_tokens": 6.3342150666999115e-06 }, "learning_rates_after_step": { "lora": 5.875463300746481e-06, "row_tokens": 5.875463300746481e-06 } }, { "update_id": 148, "row_count": 73, "row_ids_sha256": "db2c7cb1b792b740f9875c7da09e12841588c38f5a7086fb2c208670cdd7f7f2", "input_ids_vector_sha256": "30ba57509e70e4684c30417c571be5d020d05fbe18ad5b640a0523d97599d4ab", "labels_vector_sha256": "e8cefbf71ad0412029b0ce13b42528c703d0ffece9e6fffb083b86f7368f5a79", "planned_active_tokens": 65606, "realized_active_tokens": 65606, "cumulative_start": 9699337, "cumulative_end": 9764943, "loss_sum": 10693.755840301514, "token_normalized_loss": 0.16299966223061174, "grad_norm_before_clip": 0.058491721749305725, "learning_rates_at_step": { "lora": 5.875463300746481e-06, "row_tokens": 5.875463300746481e-06 }, "learning_rates_after_step": { "lora": 5.430275106263722e-06, "row_tokens": 5.430275106263722e-06 } }, { "update_id": 149, "row_count": 70, "row_ids_sha256": "c40aaf11f9833cdf5875b256c3a8caa84de91d7289272738f1be09ccdc530514", "input_ids_vector_sha256": "2a9a756e908012930d575bfb8c7395784ad676350d7afef7bbbbca3f4ad8fb52", "labels_vector_sha256": "6d8e6631600aa5a62a76d2514308ab69acefb3c11128ffd10e8397bf972b1bf9", "planned_active_tokens": 66005, "realized_active_tokens": 66005, "cumulative_start": 9764943, "cumulative_end": 9830948, "loss_sum": 9985.945083618164, "token_normalized_loss": 0.151290736817183, "grad_norm_before_clip": 0.05425907298922539, "learning_rates_at_step": { "lora": 5.430275106263722e-06, "row_tokens": 5.430275106263722e-06 }, "learning_rates_after_step": { "lora": 5.003680267842576e-06, "row_tokens": 5.003680267842576e-06 } }, { "update_id": 150, "row_count": 71, "row_ids_sha256": "d0add24e0f8fd32ed70f07052499180db62d36bb6a8cdc896363d81748c8cac7", "input_ids_vector_sha256": "cfe568715e5a624c3ba8c6636860ad90bd05c77e389d8042687d59433baf062d", "labels_vector_sha256": "6bbb131129f0bf5b09929addfcb1e06304f76b6a65e8ccc73d53e6c78e4f134a", "planned_active_tokens": 65687, "realized_active_tokens": 65687, "cumulative_start": 9830948, "cumulative_end": 9896635, "loss_sum": 9866.779663085938, "token_normalized_loss": 0.15020901644291773, "grad_norm_before_clip": 0.05642203614115715, "learning_rates_at_step": { "lora": 5.003680267842576e-06, "row_tokens": 5.003680267842576e-06 }, "learning_rates_after_step": { "lora": 4.5950826333247485e-06, "row_tokens": 4.5950826333247485e-06 } }, { "update_id": 151, "row_count": 75, "row_ids_sha256": "30c7fe28901f091fd0bed8b72cc420edf131d0f971828e1eeb0a10e199c97bfd", "input_ids_vector_sha256": "8fdbfbdf00349a39da8e754341249087129497c0cb90bab638d2da4d76857d79", "labels_vector_sha256": "c06355ebcf42be26a6d80e34b744f654ee5c23361b257b36c022be945d4090f2", "planned_active_tokens": 65453, "realized_active_tokens": 65453, "cumulative_start": 9896635, "cumulative_end": 9962088, "loss_sum": 9486.58267211914, "token_normalized_loss": 0.1449373240664162, "grad_norm_before_clip": 0.05477852746844292, "learning_rates_at_step": { "lora": 4.5950826333247485e-06, "row_tokens": 4.5950826333247485e-06 }, "learning_rates_after_step": { "lora": 4.205697285113325e-06, "row_tokens": 4.205697285113325e-06 } }, { "update_id": 152, "row_count": 71, "row_ids_sha256": "9baa80dd58fb86c68a6aa64cd9fb2a6d604a02f1228cd66d32c87f27ee285489", "input_ids_vector_sha256": "7fc0d1977e625741ac5ed1b419d62495fa1900b6be50c72a7be92d654bf0385e", "labels_vector_sha256": "8490a3414cc0be3d95edc5ea8e6fc9fe8d436a52c6561821c109bcbd379a5c4e", "planned_active_tokens": 65007, "realized_active_tokens": 65007, "cumulative_start": 9962088, "cumulative_end": 10027095, "loss_sum": 9435.955062866211, "token_normalized_loss": 0.1451529075771257, "grad_norm_before_clip": 