| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9985141158989599, | |
| "eval_steps": 42, | |
| "global_step": 168, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.30475762367257353, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6274, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 1.0297596454620361, | |
| "eval_runtime": 150.6014, | |
| "eval_samples_per_second": 1.653, | |
| "eval_steps_per_second": 0.83, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.2798137150913395, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6362, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.3178684451319545, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6299, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.33651284916847835, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6391, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.23984915388648712, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6071, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.20514040410017348, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5996, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.1950018128286362, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6298, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.14246019947393238, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5108, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.15792014279750227, | |
| "learning_rate": 0.00018, | |
| "loss": 0.5529, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.1517889177511264, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5433, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.1372121219437277, | |
| "learning_rate": 0.00019998023297700658, | |
| "loss": 0.5856, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.16740807394942855, | |
| "learning_rate": 0.00019992093972273018, | |
| "loss": 0.5546, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.13512320693394078, | |
| "learning_rate": 0.00019982214367819328, | |
| "loss": 0.6193, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.16169796294070152, | |
| "learning_rate": 0.0001996838839014696, | |
| "loss": 0.5495, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.16796913812281988, | |
| "learning_rate": 0.00019950621505224273, | |
| "loss": 0.5035, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.1800514764162192, | |
| "learning_rate": 0.00019928920737019733, | |
| "loss": 0.5083, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.170432124866908, | |
| "learning_rate": 0.0001990329466472502, | |
| "loss": 0.632, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.19129325489749488, | |
| "learning_rate": 0.00019873753419363336, | |
| "loss": 0.4813, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.1459357988760762, | |
| "learning_rate": 0.00019840308679784207, | |
| "loss": 0.4973, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.192594730984382, | |
| "learning_rate": 0.00019802973668046363, | |
| "loss": 0.5291, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.859025467969139, | |
| "learning_rate": 0.0001976176314419051, | |
| "loss": 0.5296, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.13366297885670222, | |
| "learning_rate": 0.000197166934004041, | |
| "loss": 0.4819, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.15698714419747645, | |
| "learning_rate": 0.00019667782254580374, | |
| "loss": 0.5409, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.10995943735837355, | |
| "learning_rate": 0.00019615049043274205, | |
| "loss": 0.5108, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.10796742192788925, | |
| "learning_rate": 0.00019558514614057609, | |
| "loss": 0.5215, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.11641740089490231, | |
| "learning_rate": 0.00019498201317277828, | |
| "loss": 0.5012, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.1120175962893241, | |
| "learning_rate": 0.00019434132997221345, | |
| "loss": 0.474, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.1218171278782483, | |
| "learning_rate": 0.0001936633498268728, | |
| "loss": 0.5216, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.11718521115928844, | |
| "learning_rate": 0.0001929483407697387, | |
| "loss": 0.4856, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.12611471038571026, | |
| "learning_rate": 0.00019219658547282067, | |
| "loss": 0.4823, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.11106871615269753, | |
| "learning_rate": 0.00019140838113540346, | |
| "loss": 0.4869, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.1416503230360699, | |
| "learning_rate": 0.00019058403936655233, | |
| "loss": 0.5341, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.10761396399791698, | |
| "learning_rate": 0.00018972388606192125, | |
| "loss": 0.4304, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.10975376180434356, | |
| "learning_rate": 0.0001888282612749132, | |
