{ "schema": "ddc-token-boundary-ledger/1", "runner_id": "THINK_B_2EP_SIX_ARMS_RUNNER_V3", "bindings": { "config_sha256": "ed1535bfd150f0cf7f117d9268aed978a2171136d03cd7289923b775e136e4f5", "runner_code_sha256": "fd6d1c6403810acc4460e2ce0fb5f4418881387b11f4bd21f7cd21067faa3a0a", "config_schema_sha256": "d6d13eeee42bf04cf34fc68b2fa7a61b316c4543b77d7d779815b8eddf10c099", "python_binary_sha256": "5700a6a7804feb44da6c963ebd1323982478b7ca57478e4e81981c47818453bd", "environment_lock_sha256": "627d4fe088b8f64841b0520ae5bd9dc5c81e9dcd5969b4bd45c9a9910b2f3c01", "base_composite_sha256": "26cbdb468ced4891fee1fffa2be7b7722e5d8d836d0bc986eecd97fa12cd1747", "base_lock_sha256": "9a145eebf5d9df466a5f554c4850290be671acd284e0f703b7dfdb087555858b", "frozen_rows_sha256": "1a99ada6e2afb71bf5b7a7af45422104b5e4745e6601a4081ad462473cebe804", "ordered_ids_sha256": "0152bfafdb38a00d960507db14293e4ff02830745fd63587320e6990a9b5b51f", "update_plan_sha256": "b16e590277206f95b60c22e04883dc80fe66e93201a39d09cc1e7feb5ca110e6", "data_receipt_sha256": "e66346fb413557929ab53648bc656fb3102557cf3460abb49c7b1c6e7281420d", "decontamination_receipt_sha256": "66c360871b06208ed20e473e57ad0f09c79e4dac1b3124a9461a5b729b487153", "render_receipt_sha256": "e66346fb413557929ab53648bc656fb3102557cf3460abb49c7b1c6e7281420d", "token_bake_sha256": "74234e98afe7498fb5daf1f36ac2d78acc339464f950703b8c019892f982b90b", "continuation_sha256": "74234e98afe7498fb5daf1f36ac2d78acc339464f950703b8c019892f982b90b" }, "completed_optimizer_updates": 165, "next_update_index": 165, "next_ordered_position": 11750, "cumulative_planned_active_tokens": 10788766, "cumulative_realized_active_tokens": 10788766, "optimizer_step_count": 165, "scheduler_step_count": 165, "token_bake": { "enabled": false, "positive_lr_active_tokens": null, "frozen": null, "freeze_digest": null }, "updates": [ { "update_id": 0, "row_count": 66, "row_ids_sha256": "7c44588abeaa5362afad4783a93c894465d620117cc789de76e0f5a20f633c6d", "input_ids_vector_sha256": "ac772df88d7486e8f10af3b39cca7ca9f9c1f3428c85c3086ffe7d2fd9c8a902", "labels_vector_sha256": "34219b378c552f9fbc7e597232e28d7c4ef48d5ad475d1de2c43c7825b735886", "planned_active_tokens": 65856, "realized_active_tokens": 65856, "cumulative_start": 0, "cumulative_end": 65856, "loss_sum": 36830.20881652832, "token_normalized_loss": 0.5592536567135618, "grad_norm_before_clip": 1.0257219076156616, "learning_rates_at_step": { "lora": 1.0173544705449805e-05, "row_tokens": 1.0173544705449805e-05 }, "learning_rates_after_step": { "lora": 2.037551403774914e-05, "row_tokens": 2.037551403774914e-05 } }, { "update_id": 1, "row_count": 68, "row_ids_sha256": "bc5ce5694ef773bf5e7b3ee908bb536e6ddfaaf13a954de1ba372de578657c23", "input_ids_vector_sha256": "45052c55a7cae853bd9ecad5f3b17f3d2789a22e0c9eabce8474b1d40cb9b182", "labels_vector_sha256": "d1e46ad97a7cdae3bd5da9940c80cce92600999876dd3a8d4f848a5589f9c1ee", "planned_active_tokens": 66040, "realized_active_tokens": 66040, "cumulative_start": 65856, "cumulative_end": 131896, "loss_sum": 31433.324615478516, "token_normalized_loss": 0.4759740250678152, "grad_norm_before_clip": 0.862107515335083, "learning_rates_at_step": { "lora": 2.037551403774914e-05, "row_tokens": 2.037551403774914e-05 }, "learning_rates_after_step": { "lora": 3.0515536221316615e-05, "row_tokens": 3.0515536221316615e-05 } }, { "update_id": 2, "row_count": 65, "row_ids_sha256": "14d2924059f05ca267372f880131d6848e89396dd131bf376325b873a6a68871", "input_ids_vector_sha256": "0fae4708cca539198b1e17d790429489f9afcf50b49176d0203ba65f62b97bd7", "labels_vector_sha256": "77a32895a8eae33ecb1b228b6559fb79b1a7e6b0c418efc260c827bcb0c3c5d9", "planned_active_tokens": 65639, "realized_active_tokens": 65639, "cumulative_start": 131896, "cumulative_end": 197535, "loss_sum": 31775.35385131836, "token_normalized_loss": 0.48409259512360575, "grad_norm_before_clip": 0.639792799949646, "learning_rates_at_step": { "lora": 3.0515536221316615e-05, "row_tokens": 3.0515536221316615e-05 }, "learning_rates_after_step": { "lora": 4.062095451132814e-05, "row_tokens": 4.062095451132814e-05 } }, { "update_id": 3, "row_count": 65, "row_ids_sha256": "c4bb70634d254a0cdcf2ab98cc3cb3cfc57dcf58a7108a93c10739d0475fa39b", "input_ids_vector_sha256": "d484b8e6cd6354ee6ee71ca9f7e5c3f6dc2ab85b70a2d1f31cfb1bfb262945c5", "labels_vector_sha256": "7136994a58b8f373de5ae499b0f84319898b6adfe632fecfb24b4c4c93750c96", "planned_active_tokens": 65415, "realized_active_tokens": 65415, "cumulative_start": 197535, "cumulative_end": 262950, "loss_sum": 30946.996154785156, "token_normalized_loss": 0.4730871536312032, "grad_norm_before_clip": 0.5711080431938171, "learning_rates_at_step": { "lora": 4.062095451132814e-05, "row_tokens": 4.062095451132814e-05 }, "learning_rates_after_step": { "lora": 5.0699184027831416e-05, "row_tokens": 5.0699184027831416e-05 } }, { "update_id": 4, "row_count": 72, "row_ids_sha256": "ffa1d4fc9434a37283f6bc847559aa45f8f711a7657ba1c41ccfde7e39ab4e8a", "input_ids_vector_sha256": "3a52376d2b50b9ef8bb74bbfd8c0e3c55f2e1c5a0643d084163392c7a93a9fa8", "labels_vector_sha256": "994f821416a554221e387e28ad8c19554a924667598094b7a92dece1289a4c5c", "planned_active_tokens": 65239, "realized_active_tokens": 65239, "cumulative_start": 262950, "cumulative_end": 328189, "loss_sum": 31104.10710144043, "token_normalized_loss": 0.4767716718748054, "grad_norm_before_clip": 0.5920698046684265, "learning_rates_at_step": { "lora": 5.0699184027831416e-05, "row_tokens": 5.0699184027831416e-05 }, "learning_rates_after_step": { "lora": 6.083951517473422e-05, "row_tokens": 6.083951517473422e-05 } }, { "update_id": 5, "row_count": 70, "row_ids_sha256": "a8a960c4d84152a7ca4903054fda140000fe73b40d311d7ab1f911d0d88c6b87", "input_ids_vector_sha256": "b0579abd2297eebc0794e4fe02890bb29959bb04403d97d6952c4c2eb4523c92", "labels_vector_sha256": "e06400cc0569c43afe9714c2d157b89a535fabc4d7a33302b797937bdb60462b", "planned_active_tokens": 65641, "realized_active_tokens": 65641, "cumulative_start": 328189, "cumulative_end": 393830, "loss_sum": 28283.458404541016, "token_normalized_loss": 0.4308809799445623, "grad_norm_before_clip": 0.49931854009628296, "learning_rates_at_step": { "lora": 6.083951517473422e-05, "row_tokens": 6.083951517473422e-05 }, "learning_rates_after_step": { "lora": 7.103839487368035e-05, "row_tokens": 7.103839487368035e-05 } }, { "update_id": 6, "row_count": 64, "row_ids_sha256": "a0899f5fe157237d2a789b1c5e723605f2b4b6fc0dd4d7bd19beb6148333bc8c", "input_ids_vector_sha256": "67e2c47053fec51e38de99705780465bc0a397c3979e37236db6a1acb703fbdc", "labels_vector_sha256": "eff6975e339d805ff00f60ad4ac9d9c44631b02d818f0f8376e2ca045b8a8162", "planned_active_tokens": 66020, "realized_active_tokens": 66020, "cumulative_start": 393830, "cumulative_end": 459850, "loss_sum": 26027.433563232422, "token_normalized_loss": 0.3942355886584735, "grad_norm_before_clip": 0.44088059663772583, "learning_rates_at_step": { "lora": 7.103839487368035e-05, "row_tokens": 7.103839487368035e-05 }, "learning_rates_after_step": { "lora": 8.100477966279742e-05, "row_tokens": 8.100477966279742e-05 } }, { "update_id": 7, "row_count": 65, "row_ids_sha256": "75ccdf18c141f9b4ef56cd6b9566856227e58096d95ee2865a2167c56e70750f", "input_ids_vector_sha256": "747f2d23e5f55f51d428c034e758919655e646833b760371a4e22d022acd5917", "labels_vector_sha256": "302775b18e739e31a7e86657308ad66b5be3aca020e2e6250e761e7b19bac592", "planned_active_tokens": 64515, "realized_active_tokens": 64515, "cumulative_start": 459850, "cumulative_end": 524365, "loss_sum": 24155.72621154785, "token_normalized_loss": 0.3744203086343928, "grad_norm_before_clip": 0.37404394149780273, "learning_rates_at_step": { "lora": 8.100477966279742e-05, "row_tokens": 8.100477966279742e-05 }, "learning_rates_after_step": { "lora": 9.1225750240219e-05, "row_tokens": 9.1225750240219e-05 } }, { "update_id": 8, "row_count": 66, "row_ids_sha256": "c127a67010e1dd2405b427436af5dc4a4572c8908b2ecbdcef251f0a874483e6", "input_ids_vector_sha256": "28b334d1b09d62a4aab0125423836af7a4e42e4fa7a7fec697ce81b4afd11037", "labels_vector_sha256": "14a9985e765bc206883c90e15a48417dfc63e6007fc82fe84ece9ec4092f90f1", "planned_active_tokens": 66163, "realized_active_tokens": 66163, "cumulative_start": 524365, "cumulative_end": 590528, "loss_sum": 24134.758758544922, "token_normalized_loss": 0.3647772736808325, "grad_norm_before_clip": 0.3161574602127075, "learning_rates_at_step": { "lora": 9.1225750240219e-05, "row_tokens": 9.1225750240219e-05 }, "learning_rates_after_step": { "lora": 9.999981929106849e-05, "row_tokens": 9.999981929106849e-05 } }, { "update_id": 9, "row_count": 68, "row_ids_sha256": "789b8247a96b562c3dacd6c5ee08332fbf2949f1f54c410a3033ded22a38d753", "input_ids_vector_sha256": "e9d9ddcb962f796132968ec9f502582e3447c477f7f5a9729a836a7adb2c47fc", "labels_vector_sha256": "cc8a4a873ce6eeb85998c601fea8c66fc7c7b231941180affdae5f2e247748ec", "planned_active_tokens": 65477, "realized_active_tokens": 65477, "cumulative_start": 590528, "cumulative_end": 656005, "loss_sum": 23961.598136901855, "token_normalized_loss": 0.3659544288361082, "grad_norm_before_clip": 0.2811216115951538, "learning_rates_at_step": { "lora": 9.999981929106849e-05, "row_tokens": 9.999981929106849e-05 }, "learning_rates_after_step": { "lora": 9.998695017928532e-05, "row_tokens": 9.998695017928532e-05 } }, { "update_id": 10, "row_count": 67, "row_ids_sha256": "765a208a37dd39162f5ab59170bab24c50370c1a0f4ab6245819f04fa3f68390", "input_ids_vector_sha256": "05fb005990abb6c81374b0dbfacb52f162f21ee57239b6dc0016e7f17efe0db3", "labels_vector_sha256": "4831e826c916cd464771f0818727cbe768ce7115996b50842df79d470ddc4cbb", "planned_active_tokens": 65076, "realized_active_tokens": 65076, "cumulative_start": 656005, "cumulative_end": 721081, "loss_sum": 20343.693572998047, "token_normalized_loss": 0.3126143827678107, "grad_norm_before_clip": 0.2559735178947449, "learning_rates_at_step": { "lora": 9.998695017928532e-05, "row_tokens": 9.998695017928532e-05 }, "learning_rates_after_step": { "lora": 9.99534681150285e-05, "row_tokens": 9.99534681150285e-05 } }, { "update_id": 11, "row_count": 69, "row_ids_sha256": "a69f441698a98646ac448cc6a3d53ddaa861204d2d43f41507103ea86281094d", "input_ids_vector_sha256": "4ef77a733ef09b71eae0b6c38708240d8df4b1faddee85034e60dadd3031a226", "labels_vector_sha256": "5df729b4ebdb82b4023e38c1864be555a4d47d658ef7e1575da142ab37814809", "planned_active_tokens": 65525, "realized_active_tokens": 65525, "cumulative_start": 721081, "cumulative_end": 786606, "loss_sum": 18426.090522766113, "token_normalized_loss": 0.2812070281994065, "grad_norm_before_clip": 0.18362288177013397, "learning_rates_at_step": { "lora": 9.99534681150285e-05, "row_tokens": 9.99534681150285e-05 }, "learning_rates_after_step": { "lora": 9.989860317157319e-05, "row_tokens": 9.989860317157319e-05 } }, { "update_id": 12, "row_count": 68, "row_ids_sha256": "e459890cbdcbd788d629f668a4e130767c501daffe5c7c4efc0a290091c65084", "input_ids_vector_sha256": "4184b311a541f7f76cf4082026766fe851671fa1cfb0b0100649ae7c84c26e7e", "labels_vector_sha256": "596c549ea9e2087dfab353392b2836cb1b0f1741510adc0d392c620f7e2743c6", "planned_active_tokens": 66340, "realized_active_tokens": 66340, "cumulative_start": 786606, "cumulative_end": 852946, "loss_sum": 17425.548179626465, "token_normalized_loss": 0.2626703071996754, "grad_norm_before_clip": 0.16265520453453064, "learning_rates_at_step": { "lora": 9.989860317157319e-05, "row_tokens": 9.989860317157319e-05 }, "learning_rates_after_step": { "lora": 9.982417788192085e-05, "row_tokens": 9.982417788192085e-05 } }, { "update_id": 13, "row_count": 73, "row_ids_sha256": "2c0d0d07351190245b6ac2e722f6c2684a2b343b5e6c6347c77fdb0e46914c54", "input_ids_vector_sha256": "46c7cc3f1ce117472a66bd6943afa933e5f9cc1b682b6d8cdd5a54cf56b3ab74", "labels_vector_sha256": "ebb2caa890d811d6f649bedf45050e2894ee88b0d02ed17e7ac7ef2e8c10ff0e", "planned_active_tokens": 65177, "realized_active_tokens": 65177, "cumulative_start": 852946, "cumulative_end": 918123, "loss_sum": 21065.071113586426, "token_normalized_loss": 0.3231979243227891, "grad_norm_before_clip": 0.2028338760137558, "learning_rates_at_step": { "lora": 9.982417788192085e-05, "row_tokens": 9.982417788192085e-05 }, "learning_rates_after_step": { "lora": 9.972797124583697e-05, "row_tokens": 9.972797124583697e-05 } }, { "update_id": 14, "row_count": 74, "row_ids_sha256": "a760432725273ddbef43fd0bf0ae27cb2700d3d894a5f3980dce26439562dba6", "input_ids_vector_sha256": "2fc9c780b52c11c4db417c3fc2b259eea78ac86dae41bd25f2aaf10c1c8c9831", "labels_vector_sha256": "79f502210b4b928525aaa0e6ddf29fc43f70448ca6ec48616cafbc25cd849fc6", "planned_active_tokens": 66090, "realized_active_tokens": 66090, "cumulative_start": 918123, "cumulative_end": 984213, "loss_sum": 20900.96273803711, "token_normalized_loss": 0.31625000360171146, "grad_norm_before_clip": 0.19721658527851105, "learning_rates_at_step": { "lora": 9.972797124583697e-05, "row_tokens": 9.972797124583697e-05 }, "learning_rates_after_step": { "lora": 9.961416259995746e-05, "row_tokens": 9.961416259995746e-05 } }, { "update_id": 15, "row_count": 66, "row_ids_sha256": "b2f2f89441cba32935bcb3b8fad00d06d3f52b73148a4cac9af440e3f6582d03", "input_ids_vector_sha256": "aa959556b9e93e0f73ee7df8adf1c699e9c4cc93eaa57c6cc444178128b56ec4", "labels_vector_sha256": "7059e08364a6a48a04433f7252bca1c6f2487458a0ec562097a55327b5db1e7e", "planned_active_tokens": 64406, "realized_active_tokens": 64406, "cumulative_start": 984213, "cumulative_end": 1048619, "loss_sum": 17135.82063293457, "token_normalized_loss": 0.26605938317756994, "grad_norm_before_clip": 0.1514115184545517, "learning_rates_at_step": { "lora": 9.961416259995746e-05, "row_tokens": 9.961416259995746e-05 }, "learning_rates_after_step": { "lora": 9.947573271204726e-05, "row_tokens": 9.947573271204726e-05 } }, { "update_id": 16, "row_count": 71, "row_ids_sha256": "e9d9cf66952683200e1ad3f8adf1750af457a966cddf782fda4484b26406fe33", "input_ids_vector_sha256": "c80ae9b3d6d64daa8be508e6cad65979af9263fbd15267dcb7ff47e278710b45", "labels_vector_sha256": "ef5b03b804483778c1893f6f361953e5c543a01d68c078550ec09631ff296404", "planned_active_tokens": 66589, "realized_active_tokens": 66589, "cumulative_start": 1048619, "cumulative_end": 1115208, "loss_sum": 16009.054412841797, "token_normalized_loss": 0.24041590071696223, "grad_norm_before_clip": 0.14175383746623993, "learning_rates_at_step": { "lora": 9.947573271204726e-05, "row_tokens": 9.947573271204726e-05 }, "learning_rates_after_step": { "lora": 9.932099637986787e-05, "row_tokens": 9.932099637986787e-05 } }, { "update_id": 17, "row_count": 70, "row_ids_sha256": "a438bbb67705db346ae46a80139226c526574769aa4ab4cb75cb8861191cceaa", "input_ids_vector_sha256": "b289bc4aad951d1f91dde1fa68a9b4fbefcdb14d73c0ff82dfcaa34c30b5d6d5", "labels_vector_sha256": "4448ba67738fce5b90bb00c3a50a514a4286cabf3d14a9e7ce26e0b2ce1fcbcd", "planned_active_tokens": 64727, "realized_active_tokens": 64727, "cumulative_start": 1115208, "cumulative_end": 1179935, "loss_sum": 20681.138862609863, "token_normalized_loss": 0.31951332307398556, "grad_norm_before_clip": 0.14233136177062988, "learning_rates_at_step": { "lora": 9.932099637986787e-05, "row_tokens": 9.932099637986787e-05 }, "learning_rates_after_step": { "lora": 9.914397335538702e-05, "row_tokens": 9.914397335538702e-05 } }, { "update_id": 18, "row_count": 72, "row_ids_sha256": "ec8d6fd5d89e71f13963cc93d8fd6b594fda845b13bb19c3d5ad22bfd44b26d4", "input_ids_vector_sha256": "3e87be022a9be199ca5970f488d4a2d55cfa0dadd40f32ae50f9a25f72562ba0", "labels_vector_sha256": "254fda0c0386cabe4ec430f792197b8c015458681674afc8d9f95205a79d5e2a", "planned_active_tokens": 65589, "realized_active_tokens": 65589, "cumulative_start": 1179935, "cumulative_end": 1245524, "loss_sum": 18849.065032958984, "token_normalized_loss": 0.2873814973998534, "grad_norm_before_clip": 0.12853355705738068, "learning_rates_at_step": { "lora": 9.914397335538702e-05, "row_tokens": 9.914397335538702e-05 }, "learning_rates_after_step": { "lora": 9.894585989671054e-05, "row_tokens": 9.894585989671054e-05 } }, { "update_id": 19, "row_count": 68, "row_ids_sha256": "52764c066f56c4c2cc729a935807b00db30a44907d980c392c83a2ba1d322b64", "input_ids_vector_sha256": "6f80ee0d1aa66c5d04b4af633f7e8041fefe6053024cea8a63486cfb3901c0e3", "labels_vector_sha256": "98b434fa6f571193cf28f2218db15ca71ccab108816025ad6f37ea2ad7f2a546", "planned_active_tokens": 65843, "realized_active_tokens": 65843, "cumulative_start": 1245524, "cumulative_end": 1311367, "loss_sum": 15803.549369812012, "token_normalized_loss": 0.24001867123022966, "grad_norm_before_clip": 0.11425246298313141, "learning_rates_at_step": { "lora": 9.894585989671054e-05, "row_tokens": 9.894585989671054e-05 }, "learning_rates_after_step": { "lora": 9.872800589662978e-05, "row_tokens": 9.872800589662978e-05 } }, { "update_id": 20, "row_count": 69, "row_ids_sha256": "8cda969b222e5c66315f41f28da1d4425e4ece6f14e54c5b712207a55e875aab", "input_ids_vector_sha256": "ccc5a1e4708f9277527c60702915748027d27cbeda3a0045935ffb661787c5e3", "labels_vector_sha256": "c3e8ded291e90ad8b76c83820a16a6e56822bf7afc1ad040e507306b7f2c471f", "planned_active_tokens": 65664, "realized_active_tokens": 65664, "cumulative_start": 1311367, "cumulative_end": 1377031, "loss_sum": 16967.58219909668, "token_normalized_loss": 0.25840007003984955, "grad_norm_before_clip": 0.11823660880327225, "learning_rates_at_step": { "lora": 9.872800589662978e-05, "row_tokens": 9.872800589662978e-05 }, "learning_rates_after_step": { "lora": 9.84917845011258e-05, "row_tokens": 9.84917845011258e-05 } }, { "update_id": 21, "row_count": 75, "row_ids_sha256": "eaf2019de3060a975db02f73f3a399cad30d9e6f3887fa745d717bb5167ac726", "input_ids_vector_sha256": "c855e2db4f14316cbc9e7d7de27224224bf5d3a939d3db0fafb6ee75ade59412", "labels_vector_sha256": "3e30b06286b6c55f7aac8adf5edd5cbf0da7f5cb3537cec96874e7bb373e9fb8", "planned_active_tokens": 65189, "realized_active_tokens": 