0.055663641542196274, "learning_rates_at_step": { "lora": 4.205697285113325e-06, "row_tokens": 4.205697285113325e-06 }, "learning_rates_after_step": { "lora": 3.830082663750545e-06, "row_tokens": 3.830082663750545e-06 } }, { "update_id": 153, "row_count": 72, "row_ids_sha256": "f085e66940b1878f807a896563bb1a97fa46be7a398621abef91f346178c8dc3", "input_ids_vector_sha256": "c397c55c4206636b9b666c9e62b8f57590b834898841a95011b06e3d7f25d9df", "labels_vector_sha256": "39fe2feb5e802985bba0845c9836b2a149bd8808d8861ed6b23e1505fbd3ec40", "planned_active_tokens": 65496, "realized_active_tokens": 65496, "cumulative_start": 10027095, "cumulative_end": 10092591, "loss_sum": 8841.553123474121, "token_normalized_loss": 0.13499378776526996, "grad_norm_before_clip": 0.05537968873977661, "learning_rates_at_step": { "lora": 3.830082663750545e-06, "row_tokens": 3.830082663750545e-06 }, "learning_rates_after_step": { "lora": 3.469138093628005e-06, "row_tokens": 3.469138093628005e-06 } }, { "update_id": 154, "row_count": 77, "row_ids_sha256": "94f564568bc787f9b815fc02f741aec791276e32dc7f316a41dbab9b2888bf85", "input_ids_vector_sha256": "b6c985e22ce9ad24061fbb1b2b3f0e3df943e418a1d03a200baa868d31e2ec5e", "labels_vector_sha256": "8e9cf05d8dc30b891ff64a1b1ecb53fbc953a158861ee98248679b74f03151ce", "planned_active_tokens": 65913, "realized_active_tokens": 65913, "cumulative_start": 10092591, "cumulative_end": 10158504, "loss_sum": 8792.76948928833, "token_normalized_loss": 0.1333996251010928, "grad_norm_before_clip": 0.05523339658975601, "learning_rates_at_step": { "lora": 3.469138093628005e-06, "row_tokens": 3.469138093628005e-06 }, "learning_rates_after_step": { "lora": 3.12688638989036e-06, "row_tokens": 3.12688638989036e-06 } }, { "update_id": 155, "row_count": 74, "row_ids_sha256": "4d8e009b023a69b969a609387ee28c9395a0923260ba237e9253513ac54ee48d", "input_ids_vector_sha256": "51165986147a8f260ffa873cab386cdbbe4dedca5536fc4265bc090dcb312bf9", "labels_vector_sha256": "4398ab37c31d15b0f81e47e6ef04799bd5558fd886c38b6c6137e4e1c710ff82", "planned_active_tokens": 65629, "realized_active_tokens": 65629, "cumulative_start": 10158504, "cumulative_end": 10224133, "loss_sum": 8488.09546661377, "token_normalized_loss": 0.12933452386313626, "grad_norm_before_clip": 0.05546818673610687, "learning_rates_at_step": { "lora": 3.12688638989036e-06, "row_tokens": 3.12688638989036e-06 }, "learning_rates_after_step": { "lora": 2.8007816038614353e-06, "row_tokens": 2.8007816038614353e-06 } }, { "update_id": 156, "row_count": 76, "row_ids_sha256": "681349cd839f29caec6b144b22ab2bd59145bc77c5ccf7ea32dccdb5334bbf61", "input_ids_vector_sha256": "93a368dc2c859df7aac569350e8126eac97ab75366b68c771c3406bcd7d24de6", "labels_vector_sha256": "1e28ede0f1c1732793620ebe265013c17c87c6b51c4e1c8d68144281c6398d7a", "planned_active_tokens": 65853, "realized_active_tokens": 65853, "cumulative_start": 10224133, "cumulative_end": 10289986, "loss_sum": 12254.924278259277, "token_normalized_loss": 0.1860951555473445, "grad_norm_before_clip": 0.06253273785114288, "learning_rates_at_step": { "lora": 2.8007816038614353e-06, "row_tokens": 2.8007816038614353e-06 }, "learning_rates_after_step": { "lora": 2.4958453288565443e-06, "row_tokens": 2.4958453288565443e-06 } }, { "update_id": 157, "row_count": 70, "row_ids_sha256": "2a9128ef271b114f55d051fd9ba533749271284a4e465c774f3db22df263db51", "input_ids_vector_sha256": "4412623e4360c881dabb66d8f620eaa424fc3565a5d4be90253a5161546fc124", "labels_vector_sha256": "6936e4d72d49bc5c6799a71a96a3ec0e77225b9f8de807585efd3d8b1050e895", "planned_active_tokens": 65040, "realized_active_tokens": 65040, "cumulative_start": 10289986, "cumulative_end": 