| { | |
| "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": 0.29289755909939047, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6274, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 1.0297596454620361, | |
| "eval_runtime": 153.0715, | |
| "eval_samples_per_second": 1.627, | |
| "eval_steps_per_second": 0.817, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.27956410941469395, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6362, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.3009590515152092, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6299, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.34354546270515235, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6395, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.2470961998205002, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6068, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.2071993792912813, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5993, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.20237019487582247, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6293, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.13810925455451734, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5101, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.22762469698117493, | |
| "learning_rate": 0.00018, | |
| "loss": 0.5527, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.15718586910417978, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5437, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.13825435646308676, | |
| "learning_rate": 0.00019998023297700658, | |
| "loss": 0.5858, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.15072585803075744, | |
| "learning_rate": 0.00019992093972273018, | |
| "loss": 0.554, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.13616736806741778, | |
| "learning_rate": 0.00019982214367819328, | |
| "loss": 0.6199, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.15841398198140327, | |
| "learning_rate": 0.0001996838839014696, | |
| "loss": 0.5495, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.16548492090565414, | |
| "learning_rate": 0.00019950621505224273, | |
| "loss": 0.5043, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.16965923620151113, | |
| "learning_rate": 0.00019928920737019733, | |
| "loss": 0.5083, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.1853247768877015, | |
| "learning_rate": 0.0001990329466472502, | |
| "loss": 0.6318, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.1824469971767483, | |
| "learning_rate": 0.00019873753419363336, | |
| "loss": 0.4809, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.13288488939703025, | |
| "learning_rate": 0.00019840308679784207, | |
| "loss": 0.4974, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.18507602579352214, | |
| "learning_rate": 0.00019802973668046363, | |
| "loss": 0.5288, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.13591012925799634, | |
| "learning_rate": 0.0001976176314419051, | |
| "loss": 0.5291, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.13704233024715257, | |
| "learning_rate": 0.000197166934004041, | |
| "loss": 0.4819, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.15362008062157553, | |
| "learning_rate": 0.00019667782254580374, | |
| "loss": 0.5408, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.11171032733778614, | |
| "learning_rate": 0.00019615049043274205, | |
| "loss": 0.5101, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.11449051252849428, | |
| "learning_rate": 0.00019558514614057609, | |
| "loss": 0.5209, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.11611407915744347, | |
| "learning_rate": 0.00019498201317277828, | |
| "loss": 0.5005, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.12735633641627706, | |
| "learning_rate": 0.00019434132997221345, | |
| "loss": 0.4741, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.11871518327376328, | |
| "learning_rate": 0.0001936633498268728, | |
| "loss": 0.5213, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.11403376465541806, | |
| "learning_rate": 0.0001929483407697387, | |
| "loss": 0.4842, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.11675155934307391, | |
| "learning_rate": 0.00019219658547282067, | |
| "loss": 0.4825, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.10789338581384152, | |
| "learning_rate": 0.00019140838113540346, | |
| "loss": 0.4866, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.14336837303756964, | |
| "learning_rate": 0.00019058403936655233, | |
| "loss": 0.5325, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.10401694793599091, | |
| "learning_rate": 0.00018972388606192125, | |
| "loss": 0.4292, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.10614245315138679, | |
| "learning_rate": 0.0001888282612749132, | |
| "loss": 0.4638, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.1250509143492961, | |
| "learning_rate": 0.00018789751908224338, | |
| "loss": 0.4963, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.12668831423294083, | |
| "learning_rate": 0.00018693202744395827, | |
| "loss": 0.5043, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.11594924793976216, | |
| "learning_rate": 0.00018593216805796612, | |
| "loss": 0.5396, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.12365894749489846, | |
| "learning_rate": 0.00018489833620913642, | |
| "loss": 0.4899, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.13164318970805183, | |
| "learning_rate": 0.00018383094061302766, | |
| "loss": 0.5065, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.10587870431925453, | |
| "learning_rate": 0.00018273040325430574, | |
| "loss": 0.4805, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.12134503625595237, | |
| "learning_rate": 0.00018159715921991612, | |
| "loss": 0.5103, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.10855490758268896, | |
| "learning_rate": 0.00018043165652707649, | |
| "loss": 0.44, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.9770342707633972, | |
| "eval_runtime": 154.3933, | |