65189, "cumulative_start": 1377031, "cumulative_end": 1442220, "loss_sum": 19739.002899169922, "token_normalized_loss": 0.30279652854269773, "grad_norm_before_clip": 0.12580636143684387, "learning_rates_at_step": { "lora": 9.84917845011258e-05, "row_tokens": 9.84917845011258e-05 }, "learning_rates_after_step": { "lora": 9.823488736250502e-05, "row_tokens": 9.823488736250502e-05 } }, { "update_id": 22, "row_count": 72, "row_ids_sha256": "af1fecc4134b44c5388a8c1da6a65217d768d53be6160202de881876545ce2f5", "input_ids_vector_sha256": "5778057573d37d813f4be20531fb19fdf01eeb4a41421134397322a18eba8796", "labels_vector_sha256": "2d86d41282b1ccb6ba5292fc3cefdb7ffeed9637c1830fa972b999afac86cd1d", "planned_active_tokens": 65410, "realized_active_tokens": 65410, "cumulative_start": 1442220, "cumulative_end": 1507630, "loss_sum": 17475.113288879395, "token_normalized_loss": 0.2671627165399693, "grad_norm_before_clip": 0.10962013155221939, "learning_rates_at_step": { "lora": 9.823488736250502e-05, "row_tokens": 9.823488736250502e-05 }, "learning_rates_after_step": { "lora": 9.795757363905944e-05, "row_tokens": 9.795757363905944e-05 } }, { "update_id": 23, "row_count": 72, "row_ids_sha256": "08c36a6eb8796f9bae40d6e18604e3dd24abf3e5e695ac98f654ff6f77be58a7", "input_ids_vector_sha256": "21d3e92e51f722cdb8985e4c71d16597f4f168c647a2b09f9b0c90b0d64d6ad9", "labels_vector_sha256": "e7a54de2bddca1262546b9e7defc611e5acfeb3e4d6f2612c4ca9de01d79832f", "planned_active_tokens": 65550, "realized_active_tokens": 65550, "cumulative_start": 1507630, "cumulative_end": 1573180, "loss_sum": 20593.016010284424, "token_normalized_loss": 0.31415737620571205, "grad_norm_before_clip": 0.12404319643974304, "learning_rates_at_step": { "lora": 9.795757363905944e-05, "row_tokens": 9.795757363905944e-05 }, "learning_rates_after_step": { "lora": 9.766145069157434e-05, "row_tokens": 9.766145069157434e-05 } }, { "update_id": 24, "row_count": 67, "row_ids_sha256": "b67fa36c6eaa9a27ae13afd254ddce5dc151e2a52f14274f7758502e5bfdea43", "input_ids_vector_sha256": "cc5351aa2aa91c475a0b0158b25265ee83f5c9b7b6e4e9a39cccf9f3bb0c5494", "labels_vector_sha256": "4ea72a621ea22e5d831f108254602c3ae9889903990d870a76958b71ed20dfdd", "planned_active_tokens": 65344, "realized_active_tokens": 65344, "cumulative_start": 1573180, "cumulative_end": 1638524, "loss_sum": 25915.75656890869, "token_normalized_loss": 0.39660499156630585, "grad_norm_before_clip": 0.13470543920993805, "learning_rates_at_step": { "lora": 9.766145069157434e-05, "row_tokens": 9.766145069157434e-05 }, "learning_rates_after_step": { "lora": 9.734343372735864e-05, "row_tokens": 9.734343372735864e-05 } }, { "update_id": 25, "row_count": 70, "row_ids_sha256": "48074a75c08516869c32ee83176b555d7a5ae7e1cbe6a6944dddab9d1f5ed280", "input_ids_vector_sha256": "99fa0c5e46d2755cb78b680d63b136918e027c919fb4841920293235229eaaca", "labels_vector_sha256": "c895331428b2e140796543038fa876388c5b40c643b75ec58930cda0bac411a5", "planned_active_tokens": 65819, "realized_active_tokens": 65819, "cumulative_start": 1638524, "cumulative_end": 1704343, "loss_sum": 18845.561157226562, "token_normalized_loss": 0.2863240273663617, "grad_norm_before_clip": 0.11002977937459946, "learning_rates_at_step": { "lora": 9.734343372735864e-05, "row_tokens": 9.734343372735864e-05 }, "learning_rates_after_step": { "lora": 9.700732994374042e-05, "row_tokens": 9.700732994374042e-05 } }, { "update_id": 26, "row_count": 78, "row_ids_sha256": "b4e12a4eff52dc5008dcb6d5ec7b012284dc93568ddb96f99bb37233eeb4179d", "input_ids_vector_sha256": "b2f60411c1353c6c5f22bd0c864b18dbc7a30183797ee7fb2a45eaca7a34fccd", "labels_vector_sha256": "e6b9d6488afc0c2a3bc1520eb5fce76ed3a2e0736b2c91a2cc441395a967ca4a", "planned_active_tokens": 65517, "realized_active_tokens": 65517, "cumulative_start": 1704343, "cumulative_end": 1769860, "loss_sum": 16546.60173034668, "token_normalized_loss": 0.25255432529491095, "grad_norm_before_clip": 0.11002335697412491, "learning_rates_at_step": { "lora": 9.700732994374042e-05, "row_tokens": 9.700732994374042e-05 }, "learning_rates_after_step": { "lora": 9.664667931800648e-05, "row_tokens": 9.664667931800648e-05 } }, { "update_id": 27, "row_count": 78, "row_ids_sha256": "08eccec02014f502d66863072f3ecbc21acdee01f3891d6c263188fc8a829311", "input_ids_vector_sha256": "97ee1d9078292f534ae076920d59cde7411bed25c3961b4103b9ece52fb00e99", "labels_vector_sha256": "7e27ecf50d5bf74964285745adcab48f3f609ede6a3563a7bf41deb6449e1144", "planned_active_tokens": 66447, "realized_active_tokens": 66447, "cumulative_start": 1769860, "cumulative_end": 1836307, "loss_sum": 15863.474914550781, "token_normalized_loss": 0.2387387679586856, "grad_norm_before_clip": 0.1050301045179367, "learning_rates_at_step": { "lora": 9.664667931800648e-05, "row_tokens": 9.664667931800648e-05 }, "learning_rates_after_step": { "lora": 9.627604962003078e-05, "row_tokens": 9.627604962003078e-05 } }, { "update_id": 28, "row_count": 76, "row_ids_sha256": "0a29dffd60f60e28f932e337139b2ce0d215a89806ee0f448df2bb1a1bc26af0", "input_ids_vector_sha256": "0c2b7d2bef71a10c86ffd2a3d06e94629e117d49b7e3bf7ee46a938ab83f4c7a", "labels_vector_sha256": "929874c5de22f6c7f1530c9f0896a2c96a6823dbf2bf785321bc0019e9d01f7e", "planned_active_tokens": 64780, "realized_active_tokens": 64780, "cumulative_start": 1836307, "cumulative_end": 1901087, "loss_sum": 15431.188507080078, "token_normalized_loss": 0.2382091464507576, "grad_norm_before_clip": 0.10523726046085358, "learning_rates_at_step": { "lora": 9.627604962003078e-05, "row_tokens": 9.627604962003078e-05 }, "learning_rates_after_step": { "lora": 9.588492702807173e-05, "row_tokens": 9.588492702807173e-05 } }, { "update_id": 29, "row_count": 71, "row_ids_sha256": "f811df3db1831a2924fe79862bba3047f155ac3957bf16955db407c4c3fba46e", "input_ids_vector_sha256": "0739d5590b7cdf1a474bffb2d3698cbea410617d16c6df49a27824c7eeb52850", "labels_vector_sha256": "e8232c984d7a08caccfe8e834ee6c72edd7230c08116947fe87112c6d26004bf", "planned_active_tokens": 65082, "realized_active_tokens": 65082, "cumulative_start": 1901087, "cumulative_end": 1966169, "loss_sum": 13869.645198822021, "token_normalized_loss": 0.21311031005227285, "grad_norm_before_clip": 0.09779830276966095, "learning_rates_at_step": { "lora": 9.588492702807173e-05, "row_tokens": 9.588492702807173e-05 }, "learning_rates_after_step": { "lora": 9.547212782391059e-05, "row_tokens": 9.547212782391059e-05 } }, { "update_id": 30, "row_count": 75, "row_ids_sha256": "413690534e81c5defa21d2634e9ff007cb345214c4232903088db03461415cda", "input_ids_vector_sha256": "56e9fac5ebc2c42b30524ee7e6cfb85071c5c820a29d10471ac9d7d8912b0a6e", "labels_vector_sha256": "f902245fce0e941f9556df177c7fa02edc90fcbb40c3de5ccf7542c1a4f7ad8e", "planned_active_tokens": 65552, "realized_active_tokens": 65552, "cumulative_start": 1966169, "cumulative_end": 2031721, "loss_sum": 14763.468719482422, "token_normalized_loss": 0.22521767023862616, "grad_norm_before_clip": 0.09477154165506363, "learning_rates_at_step": { "lora": 9.547212782391059e-05, "row_tokens": 9.547212782391059e-05 }, "learning_rates_after_step": { "lora": 9.503573512385892e-05, "row_tokens": 9.503573512385892e-05 } }, { "update_id": 31, "row_count": 77, "row_ids_sha256": "704a3203aeb9a2a26ff26df66b524e749cc8a41e67f330ad7bce30e00571dea2", "input_ids_vector_sha256": "c30443b7a3b1dabd564531800e96ac55f897b7c1ad991d48ca7e46677a85253f", "labels_vector_sha256": "9368ed31860b1c00c650601a4ac601e0794c017400489de18f8dfc3ed6b782cb", "planned_active_tokens": 66272, "realized_active_tokens": 66272, "cumulative_start": 2031721, "cumulative_end": 2097993, "loss_sum": 12918.119758605957, "token_normalized_loss": 0.19492575685969876, "grad_norm_before_clip": 0.09158577769994736, "learning_rates_at_step": { "lora": 9.503573512385892e-05, "row_tokens": 9.503573512385892e-05 }, "learning_rates_after_step": { "lora": 9.458590849592805e-05, "row_tokens": 9.458590849592805e-05 } }, { "update_id": 32, "row_count": 82, "row_ids_sha256": "6d81ffb902d49e4f75423a9a8b4fbdd0a9434ee2ac549722d2de0c13b14031aa", "input_ids_vector_sha256": "2eee393358898a53d913f2d8d9541130a6dd5358c6db90285ae5daf272a3b35f", "labels_vector_sha256": "f535e38ff7f3d8d955542a70eb597690feec84335ee337c793e2fd112659c0bc", "planned_active_tokens": 65481, "realized_active_tokens": 65481, "cumulative_start": 2097993, "cumulative_end": 2163474, "loss_sum": 13965.123626708984, "token_normalized_loss": 0.21326985884010605, "grad_norm_before_clip": 0.10092455893754959, "learning_rates_at_step": { "lora": 9.458590849592805e-05, "row_tokens": 9.458590849592805e-05 }, "learning_rates_after_step": { "lora": 9.412241559263748e-05, "row_tokens": 9.412241559263748e-05 } }, { "update_id": 33, "row_count": 73, "row_ids_sha256": "47447ab73c40cd79a10e9d92febb3cfbd055786b20f778b61cf8e7792a704dd1", "input_ids_vector_sha256": "efad90d2017d7b773b61ee304bae53830b22af6ecd6d6932680f58886df41504", "labels_vector_sha256": "e56eb9bd25fd276f739aca6a62cc50cc63ff2c3a627d64729772cdc076e99558", "planned_active_tokens": 64839, "realized_active_tokens": 64839, "cumulative_start": 2163474, "cumulative_end": 2228313, "loss_sum": 12883.099308013916, "token_normalized_loss": 0.1986936767688261, "grad_norm_before_clip": 0.09143892675638199, "learning_rates_at_step": { "lora": 9.412241559263748e-05, "row_tokens": 9.412241559263748e-05 }, "learning_rates_after_step": { "lora": 9.36332060399158e-05, "row_tokens": 9.36332060399158e-05 } }, { "update_id": 34, "row_count": 73, "row_ids_sha256": "b1cd174bfb32465bc45604243d5af374c458a770f5a0d3324730a114f045fcaf", "input_ids_vector_sha256": "3b246eafe77e2d97cca6dd71094b69942b27ccd05fd24c399ab9ffd9dbbce2f5", "labels_vector_sha256": "d596e6ccfddab59b89944b70f516893459eccff1c01d15c76c43672a6ba6d136", "planned_active_tokens": 65886, "realized_active_tokens": 65886, "cumulative_start": 2228313, "cumulative_end": 2294199, "loss_sum": 12673.125274658203, "token_normalized_loss": 0.19234928929754733, "grad_norm_before_clip": 0.09066075831651688, "learning_rates_at_step": { "lora": 9.36332060399158e-05, "row_tokens": 9.36332060399158e-05 }, "learning_rates_after_step": { "lora": 9.313121233038475e-05, "row_tokens": 9.313121233038475e-05 } }, { "update_id": 35, "row_count": 75, "row_ids_sha256": "d89903736afacf98cc20773015175bb55d2d5ba2ad69468004a628f27b7d5806", "input_ids_vector_sha256": "9a178c40345ae25ffdeac0dadc3bb2cc9f1e4f814940e8fdf0359e5ffbf18240", "labels_vector_sha256": "d6937f845c1d3d3b6390c6be0f064a9c19885e6166df8dfd00f1d4130fdd20a3", "planned_active_tokens": 65198, "realized_active_tokens": 65198, "cumulative_start": 2294199, "cumulative_end": 2359397, "loss_sum": 11437.058609008789, "token_normalized_loss": 0.17542039033419413, "grad_norm_before_clip": 0.08434521406888962, "learning_rates_at_step": { "lora": 9.313121233038475e-05, "row_tokens": 9.313121233038475e-05 }, "learning_rates_after_step": { "lora": 9.26034957930445e-05, "row_tokens": 9.26034957930445e-05 } }, { "update_id": 36, "row_count": 76, "row_ids_sha256": "38ab5a1c101a70628683e0d06bedde6aef5b63539306662a1ae8ca9927d08da1", "input_ids_vector_sha256": "b06c1059260edbc71687f5984648bc5288ded0536cbbdb0ab2acd422679228fe", "labels_vector_sha256": "af971c5d609c0eca68f366ece123431a8dc28df521b2893e8939596488cbb9ff", "planned_active_tokens": 66201, "realized_active_tokens": 66201, "cumulative_start": 2359397, "cumulative_end": 2425598, "loss_sum": 11934.491706848145, "token_normalized_loss": 0.18027660770755946, "grad_norm_before_clip": 0.09112250059843063, "learning_rates_at_step": { "lora": 9.26034957930445e-05, "row_tokens": 9.26034957930445e-05 }, "learning_rates_after_step": { "lora": 9.206316384492652e-05, "row_tokens": 9.206316384492652e-05 } }, { "update_id": 37, "row_count": 79, "row_ids_sha256": "040addb0c10bbba569084c7db5761b67e5ef1b49694672d72613eb99bab9ea58", "input_ids_vector_sha256": "7753a6dcb6ce9f5e807c5158989d18a7494bccde506bed84bc5b65783f03d7ae", "labels_vector_sha256": "e8c0ee6491c79c7d071db3282860f66cff0252eab435e1189aaa249cd33d9c8a", "planned_active_tokens": 65568, "realized_active_tokens": 65568, "cumulative_start": 2425598, "cumulative_end": 2491166, "loss_sum": 11678.807796478271, "token_normalized_loss": 0.1781174932356984, "grad_norm_before_clip": 0.08789415657520294, "learning_rates_at_step": { "lora": 9.206316384492652e-05, "row_tokens": 9.206316384492652e-05 }, "learning_rates_after_step": { "lora": 9.150597994105091e-05, "row_tokens": 9.150597994105091e-05 } }, { "update_id": 38, "row_count": 73, "row_ids_sha256": "1d75077f7568c6d720c339d61a29477bbaffda542b168e18ecd387f07c8b1186", "input_ids_vector_sha256": "ef6858d671ee40037dd1bd518afda57f8c4ba273c84a7d8f7987566c1a0d208d", "labels_vector_sha256": "55f77698db3ced6be11fd9abe586adbff2b2c4b48ae0563cd55614dae8f1c90d", "planned_active_tokens": 65510, "realized_active_tokens": 65510, "cumulative_start": 2491166, "cumulative_end": 2556676, "loss_sum": 11348.690246582031, "token_normalized_loss": 0.1732359982686923, "grad_norm_before_clip": 0.09103299677371979, "learning_rates_at_step": { "lora": 9.150597994105091e-05, "row_tokens": 9.150597994105091e-05 }, "learning_rates_after_step": { "lora": 9.093607942075762e-05, "row_tokens": 9.093607942075762e-05 } }, { "update_id": 39, "row_count": 73, "row_ids_sha256": "3334538ba5751bf7be53ffca9d6b1f3f279963c46067ed23d3d336435d757d71", "input_ids_vector_sha256": "fd9716fb5ca92aa332ac30815f7dccb13781148b3bebee3eed113213ebda829a", "labels_vector_sha256": "84a4d502e12644bbc8021a3b0f278ff22069ceebe66a2bbba8a9123925a5517a", "planned_active_tokens": 65018, "realized_active_tokens": 65018, "cumulative_start": 2556676, "cumulative_end": 2621694, "loss_sum": 12509.35122680664, "token_normalized_loss": 0.19239827781240026, "grad_norm_before_clip": 0.0945063978433609, "learning_rates_at_step": { "lora": 9.093607942075762e-05, "row_tokens": 9.093607942075762e-05 }, "learning_rates_after_step": { "lora": 9.034321490530057e-05, "row_tokens": 9.034321490530057e-05 } }, { "update_id": 40, "row_count": 74, "row_ids_sha256": "5d2c22b4d7460dc75e5548e14c65ff83589121d4930b6cdfd91d4bd9b7a934e4", "input_ids_vector_sha256": "de46fac2f40bf6bc498f1630df5fe8356c449d9b2941ac5a55bdb5d306f0e0a2", "labels_vector_sha256": "5cd280280105925f0d473704f3c025e166f53fb9e091348c1f62a97a198d4559", "planned_active_tokens": 65713, "realized_active_tokens": 65713, "cumulative_start": 2621694, "cumulative_end": 2687407, "loss_sum": 14710.798637390137, "token_normalized_loss": 0.2238643592194868, "grad_norm_before_clip": 0.1028904914855957, "learning_rates_at_step": { "lora": 9.034321490530057e-05, "row_tokens": 9.034321490530057e-05 }, "learning_rates_after_step": { "lora": 8.973337951557174e-05, "row_tokens": 8.973337951557174e-05 } }, { "update_id": 41, "row_count": 78, "row_ids_sha256": "b20ae6007b7b814501c9c4cb9e0f1b179cf35fa5c405abc8b6dbc3517aa97425", "input_ids_vector_sha256": "8e89502a2f44ba6499972eb212b0428e927bc075bd989604893aa5d7bd5b8189", "labels_vector_sha256": "a64c9ec36264f51c3f4ab337cc9772e23df1931196e07cee37c027c4870e1e60", "planned_active_tokens": 65740, "realized_active_tokens": 65740, "cumulative_start": 2687407, "cumulative_end": 2753147, "loss_sum": 15293.534210205078, "token_normalized_loss": 0.23263666276551687, "grad_norm_before_clip": 0.09642995148897171, "learning_rates_at_step": { "lora": 8.973337951557174e-05, "row_tokens": 8.973337951557174e-05 }, "learning_rates_after_step": { "lora": 8.910445068418351e-05, "row_tokens": 8.910445068418351e-05 } }, { "update_id": 42, "row_count": 78, "row_ids_sha256": "781f65aa90845d4b7cff9d07e67b57d223309b5e5a2d9d6166cd81d7ea980989", "input_ids_vector_sha256": "f54fc2d9f3e285acf4a998cbabc6b9fb7506041caddbe6b336a40ef41c9566d0", "labels_vector_sha256": "ef2269e9a073c8afe7959c4f742ae3a1d159aaf775caf537d80904a05a638e85", "planned_active_tokens": 66011, "realized_active_tokens": 66011, "cumulative_start": 2753147, "cumulative_end": 2819158, "loss_sum": 13857.143394470215, "token_normalized_loss": 0.20992173114284307, "grad_norm_before_clip": 0.13601583242416382, "learning_rates_at_step": { "lora": 8.910445068418351e-05, "row_tokens": 8.910445068418351e-05 }, "learning_rates_after_step": { "lora": 8.84735880851911e-05, "row_tokens": 8.84735880851911e-05 } }, { "update_id": 43, "row_count": 74, "row_ids_sha256": "67c638bb44c1f05f1edc919b100d5374b41c805b6193f718c0500dba7d631a14", "input_ids_vector_sha256": "4cee4e5682959d2d2cbc8a807f824f7872f18f9a23a04ac1c6fef0edf6a37452", "labels_vector_sha256": "91c43b511b61c108f9b3bad194696bb21d9d698a5d9b69ce59265f0fbb7d9e79", "planned_active_tokens": 64554, "realized_active_tokens": 64554, "cumulative_start": 2819158, "cumulative_end": 2883712, "loss_sum": 13128.328540802002, "token_normalized_loss": 0.20336971436010165, "grad_norm_before_clip": 0.09796834737062454, "learning_rates_at_step": { "lora": 8.84735880851911e-05, "row_tokens": 8.84735880851911e-05 }, "learning_rates_after_step": { "lora": 8.781481860875498e-05, "row_tokens": 8.781481860875498e-05 } }, { "update_id": 44, "row_count": 78, "row_ids_sha256": "70072bf5ef82ec9febf93fe494b1e4f491c6abc9f29b61123955d540646abb95", "input_ids_vector_sha256": "2e7ab9e8737e1f76a1ad3dba0d516e29bc699b56f862f1e5a5f4d84cbcd38bca", "labels_vector_sha256": "96a3c80c8c8f27e304f5142790d816d09137aeb45c3cb4d0e504c721cd7293a8", "planned_active_tokens": 65790, "realized_active_tokens": 65790, "cumulative_start": 2883712, "cumulative_end": 2949502, "loss_sum": 12356.7611618042, "token_normalized_loss": 0.18782126708928712, "grad_norm_before_clip": 0.0956386849284172, "learning_rates_at_step": { "lora": 8.781481860875498e-05, "row_tokens": 8.781481860875498e-05 }, "learning_rates_after_step": { "lora": 8.714022900305599e-05, "row_tokens": 8.714022900305599e-05 } }, { "update_id": 45, "row_count": 81, "row_ids_sha256": "464929e273a57927d0a61338c9b07bb0008b392c040b212a3a3032f5d150879d", "input_ids_vector_sha256": "086b240f0a1ed2ec2e373c24cb1e60048c1164aac8cc6aa92466680a34c65018", "labels_vector_sha256": "2a2dcd010bdf7266d63ca08e407458d990224982b8fd9866c7e0b44f495534fb", "planned_active_tokens": 