10355026, "loss_sum": 10849.36455154419, "token_normalized_loss": 0.16681064808647278, "grad_norm_before_clip": 0.05840596184134483, "learning_rates_at_step": { "lora": 2.4958453288565443e-06, "row_tokens": 2.4958453288565443e-06 }, "learning_rates_after_step": { "lora": 2.2041940916720904e-06, "row_tokens": 2.2041940916720904e-06 } }, { "update_id": 158, "row_count": 54, "row_ids_sha256": "cd7bc361ac7b3f529f84673f33283d7494a2562fc11b17c80b254f5f58c6ea68", "input_ids_vector_sha256": "96686348241aa0e6210e18a872edf12acb2d234edeb8682bbed792883d42abc8", "labels_vector_sha256": "36ad8dec2fb255565334e0abc1111450df2ad547997a44925999d9abaa5fe73b", "planned_active_tokens": 65940, "realized_active_tokens": 65940, "cumulative_start": 10355026, "cumulative_end": 10420966, "loss_sum": 13241.424263000488, "token_normalized_loss": 0.20081019507128434, "grad_norm_before_clip": 0.06094012036919594, "learning_rates_at_step": { "lora": 2.2041940916720904e-06, "row_tokens": 2.2041940916720904e-06 }, "learning_rates_after_step": { "lora": 1.934385097939129e-06, "row_tokens": 1.934385097939129e-06 } }, { "update_id": 159, "row_count": 47, "row_ids_sha256": "89483fc7ff9d1a1891457360522a66c860312ac2da57679366f2a96b1e09f4eb", "input_ids_vector_sha256": "a7bc58ec16db890a82a81e68e823f437496030610e25c067ff39a7a84e2eced2", "labels_vector_sha256": "f996d848d29d55c4f8b5e265b49d77cf5075f2b1b39b0022bc936f56e6f1620d", "planned_active_tokens": 64918, "realized_active_tokens": 64918, "cumulative_start": 10420966, "cumulative_end": 10485884, "loss_sum": 12444.060203552246, "token_normalized_loss": 0.19168890297840732, "grad_norm_before_clip": 0.062198083847761154, "learning_rates_at_step": { "lora": 1.934385097939129e-06, "row_tokens": 1.934385097939129e-06 }, "learning_rates_after_step": { "lora": 1.6767780761615837e-06, "row_tokens": 1.6767780761615837e-06 } }, { "update_id": 160, "row_count": 51, "row_ids_sha256": "ea4afacb1a85d69861e9362364a10a7abf195e18f37aec330f00e54cf8daf624", "input_ids_vector_sha256": "141df28dfb21591f679f3c59167e25c37e1523d787cb1ae6206cab431fbf2384", "labels_vector_sha256": "3fefbc662316889b0cdd37a964ca59ad6c708713301e67963b1e469dc58d880a", "planned_active_tokens": 66272, "realized_active_tokens": 66272, "cumulative_start": 10485884, "cumulative_end": 10552156, "loss_sum": 12960.020454406738, "token_normalized_loss": 0.19555801023670236, "grad_norm_before_clip": 0.06344305723905563, "learning_rates_at_step": { "lora": 1.6767780761615837e-06, "row_tokens": 1.6767780761615837e-06 }, "learning_rates_after_step": { "lora": 1.4426882244596429e-06, "row_tokens": 1.4426882244596429e-06 } }, { "update_id": 161, "row_count": 54, "row_ids_sha256": "d8c96930ea71ca5a08dea77e4a0bcd693da6a11a481bfc438e58465b4668f75a", "input_ids_vector_sha256": "748e485f5bba8149629a17f208b4cdd48c8eb0b0b472d74e224efdbf3f5d5afd", "labels_vector_sha256": "ef40ad7a22bf2dcee4f4cc9b8ef5e9f2e184877168f5a11144500395fbec547f", "planned_active_tokens": 64718, "realized_active_tokens": 64718, "cumulative_start": 10552156, "cumulative_end": 10616874, "loss_sum": 13422.53991317749, "token_normalized_loss": 0.20740041276271656, "grad_norm_before_clip": 0.06300444900989532, "learning_rates_at_step": { "lora": 1.4426882244596429e-06, "row_tokens": 1.4426882244596429e-06 }, "learning_rates_after_step": { "lora": 1.2220013654773987e-06, "row_tokens": 1.2220013654773987e-06 } }, { "update_id": 162, "row_count": 47, "row_ids_sha256": "b1accbe0a2358d4069422ecc47dd5b7268e85c8baeaaf436c62a84e43cd04367", "input_ids_vector_sha256": "b3be91775068b7eba9543df474087e310e47630a32d392cfd432320575405a23", "labels_vector_sha256": "76e9cbebc7ace364e65ee41ce3acd9c4d507be5be9754034259c9132212c1167", "planned_active_tokens": 