| "eval_samples_per_second": 1.613, | |
| "eval_steps_per_second": 0.81, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.11753509152196469, | |
| "learning_rate": 0.00017923435594615744, | |
| "loss": 0.4819, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.13774603727779988, | |
| "learning_rate": 0.00017800573081852122, | |
| "loss": 0.5451, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.11994636996852912, | |
| "learning_rate": 0.0001767462668693908, | |
| "loss": 0.5079, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.11803018063108017, | |
| "learning_rate": 0.00017545646201582303, | |
| "loss": 0.5183, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.12122026879022209, | |
| "learning_rate": 0.00017413682616986185, | |
| "loss": 0.4692, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.12840154129375927, | |
| "learning_rate": 0.00017278788103694943, | |
| "loss": 0.4538, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.10754191454075242, | |
| "learning_rate": 0.000171410159909675, | |
| "loss": 0.4745, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.10980392972154758, | |
| "learning_rate": 0.00017000420745694254, | |
| "loss": 0.5077, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.1018314102740997, | |
| "learning_rate": 0.00016857057950864132, | |
| "loss": 0.5077, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.10572512690181787, | |
| "learning_rate": 0.0001671098428359037, | |
| "loss": 0.4637, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.13106813432864392, | |
| "learning_rate": 0.00016562257492703757, | |
| "loss": 0.4718, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.11024484284605006, | |
| "learning_rate": 0.000164109363759222, | |
| "loss": 0.5115, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.12388815222110366, | |
| "learning_rate": 0.000162570807566056, | |
| "loss": 0.4756, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.12702188830349206, | |
| "learning_rate": 0.00016100751460105243, | |
| "loss": 0.4881, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.11086308659932648, | |
| "learning_rate": 0.00015942010289717105, | |
| "loss": 0.4701, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.11517890306226485, | |
| "learning_rate": 0.00015780920002248484, | |
| "loss": 0.4835, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.1253541013817757, | |
| "learning_rate": 0.0001561754428320771, | |
| "loss": 0.5147, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.11139862438389284, | |
| "learning_rate": 0.00015451947721626676, | |
| "loss": 0.4552, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.1152466912880715, | |
| "learning_rate": 0.00015284195784526195, | |
| "loss": 0.5053, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.11719477948785388, | |
| "learning_rate": 0.00015114354791034225, | |
| "loss": 0.5079, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.12730957754598396, | |
| "learning_rate": 0.0001494249188616723, | |
| "loss": 0.58, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.11907511614241814, | |
| "learning_rate": 0.00014768675014285062, | |
| "loss": 0.4579, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.13613222195950112, | |
| "learning_rate": 0.00014592972892229778, | |
| "loss": 0.4965, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.11610538032286319, | |
| "learning_rate": 0.0001441545498215912, | |
| "loss": 0.4467, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.1046724680527729, | |
| "learning_rate": 0.00014236191464085286, | |
| "loss": 0.4463, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.11631551012538931, | |
| "learning_rate": 0.00014055253208129938, | |
| "loss": 0.5304, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.11792788666231199, | |
| "learning_rate": 0.00013872711746506413, | |
| "loss": 0.453, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.11566019183159004, | |
| "learning_rate": 0.00013688639245240078, | |
| "loss": 0.5192, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.12967661214418025, | |
| "learning_rate": 0.00013503108475638244, | |
| "loss": 0.5625, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.10928605499181634, | |
| "learning_rate": 0.0001331619278552068, | |
| "loss": 0.4861, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.12485741141890881, | |
| "learning_rate": 0.00013127966070222274, | |
| "loss": 0.4782, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.1271574869759, | |
| "learning_rate": 0.00012938502743379212, | |
| "loss": 0.4819, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.14528941719728583, | |
| "learning_rate": 0.00012747877707510252, | |
| "loss": 0.5132, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.12760881412243183, | |
| "learning_rate": 0.0001255616632440475, | |
| "loss": 0.4817, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.1239773484446177, | |
| "learning_rate": 0.0001236344438532905, | |
| "loss": 0.5045, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.1307497276864945, | |
| "learning_rate": 0.0001216978808106318, | |
| "loss": 0.5091, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.1355749101256534, | |
| "learning_rate": 0.00011975273971779528, | |
| "loss": 0.5145, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.12394948316350356, | |
| "learning_rate": 0.00011779978956775506, | |
| "loss": 0.5057, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.13424789799433426, | |
| "learning_rate": 0.0001158398024407215, | |
| "loss": 0.5058, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.13054427599188898, | |
| "learning_rate": 0.00011387355319890685, | |
| "loss": 0.4683, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.11996704114604598, | |
| "learning_rate": 0.00011190181918019049, | |
| "loss": 0.4748, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.13071208464837492, | |
| "learning_rate": 0.00010992537989080618, | |
| "loss": 0.4406, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.9652944803237915, | |
| "eval_runtime": 153.5377, | |
| "eval_samples_per_second": 1.622, | |