65801, "realized_active_tokens": 65801, "cumulative_start": 2949502, "cumulative_end": 3015303, "loss_sum": 12519.497184753418, "token_normalized_loss": 0.19026302312660018, "grad_norm_before_clip": 0.0913286954164505, "learning_rates_at_step": { "lora": 8.714022900305599e-05, "row_tokens": 8.714022900305599e-05 }, "learning_rates_after_step": { "lora": 8.645660090700178e-05, "row_tokens": 8.645660090700178e-05 } }, { "update_id": 46, "row_count": 80, "row_ids_sha256": "e4da090f4c15534838d5a488fc78621c519c11c40440a9a927a2a183bc4a6166", "input_ids_vector_sha256": "bfb3bf4a4efba9f58ba87e124745c771d9af58be2a1f3e7746521e480f07c3e3", "labels_vector_sha256": "db043acaee8a11c7af65e995c0aca00cb297c545a4ed42066040a0cb1c1e937f", "planned_active_tokens": 65198, "realized_active_tokens": 65198, "cumulative_start": 3015303, "cumulative_end": 3080501, "loss_sum": 12186.700733184814, "token_normalized_loss": 0.18691832162312977, "grad_norm_before_clip": 0.09290819615125656, "learning_rates_at_step": { "lora": 8.645660090700178e-05, "row_tokens": 8.645660090700178e-05 }, "learning_rates_after_step": { "lora": 8.575357653709422e-05, "row_tokens": 8.575357653709422e-05 } }, { "update_id": 47, "row_count": 75, "row_ids_sha256": "d907e892106ae06d8461e74749854d69998bfb1d90c808ac1c861a5383a64a4c", "input_ids_vector_sha256": "e6d81c69edfaa0e8389d29cf14054203b054b9f7f3b3a622989728a1d8eb3e0f", "labels_vector_sha256": "4878754331de1ac00c33841f9877f5c9730695e6b82838cfb13668e440d24b14", "planned_active_tokens": 65616, "realized_active_tokens": 65616, "cumulative_start": 3080501, "cumulative_end": 3146117, "loss_sum": 12683.943115234375, "token_normalized_loss": 0.19330564367279893, "grad_norm_before_clip": 0.09388837218284607, "learning_rates_at_step": { "lora": 8.575357653709422e-05, "row_tokens": 8.575357653709422e-05 }, "learning_rates_after_step": { "lora": 8.503044293229835e-05, "row_tokens": 8.503044293229835e-05 } }, { "update_id": 48, "row_count": 77, "row_ids_sha256": "2bb0dd39e407a5e1e27291f99fa6f947aedfd934d76656326233e0cd29179d7d", "input_ids_vector_sha256": "f37a8bbdb36757c1accdc101c742b76b22f55370f5277f505bfb76a687f661f7", "labels_vector_sha256": "5a624de5115adb7382a1c1f200ed33c331346b32b3d04853a618b157e9f214d7", "planned_active_tokens": 66099, "realized_active_tokens": 66099, "cumulative_start": 3146117, "cumulative_end": 3212216, "loss_sum": 12352.946899414062, "token_normalized_loss": 0.18688553381161685, "grad_norm_before_clip": 0.08651527762413025, "learning_rates_at_step": { "lora": 8.503044293229835e-05, "row_tokens": 8.503044293229835e-05 }, "learning_rates_after_step": { "lora": 8.430572973181359e-05, "row_tokens": 8.430572973181359e-05 } }, { "update_id": 49, "row_count": 73, "row_ids_sha256": "29b8e2f8483cee300dcdfee06682715a7bb6a5f8f8b0ef3ae592936b6f3b7237", "input_ids_vector_sha256": "996ac556b54e2126d651e745dc6bfa14544fc8f30cb580c6c893cb9497e9d771", "labels_vector_sha256": "260cac8865a501bf734ee45e87032e90ebf058a8d01c4aad2691b5f14261c128", "planned_active_tokens": 64936, "realized_active_tokens": 64936, "cumulative_start": 3212216, "cumulative_end": 3277152, "loss_sum": 10457.10620880127, "token_normalized_loss": 0.16103711668105936, "grad_norm_before_clip": 0.08751116693019867, "learning_rates_at_step": { "lora": 8.430572973181359e-05, "row_tokens": 8.430572973181359e-05 }, "learning_rates_after_step": { "lora": 8.355603321601993e-05, "row_tokens": 8.355603321601993e-05 } }, { "update_id": 50, "row_count": 79, "row_ids_sha256": "33cb7e87edc6cc4d0991090de440080183d0a97a8ed9486f463b2985cff9edd2", "input_ids_vector_sha256": "956d525ebcd7b138567c41e7e30f9eadd1554eeb9824700f7c86303924c3148c", "labels_vector_sha256": "b52d38abc9b7bc26be34249aa7cb81504824cc1531d0a64d3a75ba7843f5ce77", "planned_active_tokens": 65903, "realized_active_tokens": 65903, "cumulative_start": 3277152, "cumulative_end": 3343055, "loss_sum": 11601.571437835693, "token_normalized_loss": 0.1760401110394928, "grad_norm_before_clip": 0.09314046800136566, "learning_rates_at_step": { "lora": 8.355603321601993e-05, "row_tokens": 8.355603321601993e-05 }, "learning_rates_after_step": { "lora": 8.279957691690663e-05, "row_tokens": 8.279957691690663e-05 } }, { "update_id": 51, "row_count": 76, "row_ids_sha256": "0474843e91bc842f3182e84a5de38c012854629a37788dc7c1d70a8cbf7bb898", "input_ids_vector_sha256": "7035cdb3889c07dd82b33414c6ea5737d60abcf4938052753dac1b710e9a3fe6", "labels_vector_sha256": "90a47ffa94ea148a12213acf415f6015efaf12d4572f7bba6a3e8f952ba79e18", "planned_active_tokens": 65285, "realized_active_tokens": 65285, "cumulative_start": 3343055, "cumulative_end": 3408340, "loss_sum": 11881.37608718872, "token_normalized_loss": 0.18199243451311511, "grad_norm_before_clip": 0.0944008007645607, "learning_rates_at_step": { "lora": 8.279957691690663e-05, "row_tokens": 8.279957691690663e-05 }, "learning_rates_after_step": { "lora": 8.202607823271779e-05, "row_tokens": 8.202607823271779e-05 } }, { "update_id": 52, "row_count": 74, "row_ids_sha256": "76e6b933c37024ea663425bd15244ddf2668ea16a8289f427965d72bb7f2cf1e", "input_ids_vector_sha256": "864e5e2fbbccc82648038ae9e1451cd112fa9cb2a2f2d132223779bec8a8e7ce", "labels_vector_sha256": "3ee0f4423f604b7aecfc96a4d0cdcbd5edfdfdc5313903e744075fc1dccdea7b", "planned_active_tokens": 65590, "realized_active_tokens": 65590, "cumulative_start": 3408340, "cumulative_end": 3473930, "loss_sum": 13664.27742767334, "token_normalized_loss": 0.2083286694263354, "grad_norm_before_clip": 0.09272623062133789, "learning_rates_at_step": { "lora": 8.202607823271779e-05, "row_tokens": 8.202607823271779e-05 }, "learning_rates_after_step": { "lora": 8.124524779147817e-05, "row_tokens": 8.124524779147817e-05 } }, { "update_id": 53, "row_count": 77, "row_ids_sha256": "88cdb567e625ba17fa837718ab5fc3b952f6352061a13ac8f26b6bc2526a893d", "input_ids_vector_sha256": "eaa02c8e6b4519882a029edf35447bcefacddc756287b0d9586f40b451add6c7", "labels_vector_sha256": "4e93cbc2041116241ff4ed0cf6a473f67c0aa24bfc5ed6edd5031efb436e393a", "planned_active_tokens": 65103, "realized_active_tokens": 65103, "cumulative_start": 3473930, "cumulative_end": 3539033, "loss_sum": 13818.952030181885, "token_normalized_loss": 0.21226290693488603, "grad_norm_before_clip": 0.09860926121473312, "learning_rates_at_step": { "lora": 8.124524779147817e-05, "row_tokens": 8.124524779147817e-05 }, "learning_rates_after_step": { "lora": 8.044707788795987e-05, "row_tokens": 8.044707788795987e-05 } }, { "update_id": 54, "row_count": 73, "row_ids_sha256": "bdb2e91f9f7fdc303689d1f9a30b62eff13469f0bbfc45ff7a105e8a0b791b81", "input_ids_vector_sha256": "75b9e95edb2587f6abacddadc4922f470692429bbe56451a61243b58543ef7c2", "labels_vector_sha256": "cddb8f2bf1402e8d743603eca62f93fe68638eec427bce6094c58b27ad69a430", "planned_active_tokens": 65480, "realized_active_tokens": 65480, "cumulative_start": 3539033, "cumulative_end": 3604513, "loss_sum": 12728.849304199219, "token_normalized_loss": 0.19439293378434971, "grad_norm_before_clip": 0.09796017408370972, "learning_rates_at_step": { "lora": 8.044707788795987e-05, "row_tokens": 8.044707788795987e-05 }, "learning_rates_after_step": { "lora": 7.963414790736901e-05, "row_tokens": 7.963414790736901e-05 } }, { "update_id": 55, "row_count": 76, "row_ids_sha256": "8a9550045c4a7a0a1765a864f41b83441df23a9ea36aae7c983041d7ce2e59e6", "input_ids_vector_sha256": "6a71cae469c1cf40c37654feb90b50e728daab49d30e09a3c960926f9e6e4a30", "labels_vector_sha256": "4ed22cdd7b60de88e5241fe7d430118b1e552cae40679006f740f3c1c62da848", "planned_active_tokens": 65659, "realized_active_tokens": 65659, "cumulative_start": 3604513, "cumulative_end": 3670172, "loss_sum": 12996.634384155273, "token_normalized_loss": 0.19794140002368713, "grad_norm_before_clip": 0.09664937853813171, "learning_rates_at_step": { "lora": 7.963414790736901e-05, "row_tokens": 7.963414790736901e-05 }, "learning_rates_after_step": { "lora": 7.880992071238274e-05, "row_tokens": 7.880992071238274e-05 } }, { "update_id": 56, "row_count": 78, "row_ids_sha256": "9d5a01d7e6894b25715e764a532e590dc5ff504c55830b438edd8401b966a063", "input_ids_vector_sha256": "fe4e3339e910545a9e243649e91a8ae38f7522929bc9dbfbd8f4c2969c5f1611", "labels_vector_sha256": "5c633432813d36658bae7c9b8495b9b4971f65cb14229e52e4a2056b3a95a19b", "planned_active_tokens": 65583, "realized_active_tokens": 65583, "cumulative_start": 3670172, "cumulative_end": 3735755, "loss_sum": 11807.507007598877, "token_normalized_loss": 0.1800391413567369, "grad_norm_before_clip": 0.08626383543014526, "learning_rates_at_step": { "lora": 7.880992071238274e-05, "row_tokens": 7.880992071238274e-05 }, "learning_rates_after_step": { "lora": 7.797004111875818e-05, "row_tokens": 7.797004111875818e-05 } }, { "update_id": 57, "row_count": 77, "row_ids_sha256": "1d66c19478d4c4b53b72445a8dff78952eefb6d4b5744331da5a57430d80f173", "input_ids_vector_sha256": "3d847804cb5680385e41319f68e2052f191a18581c7d26c17659a8aaebda0c2a", "labels_vector_sha256": "8f047c4790f991a78c20db387a4bd5a6184d7a79a3204c4b5c282658d2cdf395", "planned_active_tokens": 65876, "realized_active_tokens": 65876, "cumulative_start": 3735755, "cumulative_end": 3801631, "loss_sum": 12726.984943389893, "token_normalized_loss": 0.19319607965556337, "grad_norm_before_clip": 0.0934160128235817, "learning_rates_at_step": { "lora": 7.797004111875818e-05, "row_tokens": 7.797004111875818e-05 }, "learning_rates_after_step": { "lora": 7.71265424378469e-05, "row_tokens": 7.71265424378469e-05 } }, { "update_id": 58, "row_count": 74, "row_ids_sha256": "b2a27c86509c02d906f6797785b32df5bf8639143966f89e7cd40145bc1fb817", "input_ids_vector_sha256": "1e705348900a40d0567c07c9ffe0eca927cf71faad733e4e5baeeda6c56394cf", "labels_vector_sha256": "2f19cab70807d8119bcaf93d9858243496446ab248931f3c9babd1af4dc4ab50", "planned_active_tokens": 65259, "realized_active_tokens": 65259, "cumulative_start": 3801631, "cumulative_end": 3866890, "loss_sum": 12094.33895111084, "token_normalized_loss": 0.18532829113395607, "grad_norm_before_clip": 0.08985763043165207, "learning_rates_at_step": { "lora": 7.71265424378469e-05, "row_tokens": 7.71265424378469e-05 }, "learning_rates_after_step": { "lora": 7.626774088534177e-05, "row_tokens": 7.626774088534177e-05 } }, { "update_id": 59, "row_count": 76, "row_ids_sha256": "291b5e1d761c8af31f2556fd610a473736d83db87cc408b168ea7f7e435ca115", "input_ids_vector_sha256": "538debee29c9cc81d926a451f66419171dd17e7cc3323867c1067032c1bb5e0b", "labels_vector_sha256": "037337f2eae44028186feb35bab3cda87af68809ba1644f8e47b0a8ae28ae83a", "planned_active_tokens": 65579, "realized_active_tokens": 65579, "cumulative_start": 3866890, "cumulative_end": 3932469, "loss_sum": 11171.31481552124, "token_normalized_loss": 0.17034896560669177, "grad_norm_before_clip": 0.08831323683261871, "learning_rates_at_step": { "lora": 7.626774088534177e-05, "row_tokens": 7.626774088534177e-05 }, "learning_rates_after_step": { "lora": 7.540264853446332e-05, "row_tokens": 7.540264853446332e-05 } }, { "update_id": 60, "row_count": 77, "row_ids_sha256": "ed73c238f727339f16dbedb471cc2e5957df7bacdf54aa4398f345baa9599c41", "input_ids_vector_sha256": "8a573fbf34ee026583e81ecbfa77e2831adb9d6e516143fa035dd5668f6a201c", "labels_vector_sha256": "1d197d9f005a7b8e658223930659e71c93e1fbd7ae51e6ee5349f4e4bc782f16", "planned_active_tokens": 65238, "realized_active_tokens": 65238, "cumulative_start": 3932469, "cumulative_end": 3997707, "loss_sum": 11039.529731750488, "token_normalized_loss": 0.1692193159163446, "grad_norm_before_clip": 0.08440037071704865, "learning_rates_at_step": { "lora": 7.540264853446332e-05, "row_tokens": 7.540264853446332e-05 }, "learning_rates_after_step": { "lora": 7.451618065124546e-05, "row_tokens": 7.451618065124546e-05 } }, { "update_id": 61, "row_count": 76, "row_ids_sha256": "b0b0db7b098c205ef08ed9ea3f2550aefe9db6ce460156bc246f2fd3cd00a5fc", "input_ids_vector_sha256": "6a5025b5c56629bdb10ef8df2ce697393e2573f14e99a8993689c35d3795e4d3", "labels_vector_sha256": "14b8f0baeb9f74b058232af7f1b8b555577c8d473fb47ac409c7c1fe0843989c", "planned_active_tokens": 66053, "realized_active_tokens": 66053, "cumulative_start": 3997707, "cumulative_end": 4063760, "loss_sum": 10913.388122558594, "token_normalized_loss": 0.16522168747155458, "grad_norm_before_clip": 0.0854608342051506, "learning_rates_at_step": { "lora": 7.451618065124546e-05, "row_tokens": 7.451618065124546e-05 }, "learning_rates_after_step": { "lora": 7.362763925643086e-05, "row_tokens": 7.362763925643086e-05 } }, { "update_id": 62, "row_count": 77, "row_ids_sha256": "b35103eb713f1bccd2ef5de035bb819f2793cc185c6954cb9e76e1b9b82a4389", "input_ids_vector_sha256": "93446d7f07e51355c03f8a827efd4805e98ca96d68a814ca9151894945963260", "labels_vector_sha256": "c115ccd24098559483ee15784b887616e12eebb0fefb17bdfe8916c3f43547d1", "planned_active_tokens": 65453, "realized_active_tokens": 65453, "cumulative_start": 4063760, "cumulative_end": 4129213, "loss_sum": 10250.972137451172, "token_normalized_loss": 0.1566157721945697, "grad_norm_before_clip": 0.08557523041963577, "learning_rates_at_step": { "lora": 7.362763925643086e-05, "row_tokens": 7.362763925643086e-05 }, "learning_rates_after_step": { "lora": 7.273275075565455e-05, "row_tokens": 7.273275075565455e-05 } }, { "update_id": 63, "row_count": 80, "row_ids_sha256": "4f68e702212d9269aa197df4721994bb91fb228154e4baf417f3004b11f85bf5", "input_ids_vector_sha256": "927258344a497f39e35f93f628e5d2c97558dee679b363da3ec3a1e66e544ba7", "labels_vector_sha256": "6d8de7fdd4b11ca1121e3b7c1fa74572ac5a8a39bda28b558f96ea673bedb761", "planned_active_tokens": 65209, "realized_active_tokens": 65209, "cumulative_start": 4129213, "cumulative_end": 4194422, "loss_sum": 9857.580814361572, "token_normalized_loss": 0.15116902290115738, "grad_norm_before_clip": 0.08248785138130188, "learning_rates_at_step": { "lora": 7.273275075565455e-05, "row_tokens": 7.273275075565455e-05 }, "learning_rates_after_step": { "lora": 7.181644834980592e-05, "row_tokens": 7.181644834980592e-05 } }, { "update_id": 64, "row_count": 78, "row_ids_sha256": "1e898517c6bdb37cd696b9dddf7a96c0663d8590f34433b6b4dc7345151854f7", "input_ids_vector_sha256": "289c90fdf6a628b5146548735211f031230181d1785021c4aa62e6cc439e0638", "labels_vector_sha256": "c888bcffff5a6c91f605e7c8fe7c08c9b80d5f6b6b5c8869a1606da96de17c49", "planned_active_tokens": 66080, "realized_active_tokens": 66080, "cumulative_start": 4194422, "cumulative_end": 4260502, "loss_sum": 10932.211711883545, "token_normalized_loss": 0.16543903922341927, "grad_norm_before_clip": 0.0877639502286911, "learning_rates_at_step": { "lora": 7.181644834980592e-05, "row_tokens": 7.181644834980592e-05 }, "learning_rates_after_step": { "lora": 7.089782935335964e-05, "row_tokens": 7.089782935335964e-05 } }, { "update_id": 65, "row_count": 83, "row_ids_sha256": "5ee8aac9cd91275b400a4fd24bd23561429895531ec0fb88eaff5e9aeb322446", "input_ids_vector_sha256": "411aed6d4ccd152950336043395e83e7776153cb239b0372e417a2758c6108e9", "labels_vector_sha256": "88e3650664b65cdd6968d2c62be078075f1a7bfacbc3f16e83c041d901ec247d", "planned_active_tokens": 65595, "realized_active_tokens": 65595, "cumulative_start": 4260502, "cumulative_end": 4326097, "loss_sum": 13813.481086730957, "token_normalized_loss": 0.21058740889901603, "grad_norm_before_clip": 0.09743442386388779, "learning_rates_at_step": { "lora": 7.089782935335964e-05, "row_tokens": 7.089782935335964e-05 }, "learning_rates_after_step": { "lora": 6.996525694782018e-05, "row_tokens": 6.996525694782018e-05 } }, { "update_id": 66, "row_count": 65, "row_ids_sha256": "96f3b0832e4e789a574a75fb64bb4a01c93510db20275cf03a90fe9c4b2a9199", "input_ids_vector_sha256": "abf4359885fe03676b612e4ebaad505d0aa4759c95550f39f1b9d6611ca06201", "labels_vector_sha256": "c823035941cbcabc4534a1e2c62bb7c39d175efe9b828aa786bcb6262c4f4fd7", "planned_active_tokens": 65970, "realized_active_tokens": 65970, "cumulative_start": 4326097, "cumulative_end": 4392067, "loss_sum": 13287.478260040283, "token_normalized_loss": 0.20141698135577207, "grad_norm_before_clip": 0.08956732600927353, "learning_rates_at_step": { "lora": 6.996525694782018e-05, "row_tokens": 6.996525694782018e-05 }, "learning_rates_after_step": { "lora": 6.904545817010007e-05, "row_tokens": 6.904545817010007e-05 } }, { "update_id": 67, "row_count": 53, "row_ids_sha256": "0a686ee3ceef3fde74c1f8d3d7dc6cd99630386f55323c77ca123191a1593bad", "input_ids_vector_sha256": "403feb200fba9e54fa7c94e0259b01b411f371b57c3e58a542d7650db0fca1ad", "labels_vector_sha256": "41c4df5b4e3960949c095f6295d25b359b22f3eb4526ca425af4eaf6dd3eaf86", "planned_active_tokens": 64496, "realized_active_tokens": 64496, "cumulative_start": 4392067, "cumulative_end": 4456563, "loss_sum": 14988.91750717163, "token_normalized_loss": 0.23240073038904166, "grad_norm_before_clip": 0.09937790781259537, "learning_rates_at_step": { "lora": 6.904545817010007e-05, "row_tokens": 6.904545817010007e-05 }, "learning_rates_after_step": { "lora": 6.80941050313557e-05, "row_tokens": 6.80941050313557e-05 } }, { "update_id": 68, "row_count": 55, "row_ids_sha256": "4537462ed234e6592c8aa896aa9a448ced67d05804bd44725f1bd071c68eb86c", "input_ids_vector_sha256": "ce570f4e5a9a72bce98047f016fa58a62ef3f07b4b6c7239dbfae94cc03ba24b", "labels_vector_sha256": "a8df5636350c031839c73497a86da3ba7974750b8dbdf4a77f70454e60393a49", "planned_active_tokens": 66155, "realized_active_tokens": 66155, "cumulative_start": 4456563, "cumulative_end": 4522718, "loss_sum": 14475.607089996338, "token_normalized_loss": 0.2188134999621546, "grad_norm_before_clip": 0.09499987959861755, "learning_rates_at_step": { "lora": 6.80941050313557e-05, "row_tokens": 6.80941050313557e-05 }, "learning_rates_after_step": { "lora": 6.713660813051916e-05, "row_tokens": 6.713660813051916e-05 } }, { "update_id": 69, "row_count": 