| "loss": 0.4646, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.12848879670359908, | |
| "learning_rate": 0.00018789751908224338, | |
| "loss": 0.4972, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.11904721819683833, | |
| "learning_rate": 0.00018693202744395827, | |
| "loss": 0.505, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.12249852034224981, | |
| "learning_rate": 0.00018593216805796612, | |
| "loss": 0.5396, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.12453395046646995, | |
| "learning_rate": 0.00018489833620913642, | |
| "loss": 0.4917, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.12585770374422164, | |
| "learning_rate": 0.00018383094061302766, | |
| "loss": 0.5079, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.11095271476322731, | |
| "learning_rate": 0.00018273040325430574, | |
| "loss": 0.4812, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.12968115101635422, | |
| "learning_rate": 0.00018159715921991612, | |
| "loss": 0.5106, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.10933018515590627, | |
| "learning_rate": 0.00018043165652707649, | |
| "loss": 0.4403, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.9767947196960449, | |
| "eval_runtime": 152.0343, | |
| "eval_samples_per_second": 1.638, | |
| "eval_steps_per_second": 0.822, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.11525719626792096, | |
| "learning_rate": 0.00017923435594615744, | |
| "loss": 0.482, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.12962154411778218, | |
| "learning_rate": 0.00017800573081852122, | |
| "loss": 0.5452, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.12555700120045588, | |
| "learning_rate": 0.0001767462668693908, | |
| "loss": 0.5084, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.11427565378293324, | |
| "learning_rate": 0.00017545646201582303, | |
| "loss": 0.5191, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.10974901402857151, | |
| "learning_rate": 0.00017413682616986185, | |
| "loss": 0.4703, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.11781465084480325, | |
| "learning_rate": 0.00017278788103694943, | |
| "loss": 0.4548, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.10781807228559999, | |
| "learning_rate": 0.000171410159909675, | |
| "loss": 0.476, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.12502639462035098, | |
| "learning_rate": 0.00017000420745694254, | |
| "loss": 0.5084, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.10718920826593327, | |
| "learning_rate": 0.00016857057950864132, | |
| "loss": 0.5093, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.10040549880547282, | |
| "learning_rate": 0.0001671098428359037, | |
| "loss": 0.4644, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.11778478994740472, | |
| "learning_rate": 0.00016562257492703757, | |
| "loss": 0.4725, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.1008386031049932, | |
| "learning_rate": 0.000164109363759222, | |
| "loss": 0.5121, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.1170302528140235, | |
| "learning_rate": 0.000162570807566056, | |
| "loss": 0.4766, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.1104526773884303, | |
| "learning_rate": 0.00016100751460105243, | |
| "loss": 0.4886, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.10467920768691032, | |
| "learning_rate": 0.00015942010289717105, | |
| "loss": 0.4703, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.11551406829220555, | |
| "learning_rate": 0.00015780920002248484, | |
| "loss": 0.4837, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.11133818831887894, | |
| "learning_rate": 0.0001561754428320771, | |
| "loss": 0.5148, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.11281448423273216, | |
| "learning_rate": 0.00015451947721626676, | |
| "loss": 0.4561, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.13934126997471205, | |
| "learning_rate": 0.00015284195784526195, | |
| "loss": 0.5069, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.11851655387640142, | |
| "learning_rate": 0.00015114354791034225, | |
| "loss": 0.5094, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.12909148374566123, | |
| "learning_rate": 0.0001494249188616723, | |
| "loss": 0.581, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.11070161341925377, | |
| "learning_rate": 0.00014768675014285062, | |
| "loss": 0.4585, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.13308674882888374, | |
| "learning_rate": 0.00014592972892229778, | |
| "loss": 0.4974, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.12124588853708144, | |
| "learning_rate": 0.0001441545498215912, | |
| "loss": 0.4463, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.1183570515369953, | |
| "learning_rate": 0.00014236191464085286, | |
| "loss": 0.447, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.13520024884417237, | |