| "eval_steps_per_second": 0.814, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.11393059638442392, | |
| "learning_rate": 0.00010794501669717145, | |
| "loss": 0.4877, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.11364636282385192, | |
| "learning_rate": 0.00010596151251698199, | |
| "loss": 0.4597, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.12548992883216656, | |
| "learning_rate": 0.0001039756515096926, | |
| "loss": 0.4685, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.12825123946992636, | |
| "learning_rate": 0.00010198821876650701, | |
| "loss": 0.4924, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.12264617890724591, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4678, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.1238990144553216, | |
| "learning_rate": 9.801178123349298e-05, | |
| "loss": 0.4854, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.13533454171482565, | |
| "learning_rate": 9.602434849030745e-05, | |
| "loss": 0.4784, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.13623990945679543, | |
| "learning_rate": 9.403848748301802e-05, | |
| "loss": 0.5322, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.16411366119133766, | |
| "learning_rate": 9.205498330282856e-05, | |
| "loss": 0.5258, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.11646811294381437, | |
| "learning_rate": 9.007462010919386e-05, | |
| "loss": 0.4599, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.12610649830308363, | |
| "learning_rate": 8.809818081980953e-05, | |
| "loss": 0.4891, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.12324596375061997, | |
| "learning_rate": 8.612644680109319e-05, | |
| "loss": 0.4771, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.12390720748290898, | |
| "learning_rate": 8.416019755927851e-05, | |
| "loss": 0.4814, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.11156155557793686, | |
| "learning_rate": 8.2200210432245e-05, | |
| "loss": 0.5041, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.11235071151397323, | |
| "learning_rate": 8.024726028220474e-05, | |
| "loss": 0.5023, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.12382934807374943, | |
| "learning_rate": 7.83021191893682e-05, | |
| "loss": 0.4917, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.13051565063971357, | |
| "learning_rate": 7.636555614670953e-05, | |
| "loss": 0.4567, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.11975246976802223, | |
| "learning_rate": 7.443833675595255e-05, | |
| "loss": 0.4584, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.1390325061190883, | |
| "learning_rate": 7.252122292489747e-05, | |
| "loss": 0.4853, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.13092284300351015, | |
| "learning_rate": 7.061497256620793e-05, | |
| "loss": 0.4623, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.1291139691922856, | |
| "learning_rate": 6.87203392977773e-05, | |
| "loss": 0.4826, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.12630345566649634, | |
| "learning_rate": 6.683807214479323e-05, | |
| "loss": 0.4597, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.13061642512554117, | |
| "learning_rate": 6.496891524361757e-05, | |
| "loss": 0.4415, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.11414947835097713, | |
| "learning_rate": 6.311360754759923e-05, | |
| "loss": 0.4011, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.11453014913188615, | |
| "learning_rate": 6.127288253493591e-05, | |
| "loss": 0.5118, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.15000233747727326, | |
| "learning_rate": 5.9447467918700614e-05, | |
| "loss": 0.482, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.14402898946913723, | |
| "learning_rate": 5.763808535914723e-05, | |
| "loss": 0.4881, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.1371953423024685, | |
| "learning_rate": 5.584545017840885e-05, | |
| "loss": 0.5178, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.14105581149485616, | |
| "learning_rate": 5.407027107770219e-05, | |
| "loss": 0.5584, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.13646035299068404, | |
| "learning_rate": 5.2313249857149414e-05, | |
| "loss": 0.4535, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.14741390311838884, | |
| "learning_rate": 5.0575081138327715e-05, | |
| "loss": 0.5, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.12313483515418376, | |
| "learning_rate": 4.885645208965779e-05, | |
| "loss": 0.4982, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.13192008123654636, | |
| "learning_rate": 4.715804215473809e-05, | |
| "loss": 0.4698, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.11197839688620534, | |
| "learning_rate": 4.548052278373327e-05, | |
| "loss": 0.4737, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.12574619847347493, | |
| "learning_rate": 4.382455716792291e-05, | |
| "loss": 0.472, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.11785384512597769, | |
| "learning_rate": 4.219079997751515e-05, | |
| "loss": 0.4954, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1371137498771671, | |
| "learning_rate": 4.0579897102828966e-05, | |
| "loss": 0.4645, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.1562115085397725, | |
| "learning_rate": 3.899248539894757e-05, | |
| "loss": 0.4798, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.14251919403809987, | |
| "learning_rate": 3.7429192433944014e-05, | |
| "loss": 0.4794, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.13567896269328303, | |
| "learning_rate": 3.589063624077802e-05, | |
| "loss": 0.4441, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.14154096161651117, | |
| "learning_rate": 3.4377425072962465e-05, | |
| "loss": 0.4583, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.12685010104316322, | |
| "learning_rate": 3.289015716409631e-05, | |
| "loss": 0.4445, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.9644750952720642, | |
| "eval_runtime": 155.1633, | |
| "eval_samples_per_second": 1.605, | |