50, "row_ids_sha256": "2af5b7edb6bc48f50577d2a9c47ce6580b066067b61c32a8b6a2085e7df1d4a9", "input_ids_vector_sha256": "7479db5c28ad121602e9788c87515a5d50437523d4598ebf6464f0ebbe21faf5", "labels_vector_sha256": "d1af197bb65fe7bba93546d5ef51bef49f2ccbaad680f3584a3c9ee56472d0b1", "planned_active_tokens": 66055, "realized_active_tokens": 66055, "cumulative_start": 4522718, "cumulative_end": 4588773, "loss_sum": 15619.015155792236, "token_normalized_loss": 0.23645469920206247, "grad_norm_before_clip": 0.09883268177509308, "learning_rates_at_step": { "lora": 6.713660813051916e-05, "row_tokens": 6.713660813051916e-05 }, "learning_rates_after_step": { "lora": 6.61921605152556e-05, "row_tokens": 6.61921605152556e-05 } }, { "update_id": 70, "row_count": 55, "row_ids_sha256": "18c8d9d2ebd7fdc130ba799e33ec70af17b7e80495cd7269264154fca9b3970a", "input_ids_vector_sha256": "89f4f54f9d87fcf7bc52cd4fab60bc9cc2a6d34f9f9a70271ca88599d261f2c9", "labels_vector_sha256": "c9b9031a3c59ff433b4ea94cb2f3f384d566a958ba6b3c2245fdd57a39e04a81", "planned_active_tokens": 64675, "realized_active_tokens": 64675, "cumulative_start": 4588773, "cumulative_end": 4653448, "loss_sum": 15771.732048034668, "token_normalized_loss": 0.24386133819922176, "grad_norm_before_clip": 0.1044044941663742, "learning_rates_at_step": { "lora": 6.61921605152556e-05, "row_tokens": 6.61921605152556e-05 }, "learning_rates_after_step": { "lora": 6.521793356133153e-05, "row_tokens": 6.521793356133153e-05 } }, { "update_id": 71, "row_count": 51, "row_ids_sha256": "89bc565cea6f218c2c697029aac4cba2ddb8ead6064e7c60de5ebb8a705fb787", "input_ids_vector_sha256": "cb0f174c1f0195a57117a01796cc5196007fda64af5ed417544427de9e2a0c58", "labels_vector_sha256": "446c6eb3714ef0ff99b6b9d44665987dc7bb781aab29ec9090f5d141eb389f57", "planned_active_tokens": 66253, "realized_active_tokens": 66253, "cumulative_start": 4653448, "cumulative_end": 4719701, "loss_sum": 15187.903076171875, "token_normalized_loss": 0.22924098646358468, "grad_norm_before_clip": 0.09330666810274124, "learning_rates_at_step": { "lora": 6.521793356133153e-05, "row_tokens": 6.521793356133153e-05 }, "learning_rates_after_step": { "lora": 6.423805392663798e-05, "row_tokens": 6.423805392663798e-05 } }, { "update_id": 72, "row_count": 45, "row_ids_sha256": "ea631aa897a5fcd08c1af65cf09e2b66e7c4958ed1c3fbf543fc53e9b6b23850", "input_ids_vector_sha256": "141ae648e150f8762fda750dc2170355b726d0efe48e888e63f687ec8ea08676", "labels_vector_sha256": "44f117e6d48f53b890b5abf8d0536c570b1a2212268752980f706b78468fcf07", "planned_active_tokens": 66202, "realized_active_tokens": 66202, "cumulative_start": 4719701, "cumulative_end": 4785903, "loss_sum": 16359.362274169922, "token_normalized_loss": 0.2471128104010441, "grad_norm_before_clip": 0.09602040797472, "learning_rates_at_step": { "lora": 6.423805392663798e-05, "row_tokens": 6.423805392663798e-05 }, "learning_rates_after_step": { "lora": 6.32873994158059e-05, "row_tokens": 6.32873994158059e-05 } }, { "update_id": 73, "row_count": 53, "row_ids_sha256": "95d847760fd980848891cb367523d1a484cd29618981cb55caa8af9b54d5d2c2", "input_ids_vector_sha256": "c66f0996281cbfbe958c8ac17de445fabb3141385ac837f9ad729753ad84a7d4", "labels_vector_sha256": "54b69c06ec9ed9976edf5bb4e33fc0e9085a6ed6c3e4fc3e250092e7f5f25407", "planned_active_tokens": 63844, "realized_active_tokens": 63844, "cumulative_start": 4785903, "cumulative_end": 4849747, "loss_sum": 14921.027500152588, "token_normalized_loss": 0.2337107245810505, "grad_norm_before_clip": 0.09583679586648941, "learning_rates_at_step": { "lora": 6.32873994158059e-05, "row_tokens": 6.32873994158059e-05 }, "learning_rates_after_step": { "lora": 6.230350619643327e-05, "row_tokens": 6.230350619643327e-05 } }, { "update_id": 74, "row_count": 67, "row_ids_sha256": "cf21a0ba5606155baa286cf4b628cdb640fbaa9f89ea5a73374b72a2976623c7", "input_ids_vector_sha256": "2415abf5248f79d40002eef0bc2e7d8a4adf069312df7da6676ccaf875c2ba92", "labels_vector_sha256": "f128d625a531f438b44021239cdb2f608c50b30d613fb58cf895cde4ba999798", "planned_active_tokens": 65712, "realized_active_tokens": 65712, "cumulative_start": 4849747, "cumulative_end": 4915459, "loss_sum": 15804.160875320435, "token_normalized_loss": 0.2405064657189012, "grad_norm_before_clip": 0.10158081352710724, "learning_rates_at_step": { "lora": 6.230350619643327e-05, "row_tokens": 6.230350619643327e-05 }, "learning_rates_after_step": { "lora": 6.131142205154966e-05, "row_tokens": 6.131142205154966e-05 } }, { "update_id": 75, "row_count": 77, "row_ids_sha256": "041a319985bdebcc04b941d9035855821dee54ee2524fa054d1115a9a23558ca", "input_ids_vector_sha256": "78e04c040aed017d05a75f80ea4a2253c58d82008ae293536c4eb05cf5a1a842", "labels_vector_sha256": "9d8ef551c5f1ee7df0c0c87b50b2db18250d87e37bd01c160ca29f33caf199b2", "planned_active_tokens": 65917, "realized_active_tokens": 65917, "cumulative_start": 4915459, "cumulative_end": 4981376, "loss_sum": 15155.80924987793, "token_normalized_loss": 0.2299226185942614, "grad_norm_before_clip": 0.10229898989200592, "learning_rates_at_step": { "lora": 6.131142205154966e-05, "row_tokens": 6.131142205154966e-05 }, "learning_rates_after_step": { "lora": 6.031345465547857e-05, "row_tokens": 6.031345465547857e-05 } }, { "update_id": 76, "row_count": 80, "row_ids_sha256": "e5c45805a118764f7c51ca741279a4a09be9156c659077145af3c3cf670597ad", "input_ids_vector_sha256": "9ea8a41417731614a3ad3f38ef1d875377b84b9f0f118e37f3113488f0390c70", "labels_vector_sha256": "560e982b577e7f6b36a7341fe1aad5c0330ab064034a91f86748e7f3bd6c952e", "planned_active_tokens": 65994, "realized_active_tokens": 65994, "cumulative_start": 4981376, "cumulative_end": 5047370, "loss_sum": 15129.338356018066, "token_normalized_loss": 0.22925324053729226, "grad_norm_before_clip": 0.09460914134979248, "learning_rates_at_step": { "lora": 6.031345465547857e-05, "row_tokens": 6.031345465547857e-05 }, "learning_rates_after_step": { "lora": 5.933223783230346e-05, "row_tokens": 5.933223783230346e-05 } }, { "update_id": 77, "row_count": 71, "row_ids_sha256": "417ef7782a96b17308bc64c9d8c1a1609eaa832b7e71016b8a28db9e1a42d701", "input_ids_vector_sha256": "05a7e8ca9a1414ac6fd570287fd14bd3e35b3ba0e0a4259885423d2635591edd", "labels_vector_sha256": "d4891bd7fcb8c5f84b0b149d9744a5c79af29bf4c6cb60f62cfc2eb4406e26b1", "planned_active_tokens": 64610, "realized_active_tokens": 64610, "cumulative_start": 5047370, "cumulative_end": 5111980, "loss_sum": 18080.693632125854, "token_normalized_loss": 0.2798435788906648, "grad_norm_before_clip": 0.1036377027630806, "learning_rates_at_step": { "lora": 5.933223783230346e-05, "row_tokens": 5.933223783230346e-05 }, "learning_rates_after_step": { "lora": 5.8318952689688775e-05, "row_tokens": 5.8318952689688775e-05 } }, { "update_id": 78, "row_count": 74, "row_ids_sha256": "1d6a64e4653e5e65742c654a6da8745a23d0893975b8a869779009f25232dea2", "input_ids_vector_sha256": "1fe4e28535aa616e62574345bdefeae10b356e45dd124291bbb6835f93eddef0", "labels_vector_sha256": "923d835399e7c90146dcecee38f8288c7335dd8a2f3ac6cd983cbdae542395d5", "planned_active_tokens": 66465, "realized_active_tokens": 66465, "cumulative_start": 5111980, "cumulative_end": 5178445, "loss_sum": 16630.191764831543, "token_normalized_loss": 0.2502097609994966, "grad_norm_before_clip": 0.09924894571304321, "learning_rates_at_step": { "lora": 5.8318952689688775e-05, "row_tokens": 5.8318952689688775e-05 }, "learning_rates_after_step": { "lora": 5.7316161167526514e-05, "row_tokens": 5.7316161167526514e-05 } }, { "update_id": 79, "row_count": 72, "row_ids_sha256": "e49fd14ce274cc37371b32a640bea12c62f1d9bab984939b141bc751bdac1882", "input_ids_vector_sha256": "353ef54a97a63a52ff48c12712789eebcc70b95f484c536aa6705516fa79016b", "labels_vector_sha256": "0975e3716862a4c2febcadacdf85e57aa3b27fc0cc90b2c7503801dd8eb20fd8", "planned_active_tokens": 65550, "realized_active_tokens": 65550, "cumulative_start": 5178445, "cumulative_end": 5243995, "loss_sum": 14916.422138214111, "token_normalized_loss": 0.2275579273564319, "grad_norm_before_clip": 0.10845252871513367, "learning_rates_at_step": { "lora": 5.7316161167526514e-05, "row_tokens": 5.7316161167526514e-05 }, "learning_rates_after_step": { "lora": 5.632287537442827e-05, "row_tokens": 5.632287537442827e-05 } }, { "update_id": 80, "row_count": 65, "row_ids_sha256": "99ec0170142018baacd130978430e67188f917515ec7da106dceb85d5547d5cd", "input_ids_vector_sha256": "ecfcd7484c34489db8d9aeb0b9bc308b24ce7d902c0af5350fa11c862ace7583", "labels_vector_sha256": "1bc3316d497827225e41b53bde6c3f180e0eda163626004a649e275a6d450fe0", "planned_active_tokens": 64735, "realized_active_tokens": 64735, "cumulative_start": 5243995, "cumulative_end": 5308730, "loss_sum": 15572.33515548706, "token_normalized_loss": 0.2405551116936288, "grad_norm_before_clip": 0.09147574007511139, "learning_rates_at_step": { "lora": 5.632287537442827e-05, "row_tokens": 5.632287537442827e-05 }, "learning_rates_after_step": { "lora": 5.531347873464897e-05, "row_tokens": 5.531347873464897e-05 } }, { "update_id": 81, "row_count": 71, "row_ids_sha256": "a9d382a42434498ddb4a9a71d738d6c270e66f76c580b7a99287617231a81eb6", "input_ids_vector_sha256": "d83c61deeddad01eb9fe1a9467624361151e3b087e291ae308ebd4830bdb5bc4", "labels_vector_sha256": "d4aa6114d343158074a4f18846854d7f02d6cce8d11458feb719ac456403182e", "planned_active_tokens": 65616, "realized_active_tokens": 65616, "cumulative_start": 5308730, "cumulative_end": 5374346, "loss_sum": 15429.772020339966, "token_normalized_loss": 0.23515258504541522, "grad_norm_before_clip": 0.09582490473985672, "learning_rates_at_step": { "lora": 5.531347873464897e-05, "row_tokens": 5.531347873464897e-05 }, "learning_rates_after_step": { "lora": 5.4300868367991254e-05, "row_tokens": 5.4300868367991254e-05 } }, { "update_id": 82, "row_count": 65, "row_ids_sha256": "bd411b36a53b7ef237556ad46c7ce0c190a7202454e05539f8a87e7199578666", "input_ids_vector_sha256": "fdcc4093749b27d054a3d0900ac0bd488058e00ddee3dde5dc05825d5877089f", "labels_vector_sha256": "19db31d6c0c51f30f45e6877e5ac89105febf6059ef65d19251b66bc9d7369fe", "planned_active_tokens": 65682, "realized_active_tokens": 65682, "cumulative_start": 5374346, "cumulative_end": 5440028, "loss_sum": 19908.204021453857, "token_normalized_loss": 0.30309984503294446, "grad_norm_before_clip": 0.10219351947307587, "learning_rates_at_step": { "lora": 5.4300868367991254e-05, "row_tokens": 5.4300868367991254e-05 }, "learning_rates_after_step": { "lora": 5.32881776237621e-05, "row_tokens": 5.32881776237621e-05 } }, { "update_id": 83, "row_count": 70, "row_ids_sha256": "f2295528263a9d341a8457408e0ad56ba7574b71238cfc340d5f233ebc5f1a95", "input_ids_vector_sha256": "df7081f66d39053bd5f866ea84c1bfa8ac8721f512dbcc82ddbb5eabf4e54928", "labels_vector_sha256": "cb94778cd022a20b4aaa968509ee7a01a380f3cde3509c0b37efcc8071a5abd5", "planned_active_tokens": 65572, "realized_active_tokens": 65572, "cumulative_start": 5440028, "cumulative_end": 5505600, "loss_sum": 15679.376670837402, "token_normalized_loss": 0.23911695038793085, "grad_norm_before_clip": 0.10924366116523743, "learning_rates_at_step": { "lora": 5.32881776237621e-05, "row_tokens": 5.32881776237621e-05 }, "learning_rates_after_step": { "lora": 5.227406830725395e-05, "row_tokens": 5.227406830725395e-05 } }, { "update_id": 84, "row_count": 64, "row_ids_sha256": "c301509c35e527601872fd9220f0deaa5c4efc647fdd54124846fac70bf96079", "input_ids_vector_sha256": "12d061213c2f3d3d17ec5437e470a9b150147df9f4df0f180a1181e7bf376621", "labels_vector_sha256": "67b64e9bf531d6033e9d44335b1ddac3f8e8fc21a3445d68dee5f5142721819c", "planned_active_tokens": 65576, "realized_active_tokens": 65576, "cumulative_start": 5505600, "cumulative_end": 5571176, "loss_sum": 15544.592655181885, "token_normalized_loss": 0.23704697839425834, "grad_norm_before_clip": 0.11369714885950089, "learning_rates_at_step": { "lora": 5.227406830725395e-05, "row_tokens": 5.227406830725395e-05 }, "learning_rates_after_step": { "lora": 5.125075214933987e-05, "row_tokens": 5.125075214933987e-05 } }, { "update_id": 85, "row_count": 66, "row_ids_sha256": "61829ee69b041b867b1305a754493188c2b4bbdaba30d08c68cdf1f576e2187f", "input_ids_vector_sha256": "7860168e34338bc71586e1343c427e0d581556ebdd60c193e4f90b9144789a24", "labels_vector_sha256": "16af4628f24ccc7fe89dd483a1e18b51c2ac62a0a71c341f4f6e33aaacc3f781", "planned_active_tokens": 66110, "realized_active_tokens": 66110, "cumulative_start": 5571176, "cumulative_end": 5637286, "loss_sum": 17178.862998962402, "token_normalized_loss": 0.2598527151559885, "grad_norm_before_clip": 0.10277950763702393, "learning_rates_at_step": { "lora": 5.125075214933987e-05, "row_tokens": 5.125075214933987e-05 }, "learning_rates_after_step": { "lora": 5.0248580146382496e-05, "row_tokens": 5.0248580146382496e-05 } }, { "update_id": 86, "row_count": 69, "row_ids_sha256": "4054adf52366efb5fd88ae18bcefc5bf622932ba0da57ad385c9289d15c5223b", "input_ids_vector_sha256": "2bb90e148690fc8636a60ef578d49741b48025b45123f7ba0de58d595934b7fc", "labels_vector_sha256": "1147a8276226d710a63cfa9b5f3342dc197495bc01fa1cb6f231688f1d0440b8", "planned_active_tokens": 64711, "realized_active_tokens": 64711, "cumulative_start": 5637286, "cumulative_end": 5701997, "loss_sum": 17475.53636932373, "token_normalized_loss": 0.2700551122579427, "grad_norm_before_clip": 0.10534842312335968, "learning_rates_at_step": { "lora": 5.0248580146382496e-05, "row_tokens": 5.0248580146382496e-05 }, "learning_rates_after_step": { "lora": 4.9225540082638196e-05, "row_tokens": 4.9225540082638196e-05 } }, { "update_id": 87, "row_count": 72, "row_ids_sha256": "e5f03e6fa2dd278924fddea18184ef99930fa15aa669ab3764d4385e3fca061c", "input_ids_vector_sha256": "10ff8db62762c22dca84748024c8c1bd665acd435e54cce19640bfb69eecff1f", "labels_vector_sha256": "ae18249d38b27604497ade0d948d3330614de21e8c430156b6dc5bf27664ecee", "planned_active_tokens": 66052, "realized_active_tokens": 66052, "cumulative_start": 5701997, "cumulative_end": 5768049, "loss_sum": 15695.820182800293, "token_normalized_loss": 0.23762823506934375, "grad_norm_before_clip": 0.1052853986620903, "learning_rates_at_step": { "lora": 4.9225540082638196e-05, "row_tokens": 4.9225540082638196e-05 }, "learning_rates_after_step": { "lora": 4.821065658641943e-05, "row_tokens": 4.821065658641943e-05 } }, { "update_id": 88, "row_count": 65, "row_ids_sha256": "8ed411546e823e7207236e1aeefb156584a270caf6f52354fa5a71489fe11bd3", "input_ids_vector_sha256": "b76c86e4247a2b637b2b04ed892432bfb4bcc3b3b4d9b1c02a38688022235332", "labels_vector_sha256": "e0b9dd3936f55662dfa3409daf2728af882d46e2bc94f4644182a73ff0e9cb39", "planned_active_tokens": 65546, "realized_active_tokens": 65546, "cumulative_start": 5768049, "cumulative_end": 5833595, "loss_sum": 15161.390075683594, "token_normalized_loss": 0.23130915808262278, "grad_norm_before_clip": 0.10202765464782715, "learning_rates_at_step": { "lora": 4.821065658641943e-05, "row_tokens": 4.821065658641943e-05 }, "learning_rates_after_step": { "lora": 4.7205402921047255e-05, "row_tokens": 4.7205402921047255e-05 } }, { "update_id": 89, "row_count": 65, "row_ids_sha256": "e592489b201b1f68cf745bc3b28fa54617784d0ae326a92e8b9a7a5fb410427f", "input_ids_vector_sha256": "338e0f9e5f421244fbe656baf26495197ea49b6e04680bde3272887ac4cf5ded", "labels_vector_sha256": "e0d40c599ff1ad5f63e33d024bed52db00ed8e5b7a9983a3515fe1387c8cf335", "planned_active_tokens": 64971, "realized_active_tokens": 64971, "cumulative_start": 5833595, "cumulative_end": 5898566, "loss_sum": 15275.844398498535, "token_normalized_loss": 0.2351178894968299, "grad_norm_before_clip": 0.10079793632030487, "learning_rates_at_step": { "lora": 4.7205402921047255e-05, "row_tokens": 4.7205402921047255e-05 }, "learning_rates_after_step": { "lora": 4.6189281277078656e-05, "row_tokens": 4.6189281277078656e-05 } }, { "update_id": 90, "row_count": 66, "row_ids_sha256": "d8c36c6e1ad51cf7686f48dddfff290b14cc1c123657486e46b4e77bc1259b72", "input_ids_vector_sha256": "6f122293e240a85417b367b81a4b858e07199c1b4023f3f722aeab28d2d8a68e", "labels_vector_sha256": "95894a122da73d6b166678064e944fec2bc64bb4492b2c5fee76186296f110e8", "planned_active_tokens": 65748, "realized_active_tokens": 65748, "cumulative_start": 5898566, "cumulative_end": 5964314, "loss_sum": 13765.040588378906, "token_normalized_loss": 0.209360597864253, "grad_norm_before_clip": 0.09322769194841385, "learning_rates_at_step": { "lora": 4.6189281277078656e-05, "row_tokens": 4.6189281277078656e-05 }, "learning_rates_after_step": { "lora": 4.518032119954603e-05, "row_tokens": 4.518032119954603e-05 } }, { "update_id": 91, "row_count": 68, "row_ids_sha256": "10b97a507383715daffe32ebf96e2ca0fd7d90544eef4372586bf74a8fa3e4b0", "input_ids_vector_sha256": "e0217988ecee74b7743cd5eb778bfba705f6d854cc632f0985c92a8d09e693e4", "labels_vector_sha256": "90a858f3cf42edb89e2be069944585a0c88f819056347cc21ff85f4d778cf404", "planned_active_tokens": 65386, "realized_active_tokens": 65386, "cumulative_start": 5964314, "cumulative_end": 6029700, "loss_sum": 16257.521995544434, "token_normalized_loss": 0.24863918874903548, "grad_norm_before_clip": 0.10768872499465942, "learning_rates_at_step": { "lora": 4.518032119954603e-05, "row_tokens": 4.518032119954603e-05 }, "learning_rates_after_step": { "lora": 4.416675439867134e-05, "row_tokens": 4.416675439867134e-05 } }, { "update_id": 92, "row_count": 67, "row_ids_sha256": "01d54ae5cfd27510462cf525eb504d48c1d7a900dc3fce058315edd24b5a5923", "input_ids_vector_sha256": "3f48ab072ffbaf181dee333ccdc9a27ddbe4807529952c55601a873898851c87", "labels_vector_sha256": "9cbaf803fc7a49a613416b4eddbedef34f7e2486828b25e3a3b0fef4b9fee0a3", "planned_active_tokens": 65814, "realized_active_tokens": 65814, "cumulative_start": 6029700, "cumulative_end": 6095514, "loss_sum": 14596.075248718262, "token_normalized_loss": 0.2217776650669806, "grad_norm_before_clip": 0.10696783661842346, "learning_rates_at_step": { "lora": 4.416675439867134e-05, "row_tokens": 4.416675439867134e-05 }, "learning_rates_after_step": { "lora": 4.316458797110751e-05, "row_tokens": 4.316458797110751e-05 } }, { "update_id": 93, "row_count": 