65948, "realized_active_tokens": 65948, "cumulative_start": 10616874, "cumulative_end": 10682822, "loss_sum": 12408.396781921387, "token_normalized_loss": 0.18815425459333698, "grad_norm_before_clip": 0.05994779244065285, "learning_rates_at_step": { "lora": 1.2220013654773987e-06, "row_tokens": 1.2220013654773987e-06 }, "learning_rates_after_step": { "lora": 1.0214973507790492e-06, "row_tokens": 1.0214973507790492e-06 } }, { "update_id": 163, "row_count": 45, "row_ids_sha256": "ad049378685e7fd74a9399e5e3c7d08c35d2b7cd5e1d810ec1b81963a31afc26", "input_ids_vector_sha256": "2bcc4951135c0dc1ea3210fe6de63d61bae421bc3f3b5d71ffc4c99a72120336", "labels_vector_sha256": "64f69e4644f1443ed8c5d11e838485f9765ad934c4d95ef6fa7f38876af52d11", "planned_active_tokens": 65239, "realized_active_tokens": 65239, "cumulative_start": 10682822, "cumulative_end": 10748061, "loss_sum": 13417.02490234375, "token_normalized_loss": 0.20565957329731832, "grad_norm_before_clip": 0.06616192311048508, "learning_rates_at_step": { "lora": 1.0214973507790492e-06, "row_tokens": 1.0214973507790492e-06 }, "learning_rates_after_step": { "lora": 8.336067759043442e-07, "row_tokens": 8.336067759043442e-07 } }, { "update_id": 164, "row_count": 46, "row_ids_sha256": "5fd7f496b484074b274d86d181e03a39d4f9c0d280c747161413fd52fafd7c1c", "input_ids_vector_sha256": "5ec1fb26d03d5372925447ed8cb0859751cf6a5e74df8ea081f7646fa6440f84", "labels_vector_sha256": "c57fc4987901c0cc59d6f9b9ef997bd7c79fef9f425f01bea4747832a59c2a91", "planned_active_tokens": 67184, "realized_active_tokens": 67184, "cumulative_start": 10748061, "cumulative_end": 10815245, "loss_sum": 13560.153129577637, "token_normalized_loss": 0.20183604920185813, "grad_norm_before_clip": 0.06180329993367195, "learning_rates_at_step": { "lora": 8.336067759043442e-07, "row_tokens": 8.336067759043442e-07 }, "learning_rates_after_step": { "lora": 6.71569728545629e-07, "row_tokens": 6.71569728545629e-07 } }, { "update_id": 165, "row_count": 53, "row_ids_sha256": "313985a8e1f48e24c59635edab0e3b1267eff174cb1d51ad2defb54bd7c67d3d", "input_ids_vector_sha256": "6f5a923568b74e36a6feafb5ee2f8cae28f5396e929b9fae48a7613a6f2e5cfc", "labels_vector_sha256": "b405f8c4f167133a621ea3a7fc7a00d609f2f82ae648fb58cc3861e1fa01ff39", "planned_active_tokens": 64277, "realized_active_tokens": 64277, "cumulative_start": 10815245, "cumulative_end": 10879522, "loss_sum": 13095.276664733887, "token_normalized_loss": 0.2037319206673287, "grad_norm_before_clip": 0.06287156790494919, "learning_rates_at_step": { "lora": 6.71569728545629e-07, "row_tokens": 6.71569728545629e-07 }, "learning_rates_after_step": { "lora": 5.247210988667428e-07, "row_tokens": 5.247210988667428e-07 } }, { "update_id": 166, "row_count": 65, "row_ids_sha256": "30d65f1f73d6d68fe045914a3fa48a872fbe2fd5b1f5f1a550175a002b37fd69", "input_ids_vector_sha256": "fbdac94b5e791a13648b99793fa45962e07ba9eeb3120c18adf97a79d983e251", "labels_vector_sha256": "a1e80c5905fbb028fedbf2d32c48c6596b42dea5fec58f82951280e7072394f6", "planned_active_tokens": 65319, "realized_active_tokens": 65319, "cumulative_start": 10879522, "cumulative_end": 10944841, "loss_sum": 13498.469079971313, "token_normalized_loss": 0.20665455809138708, "grad_norm_before_clip": 0.06200416758656502, "learning_rates_at_step": { "lora": 5.247210988667428e-07, "row_tokens": 5.247210988667428e-07 }, "learning_rates_after_step": { "lora": 3.9554846764645604e-07, "row_tokens": 3.9554846764645604e-07 } }, { "update_id": 167, "row_count": 73, "row_ids_sha256": "504e8aaa21bfa5f4b93838aeae59ee59343c78431249ee7badd5cd4727b75ece", "input_ids_vector_sha256": "7f0db0a19e328fb198287f0d3673578e6737183fa9f264b528d4a310ffdfc649", "labels_vector_sha256": "57e5e4a56cfae16c78bcd8a95a8f41d5576f2b2a49ec35e2307fd28631389494", "planned_active_tokens": 65502, "realized_active_tokens": 65502, "cumulative_start": 10944841, "cumulative_end": 11010343, "loss_sum": 12827.26448059082, "token_normalized_loss": 0.1958301193946875, "grad_norm_before_clip": 0.06053180620074272, "learning_rates_at_step": { "lora": 3.9554846764645604e-07, "row_tokens": 3.9554846764645604e-07 }, "learning_rates_after_step": { "lora": 2.841453809859784e-07, "row_tokens": 2.841453809859784e-07 } }, { "update_id": 168, "row_count": 76, "row_ids_sha256": "26ee1682a5079c64ea8830aaa203c60121fd2a36a4324511313e04825113e751", "input_ids_vector_sha256": "9e6efd86726fccaa2f0926b63eb5c61b7596afd8e750e7469e448dfbc5d24305", "labels_vector_sha256": "90f659b16e8bbdb2ff402e18e10377186e7437660b1e52d80de1d4c4a8f07302", "planned_active_tokens": 65753, "realized_active_tokens": 65753, "cumulative_start": 11010343, "cumulative_end": 11076096, "loss_sum": 13842.160583496094, "token_normalized_loss": 0.21051755179985845, "grad_norm_before_clip": 0.06377389281988144, "learning_rates_at_step": { "lora": 2.841453809859784e-07, "row_tokens": 2.841453809859784e-07 }, "learning_rates_after_step": { "lora": 1.9136098821897485e-07, "row_tokens": 1.9136098821897485e-07 } }, { "update_id": 169, "row_count": 64, "row_ids_sha256": "fda6835eb04d2bb066f04ad93e6fed3aac50d7d1eacacf6908b0c4b41205582a", "input_ids_vector_sha256": "a6bb99dd922d1505ff086d2672c8599aaea1090b3dfe9ec5297fca7920ac09b7", "labels_vector_sha256": "a8bc73cfe12d034ebaa21947080ef2571e196f5defc90bc94497d5c8e077e835", "planned_active_tokens": 65535, "realized_active_tokens": 65535, "cumulative_start": 11076096, "cumulative_end": 11141631, "loss_sum": 15618.952213287354, "token_normalized_loss": 0.23832993382600676, "grad_norm_before_clip": 0.0692538321018219, "learning_rates_at_step": { "lora": 1.9136098821897485e-07, "row_tokens": 1.9136098821897485e-07 }, "learning_rates_after_step": { "lora": 1.1648965763972897e-07, "row_tokens": 1.1648965763972897e-07 } }, { "update_id": 170, "row_count": 69, "row_ids_sha256": "db4456e9f309327dfbd78a2d26ef68eb4eafbc985009b2350009fb3832067a04", "input_ids_vector_sha256": "a32b0eb93dbf477b40daae5de03db8cd3ed394db95e61415e8a2222fb98cf33a", "labels_vector_sha256": "324fa78372d35f84b887b9a9377e99b651109c9cbcc903c42acf582421ee6ffe", "planned_active_tokens": 65876, "realized_active_tokens": 65876, "cumulative_start": 11141631, "cumulative_end": 11207507, "loss_sum": 14030.309116363525, "token_normalized_loss": 0.212980586501359, "grad_norm_before_clip": 0.06223655492067337, "learning_rates_at_step": { "lora": 1.1648965763972897e-07, "row_tokens": 1.1648965763972897e-07 }, "learning_rates_after_step": { "lora": 6.07233308663413e-08, "row_tokens": 6.07233308663413e-08 } }, { "update_id": 171, "row_count": 67, "row_ids_sha256": "7916910c37554f8bf5e6574138968ffea5248b1cf9d3b47f0dbf0844ff2aeed8", "input_ids_vector_sha256": "0d13b1e0a9efd72e511632dcb352cfb18d76220c74d05f1c6b47ce5236fd6f39", "labels_vector_sha256": "e6e5c2331191e1459ca1ea3ee6ff098a63ca44137359513a2ffc909ec37a9d08", "planned_active_tokens": 64993, "realized_active_tokens": 64993, "cumulative_start": 11207507, "cumulative_end": 11272500, "loss_sum": 13533.568244934082, "token_normalized_loss": 0.20823116712467624, "grad_norm_before_clip": 0.059106215834617615, "learning_rates_at_step": { "lora": 6.07233308663413e-08, "row_tokens": 6.07233308663413e-08 }, "learning_rates_after_step": { "lora": 2.220339418241846e-08, "row_tokens": 2.220339418241846e-08 } }, { "update_id": 172, "row_count": 