| "eval_steps_per_second": 0.806, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.127123263865781, | |
| "learning_rate": 3.14294204913587e-05, | |
| "loss": 0.4723, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.1220769041189061, | |
| "learning_rate": 2.9995792543057478e-05, | |
| "loss": 0.478, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.11180771873074949, | |
| "learning_rate": 2.8589840090325027e-05, | |
| "loss": 0.4909, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.13041544813383119, | |
| "learning_rate": 2.7212118963050592e-05, | |
| "loss": 0.4866, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.1405254912635375, | |
| "learning_rate": 2.586317383013821e-05, | |
| "loss": 0.4644, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.1344866421919383, | |
| "learning_rate": 2.4543537984176978e-05, | |
| "loss": 0.4456, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.11186073290835527, | |
| "learning_rate": 2.325373313060919e-05, | |
| "loss": 0.4994, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.11996543804410408, | |
| "learning_rate": 2.19942691814788e-05, | |
| "loss": 0.5109, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.1213151845276305, | |
| "learning_rate": 2.076564405384258e-05, | |
| "loss": 0.4519, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.13448384445509487, | |
| "learning_rate": 1.9568343472923524e-05, | |
| "loss": 0.5102, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.12446550745886309, | |
| "learning_rate": 1.840284078008393e-05, | |
| "loss": 0.5064, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.14018404224721664, | |
| "learning_rate": 1.7269596745694295e-05, | |
| "loss": 0.4805, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.1218286028602253, | |
| "learning_rate": 1.616905938697234e-05, | |
| "loss": 0.4388, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.1293899895212843, | |
| "learning_rate": 1.5101663790863596e-05, | |
| "loss": 0.4935, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.1338746361041889, | |
| "learning_rate": 1.4067831942033904e-05, | |
| "loss": 0.4892, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.12837018186336283, | |
| "learning_rate": 1.3067972556041752e-05, | |
| "loss": 0.4566, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.1359601843202103, | |
| "learning_rate": 1.210248091775663e-05, | |
| "loss": 0.5277, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.14494892289523706, | |
| "learning_rate": 1.1171738725086833e-05, | |
| "loss": 0.4419, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.13277965728610747, | |
| "learning_rate": 1.0276113938078769e-05, | |
| "loss": 0.4662, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.12880953859990588, | |
| "learning_rate": 9.415960633447674e-06, | |
| "loss": 0.4452, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.13140413424345554, | |
| "learning_rate": 8.59161886459654e-06, | |
| "loss": 0.485, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.1319983788489183, | |
| "learning_rate": 7.803414527179343e-06, | |
| "loss": 0.4601, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.12708833536742306, | |
| "learning_rate": 7.051659230261298e-06, | |
| "loss": 0.498, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.13606238849102634, | |
| "learning_rate": 6.336650173127223e-06, | |
| "loss": 0.504, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1295595089005434, | |
| "learning_rate": 5.658670027786561e-06, | |
| "loss": 0.4704, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1426317101207339, | |
| "learning_rate": 5.017986827221733e-06, | |
| "loss": 0.4979, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.14306220741750347, | |
| "learning_rate": 4.4148538594239174e-06, | |
| "loss": 0.5164, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.11966060296362775, | |
| "learning_rate": 3.849509567257959e-06, | |
| "loss": 0.508, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.13410648920269255, | |
| "learning_rate": 3.3221774541962845e-06, | |
| "loss": 0.4647, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.14279456888939884, | |
| "learning_rate": 2.8330659959589946e-06, | |
| "loss": 0.4446, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.12867690404851015, | |
| "learning_rate": 2.3823685580949273e-06, | |
| "loss": 0.453, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.12934543469906262, | |
| "learning_rate": 1.9702633195363917e-06, | |
| "loss": 0.4782, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.12938497975922164, | |
| "learning_rate": 1.5969132021579347e-06, | |
| "loss": 0.5004, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.12309207540187947, | |
| "learning_rate": 1.2624658063666639e-06, | |
| "loss": 0.5235, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.11548916080113472, | |
| "learning_rate": 9.670533527498137e-07, | |
| "loss": 0.4436, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.12421128547481033, | |
| "learning_rate": 7.10792629802659e-07, | |
| "loss": 0.493, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.12845932182275463, | |
| "learning_rate": 4.937849477572587e-07, | |
| "loss": 0.4322, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.12044022703621939, | |
| "learning_rate": 3.161160985304168e-07, | |
| "loss": 0.4261, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.12854436448690823, | |
| "learning_rate": 1.7785632180670198e-07, | |
| "loss": 0.436, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.12592400334168094, | |
| "learning_rate": 7.906027726981568e-08, | |
| "loss": 0.5008, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.13604136642444342, | |
| "learning_rate": 1.976702299344435e-08, | |
| "loss": 0.5117, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.12600479253243235, | |
| "learning_rate": 0.0, | |
| "loss": 0.4609, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.9640664458274841, | |
| "eval_runtime": 154.1049, | |
| "eval_samples_per_second": 1.616, | |
| "eval_steps_per_second": 0.811, | |
| "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 | |
| } | |