68, "row_ids_sha256": "4a3100ca874fe8d01ffeb77fe6162c7d2d9bd5a56e49c52b9ccb4604515b83ad", "input_ids_vector_sha256": "848daa644ac00b42a7e6073e99e9e42ae5c9089fa2773fe8c824091bd96e762a", "labels_vector_sha256": "cb61f49b8030253c000800f5a6ae944eeeb7ab78cc82aed7223f1a39e6ca12bb", "planned_active_tokens": 65229, "realized_active_tokens": 65229, "cumulative_start": 6095514, "cumulative_end": 6160743, "loss_sum": 12031.032150268555, "token_normalized_loss": 0.18444299545092757, "grad_norm_before_clip": 0.10376222431659698, "learning_rates_at_step": { "lora": 4.316458797110751e-05, "row_tokens": 4.316458797110751e-05 }, "learning_rates_after_step": { "lora": 4.215029760717739e-05, "row_tokens": 4.215029760717739e-05 } }, { "update_id": 94, "row_count": 68, "row_ids_sha256": "e692f3452fdeeb69b9338f091b9f6793613fca4ba981a8f8da51158857c632ae", "input_ids_vector_sha256": "4d65214754c5b2375c5389d2d26b341a50c3d46f6581de039e0c1120d8af1964", "labels_vector_sha256": "a3f6897cd3d20392edda12d7defe0871e83b2bbdb75f742c0fbfe86a75767be8", "planned_active_tokens": 66204, "realized_active_tokens": 66204, "cumulative_start": 6160743, "cumulative_end": 6226947, "loss_sum": 12376.00821685791, "token_normalized_loss": 0.18693746928973945, "grad_norm_before_clip": 0.09526067972183228, "learning_rates_at_step": { "lora": 4.215029760717739e-05, "row_tokens": 4.215029760717739e-05 }, "learning_rates_after_step": { "lora": 4.1156824323634346e-05, "row_tokens": 4.1156824323634346e-05 } }, { "update_id": 95, "row_count": 71, "row_ids_sha256": "61f7647608241b0fbd8f082327d94d8931b93e27f82378861e10142b29fb5401", "input_ids_vector_sha256": "c625a4993ecdd89151f3ece2cfa42a67cdf97880a59b7739439b28437f80fdb4", "labels_vector_sha256": "30e26b974837d9b03b54a2d916d4c7af875efce4ec21d9854326403a1e10f66c", "planned_active_tokens": 65055, "realized_active_tokens": 65055, "cumulative_start": 6226947, "cumulative_end": 6292002, "loss_sum": 13095.389125823975, "token_normalized_loss": 0.2012971966155403, "grad_norm_before_clip": 0.0949740782380104, "learning_rates_at_step": { "lora": 4.1156824323634346e-05, "row_tokens": 4.1156824323634346e-05 }, "learning_rates_after_step": { "lora": 4.0166380474230566e-05, "row_tokens": 4.0166380474230566e-05 } }, { "update_id": 96, "row_count": 72, "row_ids_sha256": "cd234affc2a172e89b147c1fe7b7b3ba2cfa70ad3cc1bcbea5b22d1b694fd964", "input_ids_vector_sha256": "e22ba7b734c43a6855664463a0f1cc4f817e0ccdeba99c8c2f5db3bfb7561ad6", "labels_vector_sha256": "cc45edf25868e7e140434925585d833ef3bc794690a97fee7dc6442955960d44", "planned_active_tokens": 65092, "realized_active_tokens": 65092, "cumulative_start": 6292002, "cumulative_end": 6357094, "loss_sum": 16478.495433807373, "token_normalized_loss": 0.253156999843412, "grad_norm_before_clip": 0.1188759133219719, "learning_rates_at_step": { "lora": 4.0166380474230566e-05, "row_tokens": 4.0166380474230566e-05 }, "learning_rates_after_step": { "lora": 3.915843248107764e-05, "row_tokens": 3.915843248107764e-05 } }, { "update_id": 97, "row_count": 70, "row_ids_sha256": "102a6b4605fc4271ac9a73460ab259c22996b89803e1f1dd8d082fc73ffd84fb", "input_ids_vector_sha256": "4955006dd4c00ac43119b0a9f39d4dcbf32aa5c012eba1634b0a71366bd44aa8", "labels_vector_sha256": "e52679358f3bddc7d3fcc4268dca7823eba0e59ce7dea1a65824c448dc1a4db4", "planned_active_tokens": 66514, "realized_active_tokens": 66514, "cumulative_start": 6357094, "cumulative_end": 6423608, "loss_sum": 13836.08065032959, "token_normalized_loss": 0.20801756999022145, "grad_norm_before_clip": 0.09955458343029022, "learning_rates_at_step": { "lora": 3.915843248107764e-05, "row_tokens": 3.915843248107764e-05 }, "learning_rates_after_step": { "lora": 3.817898472769514e-05, "row_tokens": 3.817898472769514e-05 } }, { "update_id": 98, "row_count": 70, "row_ids_sha256": "6c995d6771130d8b1527ca1d90412236f0412b2ccd4e84898b06e47fa5aaa0c4", "input_ids_vector_sha256": "1bd08d7755998dedb2affe120a212c8e16eac69a62cc588bd4928d5476c1a9c6", "labels_vector_sha256": "682e4b9769f9e1a109532355d32208ac1eba33f3073a95417c73d1c2bf53b2d7", "planned_active_tokens": 64926, "realized_active_tokens": 64926, "cumulative_start": 6423608, "cumulative_end": 6488534, "loss_sum": 10853.524011611938, "token_normalized_loss": 0.16716760637667405, "grad_norm_before_clip": 0.0923881083726883, "learning_rates_at_step": { "lora": 3.817898472769514e-05, "row_tokens": 3.817898472769514e-05 }, "learning_rates_after_step": { "lora": 3.71998566957128e-05, "row_tokens": 3.71998566957128e-05 } }, { "update_id": 99, "row_count": 68, "row_ids_sha256": "c185d67bacda6a910800f44e2c2700d2bf4c09a9be5c2f39375315cf5b58ef65", "input_ids_vector_sha256": "f34b09c35ee067a0dbe05dbdd774e32b47379ff163c5f54062edba50a5ddfe4c", "labels_vector_sha256": "76ddccc96332735edafb3a5b554ef5fe925d0eee86cdc10af8c72658f6e3cbed", "planned_active_tokens": 65224, "realized_active_tokens": 65224, "cumulative_start": 6488534, "cumulative_end": 6553758, "loss_sum": 13689.52279663086, "token_normalized_loss": 0.2098847478938866, "grad_norm_before_clip": 0.09752392023801804, "learning_rates_at_step": { "lora": 3.71998566957128e-05, "row_tokens": 3.71998566957128e-05 }, "learning_rates_after_step": { "lora": 3.6218985840166206e-05, "row_tokens": 3.6218985840166206e-05 } }, { "update_id": 100, "row_count": 73, "row_ids_sha256": "96e550ffb8d443754a3acb6fa314fb3c22df75a9737f0b5a7e533c40b74d0a87", "input_ids_vector_sha256": "79472625f7857589f742c04c400bf721712146888cc728734b050f09ef547915", "labels_vector_sha256": "be2cef3d3789d5e36e5aa33573f9afd77fb2b9ac375001f245d5ef0fbf00c027", "planned_active_tokens": 65692, "realized_active_tokens": 65692, "cumulative_start": 6553758, "cumulative_end": 6619450, "loss_sum": 15819.651027679443, "token_normalized_loss": 0.24081548784752244, "grad_norm_before_clip": 0.10943400114774704, "learning_rates_at_step": { "lora": 3.6218985840166206e-05, "row_tokens": 3.6218985840166206e-05 }, "learning_rates_after_step": { "lora": 3.52409951884604e-05, "row_tokens": 3.52409951884604e-05 } }, { "update_id": 101, "row_count": 71, "row_ids_sha256": "05898780d575f0ac06dfe4d23463136a1a2409945931a9aa57b443746ced1f4c", "input_ids_vector_sha256": "4d5a02eece23538148209bccacf9db404b47b9e00d90f82b33f4c4a6b8f772a3", "labels_vector_sha256": "186cd92bdb9767a227a7132f65a970a62f51bc0ef5f91f52704b700db2f4c1ed", "planned_active_tokens": 65883, "realized_active_tokens": 65883, "cumulative_start": 6619450, "cumulative_end": 6685333, "loss_sum": 12619.36516571045, "token_normalized_loss": 0.19154205433435711, "grad_norm_before_clip": 0.09776115417480469, "learning_rates_at_step": { "lora": 3.52409951884604e-05, "row_tokens": 3.52409951884604e-05 }, "learning_rates_after_step": { "lora": 3.427348916175999e-05, "row_tokens": 3.427348916175999e-05 } }, { "update_id": 102, "row_count": 68, "row_ids_sha256": "0a4981dc6da64a32e48417baead42f3a8aafe5be38e992392159bdbec491fe5e", "input_ids_vector_sha256": "188b8c5287838655610879777930953f011bbc3d823f0097c31d5ae3fe6fcb9a", "labels_vector_sha256": "f3de558047584fa65311389bd14a7becf9fda3261d703e8be5896530f2b71fbc", "planned_active_tokens": 65588, "realized_active_tokens": 65588, "cumulative_start": 6685333, "cumulative_end": 6750921, "loss_sum": 11972.09810256958, "token_normalized_loss": 0.18253488599392542, "grad_norm_before_clip": 0.09797606617212296, "learning_rates_at_step": { "lora": 3.427348916175999e-05, "row_tokens": 3.427348916175999e-05 }, "learning_rates_after_step": { "lora": 3.3322214033691666e-05, "row_tokens": 3.3322214033691666e-05 } }, { "update_id": 103, "row_count": 73, "row_ids_sha256": "3bdaf47419aef44a8789007e545545ea6816ec4957c8141c7f0f3db690fa878f", "input_ids_vector_sha256": "388239245b4d766f2c28578114760a6263b644bce713268dec8272b6d5db322c", "labels_vector_sha256": "bcadf7e41508db86e7fad79552975fcce3c384d28b93f26f4896e510c5df81d0", "planned_active_tokens": 64921, "realized_active_tokens": 64921, "cumulative_start": 6750921, "cumulative_end": 6815842, "loss_sum": 15045.337661743164, "token_normalized_loss": 0.23174839669356856, "grad_norm_before_clip": 0.10018561035394669, "learning_rates_at_step": { "lora": 3.3322214033691666e-05, "row_tokens": 3.3322214033691666e-05 }, "learning_rates_after_step": { "lora": 3.235255231997938e-05, "row_tokens": 3.235255231997938e-05 } }, { "update_id": 104, "row_count": 73, "row_ids_sha256": "dca7fd2e2b7fb27398e28a6af7f7a6b93b0b72400df16da208508a87d1a794a6", "input_ids_vector_sha256": "8e4bafd933b0b5be4255f6aba18bfb5c4da0239a02c3d7cb6642377e47c87c6c", "labels_vector_sha256": "8ee14425ffbf4c305395025851088178c016776462ab85728c4ca3e663ab9646", "planned_active_tokens": 66655, "realized_active_tokens": 66655, "cumulative_start": 6815842, "cumulative_end": 6882497, "loss_sum": 14982.038917541504, "token_normalized_loss": 0.22476991849885986, "grad_norm_before_clip": 0.09786437451839447, "learning_rates_at_step": { "lora": 3.235255231997938e-05, "row_tokens": 3.235255231997938e-05 }, "learning_rates_after_step": { "lora": 3.141420921772224e-05, "row_tokens": 3.141420921772224e-05 } }, { "update_id": 105, "row_count": 72, "row_ids_sha256": "c3b9e6393aafc6afc5c41eb383b33865c0ae5be2a1e68b4fe2b4a1defc829cfd", "input_ids_vector_sha256": "4e0f938f18fb078a805f216a3c260c0b35b892300aa4a91f41fb749f4f68be86", "labels_vector_sha256": "0c5e5f318786febc56dffd655d66eea85425eccbf05399c21bb27f33aebc756c", "planned_active_tokens": 65000, "realized_active_tokens": 65000, "cumulative_start": 6882497, "cumulative_end": 6947497, "loss_sum": 14175.416707992554, "token_normalized_loss": 0.21808333396911622, "grad_norm_before_clip": 0.09297284483909607, "learning_rates_at_step": { "lora": 3.141420921772224e-05, "row_tokens": 3.141420921772224e-05 }, "learning_rates_after_step": { "lora": 3.0472977574431173e-05, "row_tokens": 3.0472977574431173e-05 } }, { "update_id": 106, "row_count": 70, "row_ids_sha256": "195660f78175379feb06e22546a4fb919cc33d22550e28cc19d5d97b3f22e5c9", "input_ids_vector_sha256": "0cabdac17256320692334cae8d2f8e850a9b6121c9afc7cb9a25ccd6d7d74367", "labels_vector_sha256": "62fb608ca1c71a72493048a303399e4f6fa545025824ea361cd770ea0e63ea81", "planned_active_tokens": 65731, "realized_active_tokens": 65731, "cumulative_start": 6947497, "cumulative_end": 7013228, "loss_sum": 21088.593719482422, "token_normalized_loss": 0.32083177982203864, "grad_norm_before_clip": 0.10964761674404144, "learning_rates_at_step": { "lora": 3.0472977574431173e-05, "row_tokens": 3.0472977574431173e-05 }, "learning_rates_after_step": { "lora": 2.9546724646057855e-05, "row_tokens": 2.9546724646057855e-05 } }, { "update_id": 107, "row_count": 67, "row_ids_sha256": "003a7b1d9c7fe0312bafdafbfb94b779230089ba9cc37b025be9377a4beff343", "input_ids_vector_sha256": "04a7f46b8c3f67b8e018ef0e87a472e3b5c4578e82f23fa9260fd8bca778f305", "labels_vector_sha256": "5398a5fec01fb448474078ae9f1064a7a9b13ada883e7f59b814d776ffe43c34", "planned_active_tokens": 65244, "realized_active_tokens": 65244, "cumulative_start": 7013228, "cumulative_end": 7078472, "loss_sum": 16678.89865875244, "token_normalized_loss": 0.2556388121321875, "grad_norm_before_clip": 0.1031094640493393, "learning_rates_at_step": { "lora": 2.9546724646057855e-05, "row_tokens": 2.9546724646057855e-05 }, "learning_rates_after_step": { "lora": 2.862795822925328e-05, "row_tokens": 2.862795822925328e-05 } }, { "update_id": 108, "row_count": 77, "row_ids_sha256": "778749060760bf142d39617cdf2d705ebcdefeab50cb618feab030d5ae60e931", "input_ids_vector_sha256": "c143160fa3fad6dc7a026b68e1ca8878c8909b7ae6546aced86f12271eefeb55", "labels_vector_sha256": "6bdac32082d0a39d32caa27e993e60c6b3da2f57dc776088f8bb568ed6114c0b", "planned_active_tokens": 65306, "realized_active_tokens": 65306, "cumulative_start": 7078472, "cumulative_end": 7143778, "loss_sum": 13721.943382263184, "token_normalized_loss": 0.2101176520115025, "grad_norm_before_clip": 0.11155574768781662, "learning_rates_at_step": { "lora": 2.862795822925328e-05, "row_tokens": 2.862795822925328e-05 }, "learning_rates_after_step": { "lora": 2.7714319474567724e-05, "row_tokens": 2.7714319474567724e-05 } }, { "update_id": 109, "row_count": 75, "row_ids_sha256": "2b0a9ccfc736c65baa34c14e4d29a7d85137444e35fa544f0ee59fa6baf68b6a", "input_ids_vector_sha256": "d1a6546efc9ca2839751f923bf48fef842c73a02166b301d86246f92f46cf7bc", "labels_vector_sha256": "d541256fa34abc35d313df16bdf1952c8359a6003485aafa7a4deeedc877fdab", "planned_active_tokens": 65567, "realized_active_tokens": 65567, "cumulative_start": 7143778, "cumulative_end": 7209345, "loss_sum": 12353.876174926758, "token_normalized_loss": 0.18841606562640897, "grad_norm_before_clip": 0.10020986944437027, "learning_rates_at_step": { "lora": 2.7714319474567724e-05, "row_tokens": 2.7714319474567724e-05 }, "learning_rates_after_step": { "lora": 2.6811260199483424e-05, "row_tokens": 2.6811260199483424e-05 } }, { "update_id": 110, "row_count": 76, "row_ids_sha256": "912e31fa3aa900f25f3153159a62e964cf826378ffba0c18bdb7763a7e9bdb72", "input_ids_vector_sha256": "81296a9b0855aba2abdf7fe7389e50a73d488017ea08f7b0745eae721d0113aa", "labels_vector_sha256": "750fd4fd329257e703f98908e0d93e129ff3697980d4afbe6e5152338384e51d", "planned_active_tokens": 65466, "realized_active_tokens": 65466, "cumulative_start": 7209345, "cumulative_end": 7274811, "loss_sum": 12530.899593353271, "token_normalized_loss": 0.19141080245246803, "grad_norm_before_clip": 0.09853598475456238, "learning_rates_at_step": { "lora": 2.6811260199483424e-05, "row_tokens": 2.6811260199483424e-05 }, "learning_rates_after_step": { "lora": 2.5919570044908216e-05, "row_tokens": 2.5919570044908216e-05 } }, { "update_id": 111, "row_count": 74, "row_ids_sha256": "13f8be8983a51797143858054466ec555c02172077f78bc344ec0a9b0905fa87", "input_ids_vector_sha256": "d7aa2b1e621811953ef49e478909bbe5dd8a51273e444045323795dda3b98f29", "labels_vector_sha256": "1cb9e2d9ac94fb88716a8abb5d442eac3370aa86f2d0be3065f7bd874cc5c8ab", "planned_active_tokens": 65331, "realized_active_tokens": 65331, "cumulative_start": 7274811, "cumulative_end": 7340142, "loss_sum": 10504.901021957397, "token_normalized_loss": 0.16079504403663494, "grad_norm_before_clip": 0.09566672146320343, "learning_rates_at_step": { "lora": 2.5919570044908216e-05, "row_tokens": 2.5919570044908216e-05 }, "learning_rates_after_step": { "lora": 2.502969054711351e-05, "row_tokens": 2.502969054711351e-05 } }, { "update_id": 112, "row_count": 75, "row_ids_sha256": "81c22e8a2d97121bc75c404e3ab3ffcd7bae89e9a6489a83ce5916a0f7e5cd0a", "input_ids_vector_sha256": "e06072a2efef5a00eb3757b1224d2167e9d2deac612e4416c7dd34063b0b6053", "labels_vector_sha256": "2754869154e6a6132b12d1eb35806a9a7c4fbee8f2b5c442c2d1523feec5dca6", "planned_active_tokens": 65931, "realized_active_tokens": 65931, "cumulative_start": 7340142, "cumulative_end": 7406073, "loss_sum": 11187.109107971191, "token_normalized_loss": 0.16967904487981664, "grad_norm_before_clip": 0.09595100581645966, "learning_rates_at_step": { "lora": 2.502969054711351e-05, "row_tokens": 2.502969054711351e-05 }, "learning_rates_after_step": { "lora": 2.4156233020308626e-05, "row_tokens": 2.4156233020308626e-05 } }, { "update_id": 113, "row_count": 76, "row_ids_sha256": "5d222d8d916fe3fb9758848625e6519f6b12274d98be026798d4dbbcf2b89d38", "input_ids_vector_sha256": "65a9bf85013581ceb88a0a78826aaa265d26807a9b7f4a6d89577f081c80d433", "labels_vector_sha256": "b7fe61b11d3515854e77422e051259101aeb4de0d74c36d6bef238d79836a580", "planned_active_tokens": 65478, "realized_active_tokens": 65478, "cumulative_start": 7406073, "cumulative_end": 7471551, "loss_sum": 11062.30493927002, "token_normalized_loss": 0.16894689726732673, "grad_norm_before_clip": 0.09779825806617737, "learning_rates_at_step": { "lora": 2.4156233020308626e-05, "row_tokens": 2.4156233020308626e-05 }, "learning_rates_after_step": { "lora": 2.3296066116880822e-05, "row_tokens": 2.3296066116880822e-05 } }, { "update_id": 114, "row_count": 80, "row_ids_sha256": "e9787f03b7e8aebf1bae2904e0a5d87c9cde1358c4cee9620525d02854539a56", "input_ids_vector_sha256": "58ffb27a9470051e8f7b451f73784bc752379d99bd3a5742d1c43450811ee665", "labels_vector_sha256": "3af3459392ce1028e066857708f2a248dbc80f7ebe620dff8d400bdd2242b7f8", "planned_active_tokens": 65275, "realized_active_tokens": 65275, "cumulative_start": 7471551, "cumulative_end": 7536826, "loss_sum": 10250.395915985107, "token_normalized_loss": 0.15703402399057997, "grad_norm_before_clip": 0.09313441067934036, "learning_rates_at_step": { "lora": 2.3296066116880822e-05, "row_tokens": 2.3296066116880822e-05 }, "learning_rates_after_step": { "lora": 2.2440872327628943e-05, "row_tokens": 2.2440872327628943e-05 } }, { "update_id": 115, "row_count": 77, "row_ids_sha256": "bb9b47e0c0ff11df0c35aafccecff8dfd6ef1cfd1bc1819686e57f2e0eac2dda", "input_ids_vector_sha256": "af5c9d58beaf99776cb4d08ed40b653f2945d84f8ce3c1102bf5806f0c6e78a8", "labels_vector_sha256": "5d9a7cd79cc1e7f0c4a0eeabd32c52cfe66943a4875eae5d23360323f5b20e2a", "planned_active_tokens": 65735, "realized_active_tokens": 65735, "cumulative_start": 7536826, "cumulative_end": 7602561, "loss_sum": 10582.05969619751, "token_normalized_loss": 0.1609805993184378, "grad_norm_before_clip": 0.09096353501081467, "learning_rates_at_step": { "lora": 2.2440872327628943e-05, "row_tokens": 2.2440872327628943e-05 }, "learning_rates_after_step": { "lora": 2.159971908462149e-05, "row_tokens": 2.159971908462149e-05 } }, { "update_id": 116, "row_count": 73, "row_ids_sha256": "21086bfb09d85308a9d8c95c26a3af6ad296d94cfe006fa4a1b626891a55d84b", "input_ids_vector_sha256": "9ff27933d14a83ddc8d49f6ddce21eb9fed4734541805feea9c49b67679f63cb", "labels_vector_sha256": "ad33d0f58b6946c6877de011960fd30cf3f0ce176db434046153a0892a0c9eff", "planned_active_tokens": 65530, "realized_active_tokens": 65530, "cumulative_start": 7602561, "cumulative_end": 7668091, "loss_sum": 9879.724672317505, "token_normalized_loss": 0.1507664378501069, "grad_norm_before_clip": 0.09162057936191559, "learning_rates_at_step": { "lora": 