63, "row_ids_sha256": "bd41e2fc370108fc65a0ba4b214bb1bbc36bcbc6e3dde2279a65ff5bed3a2da6", "input_ids_vector_sha256": "5802debcb675acd691076afe668b12cdf638f0e220655b2877bd347716ea6c86", "labels_vector_sha256": "fc96e43f056223cb809bec62912760e86fa95f397fe729e70f62bc82433884c3", "planned_active_tokens": 66709, "realized_active_tokens": 66709, "cumulative_start": 11272500, "cumulative_end": 11339209, "loss_sum": 13883.325355529785, "token_normalized_loss": 0.20811772557720526, "grad_norm_before_clip": 0.06189237907528877, "learning_rates_at_step": { "lora": 2.220339418241846e-08, "row_tokens": 2.220339418241846e-08 }, "learning_rates_after_step": { "lora": 3.0609329541075872e-09, "row_tokens": 3.0609329541075872e-09 } }, { "update_id": 173, "row_count": 67, "row_ids_sha256": "446c84220a9bbd5a92a898b47fe5fa6488dfd2cb9f563a86a9ef42d45401447c", "input_ids_vector_sha256": "bc7c2c089058288ce65e96bd228d56b149903fab2c7cadfe9ddc666e8d7345ba", "labels_vector_sha256": "87120055eb584d97cf5c5ab5077f89f90e376c9443fc509ea92e7a5208ed78a9", "planned_active_tokens": 64145, "realized_active_tokens": 64145, "cumulative_start": 11339209, "cumulative_end": 11403354, "loss_sum": 13519.707010269165, "token_normalized_loss": 0.21076790100973053, "grad_norm_before_clip": 0.060558952391147614, "learning_rates_at_step": { "lora": 3.0609329541075872e-09, "row_tokens": 3.0609329541075872e-09 }, "learning_rates_after_step": { "lora": 0.0, "row_tokens": 0.0 } }, { "update_id": 174, "row_count": 34, "row_ids_sha256": "78a6dd54ddf390db5676d95ece15782cef074566cce9c15b81c096d316a03550", "input_ids_vector_sha256": "90997b4cbf8b6637a03ee22ba7b2f8649fe4c6510fe39508b50bd3097088edb5", "labels_vector_sha256": "54e88122ed05f209852401955ebd3504847545fd59bfbd20c54807c40642c725", "planned_active_tokens": 37880, "realized_active_tokens": 37880, "cumulative_start": 11403354, "cumulative_end": 11441234, "loss_sum": 8773.458435058594, "token_normalized_loss": 0.23161189110503153, "grad_norm_before_clip": 0.08213622123003006, "learning_rates_at_step": { "lora": 0.0, "row_tokens": 0.0 }, "learning_rates_after_step": { "lora": 0.0, "row_tokens": 0.0 } } ], "frozen_base_digest_start": { "scope": "frozen_base", "sha256": "5d57dc425b2edd8291c27f6806bbaf90b43e8923f1103860cd0a20346464b375", "parameter_count": 355, "bytes": 14596022272, "names_sha256": "55a15eecdc0d2c8fdc49edbfe55db4accc70117860d030d39e44f294cd5304d2" }, "optimizer_membership": { "schema": "think-b-2ep-optimizer-membership/1", "token_bake_enabled": false, "groups": [ { "group": "lora", "param_count": 448, "params": [ { "name": "base_model.model.model.layers.0.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.0.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.0.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.0.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.0.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.0.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.1.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.1.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.2.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.2.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.3.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.3.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.4.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.4.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.5.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.5.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.6.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.6.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.7.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.7.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.8.