| "learning_rate": 0.00014055253208129938, | |
| "loss": 0.5309, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.12184981458813801, | |
| "learning_rate": 0.00013872711746506413, | |
| "loss": 0.4532, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.12449299540645078, | |
| "learning_rate": 0.00013688639245240078, | |
| "loss": 0.5198, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.1383134750490429, | |
| "learning_rate": 0.00013503108475638244, | |
| "loss": 0.5629, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.246237001656926, | |
| "learning_rate": 0.0001331619278552068, | |
| "loss": 0.4869, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.13337703940933632, | |
| "learning_rate": 0.00013127966070222274, | |
| "loss": 0.4792, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.12428922033806454, | |
| "learning_rate": 0.00012938502743379212, | |
| "loss": 0.4825, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.13290774912900208, | |
| "learning_rate": 0.00012747877707510252, | |
| "loss": 0.5138, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.11185975046756892, | |
| "learning_rate": 0.0001255616632440475, | |
| "loss": 0.4815, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.1130592868215497, | |
| "learning_rate": 0.0001236344438532905, | |
| "loss": 0.5046, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.12882943465594857, | |
| "learning_rate": 0.0001216978808106318, | |
| "loss": 0.5091, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.14837896297082676, | |
| "learning_rate": 0.00011975273971779528, | |
| "loss": 0.5158, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.1265223309856292, | |
| "learning_rate": 0.00011779978956775506, | |
| "loss": 0.5068, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.14042502330520407, | |
| "learning_rate": 0.0001158398024407215, | |
| "loss": 0.5061, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.1261526695491767, | |
| "learning_rate": 0.00011387355319890685, | |
| "loss": 0.4691, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.12007305451001854, | |
| "learning_rate": 0.00011190181918019049, | |
| "loss": 0.4753, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.12809956897166885, | |
| "learning_rate": 0.00010992537989080618, | |
| "loss": 0.4417, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.9675251841545105, | |
| "eval_runtime": 152.4793, | |
| "eval_samples_per_second": 1.633, | |
| "eval_steps_per_second": 0.82, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.11858329804793687, | |
| "learning_rate": 0.00010794501669717145, | |
| "loss": 0.4868, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.10984649953887334, | |
| "learning_rate": 0.00010596151251698199, | |
| "loss": 0.4598, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.10927203986256682, | |
| "learning_rate": 0.0001039756515096926, | |
| "loss": 0.4693, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.11205046531522328, | |
| "learning_rate": 0.00010198821876650701, | |
| "loss": 0.4921, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.13232347270009215, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4695, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.12136881873560385, | |
| "learning_rate": 9.801178123349298e-05, | |
| "loss": 0.4859, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.14347476421156694, | |
| "learning_rate": 9.602434849030745e-05, | |
| "loss": 0.4796, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.13956845267055204, | |
| "learning_rate": 9.403848748301802e-05, | |
| "loss": 0.5339, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.12814010903196785, | |
| "learning_rate": 9.205498330282856e-05, | |
| "loss": 0.5267, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.12798850330908082, | |
| "learning_rate": 9.007462010919386e-05, | |
| "loss": 0.4604, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.13673366056605873, | |
| "learning_rate": 8.809818081980953e-05, | |
| "loss": 0.49, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.12607483394599764, | |
| "learning_rate": 8.612644680109319e-05, | |
| "loss": 0.4774, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.1365629261848207, | |
| "learning_rate": 8.416019755927851e-05, | |
| "loss": 0.4827, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.12122559291940836, | |
| "learning_rate": 8.2200210432245e-05, | |
| "loss": 0.5044, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.11655390642565265, | |
| "learning_rate": 8.024726028220474e-05, | |
| "loss": 0.503, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.12394574502796742, | |
| "learning_rate": 7.83021191893682e-05, | |
| "loss": 0.491, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.14922841699852962, | |
| "learning_rate": 7.636555614670953e-05, | |
| "loss": 0.457, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.11076924096187928, | |