2.159971908462149e-05, "row_tokens": 2.159971908462149e-05 }, "learning_rates_after_step": { "lora": 2.07615481117157e-05, "row_tokens": 2.07615481117157e-05 } }, { "update_id": 117, "row_count": 75, "row_ids_sha256": "68d7998564104c8f0b39879c41dca103a2144a87982bea42710858cbc07d3822", "input_ids_vector_sha256": "355fb6fac4a4194219a5eeb6b72774c3173beb6a9290adfdd5fee28e9d0f4184", "labels_vector_sha256": "67c6465eb8efa57a3d98b7fa34ee19482961d4aae9438ffed697612582c30baf", "planned_active_tokens": 66224, "realized_active_tokens": 66224, "cumulative_start": 7668091, "cumulative_end": 7734315, "loss_sum": 9537.161624908447, "token_normalized_loss": 0.14401367517680067, "grad_norm_before_clip": 0.0925937369465828, "learning_rates_at_step": { "lora": 2.07615481117157e-05, "row_tokens": 2.07615481117157e-05 }, "learning_rates_after_step": { "lora": 1.9950101985969827e-05, "row_tokens": 1.9950101985969827e-05 } }, { "update_id": 118, "row_count": 74, "row_ids_sha256": "fe8c9913bbff38e6d3e91e69025818243718ee6b078fa7b10ccccc5d67ab144e", "input_ids_vector_sha256": "923d52b8017cca065665dfc6ccc77ab21b60ea5aabfe9cfdc5135c4ed6a77438", "labels_vector_sha256": "fd08a9228ba916b121e688984eefff243583f59aeb5787b3849b5862d721f8d1", "planned_active_tokens": 65059, "realized_active_tokens": 65059, "cumulative_start": 7734315, "cumulative_end": 7799374, "loss_sum": 8868.530124664307, "token_normalized_loss": 0.1363151927429611, "grad_norm_before_clip": 0.08811455965042114, "learning_rates_at_step": { "lora": 1.9950101985969827e-05, "row_tokens": 1.9950101985969827e-05 }, "learning_rates_after_step": { "lora": 1.9148411014359487e-05, "row_tokens": 1.9148411014359487e-05 } }, { "update_id": 119, "row_count": 78, "row_ids_sha256": "6fa460ccc3514e5ef36b1cfb3c9ac7b583410362b4a028408a5e41aab4b93a6c", "input_ids_vector_sha256": "2a1c018f24db9fda46160d12fae5d064331b16a6a79f18dac9acf6eecdb9cc31", "labels_vector_sha256": "8a2bdb21632ff2eff7e9f597a7e787f40017fc22fd03907b1fbce2db12a33e45", "planned_active_tokens": 65260, "realized_active_tokens": 65260, "cumulative_start": 7799374, "cumulative_end": 7864634, "loss_sum": 9006.779441833496, "token_normalized_loss": 0.13801378243692147, "grad_norm_before_clip": 0.08877688646316528, "learning_rates_at_step": { "lora": 1.9148411014359487e-05, "row_tokens": 1.9148411014359487e-05 }, "learning_rates_after_step": { "lora": 1.8354783195532815e-05, "row_tokens": 1.8354783195532815e-05 } }, { "update_id": 120, "row_count": 75, "row_ids_sha256": "311c07f9bfbca38686c1c7539ea55300c60c982fedd6c733d50a66433bbd6401", "input_ids_vector_sha256": "3373547c0475b472b5442020bce6acb73da74402d24e478de314bd23195c02ba", "labels_vector_sha256": "075eb3e7170b3d25ca03a6db3f0973c8469ac7a371b148aec09c03504f6a2906", "planned_active_tokens": 65639, "realized_active_tokens": 65639, "cumulative_start": 7864634, "cumulative_end": 7930273, "loss_sum": 9441.074825286865, "token_normalized_loss": 0.14383331289761978, "grad_norm_before_clip": 0.09360776096582413, "learning_rates_at_step": { "lora": 1.8354783195532815e-05, "row_tokens": 1.8354783195532815e-05 }, "learning_rates_after_step": { "lora": 1.7572552721089093e-05, "row_tokens": 1.7572552721089093e-05 } }, { "update_id": 121, "row_count": 73, "row_ids_sha256": "bed6e4109ef934c230f9493e2d4d4f23052aba9a6e09c074adbff28453a6743b", "input_ids_vector_sha256": "6a06cb085cba5c7e00cb7e68df25512eb054493d2703decf5333c06040e0027a", "labels_vector_sha256": "b3842da4e569db8af6fc1e3b5e3d8793280c0751ad8fb4d7f932a5f2c528ce3a", "planned_active_tokens": 65782, "realized_active_tokens": 65782, "cumulative_start": 7930273, "cumulative_end": 7996055, "loss_sum": 8378.022651672363, "token_normalized_loss": 0.12736041244827404, "grad_norm_before_clip": 0.08622893691062927, "learning_rates_at_step": { "lora": 1.7572552721089093e-05, "row_tokens": 1.7572552721089093e-05 }, "learning_rates_after_step": { "lora": 1.680998457372267e-05, "row_tokens": 1.680998457372267e-05 } }, { "update_id": 122, "row_count": 73, "row_ids_sha256": "119663ffeb04bcd3d63ff2b8d54f97c3382d9ba87a4f67b493d6aa556a58fce5", "input_ids_vector_sha256": "ef7ca009839b1b0c057264c7fdb0c246ac039106e878e19db4e780ffd519e0ed", "labels_vector_sha256": "51869be271a9a1f07b1a21f9269da61899099038b1335bc418e8193cc52617cc", "planned_active_tokens": 65247, "realized_active_tokens": 65247, "cumulative_start": 7996055, "cumulative_end": 8061302, "loss_sum": 12875.459716796875, "token_normalized_loss": 0.19733412596436425, "grad_norm_before_clip": 0.10968410968780518, "learning_rates_at_step": { "lora": 1.680998457372267e-05, "row_tokens": 1.680998457372267e-05 }, "learning_rates_after_step": { "lora": 1.6053878185403015e-05, "row_tokens": 1.6053878185403015e-05 } }, { "update_id": 123, "row_count": 79, "row_ids_sha256": "bed0855181273f5524be8254c98bd90e78bd60db0985f86821465e1602e36e0e", "input_ids_vector_sha256": "03c0294c5cf5280074c5fcb659b6cc7be6072208501178c9d94ce827a5182e5c", "labels_vector_sha256": "ded3291fadbe8b7706bf5645b710afda4ef91ce40cbe756e38dbe8b2271af262", "planned_active_tokens": 65871, "realized_active_tokens": 65871, "cumulative_start": 8061302, "cumulative_end": 8127173, "loss_sum": 12019.394947052002, "token_normalized_loss": 0.1824686879970245, "grad_norm_before_clip": 0.09719119220972061, "learning_rates_at_step": { "lora": 1.6053878185403015e-05, "row_tokens": 1.6053878185403015e-05 }, "learning_rates_after_step": { "lora": 1.531408108579902e-05, "row_tokens": 1.531408108579902e-05 } }, { "update_id": 124, "row_count": 77, "row_ids_sha256": "ca7588d9c3c12e76a90ff33fdc4e7422d0d978715092f7eac9b66112a3b29ca8", "input_ids_vector_sha256": "875615dd748a5a62aeb4c7526dd447831991cb848ac2499d5533cf2963fbc35b", "labels_vector_sha256": "0872d512658e9ce1599b8d39961b9bdce4e54c310bacdff45c4db2fd4784f1db", "planned_active_tokens": 65676, "realized_active_tokens": 65676, "cumulative_start": 8127173, "cumulative_end": 8192849, "loss_sum": 12120.844339370728, "token_normalized_loss": 0.18455515468924308, "grad_norm_before_clip": 0.10367433726787567, "learning_rates_at_step": { "lora": 1.531408108579902e-05, "row_tokens": 1.531408108579902e-05 }, "learning_rates_after_step": { "lora": 1.4596295877080029e-05, "row_tokens": 1.4596295877080029e-05 } }, { "update_id": 125, "row_count": 73, "row_ids_sha256": "3d19330d7e655a42fc25f7639c32b7a33808de95766e53737eee160ecddb2735", "input_ids_vector_sha256": "eca2d00fa9f958abec8f119d3b0317d509d10fd6e3a8972f175a845aeb52a00a", "labels_vector_sha256": "652de3b07ad14f89e90f299ef768120339692b88e16fd80e88dd9797637d5a34", "planned_active_tokens": 64976, "realized_active_tokens": 64976, "cumulative_start": 8192849, "cumulative_end": 8257825, "loss_sum": 10622.533966064453, "token_normalized_loss": 0.1634839627872515, "grad_norm_before_clip": 0.09386587142944336, "learning_rates_at_step": { "lora": 1.4596295877080029e-05, "row_tokens": 1.4596295877080029e-05 }, "learning_rates_after_step": { "lora": 1.3881187818644182e-05, "row_tokens": 1.3881187818644182e-05 } }, { "update_id": 126, "row_count": 80, "row_ids_sha256": "651cdb67f50cd3c7b623169801c0466dc378b5ed840e0e3d1635ffd326d2a867", "input_ids_vector_sha256": "10394f77b9ef79114699844b288b496350b95d57266640aeb2dbb2c815590112", "labels_vector_sha256": "81a9432487b2bd03c3c2e15c7289264bdd8ac75587851cfd1458302c8d88efa2", "planned_active_tokens": 66065, "realized_active_tokens": 66065, "cumulative_start": 8257825, "cumulative_end": 8323890, "loss_sum": 10118.078824996948, "token_normalized_loss": 0.1531533917353659, "grad_norm_before_clip": 0.09107738733291626, "learning_rates_at_step": { "lora": 1.3881187818644182e-05, "row_tokens": 1.3881187818644182e-05 }, "learning_rates_after_step": { "lora": 1.3191698677197672e-05, "row_tokens": 1.3191698677197672e-05 } }, { "update_id": 127, "row_count": 80, "row_ids_sha256": "88dc6f10d80f4964eff6b86ac1600deb5aa125791e22f08b815dfa94a4295b38", "input_ids_vector_sha256": "6920e9f79afabd7c3cb9e1c1a0ac9a0f1c9aa82860b99fd8cb46de64bbe344ab", "labels_vector_sha256": "941aae8bbaf2dc1f4fe09d55f2131679e586f1924215b50775dc6c28fa7f0355", "planned_active_tokens": 65064, "realized_active_tokens": 65064, "cumulative_start": 8323890, "cumulative_end": 8388954, "loss_sum": 10357.185806274414, "token_normalized_loss": 0.15918458450563158, "grad_norm_before_clip": 0.09424533694982529, "learning_rates_at_step": { "lora": 1.3191698677197672e-05, "row_tokens": 1.3191698677197672e-05 }, "learning_rates_after_step": { "lora": 1.2509177603000804e-05, "row_tokens": 1.2509177603000804e-05 } }, { "update_id": 128, "row_count": 80, "row_ids_sha256": "86c22d5fab2bd9be1206e843b2109902231af25a6188b412f76af5b8f0cb88d5", "input_ids_vector_sha256": "b02e792645ea27359693b8968aab488c8b787b76ef45a9d6a06c5bba40ce76ca", "labels_vector_sha256": "07608c3303a01e99e0965234b7a8ebe1fd2e92932a4cb29d57b8d8a9e5133f98", "planned_active_tokens": 65843, "realized_active_tokens": 65843, "cumulative_start": 8388954, "cumulative_end": 8454797, "loss_sum": 10325.315029144287, "token_normalized_loss": 0.15681720196747242, "grad_norm_before_clip": 0.09206037223339081, "learning_rates_at_step": { "lora": 1.2509177603000804e-05, "row_tokens": 1.2509177603000804e-05 }, "learning_rates_after_step": { "lora": 1.1850431823927915e-05, "row_tokens": 1.1850431823927915e-05 } }, { "update_id": 129, "row_count": 75, "row_ids_sha256": "4a8cc1e6e85333aa8edf47ef100c7bfe727996d890a1c95f58fa972846d71b24", "input_ids_vector_sha256": "6fdb87ef0e6aad607dab69471750de7173284f33b59f61047878eee6de4e3dcb", "labels_vector_sha256": "f93ea95e1239b07bb911e77770c76f43b7d777e8480613853be23a54425a310c", "planned_active_tokens": 65026, "realized_active_tokens": 65026, "cumulative_start": 8454797, "cumulative_end": 8519823, "loss_sum": 9595.226421356201, "token_normalized_loss": 0.14755984408323133, "grad_norm_before_clip": 0.09254449605941772, "learning_rates_at_step": { "lora": 1.1850431823927915e-05, "row_tokens": 1.1850431823927915e-05 }, "learning_rates_after_step": { "lora": 1.1202436029041536e-05, "row_tokens": 1.1202436029041536e-05 } }, { "update_id": 130, "row_count": 76, "row_ids_sha256": "ee2fb5654e154972eae4008d64e6232e41b87d9d3e8f5a06cbbac2093a6c6f2b", "input_ids_vector_sha256": "62dbeb6f7c6f5d77a0ae055b0b2d511a27bc1d8678cf638d35fd205f3c6a3c5d", "labels_vector_sha256": "60bb0e38d7f188050a82d11dd29a5d91b7b9a699e2bf9d37f9bf0e42abb46687", "planned_active_tokens": 65510, "realized_active_tokens": 65510, "cumulative_start": 8519823, "cumulative_end": 8585333, "loss_sum": 10209.071529388428, "token_normalized_loss": 0.15583989512117888, "grad_norm_before_clip": 0.09333337843418121, "learning_rates_at_step": { "lora": 1.1202436029041536e-05, "row_tokens": 1.1202436029041536e-05 }, "learning_rates_after_step": { "lora": 1.0568855654276444e-05, "row_tokens": 1.0568855654276444e-05 } }, { "update_id": 131, "row_count": 76, "row_ids_sha256": "c89b34cb81c336999f8bbb7e73d84c839a6287cc582b48833f8d3b39a309f773", "input_ids_vector_sha256": "bffad5a679f5dd0a2bd48470e51b3d7d054c4922042d09db1d3338eeee39302a", "labels_vector_sha256": "3d63c87783ea7a0c97693313d8d6764547bf197260335db64805c18ad3649589", "planned_active_tokens": 65674, "realized_active_tokens": 65674, "cumulative_start": 8585333, "cumulative_end": 8651007, "loss_sum": 9052.057426452637, "token_normalized_loss": 0.1378331977107019, "grad_norm_before_clip": 0.08757183700799942, "learning_rates_at_step": { "lora": 1.0568855654276444e-05, "row_tokens": 1.0568855654276444e-05 }, "learning_rates_after_step": { "lora": 9.95523039816092e-06, "row_tokens": 9.95523039816092e-06 } }, { "update_id": 132, "row_count": 75, "row_ids_sha256": "06b99b2dad36fd37deaf607567557eecbc8204e0f1b3041934ecd0fdb68caa48", "input_ids_vector_sha256": "8cf13ef80e8ac1ae89eb17bc911a15abcd331f68e06dd31c622b3af3ce84e6af", "labels_vector_sha256": "72b280e67c93c6dcc2f16f4df0e3faebb20d22f54bac80b31b1ea7ee0dbc82e8", "planned_active_tokens": 65282, "realized_active_tokens": 65282, "cumulative_start": 8651007, "cumulative_end": 8716289, "loss_sum": 9104.073110580444, "token_normalized_loss": 0.1394576316684606, "grad_norm_before_clip": 0.11982797086238861, "learning_rates_at_step": { "lora": 9.95523039816092e-06, "row_tokens": 9.95523039816092e-06 }, "learning_rates_after_step": { "lora": 9.352559981614689e-06, "row_tokens": 9.352559981614689e-06 } }, { "update_id": 133, "row_count": 77, "row_ids_sha256": "b115382647baba8ebc84b72224bed69eca62cd4b4cf36f8f45b4d6ffa2f60f22", "input_ids_vector_sha256": "0182bc573f40dd4f48de15678da8792fd80b6bbf7b57e5e800ae4753c7d2c85e", "labels_vector_sha256": "5a10fb93f8662aedfdcf4d892a08b3711911f1612d06b3cc5fd432346fc05f53", "planned_active_tokens": 65883, "realized_active_tokens": 65883, "cumulative_start": 8716289, "cumulative_end": 8782172, "loss_sum": 9302.528377532959, "token_normalized_loss": 0.14119770468152573, "grad_norm_before_clip": 0.09202249348163605, "learning_rates_at_step": { "lora": 9.352559981614689e-06, "row_tokens": 9.352559981614689e-06 }, "learning_rates_after_step": { "lora": 8.769176711957122e-06, "row_tokens": 8.769176711957122e-06 } }, { "update_id": 134, "row_count": 76, "row_ids_sha256": "2e1b8635c6589d55a8297eb557c282e8c3a1f2c2ee19e779f20dae83c396feeb", "input_ids_vector_sha256": "fbaf54949853e4a7ac2e9665d3842b69c0237ac1a1a16cfe099705379b98f7a3", "labels_vector_sha256": "cf342088afb736d1d5b940cc573599742076393949ab2da8b78885692eef13b4", "planned_active_tokens": 65614, "realized_active_tokens": 65614, "cumulative_start": 8782172, "cumulative_end": 8847786, "loss_sum": 10656.816116333008, "token_normalized_loss": 0.16241680306539774, "grad_norm_before_clip": 0.09224680811166763, "learning_rates_at_step": { "lora": 8.769176711957122e-06, "row_tokens": 8.769176711957122e-06 }, "learning_rates_after_step": { "lora": 8.202333787443695e-06, "row_tokens": 8.202333787443695e-06 } }, { "update_id": 135, "row_count": 76, "row_ids_sha256": "f26d21835da605c4b4e9ced87833be727f92be0e7f743f7f1e6cad0977b981be", "input_ids_vector_sha256": "bf3325215e68569cfca739558606231798224e7e5bcc4392a00dd7eaeb212096", "labels_vector_sha256": "2567412ad4417ff141a4fecf846bbf28ed5ab7fa7d9257bd752f343a6e76e6e6", "planned_active_tokens": 65672, "realized_active_tokens": 65672, "cumulative_start": 8847786, "cumulative_end": 8913458, "loss_sum": 11642.528675079346, "token_normalized_loss": 0.17728299237238618, "grad_norm_before_clip": 0.09887941926717758, "learning_rates_at_step": { "lora": 8.202333787443695e-06, "row_tokens": 8.202333787443695e-06 }, "learning_rates_after_step": { "lora": 7.656099810649337e-06, "row_tokens": 7.656099810649337e-06 } }, { "update_id": 136, "row_count": 74, "row_ids_sha256": "b278ee3c57782a4d99bddf828d2eeebab458f11f02ab0e90458cbf57eebed3c7", "input_ids_vector_sha256": "7d6ca93261544b848f7c5957c19c035c279bfe42918e094d30f9f996f4011cdd", "labels_vector_sha256": "50176e17e9b112f0c11dcc97652e9231a303ff6deb41815eb4629f3f7de71732", "planned_active_tokens": 65270, "realized_active_tokens": 65270, "cumulative_start": 8913458, "cumulative_end": 8978728, "loss_sum": 11520.284286499023, "token_normalized_loss": 0.17650198079514362, "grad_norm_before_clip": 0.10002458840608597, "learning_rates_at_step": { "lora": 7.656099810649337e-06, "row_tokens": 7.656099810649337e-06 }, "learning_rates_after_step": { "lora": 7.126251619772989e-06, "row_tokens": 7.126251619772989e-06 } }, { "update_id": 137, "row_count": 75, "row_ids_sha256": "8133397929869918b7ba7eb96b4cd7a65496a9bf3aff79c58cd7d8058a03a2af", "input_ids_vector_sha256": "aaa8748e62acd8b25a6fe040222bd429819741dca5f579673b27841d8fdc48ad", "labels_vector_sha256": "78529179db7bea9617165779c25c7b77a5c9e2a1b771af34e95b37e2a9d72366", "planned_active_tokens": 65386, "realized_active_tokens": 65386, "cumulative_start": 8978728, "cumulative_end": 9044114, "loss_sum": 10379.742610931396, "token_normalized_loss": 0.15874564296533503, "grad_norm_before_clip": 0.09289399534463882, "learning_rates_at_step": { "lora": 7.126251619772989e-06, "row_tokens": 7.126251619772989e-06 }, "learning_rates_after_step": { "lora": 6.608350549884151e-06, "row_tokens": 6.608350549884151e-06 } }, { "update_id": 138, "row_count": 78, "row_ids_sha256": "3b58c0216a4eefb45b0378a82f95bdf4ff7228524c0abc6b20c5f5abdc57e816", "input_ids_vector_sha256": "aa43879ad25929cc4266e1bd960b4bac59c2b32d3f5de78cf1332c86377c6e05", "labels_vector_sha256": "0de03cd3d63e88628dc1b37371a70c4e0a0441735b8e8c98e2480b2251ed2768", "planned_active_tokens": 66119, "realized_active_tokens": 66119, "cumulative_start": 9044114, "cumulative_end": 9110233, "loss_sum": 9913.165687561035, "token_normalized_loss": 0.14992915330783943, "grad_norm_before_clip": 0.09011631458997726, "learning_rates_at_step": { "lora": 6.608350549884151e-06, "row_tokens": 6.608350549884151e-06 }, "learning_rates_after_step": { "lora": 6.118337690817077e-06, "row_tokens": 6.118337690817077e-06 } }, { "update_id": 139, "row_count": 77, "row_ids_sha256": "6d5de4ae282230d304403380f634cf630f08600a1190fcfe7eb698a70ae31b9d", "input_ids_vector_sha256": "1379af257cdd7b8126ce402d6317dd374d4c89da330bd5ecd51cab00d497940b", "labels_vector_sha256": "00456307cbb4021f7a4e5d7fc401e53122ee38f84b3722176ac57515207394bd", "planned_active_tokens": 64814, "realized_active_tokens": 64814, "cumulative_start": 9110233, "cumulative_end": 9175047, "loss_sum": 10984.72108078003, "token_normalized_loss": 0.1694806844320676, "grad_norm_before_clip": 0.09641729295253754, "learning_rates_at_step": { "lora": 6.118337690817077e-06, "row_tokens": 6.118337690817077e-06 }, "learning_rates_after_step": { "lora": 5.6356510547881495e-06, "row_tokens": 5.6356510547881495e-06 } }, { "update_id": 140, "row_count": 75, "row_ids_sha256": "a8afff7365f254e1d10debd9edb54952ae02f36164762554af7beb18ff065542", "input_ids_vector_sha256": "71444b872ef8929614cd954ea0333ed07e36a8b51551abc87aba691f867d7beb", "labels_vector_sha256": "81deb193a44ab148dc6183c0bb34429a1cdb959fa2738019ae5dec7acf8a0e46", "planned_active_tokens": 66264, "realized_active_tokens": 66264, "cumulative_start": 9175047, "cumulative_end": 9241311, "loss_sum": 10177.100206375122, "token_normalized_loss": 0.15358415136990103, "grad_norm_before_clip": 0.0904715359210968, "learning_rates_at_step": { "lora": 5.6356510547881495e-06, "row_tokens": 5.6356510547881495e-06 }, "learning_rates_after_step": { "lora": 5.180592804703477e-06, "row_tokens": 5.180592804703477e-06 } }, { "update_id": 141, "row_count": 75, "row_ids_sha256": "2954eb9b7d7e510ca4dffd87527643c1d2516eb457b8897759625ff03ff63088", "input_ids_vector_sha256": "6a094cb0c0eae07b96c49e9bcabaad7b21cb9922be460822a95b8e380eb57f56", "labels_vector_sha256": "2f849c54acaac5f9297939133506d726e12c33488ceba59772c2d09b50d64cac", "planned_active_tokens": 64962, "realized_active_tokens": 64962, "cumulative_start": 9241311, "cumulative_end": 9306273, "loss_sum": 9339.265491485596, "token_normalized_loss": 0.14376505482413712, "grad_norm_before_clip": 0.09172554314136505, "learning_rates_at_step": { "lora": 5.180592804703477e-06, "row_tokens": 5.180592804703477e-06 }, "learning_rates_after_step": { "lora": 4.7354038286943344e-06, "row_tokens": 4.7354038286943344e-06 } }, { "update_id": 142, "row_count": 79, "row_ids_sha256": "d464b70a3c62dfff190536089f95e0ee8d900de12811b54d0b53c78d2f1a91f0", "input_ids_vector_sha256": "606b1249d60b8b59c9a508563beeed11f959aeef6a7250ea4c1b8989efabfc1a", "labels_vector_sha256": "d5fca745c52d827c9a6f30439858b9f8292850541c8518d7cf439fbc9ea8a1bc", "planned_active_tokens": 66220, "realized_active_tokens": 66220, "cumulative_start": 9306273, "cumulative_end": 9372493, "loss_sum": 9165.808109283447, "token_normalized_loss": 0.13841449878108497, "grad_norm_before_clip": 0.08947461098432541, "learning_rates_at_step": { "lora": 4.7354038286943344e-06, "row_tokens": 4.7354038286943344e-06 }, "learning_rates_after_step": { "lora": 4.314704604618513e-06, "row_tokens": 4.314704604618513e-06 } }, { "update_id": 143, "row_count": 75, "row_ids_sha256": "dab85889e77dbf88e86443c667637812d0fe3c4f1adafc4cd766f1b698d89092", "input_ids_vector_sha256": "c829fec2b8f6b1b2c2a4ba6acd4bed8a183f9a7d31213fe70458e1c081d166b8", "labels_vector_sha256": "26eb2904a40667d2810c7b542cdf6ad2944cc95c389443bedb2fc270d251de12", "planned_active_tokens": 65355, "realized_active_tokens": 65355, "cumulative_start": 9372493, "cumulative_end": 9437848, "loss_sum": 9397.875389099121, "token_normalized_loss": 0.14379734357125118, "grad_norm_before_clip": 0.09392246603965759, "learning_rates_at_step": { "lora": 4.314704604618513e-06, "row_tokens": 4.314704604618513e-06 }, "learning_rates_after_step": { "lora": 3.91486889275604e-06, "row_tokens": 3.91486889275604e-06 } }, { "update_id": 144, "row_count": 76, "row_ids_sha256": "b506e0b34480cdb5a8f675fd502a3f3c9f484e9df857a5c928652e0ae27e5d7b", "input_ids_vector_sha256": "94c13d6d066b71140277da3b1db6a9705b386e63c0c71175a265842dfa1f1899", "labels_vector_sha256": "616d68647394fb07b601240d98fba09a4dd19542d34b71410794fbe39310b5c5", "planned_active_tokens": 64999, "realized_active_tokens": 64999, "cumulative_start": 9437848, "cumulative_end": 9502847, "loss_sum": 8090.472959518433, "token_normalized_loss": 0.12447072969612506, "grad_norm_before_clip": 0.09159792214632034, "learning_rates_at_step": { "lora": 3.91486889275604e-06, "row_tokens": 3.91486889275604e-06 }, "learning_rates_after_step": { "lora": 3.5303773612422485e-06, "row_tokens": 3.5303773612422485e-06 } }, { "update_id": 145, "row_count": 80, "row_ids_sha256": "df44f39f3766465c073b3d80f1c58651d5f4392cafec035db6160599815961b0", "input_ids_vector_sha256": "ee921169b42046001fbb170b98239e45f3521271541d41873c982fe2399ba103", "labels_vector_sha256": "bf8a4c64e5254a26d39a5d48c428da61890e1da52d8b7c17f62aa61c9718edd5", "planned_active_tokens": 65583, "realized_active_tokens": 65583, "cumulative_start": 9502847, "cumulative_end": 9568430, "loss_sum": 8754.5417137146, "token_normalized_loss": 0.13348797270198984, "grad_norm_before_clip": 0.08998316526412964, "learning_rates_at_step": { "lora": 3.5303773612422485e-06, "row_tokens": 3.5303773612422485e-06 }, "learning_rates_after_step": { "lora": 3.1621970716739003e-06, "row_tokens": 3.1621970716739003e-06 } }, { "update_id": 146, "row_count": 78, "row_ids_sha256": "b796be22ce9002269cb6ac6bb128454e72ba9d87ee2e003a3934eacdcfc2c8bf", "input_ids_vector_sha256": "75894af82e080ec7dfee100a3206df6da6cad368e5bc1a8a37344f0408b7ceb4", "labels_vector_sha256": "4abdcda7f5b290e4c188f8f3c4187787fab9bb24fbb1e5315adcfbae60a284c4", "planned_active_tokens": 66112, "realized_active_tokens": 66112, "cumulative_start": 9568430, "cumulative_end": 9634542, "loss_sum": 9071.805877685547, "token_normalized_loss": 0.1372187481498903, "grad_norm_before_clip": 0.09472867101430893, "learning_rates_at_step": { "lora": 3.1621970716739003e-06, "row_tokens": 3.1621970716739003e-06 }, "learning_rates_after_step": { "lora": 2.8200219560327977e-06, "row_tokens": 2.8200219560327977e-06 } }, { "update_id": 147, "row_count": 80, "row_ids_sha256": "b3f557ecc75d23d8b67c46ce7cd7d7dc976e0efbf76facf95a483496c5920481", "input_ids_vector_sha256": "da2baf6d3c77263e56e097c4f22231ee9b807651208852688f47235a816a9a3f", "labels_vector_sha256": "80c6f644bc24b30a179c81628177265e3955cb7248c13b44442c78d2026e5216", "planned_active_tokens": 64871, "realized_active_tokens": 64871, "cumulative_start": 9634542, "cumulative_end": 9699413, "loss_sum": 11285.813360214233, "token_normalized_loss": 0.1739731676745269, "grad_norm_before_clip": 0.09917371720075607, "learning_rates_at_step": { "lora": 2.8200219560327977e-06, "row_tokens": 2.8200219560327977e-06 }, "learning_rates_after_step": { "lora": 2.488548752468867e-06, "row_tokens": 2.488548752468867e-06 } }, { "update_id": 148, "row_count": 77, "row_ids_sha256": "64fe8dba55b7bc075f7d31fe6f06bda2faa90d5b089fa50d24e4d2d4184f03d2", "input_ids_vector_sha256": "31c0103500d8ac4db78e1c4ca65487f6e6fb01139490b877883420a6ef150866", "labels_vector_sha256": "c5d491caea8748ae1b188c39aa09585a859affd61c93b79679239557e24e9337", "planned_active_tokens": 66600, "realized_active_tokens": 66600, "cumulative_start": 9699413, "cumulative_end": 9766013, "loss_sum": 11456.953105926514, "token_normalized_loss": 0.17202632291180953, "grad_norm_before_clip": 0.09694084525108337, "learning_rates_at_step": { "lora": 2.488548752468867e-06, "row_tokens": 2.488548752468867e-06 }, "learning_rates_after_step": { "lora": 2.1832831325170888e-06, "row_tokens": 2.1832831325170888e-06 } }, { "update_id": 149, "row_count": 55, "row_ids_sha256": "6ecf7be701e3ef2e174e492c3b8dfc4d4d50c58b7b864dc814bf1b7b132ce9e1", "input_ids_vector_sha256": "c40c7baeb2df66a45e6f2828d558f0aafe9b4a9b58d3c24f4542d69f38ba2ccd", "labels_vector_sha256": "10418f995f88c0f3b4b64f0729ffe41de3c40a792b88897e0745d45c0d580d94", "planned_active_tokens": 65277, "realized_active_tokens": 65277, "cumulative_start": 9766013, "cumulative_end": 9831290, "loss_sum": 12878.851249694824, "token_normalized_loss": 0.19729539117445385, "grad_norm_before_clip": 0.1019398495554924, "learning_rates_at_step": { "lora": 2.1832831325170888e-06, "row_tokens": 2.1832831325170888e-06 }, "learning_rates_after_step": { "lora": 1.8977383378495695e-06, "row_tokens": 1.8977383378495695e-06 } }, { "update_id": 150, "row_count": 51, "row_ids_sha256": "9280fa7dc13c0850dcb896fe144d63e7931f10587dcb817f2109cc0d6d32c2da", "input_ids_vector_sha256": "82cdc62f11de90f9af96e7bf7173c706c051296563d4344fabf666bcd7f3659c", "labels_vector_sha256": "2d65d22d39f7d6339bc25d78a86c6b4e874445f335ed62d68c2ecbbd30f6cd21", "planned_active_tokens": 65237, "realized_active_tokens": 65237, "cumulative_start": 9831290, "cumulative_end": 9896527, "loss_sum": 12345.737804412842, "token_normalized_loss": 0.18924441351400037, "grad_norm_before_clip": 0.10230951011180878, "learning_rates_at_step": { "lora": 1.8977383378495695e-06, "row_tokens": 1.8977383378495695e-06 }, "learning_rates_after_step": { "lora": 1.628895677001202e-06, "row_tokens": 1.628895677001202e-06 } }, { "update_id": 151, "row_count": 51, "row_ids_sha256": "495331346443283f15f77574021e7c6a417c3897e12058408d9297aad674ea4a", "input_ids_vector_sha256": "61291e7608baacc4aba7d54fabc7cec06362513cfff2d681012ab505386103c0", "labels_vector_sha256": "d7d65d7a995636e87537b342eacc9494b79d530ef13c19a45140f16846f760f5", "planned_active_tokens": 65987, "realized_active_tokens": 65987, "cumulative_start": 9896527, "cumulative_end": 9962514, "loss_sum": 13379.218643188477, "token_normalized_loss": 0.20275537065162041, "grad_norm_before_clip": 0.10287613421678543, "learning_rates_at_step": { "lora": 1.628895677001202e-06, "row_tokens": 1.628895677001202e-06 }, "learning_rates_after_step": { "lora": 1.3843040139133178e-06, "row_tokens": 1.3843040139133178e-06 } }, { "update_id": 152, "row_count": 57, "row_ids_sha256": "388eaaa8db2c27a542ca95f0e6855b324736767b34a92a26cd77e73a95481eeb", "input_ids_vector_sha256": "d41769f25a22f606e837ee429ee8a8a0ac5964ea3823b03b612995f80b0ff18d", "labels_vector_sha256": "8f024c1f707d7825acc085964c122aefac765fa30f3658130f7baaf221e68d4c", "planned_active_tokens": 64870, "realized_active_tokens": 64870, "cumulative_start": 9962514, "cumulative_end": 10027384, "loss_sum": 14292.395889282227, "token_normalized_loss": 0.22032366100327155, "grad_norm_before_clip": 0.10882115364074707, "learning_rates_at_step": { "lora": 1.3843040139133178e-06, "row_tokens": 1.3843040139133178e-06 }, "learning_rates_after_step": { "lora": 1.1535709050413402e-06, "row_tokens": 1.1535709050413402e-06 } }, { "update_id": 153, "row_count": 51, "row_ids_sha256": "7acf16cb8753c0672ba916dff7be7010460edaf64d5bd56ddb4efaf03a153bde", "input_ids_vector_sha256": "231033f79215e61f968e808bf9cd83dffc558f528138a37390af4711fe4025d1", "labels_vector_sha256": "3f9c9318ca8f1cb7ce4577f00dd0e38a84688c797944f2664f46870e2ff75895", "planned_active_tokens": 66613, "realized_active_tokens": 66613, "cumulative_start": 10027384, "cumulative_end": 10093997, "loss_sum": 12996.549072265625, "token_normalized_loss": 0.1951052958471413, "grad_norm_before_clip": 0.10088209807872772, "learning_rates_at_step": { "lora": 1.1535709050413402e-06, "row_tokens": 1.1535709050413402e-06 }, "learning_rates_after_step": { "lora": 9.494766284842749e-07, "row_tokens": 9.494766284842749e-07 } }, { "update_id": 154, "row_count": 44, "row_ids_sha256": "a17fb6e0c8f157038d11cba48e1904a6005a9cdd87ca54869611fecbb9e2bdf3", "input_ids_vector_sha256": "bd9506524882de638eeedefb40a3eb327bed40bf6b4db90834c3cabca14fd50e", "labels_vector_sha256": "9742b730a354d4e65569ea6123e2a1550e30f2413416f9414803ee30a138f26e", "planned_active_tokens": 64666, "realized_active_tokens": 64666, "cumulative_start": 10093997, "cumulative_end": 10158663, "loss_sum": 13235.021896362305, "token_normalized_loss": 0.2046673970303143, "grad_norm_before_clip": 0.10749612003564835, "learning_rates_at_step": { "lora": 9.494766284842749e-07, "row_tokens": 9.494766284842749e-07 }, "learning_rates_after_step": { "lora": 7.639074002155078e-07, "row_tokens": 7.639074002155078e-07 } }, { "update_id": 155, "row_count": 51, "row_ids_sha256": "6bec14d3cc0781b80e28ba757043809f36c8f970db02ca874e20d7f078bca412", "input_ids_vector_sha256": "e3b9f9f0afe5b6fae326f0dca4e561442a421f973a0b540412c5e8cf0b8adb66", "labels_vector_sha256": "4b7b103b002ee5c9ae3d9cd537ce7a18c1512f38ea73011cabb1f1bb050ab17a", "planned_active_tokens": 65095, "realized_active_tokens": 65095, "cumulative_start": 10158663, "cumulative_end": 10223758, "loss_sum": 14562.182147979736, "token_normalized_loss": 0.22370661568445713, "grad_norm_before_clip": 0.1133539006114006, "learning_rates_at_step": { "lora": 7.639074002155078e-07, "row_tokens": 7.639074002155078e-07 }, "learning_rates_after_step": { "lora": 5.94963360056161e-07, "row_tokens": 5.94963360056161e-07 } }, { "update_id": 156, "row_count": 60, "row_ids_sha256": "c93c233a5b6896eac4997d2875fdf9b124aaca991918aa025975b4842c23362c", "input_ids_vector_sha256": "ca845938ff326db24f8e56b2e96aef25b2439c4f365dac4a1292af9be5d42ac4", "labels_vector_sha256": "7a8dbee9cdf6f519a81e1d8a437ae29f1f2ac083df8da67d72e3aa21432dfed0", "planned_active_tokens": 66518, "realized_active_tokens": 66518, "cumulative_start": 10223758, "cumulative_end": 10290276, "loss_sum": 13597.754341125488, "token_normalized_loss": 0.20442217657063483, "grad_norm_before_clip": 0.09895586967468262, "learning_rates_at_step": { "lora": 5.94963360056161e-07, "row_tokens": 5.94963360056161e-07 }, "learning_rates_after_step": { "lora": 4.4962415390650827e-07, "row_tokens": 4.4962415390650827e-07 } }, { "update_id": 157, "row_count": 77, "row_ids_sha256": "9c805dff4a1f842fa70c4c414a5d25a2338a08912b7134f6a9057eb7a77ed7a6", "input_ids_vector_sha256": "90e3d532e3174da8715496ca000f07a5951ac711e6c2cd10ce00f38a8172a8a9", "labels_vector_sha256": "9bce37dfd221ac713c26ceabf77c9390f75255d9423226eacd2e3eb40694be8e", "planned_active_tokens": 65248, "realized_active_tokens": 65248, "cumulative_start": 10290276, "cumulative_end": 10355524, "loss_sum": 13139.884281158447, "token_normalized_loss": 0.20138370955674423, "grad_norm_before_clip": 0.11132887005805969, "learning_rates_at_step": { "lora": 4.4962415390650827e-07, "row_tokens": 4.4962415390650827e-07 }, "learning_rates_after_step": { "lora": 3.248994415036788e-07, "row_tokens": 3.248994415036788e-07 } }, { "update_id": 158, "row_count": 81, "row_ids_sha256": "dda6a53e967c7e3d13eb506b1b15618e831d5f8e39ff70e8ca77533740ad32e1", "input_ids_vector_sha256": "48814f2003ccb7d9a2dd5ed6a45cec43c9a12c14ed83729d5a0c12ddbbdd65f7", "labels_vector_sha256": "922917cbb05706c0bd806bbbd271fe8d3a48e191724a177379d2ef924974443f", "planned_active_tokens": 65037, "realized_active_tokens": 65037, "cumulative_start": 10355524, "cumulative_end": 10420561, "loss_sum": 13388.152364730835, "token_normalized_loss": 0.20585439618572252, "grad_norm_before_clip": 0.12893030047416687, "learning_rates_at_step": { "lora": 3.248994415036788e-07, "row_tokens": 3.248994415036788e-07 }, "learning_rates_after_step": { "lora": 2.1954203272112616e-07, "row_tokens": 2.1954203272112616e-07 } }, { "update_id": 159, "row_count": 68, "row_ids_sha256": "77ceaeb3de3f3c663edc1873e7b08e02083809729833fd9d824cf0246118775c", "input_ids_vector_sha256": "d7808091cc70c02e1665f5268a8f985399ce2fd3a621337ff2ef173d12c4b5b4", "labels_vector_sha256": "67811b94bdccc04daa1a6fac837965c812a5d775eeff98e2405a5ffae9be98e0", "planned_active_tokens": 65585, "realized_active_tokens": 65585, "cumulative_start": 10420561, "cumulative_end": 10486146, "loss_sum": 15584.649702072144, "token_normalized_loss": 0.23762521463859332, "grad_norm_before_clip": 0.11219551414251328, "learning_rates_at_step": { "lora": 2.1954203272112616e-07, "row_tokens": 2.1954203272112616e-07 }, "learning_rates_after_step": { "lora": 1.3466823847039235e-07, "row_tokens": 1.3466823847039235e-07 } }, { "update_id": 160, "row_count": 78, "row_ids_sha256": "2a0f52c793a5dba1b45e8b1344eb20906c6974a542f3256c65f6ecda29dfb72f", "input_ids_vector_sha256": "c6c5bd584ee209b98b12d42b4f4e1f868d2526065bff0cb41d62128f447e7202", "labels_vector_sha256": "e013d5214ebd6fd4af73f1201a6aa4078cbcfd5caa3f1293e877eadea5fb20bd", "planned_active_tokens": 65641, "realized_active_tokens": 65641, "cumulative_start": 10486146, "cumulative_end": 10551787, "loss_sum": 15278.742835998535, "token_normalized_loss": 0.2327621888148952, "grad_norm_before_clip": 0.11134673655033112, "learning_rates_at_step": { "lora": 1.3466823847039235e-07, "row_tokens": 1.3466823847039235e-07 }, "learning_rates_after_step": { "lora": 7.053518943780368e-08, "row_tokens": 7.053518943780368e-08 } }, { "update_id": 161, "row_count": 72, "row_ids_sha256": "fb8329ebf75a27fd23348160423b616d45ea99b05de9aff12bbd155e748263ed", "input_ids_vector_sha256": "a91263468b96fbfe4f503b461f1f730c8f678aed458b63cadb97a7af049af4b6", "labels_vector_sha256": "990bb65741efd1bb9ac91d4b05cacb3ed28a4068f7a9e6a2bf7fdc929c0b4d0a", "planned_active_tokens": 65491, "realized_active_tokens": 65491, "cumulative_start": 10551787, "cumulative_end": 10617278, "loss_sum": 13820.899841308594, "token_normalized_loss": 0.21103510163699735, "grad_norm_before_clip": 0.10197628289461136, "learning_rates_at_step": { "lora": 7.053518943780368e-08, "row_tokens": 7.053518943780368e-08 }, "learning_rates_after_step": { "lora": 2.7060780601778636e-08, "row_tokens": 2.7060780601778636e-08 } }, { "update_id": 162, "row_count": 68, "row_ids_sha256": "73378617771c3f7521a62003b65c4490cfbe95085ce6c3d36fa1c8eda23c2c91", "input_ids_vector_sha256": "36858bf17babded7d47ca3904e42bfe4fb563c720d812a026d2e8365e65a08b3", "labels_vector_sha256": "ed54817bd17db09d9aaf48ceb92e1a51fcc0ec2347faaac7eb54759485427c9e", "planned_active_tokens": 65277, "realized_active_tokens": 65277, "cumulative_start": 10617278, "cumulative_end": 10682555, "loss_sum": 13081.107570648193, "token_normalized_loss": 0.2003938227959035, "grad_norm_before_clip": 0.10129038244485855, "learning_rates_at_step": { "lora": 2.7060780601778636e-08, "row_tokens": 2.7060780601778636e-08 }, "learning_rates_after_step": { "lora": 3.869017389179774e-09, "row_tokens": 3.869017389179774e-09 } }, { "update_id": 163, "row_count": 71, "row_ids_sha256": "abb946194c398e0cce5fe7a7b20cb97524e7d802219cde5cc4b3ca651e9be273", "input_ids_vector_sha256": "1845dfec45137a8d2d443b8ab3783524d520eebc169ca04f75a6ce82747c791b", "labels_vector_sha256": "a3d081b289d60db46c114821ece3fea6d2be2e8fb1c10ab5e7aff5eb1e6a44e6", "planned_active_tokens": 66052, "realized_active_tokens": 66052, "cumulative_start": 10682555, "cumulative_end": 10748607, "loss_sum": 14771.342641830444, "token_normalized_loss": 0.22363202691561868, "grad_norm_before_clip": 0.10431884974241257, "learning_rates_at_step": { "lora": 3.869017389179774e-09, "row_tokens": 3.869017389179774e-09 }, "learning_rates_after_step": { "lora": 0.0, "row_tokens": 0.0 } }, { "update_id": 164, "row_count": 40, "row_ids_sha256": "eeaa62601f796edbae0f281fc8b37a89451ebb2b6094112b6a922b6b5ff5f275", "input_ids_vector_sha256": "ee2dc1dc718e0de94cceb7bd9ebf9e6524c92bb1efe54e2d71b72043c706c687", "labels_vector_sha256": "a89c702f47009e1f641021a21d492a9fa4565062f51e883d1a13df7c32ff3bcb", "planned_active_tokens": 40159, "realized_active_tokens": 40159, "cumulative_start": 10748607, "cumulative_end": 10788766, "loss_sum": 9583.425552368164, "token_normalized_loss": 0.23863705650957853, "grad_norm_before_clip": 0.13246041536331177, "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": "e6ad2d1626ce1bba946ed9a10f7560ba5a4e101d611421670d0549ead9ec9e77", "parameter_count": 399, "bytes": 8822848512, "names_sha256": "76f6fba21da1d275c17e0470efbad7fb1d99cdeef7b4ffac8a5ac71ef82fdfa8" }, "optimizer_membership": { "schema": "think-b-2ep-optimizer-membership/1", "token_bake_enabled": false, "groups": [ { "group": "lora", "param_count": 504, "params": [ { "name": "base_model.model.model.layers.0.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.0.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.0.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.0.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.0.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.0.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.0.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.0.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.1.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.1.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.1.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.1.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.1.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.1.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.1.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.2.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.2.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.2.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.2.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.2.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.2.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.2.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.3.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.3.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.3.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.3.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.3.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.3.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.3.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.4.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.4.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.4.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.4.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.4.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.4.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.4.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.5.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.5.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.5.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.5.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.5.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.5.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.5.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.6.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.6.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.6.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.6.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.6.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.6.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.6.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.7.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.7.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.7.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.7.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.7.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.7.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.7.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.8.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.8.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.8.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.8.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.8.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.8.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.8.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.9.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.9.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.9.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.9.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.9.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.9.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.9.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.10.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.10.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.10.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.10.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.10.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.10.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.10.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.11.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.11.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.11.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.11.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.11.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.11.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.11.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.12.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.12.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.12.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.12.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.12.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.12.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.12.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.13.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.13.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.13.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.13.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.13.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.13.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.13.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.14.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.14.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.14.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.14.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.14.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.14.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.14.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.15.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.15.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.15.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.15.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.15.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.15.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.15.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.16.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.16.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.16.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.16.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.16.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.16.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.16.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.17.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.17.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.17.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.17.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.17.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.17.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.17.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.18.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.18.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.18.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.18.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.18.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.18.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.18.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.19.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.19.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.19.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.19.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.19.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.19.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.19.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.20.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.20.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.20.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.20.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.20.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.20.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.20.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.21.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.21.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.21.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.21.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.21.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.21.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.21.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.22.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.22.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.22.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.22.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.22.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.22.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.22.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.23.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.23.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.23.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.23.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.23.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.23.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.23.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.24.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.24.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.24.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.24.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.24.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.24.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.24.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.25.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.25.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.25.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.25.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.25.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.25.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.25.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.26.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.26.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.26.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.26.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.26.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.26.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.26.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.27.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.27.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.27.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.27.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.27.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.27.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.27.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.28.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.28.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.28.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.28.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.28.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.28.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.28.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.29.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.29.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.29.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.29.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.29.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.29.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.29.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.30.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.30.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.30.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.30.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.30.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.30.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.30.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "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, 2560 ] }, { "name": "base_model.model.model.layers.31.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.31.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.31.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 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": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.31.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.31.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.31.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.31.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.32.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.32.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.32.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.32.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.32.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.32.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.32.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.32.self_attn.o_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.32.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.32.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.32.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.32.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.32.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.32.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.33.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.33.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.33.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.33.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.33.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.33.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.33.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.33.self_attn.o_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.33.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.33.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.33.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.33.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.33.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.33.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.34.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.34.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.34.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.34.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.34.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.34.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.34.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.34.self_attn.o_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.34.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.34.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.34.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.34.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.34.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.34.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.35.self_attn.q_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.35.self_attn.q_proj.lora_B.default.weight", "shape": [ 4096, 64 ] }, { "name": "base_model.model.model.layers.35.self_attn.k_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.35.self_attn.k_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.35.self_attn.v_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.35.self_attn.v_proj.lora_B.default.weight", "shape": [ 1024, 64 ] }, { "name": "base_model.model.model.layers.35.self_attn.o_proj.lora_A.default.weight", "shape": [ 64, 4096 ] }, { "name": "base_model.model.model.layers.35.self_attn.o_proj.lora_B.default.weight", "shape": [ 2560, 64 ] }, { "name": "base_model.model.model.layers.35.mlp.gate_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.35.mlp.gate_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.35.mlp.up_proj.lora_A.default.weight", "shape": [ 64, 2560 ] }, { "name": "base_model.model.model.layers.35.mlp.up_proj.lora_B.default.weight", "shape": [ 9728, 64 ] }, { "name": "base_model.model.model.layers.35.mlp.down_proj.lora_A.default.weight", "shape": [ 64, 9728 ] }, { "name": "base_model.model.model.layers.35.mlp.down_proj.lora_B.default.weight", "shape": [ 2560, 64 ] } ] }, { "group": "row_tokens", "param_count": 2, "params": [ { "name": "base_model.model.model.embed_tokens.token_adapter.trainable_tokens_delta.default", "shape": [ 3, 2560 ] }, { "name": "base_model.model.lm_head.token_adapter.trainable_tokens_delta.default", "shape": [ 3, 2560 ] } ] } ] }, "tied_token_canary": null, "continuation": null, "lora_digest_start": { "scope": "lora", "sha256": "b5c8857a053d3f25970801e14e08483872c0503ca38d876902da21ea70841297", "parameter_count": 506, "bytes": 528543744, "names_sha256": "bb45cbfe1a7fcc7231a004dc60e90e0ddc33dcbcb776306b4f372164f4989122" } }