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.8.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.9.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.9.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.10.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.10.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.11.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.11.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.12.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.12.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.13.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.13.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.14.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.14.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.15.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.15.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.16.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.16.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.17.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.17.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.18.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.18.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.19.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.19.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.20.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.20.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.21.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.21.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.22.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.22.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.23.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.23.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.24.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.24.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.25.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.25.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.26.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.26.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.27.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.27.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.28.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.28.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.29.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.29.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.30.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.30.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.self_attn.k_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.self_attn.v_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.self_attn.o_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.mlp.gate_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.31.mlp.up_proj.lora_B.default.weight", "shape": [ 11008, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 11008 ] }, { "name": "base_model.model.model.layers.31.mlp.down_proj.lora_B.default.weight", "shape": [ 4096, 64 ] } ] }, { "group": "row_tokens", "param_count": 2, "params": [ { "name": "base_model.model.model.embed_tokens.token_adapter.trainable_tokens_delta.default", "shape": [ 1, 4096 ] }, { "name": "base_model.model.lm_head.token_adapter.trainable_tokens_delta.default", "shape": [ 1, 4096 ] } ] } ] }, "tied_token_canary": null, "continuation": null, "lora_digest_start": { "scope": "lora", "sha256": "dfdc674eb25a281ac54726af632cb4deba0c0f9e742bd35ff3f3308c75e0a19a", "parameter_count": 450, "bytes": 639664128, "names_sha256": "8682f1cd731b6e577fe15b72c310880afd96758ffb839b1c1159f8e04f2c050f" } }