| "learning_rate": 7.443833675595255e-05, | |
| "loss": 0.4603, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.1263594611752413, | |
| "learning_rate": 7.252122292489747e-05, | |
| "loss": 0.4859, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.11432575178505003, | |
| "learning_rate": 7.061497256620793e-05, | |
| "loss": 0.4627, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.1078119391965793, | |
| "learning_rate": 6.87203392977773e-05, | |
| "loss": 0.4829, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.12752089816514908, | |
| "learning_rate": 6.683807214479323e-05, | |
| "loss": 0.46, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.11421623043902956, | |
| "learning_rate": 6.496891524361757e-05, | |
| "loss": 0.4429, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.10432253193399477, | |
| "learning_rate": 6.311360754759923e-05, | |
| "loss": 0.402, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.12155248673662734, | |
| "learning_rate": 6.127288253493591e-05, | |
| "loss": 0.5126, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.14266947863559803, | |
| "learning_rate": 5.9447467918700614e-05, | |
| "loss": 0.4821, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.14851250761112514, | |
| "learning_rate": 5.763808535914723e-05, | |
| "loss": 0.4891, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.14264023747361737, | |
| "learning_rate": 5.584545017840885e-05, | |
| "loss": 0.5181, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.12837168363458795, | |
| "learning_rate": 5.407027107770219e-05, | |
| "loss": 0.5599, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.11874709251257598, | |
| "learning_rate": 5.2313249857149414e-05, | |
| "loss": 0.4536, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.12010754957532713, | |
| "learning_rate": 5.0575081138327715e-05, | |
| "loss": 0.5004, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.13464124440677885, | |
| "learning_rate": 4.885645208965779e-05, | |
| "loss": 0.4985, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.13701854261941088, | |
| "learning_rate": 4.715804215473809e-05, | |
| "loss": 0.4709, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.1335483738873249, | |
| "learning_rate": 4.548052278373327e-05, | |
| "loss": 0.4735, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.13603172024059101, | |
| "learning_rate": 4.382455716792291e-05, | |
| "loss": 0.4721, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.13843339239058639, | |
| "learning_rate": 4.219079997751515e-05, | |
| "loss": 0.4954, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.15011169526780793, | |
| "learning_rate": 4.0579897102828966e-05, | |
| "loss": 0.4648, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.13061595453081623, | |
| "learning_rate": 3.899248539894757e-05, | |
| "loss": 0.4801, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.14067787924603412, | |
| "learning_rate": 3.7429192433944014e-05, | |
| "loss": 0.4805, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.13420057703295998, | |
| "learning_rate": 3.589063624077802e-05, | |
| "loss": 0.4446, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.14083737654873127, | |
| "learning_rate": 3.4377425072962465e-05, | |
| "loss": 0.46, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.13231889777376862, | |
| "learning_rate": 3.289015716409631e-05, | |
| "loss": 0.4451, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.9651579260826111, | |
| "eval_runtime": 155.5959, | |
| "eval_samples_per_second": 1.6, | |
| "eval_steps_per_second": 0.803, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.1119782513388846, | |
| "learning_rate": 3.14294204913587e-05, | |
| "loss": 0.4728, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.1258610262132018, | |
| "learning_rate": 2.9995792543057478e-05, | |
| "loss": 0.4793, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.1332956040603169, | |
| "learning_rate": 2.8589840090325027e-05, | |
| "loss": 0.4919, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.13669237836174272, | |
| "learning_rate": 2.7212118963050592e-05, | |
| "loss": 0.4863, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.13718073990096244, | |
| "learning_rate": 2.586317383013821e-05, | |
| "loss": 0.4652, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.1328715725965393, | |
| "learning_rate": 2.4543537984176978e-05, | |
| "loss": 0.4464, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.12474829320490133, | |
| "learning_rate": 2.325373313060919e-05, | |
| "loss": 0.501, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.13832785688421897, | |
| "learning_rate": 2.19942691814788e-05, | |
| "loss": 0.5113, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.13446374612957607, | |
| "learning_rate": 2.076564405384258e-05, | |
| "loss": 0.452, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.15087962227561943, | |
| "learning_rate": 1.9568343472923524e-05, | |
| "loss": 0.5114, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.13471199433579678, | |
| "learning_rate": 1.840284078008393e-05, | |
| "loss": 0.5074, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.14471602871304384, | |
| "learning_rate": 1.7269596745694295e-05, | |
| "loss": 0.4812, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.12791095480012135, | |
| "learning_rate": 1.616905938697234e-05, | |
| "loss": 0.439, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.13395748535894433, | |
| "learning_rate": 1.5101663790863596e-05, | |
| "loss": 0.4942, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.1216157843375959, | |
| "learning_rate": 1.4067831942033904e-05, | |
| "loss": 0.4904, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.12513029490165878, | |
| "learning_rate": 1.3067972556041752e-05, | |
| "loss": 0.4583, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.11835922329329535, | |
| "learning_rate": 1.210248091775663e-05, | |
| "loss": 0.5281, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.14513302866752326, | |
| "learning_rate": 1.1171738725086833e-05, | |
| "loss": 0.4432, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.11934584390756656, | |
| "learning_rate": 1.0276113938078769e-05, | |
| "loss": 0.4664, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.11746932949614314, | |
| "learning_rate": 9.415960633447674e-06, | |
| "loss": 0.4452, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.1183312470782195, | |
| "learning_rate": 8.59161886459654e-06, | |
| "loss": 0.4858, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.1226342260426643, | |
| "learning_rate": 7.803414527179343e-06, | |
| "loss": 0.4609, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.12849903086517628, | |
| "learning_rate": 7.051659230261298e-06, | |
| "loss": 0.4992, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.12558798780888547, | |
| "learning_rate": 6.336650173127223e-06, | |
| "loss": 0.5055, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.11991457404362381, | |
| "learning_rate": 5.658670027786561e-06, | |
| "loss": 0.4711, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.14486189343828018, | |
| "learning_rate": 5.017986827221733e-06, | |
| "loss": 0.4984, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.14292272311516058, | |
| "learning_rate": 4.4148538594239174e-06, | |
| "loss": 0.518, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.12669351883956026, | |
| "learning_rate": 3.849509567257959e-06, | |
| "loss": 0.509, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.14745027617256945, | |
| "learning_rate": 3.3221774541962845e-06, | |
| "loss": 0.4655, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.13904171638337365, | |
| "learning_rate": 2.8330659959589946e-06, | |
| "loss": 0.445, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.12971202836694892, | |
| "learning_rate": 2.3823685580949273e-06, | |
| "loss": 0.4534, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.13437185949159883, | |
| "learning_rate": 1.9702633195363917e-06, | |
| "loss": 0.4785, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.12319276879857373, | |
| "learning_rate": 1.5969132021579347e-06, | |
| "loss": 0.5011, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.12918661435233847, | |
| "learning_rate": 1.2624658063666639e-06, | |
| "loss": 0.5236, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.12874241011432666, | |
| "learning_rate": 9.670533527498137e-07, | |
| "loss": 0.4439, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.1315361924610259, | |
| "learning_rate": 7.10792629802659e-07, | |
| "loss": 0.4929, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.1336404295007477, | |
| "learning_rate": 4.937849477572587e-07, | |
| "loss": 0.433, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.11865993758025514, | |
| "learning_rate": 3.161160985304168e-07, | |
| "loss": 0.4266, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.12559739797694514, | |
| "learning_rate": 1.7785632180670198e-07, | |
| "loss": 0.436, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.12409278477769135, | |
| "learning_rate": 7.906027726981568e-08, | |
| "loss": 0.5017, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.1398337808177678, | |
| "learning_rate": 1.976702299344435e-08, | |
| "loss": 0.5112, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.11434132222175684, | |
| "learning_rate": 0.0, | |
| "loss": 0.4616, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.9651336073875427, | |
| "eval_runtime": 154.7455, | |
| "eval_samples_per_second": 1.609, | |
| "eval_steps_per_second": 0.808, | |
| "step": 168 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 168, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 42, | |
| "total_flos": 3.406567776272253e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |