| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 30, | |
| "global_step": 1248, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01603849238171612, | |
| "grad_norm": 9.394665718078613, | |
| "learning_rate": 1.44e-05, | |
| "loss": 2.4438, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03207698476343224, | |
| "grad_norm": 0.7643827199935913, | |
| "learning_rate": 3.04e-05, | |
| "loss": 1.0643, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.048115477145148355, | |
| "grad_norm": 0.95888352394104, | |
| "learning_rate": 4.64e-05, | |
| "loss": 0.6802, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.048115477145148355, | |
| "eval_loss": 0.5278236865997314, | |
| "eval_runtime": 51.9659, | |
| "eval_samples_per_second": 5.061, | |
| "eval_steps_per_second": 2.54, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06415396952686447, | |
| "grad_norm": 0.8735707998275757, | |
| "learning_rate": 6.24e-05, | |
| "loss": 0.4503, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08019246190858059, | |
| "grad_norm": 0.648636519908905, | |
| "learning_rate": 7.840000000000001e-05, | |
| "loss": 0.3343, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09623095429029671, | |
| "grad_norm": 0.5771777629852295, | |
| "learning_rate": 9.44e-05, | |
| "loss": 0.2846, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09623095429029671, | |
| "eval_loss": 0.2555956244468689, | |
| "eval_runtime": 51.322, | |
| "eval_samples_per_second": 5.125, | |
| "eval_steps_per_second": 2.572, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11226944667201283, | |
| "grad_norm": 0.5190573930740356, | |
| "learning_rate": 0.00011040000000000001, | |
| "loss": 0.2612, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12830793905372895, | |
| "grad_norm": 0.5709512829780579, | |
| "learning_rate": 0.0001264, | |
| "loss": 0.2053, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14434643143544507, | |
| "grad_norm": 0.5518093705177307, | |
| "learning_rate": 0.0001424, | |
| "loss": 0.1978, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14434643143544507, | |
| "eval_loss": 0.18320685625076294, | |
| "eval_runtime": 51.4345, | |
| "eval_samples_per_second": 5.113, | |
| "eval_steps_per_second": 2.566, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16038492381716118, | |
| "grad_norm": 1.0181713104248047, | |
| "learning_rate": 0.00015840000000000003, | |
| "loss": 0.196, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1764234161988773, | |
| "grad_norm": 0.46093398332595825, | |
| "learning_rate": 0.0001744, | |
| "loss": 0.1986, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19246190858059342, | |
| "grad_norm": 0.4132508635520935, | |
| "learning_rate": 0.0001904, | |
| "loss": 0.1498, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.19246190858059342, | |
| "eval_loss": 0.15674413740634918, | |
| "eval_runtime": 51.4011, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 2.568, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.20850040096230954, | |
| "grad_norm": 0.8875346779823303, | |
| "learning_rate": 0.00019999373925971706, | |
| "loss": 0.1678, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.22453889334402566, | |
| "grad_norm": 0.42169156670570374, | |
| "learning_rate": 0.0001999233149342545, | |
| "loss": 0.1747, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24057738572574178, | |
| "grad_norm": 0.496420681476593, | |
| "learning_rate": 0.0001997746956522446, | |
| "loss": 0.1556, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24057738572574178, | |
| "eval_loss": 0.14000150561332703, | |
| "eval_runtime": 51.4316, | |
| "eval_samples_per_second": 5.114, | |
| "eval_steps_per_second": 2.567, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2566158781074579, | |
| "grad_norm": 0.3014488220214844, | |
| "learning_rate": 0.0001995479977156566, | |
| "loss": 0.1432, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.272654370489174, | |
| "grad_norm": 0.2620065212249756, | |
| "learning_rate": 0.00019924339852688463, | |
| "loss": 0.1459, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.28869286287089013, | |
| "grad_norm": 0.16078414022922516, | |
| "learning_rate": 0.00019886113644992158, | |
| "loss": 0.1305, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.28869286287089013, | |
| "eval_loss": 0.13560178875923157, | |
| "eval_runtime": 51.4466, | |
| "eval_samples_per_second": 5.112, | |
| "eval_steps_per_second": 2.566, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.30473135525260625, | |
| "grad_norm": 0.43932297825813293, | |
| "learning_rate": 0.00019840151062382741, | |
| "loss": 0.1319, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.32076984763432237, | |
| "grad_norm": 0.2900855541229248, | |
| "learning_rate": 0.000197864880728638, | |
| "loss": 0.1385, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3368083400160385, | |
| "grad_norm": 0.217300146818161, | |
| "learning_rate": 0.00019725166670389764, | |
| "loss": 0.1378, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3368083400160385, | |
| "eval_loss": 0.12454783171415329, | |
| "eval_runtime": 51.3956, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 2.568, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3528468323977546, | |
| "grad_norm": 0.4188595712184906, | |
| "learning_rate": 0.00019656234842003545, | |
| "loss": 0.1385, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3688853247794707, | |
| "grad_norm": 0.2888878285884857, | |
| "learning_rate": 0.00019579746530284296, | |
| "loss": 0.1325, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.38492381716118684, | |
| "grad_norm": 0.34402385354042053, | |
| "learning_rate": 0.00019495761591134677, | |
| "loss": 0.1317, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.38492381716118684, | |
| "eval_loss": 0.13386917114257812, | |
| "eval_runtime": 51.6803, | |
| "eval_samples_per_second": 5.089, | |
| "eval_steps_per_second": 2.554, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.40096230954290296, | |
| "grad_norm": 1.1451412439346313, | |
| "learning_rate": 0.00019404345746940608, | |
| "loss": 0.3843, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4170008019246191, | |
| "grad_norm": 0.2417556643486023, | |
| "learning_rate": 0.00019305570535140272, | |
| "loss": 0.1433, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4330392943063352, | |
| "grad_norm": 1.9079967737197876, | |
| "learning_rate": 0.0001919951325224249, | |
| "loss": 0.1645, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4330392943063352, | |
| "eval_loss": 0.13291145861148834, | |
| "eval_runtime": 51.4824, | |
| "eval_samples_per_second": 5.109, | |
| "eval_steps_per_second": 2.564, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4490777866880513, | |
| "grad_norm": 0.18921196460723877, | |
| "learning_rate": 0.0001908625689333838, | |
| "loss": 0.1315, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.46511627906976744, | |
| "grad_norm": 0.19202928245067596, | |
| "learning_rate": 0.00018965890087153575, | |
| "loss": 0.1368, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.48115477145148355, | |
| "grad_norm": 0.22364425659179688, | |
| "learning_rate": 0.00018838507026691832, | |
| "loss": 0.12, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.48115477145148355, | |
| "eval_loss": 0.119777150452137, | |
| "eval_runtime": 51.3871, | |
| "eval_samples_per_second": 5.118, | |
| "eval_steps_per_second": 2.569, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4971932638331997, | |
| "grad_norm": 0.85606849193573, | |
| "learning_rate": 0.00018704207395524338, | |
| "loss": 0.1343, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5132317562149158, | |
| "grad_norm": 0.22195890545845032, | |
| "learning_rate": 0.00018563096289782367, | |
| "loss": 0.1306, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.529270248596632, | |
| "grad_norm": 0.14963646233081818, | |
| "learning_rate": 0.00018415284135914333, | |
| "loss": 0.1105, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.529270248596632, | |
| "eval_loss": 0.11582240462303162, | |
| "eval_runtime": 51.3905, | |
| "eval_samples_per_second": 5.118, | |
| "eval_steps_per_second": 2.569, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.545308740978348, | |
| "grad_norm": 0.18346640467643738, | |
| "learning_rate": 0.0001826088660427161, | |
| "loss": 0.1211, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5613472333600642, | |
| "grad_norm": 0.11578662693500519, | |
| "learning_rate": 0.00018100024518590737, | |
| "loss": 0.1155, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5773857257417803, | |
| "grad_norm": 0.26045721769332886, | |
| "learning_rate": 0.00017932823761442843, | |
| "loss": 0.1244, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5773857257417803, | |
| "eval_loss": 0.11565632373094559, | |
| "eval_runtime": 51.4124, | |
| "eval_samples_per_second": 5.115, | |
| "eval_steps_per_second": 2.567, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5934242181234964, | |
| "grad_norm": 0.1874534785747528, | |
| "learning_rate": 0.00017759415175724294, | |
| "loss": 0.1143, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6094627105052125, | |
| "grad_norm": 0.2457660734653473, | |
| "learning_rate": 0.00017579934462265607, | |
| "loss": 0.1218, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6255012028869287, | |
| "grad_norm": 0.18305718898773193, | |
| "learning_rate": 0.00017394522073638832, | |
| "loss": 0.1246, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6255012028869287, | |
| "eval_loss": 0.11496119201183319, | |
| "eval_runtime": 51.454, | |
| "eval_samples_per_second": 5.111, | |
| "eval_steps_per_second": 2.565, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6415396952686447, | |
| "grad_norm": 0.07789923995733261, | |
| "learning_rate": 0.0001720332310424643, | |
| "loss": 0.1121, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6575781876503609, | |
| "grad_norm": 0.28116437792778015, | |
| "learning_rate": 0.00017006487176777685, | |
| "loss": 0.1226, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.673616680032077, | |
| "grad_norm": 0.3152724802494049, | |
| "learning_rate": 0.0001680416832512155, | |
| "loss": 0.1183, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.673616680032077, | |
| "eval_loss": 0.1126837432384491, | |
| "eval_runtime": 51.5959, | |
| "eval_samples_per_second": 5.097, | |
| "eval_steps_per_second": 2.558, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 0.14282774925231934, | |
| "learning_rate": 0.00016596524873827443, | |
| "loss": 0.104, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7056936647955092, | |
| "grad_norm": 0.11272551864385605, | |
| "learning_rate": 0.0001638371931420845, | |
| "loss": 0.1221, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.7217321571772254, | |
| "grad_norm": 0.18691876530647278, | |
| "learning_rate": 0.0001616591817718379, | |
| "loss": 0.1205, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7217321571772254, | |
| "eval_loss": 0.11374854296445847, | |
| "eval_runtime": 51.4403, | |
| "eval_samples_per_second": 5.113, | |
| "eval_steps_per_second": 2.566, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7377706495589414, | |
| "grad_norm": 0.10991021245718002, | |
| "learning_rate": 0.000159432919029601, | |
| "loss": 0.1075, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7538091419406576, | |
| "grad_norm": 0.368667870759964, | |
| "learning_rate": 0.0001571601470765352, | |
| "loss": 0.1139, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7698476343223737, | |
| "grad_norm": 0.10450360178947449, | |
| "learning_rate": 0.00015484264446956916, | |
| "loss": 0.1126, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7698476343223737, | |
| "eval_loss": 0.11422113329172134, | |
| "eval_runtime": 51.4861, | |
| "eval_samples_per_second": 5.108, | |
| "eval_steps_per_second": 2.564, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7858861267040899, | |
| "grad_norm": 0.11498294025659561, | |
| "learning_rate": 0.00015248222476958972, | |
| "loss": 0.116, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8019246190858059, | |
| "grad_norm": 0.14388644695281982, | |
| "learning_rate": 0.00015008073512224045, | |
| "loss": 0.1186, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8179631114675221, | |
| "grad_norm": 0.1251271665096283, | |
| "learning_rate": 0.00014764005481243854, | |
| "loss": 0.1176, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.8179631114675221, | |
| "eval_loss": 0.11191003024578094, | |
| "eval_runtime": 51.3165, | |
| "eval_samples_per_second": 5.125, | |
| "eval_steps_per_second": 2.572, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.8340016038492382, | |
| "grad_norm": 0.25354036688804626, | |
| "learning_rate": 0.00014516209379374076, | |
| "loss": 0.1219, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.8500400962309543, | |
| "grad_norm": 0.24119634926319122, | |
| "learning_rate": 0.0001426487911937102, | |
| "loss": 0.1222, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.8660785886126704, | |
| "grad_norm": 0.1247166320681572, | |
| "learning_rate": 0.00014010211379645256, | |
| "loss": 0.1164, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8660785886126704, | |
| "eval_loss": 0.11168860644102097, | |
| "eval_runtime": 51.5163, | |
| "eval_samples_per_second": 5.105, | |
| "eval_steps_per_second": 2.562, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8821170809943866, | |
| "grad_norm": 0.11632517725229263, | |
| "learning_rate": 0.0001375240545035099, | |
| "loss": 0.1016, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8981555733761026, | |
| "grad_norm": 0.11122917383909225, | |
| "learning_rate": 0.00013491663077431614, | |
| "loss": 0.1207, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9141940657578188, | |
| "grad_norm": 0.13945281505584717, | |
| "learning_rate": 0.00013228188304743505, | |
| "loss": 0.1169, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.9141940657578188, | |
| "eval_loss": 0.10964255779981613, | |
| "eval_runtime": 51.5075, | |
| "eval_samples_per_second": 5.106, | |
| "eval_steps_per_second": 2.563, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.9302325581395349, | |
| "grad_norm": 0.10248645395040512, | |
| "learning_rate": 0.00012962187314381548, | |
| "loss": 0.1032, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.946271050521251, | |
| "grad_norm": 0.10369589179754257, | |
| "learning_rate": 0.00012693868265331433, | |
| "loss": 0.1219, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.9623095429029671, | |
| "grad_norm": 0.09495269507169724, | |
| "learning_rate": 0.00012423441130574888, | |
| "loss": 0.1139, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9623095429029671, | |
| "eval_loss": 0.1103595569729805, | |
| "eval_runtime": 51.7198, | |
| "eval_samples_per_second": 5.085, | |
| "eval_steps_per_second": 2.552, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9783480352846833, | |
| "grad_norm": 0.10655952244997025, | |
| "learning_rate": 0.00012151117532775409, | |
| "loss": 0.1131, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.9943865276663993, | |
| "grad_norm": 0.07204799354076385, | |
| "learning_rate": 0.00011877110578673005, | |
| "loss": 0.1165, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.0096230954290297, | |
| "grad_norm": 0.1082656979560852, | |
| "learning_rate": 0.00011601634692317597, | |
| "loss": 0.1177, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.0096230954290297, | |
| "eval_loss": 0.10991596430540085, | |
| "eval_runtime": 51.588, | |
| "eval_samples_per_second": 5.098, | |
| "eval_steps_per_second": 2.559, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.0256615878107458, | |
| "grad_norm": 0.10015016794204712, | |
| "learning_rate": 0.0001132490544727155, | |
| "loss": 0.1221, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.0417000801924619, | |
| "grad_norm": 0.12009429931640625, | |
| "learning_rate": 0.00011047139397912647, | |
| "loss": 0.1096, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.057738572574178, | |
| "grad_norm": 0.09907736629247665, | |
| "learning_rate": 0.00010768553909969538, | |
| "loss": 0.1191, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.057738572574178, | |
| "eval_loss": 0.10953598469495773, | |
| "eval_runtime": 51.5236, | |
| "eval_samples_per_second": 5.104, | |
| "eval_steps_per_second": 2.562, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.0737770649558942, | |
| "grad_norm": 0.08365555107593536, | |
| "learning_rate": 0.00010489366990422264, | |
| "loss": 0.1113, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.0898155573376103, | |
| "grad_norm": 0.0640728697180748, | |
| "learning_rate": 0.00010209797116900968, | |
| "loss": 0.1085, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.1058540497193263, | |
| "grad_norm": 0.10259856283664703, | |
| "learning_rate": 9.93006306671629e-05, | |
| "loss": 0.1076, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.1058540497193263, | |
| "eval_loss": 0.10963255167007446, | |
| "eval_runtime": 51.5303, | |
| "eval_samples_per_second": 5.104, | |
| "eval_steps_per_second": 2.562, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.1218925421010426, | |
| "grad_norm": 0.0759560614824295, | |
| "learning_rate": 9.650383745655265e-05, | |
| "loss": 0.1097, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1379310344827587, | |
| "grad_norm": 0.08842267841100693, | |
| "learning_rate": 9.370978016676681e-05, | |
| "loss": 0.1247, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.1539695268644747, | |
| "grad_norm": 0.09709272533655167, | |
| "learning_rate": 9.092064528639916e-05, | |
| "loss": 0.1078, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.1539695268644747, | |
| "eval_loss": 0.10957112163305283, | |
| "eval_runtime": 51.5695, | |
| "eval_samples_per_second": 5.1, | |
| "eval_steps_per_second": 2.56, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.1700080192461908, | |
| "grad_norm": 0.11856944113969803, | |
| "learning_rate": 8.81386154520138e-05, | |
| "loss": 0.1113, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.1860465116279069, | |
| "grad_norm": 0.09656749665737152, | |
| "learning_rate": 8.536586774012336e-05, | |
| "loss": 0.114, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.2020850040096231, | |
| "grad_norm": 0.10089857876300812, | |
| "learning_rate": 8.260457196351871e-05, | |
| "loss": 0.1079, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2020850040096231, | |
| "eval_loss": 0.1081293597817421, | |
| "eval_runtime": 51.6243, | |
| "eval_samples_per_second": 5.094, | |
| "eval_steps_per_second": 2.557, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2181234963913392, | |
| "grad_norm": 0.07360345870256424, | |
| "learning_rate": 7.985688897328264e-05, | |
| "loss": 0.1044, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.2341619887730553, | |
| "grad_norm": 0.10135751962661743, | |
| "learning_rate": 7.712496896781637e-05, | |
| "loss": 0.1086, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.2502004811547716, | |
| "grad_norm": 0.07764733582735062, | |
| "learning_rate": 7.44109498102029e-05, | |
| "loss": 0.0998, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.2502004811547716, | |
| "eval_loss": 0.10887689888477325, | |
| "eval_runtime": 51.3975, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 2.568, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.2662389735364876, | |
| "grad_norm": 0.09288525581359863, | |
| "learning_rate": 7.171695535522259e-05, | |
| "loss": 0.1129, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.2822774659182037, | |
| "grad_norm": 0.0900450125336647, | |
| "learning_rate": 6.904509378733154e-05, | |
| "loss": 0.1132, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2983159582999197, | |
| "grad_norm": 0.08215230703353882, | |
| "learning_rate": 6.639745597090246e-05, | |
| "loss": 0.1078, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.2983159582999197, | |
| "eval_loss": 0.10840240865945816, | |
| "eval_runtime": 51.5192, | |
| "eval_samples_per_second": 5.105, | |
| "eval_steps_per_second": 2.562, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.3143544506816358, | |
| "grad_norm": 0.05950262397527695, | |
| "learning_rate": 6.377611381401958e-05, | |
| "loss": 0.1055, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.330392943063352, | |
| "grad_norm": 0.06437576562166214, | |
| "learning_rate": 6.118311864710748e-05, | |
| "loss": 0.1012, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.3464314354450682, | |
| "grad_norm": 0.10620200634002686, | |
| "learning_rate": 5.862049961766327e-05, | |
| "loss": 0.1177, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.3464314354450682, | |
| "eval_loss": 0.1079765185713768, | |
| "eval_runtime": 51.4348, | |
| "eval_samples_per_second": 5.113, | |
| "eval_steps_per_second": 2.566, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.3624699278267842, | |
| "grad_norm": 0.1097632423043251, | |
| "learning_rate": 5.609026210234799e-05, | |
| "loss": 0.0995, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.3785084202085005, | |
| "grad_norm": 0.09094393998384476, | |
| "learning_rate": 5.359438613767951e-05, | |
| "loss": 0.1024, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.3945469125902166, | |
| "grad_norm": 0.08209466189146042, | |
| "learning_rate": 5.1134824870555875e-05, | |
| "loss": 0.1081, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.3945469125902166, | |
| "eval_loss": 0.10702705383300781, | |
| "eval_runtime": 51.3972, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 2.568, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.4105854049719326, | |
| "grad_norm": 0.11662867665290833, | |
| "learning_rate": 4.871350302982077e-05, | |
| "loss": 0.1082, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.4266238973536487, | |
| "grad_norm": 0.08680973947048187, | |
| "learning_rate": 4.6332315420067604e-05, | |
| "loss": 0.1139, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.4426623897353648, | |
| "grad_norm": 0.1044652909040451, | |
| "learning_rate": 4.399312543886067e-05, | |
| "loss": 0.1106, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4426623897353648, | |
| "eval_loss": 0.10689909011125565, | |
| "eval_runtime": 51.5936, | |
| "eval_samples_per_second": 5.098, | |
| "eval_steps_per_second": 2.558, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.458700882117081, | |
| "grad_norm": 0.08980236202478409, | |
| "learning_rate": 4.169776361853437e-05, | |
| "loss": 0.1091, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.474739374498797, | |
| "grad_norm": 0.09599462896585464, | |
| "learning_rate": 3.944802619371041e-05, | |
| "loss": 0.1116, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.4907778668805132, | |
| "grad_norm": 0.0679604634642601, | |
| "learning_rate": 3.7245673695655015e-05, | |
| "loss": 0.1067, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.4907778668805132, | |
| "eval_loss": 0.10721538215875626, | |
| "eval_runtime": 51.5029, | |
| "eval_samples_per_second": 5.107, | |
| "eval_steps_per_second": 2.563, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.5068163592622295, | |
| "grad_norm": 0.07000099122524261, | |
| "learning_rate": 3.5092429574576115e-05, | |
| "loss": 0.0962, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.5228548516439455, | |
| "grad_norm": 0.0805450901389122, | |
| "learning_rate": 3.298997885093748e-05, | |
| "loss": 0.096, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.5388933440256616, | |
| "grad_norm": 0.08625531196594238, | |
| "learning_rate": 3.0939966796846996e-05, | |
| "loss": 0.1129, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.5388933440256616, | |
| "eval_loss": 0.10685242712497711, | |
| "eval_runtime": 51.6956, | |
| "eval_samples_per_second": 5.087, | |
| "eval_steps_per_second": 2.553, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.5549318364073779, | |
| "grad_norm": 0.09168870002031326, | |
| "learning_rate": 2.8943997648549104e-05, | |
| "loss": 0.0998, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.5709703287890937, | |
| "grad_norm": 0.07784843444824219, | |
| "learning_rate": 2.7003633351030566e-05, | |
| "loss": 0.1096, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.58700882117081, | |
| "grad_norm": 0.09954196214675903, | |
| "learning_rate": 2.5120392335720898e-05, | |
| "loss": 0.1137, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.58700882117081, | |
| "eval_loss": 0.10643722862005234, | |
| "eval_runtime": 51.7509, | |
| "eval_samples_per_second": 5.082, | |
| "eval_steps_per_second": 2.551, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.603047313552526, | |
| "grad_norm": 0.08735591918230057, | |
| "learning_rate": 2.3295748332244173e-05, | |
| "loss": 0.1096, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6190858059342421, | |
| "grad_norm": 0.12118412554264069, | |
| "learning_rate": 2.153112921515259e-05, | |
| "loss": 0.115, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.6351242983159584, | |
| "grad_norm": 0.0717737153172493, | |
| "learning_rate": 1.982791588654368e-05, | |
| "loss": 0.1015, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.6351242983159584, | |
| "eval_loss": 0.10614278167486191, | |
| "eval_runtime": 51.7761, | |
| "eval_samples_per_second": 5.08, | |
| "eval_steps_per_second": 2.549, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.6511627906976745, | |
| "grad_norm": 0.07388223707675934, | |
| "learning_rate": 1.8187441195435817e-05, | |
| "loss": 0.1004, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.6672012830793905, | |
| "grad_norm": 0.1101582869887352, | |
| "learning_rate": 1.6610988894747626e-05, | |
| "loss": 0.1051, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.6832397754611068, | |
| "grad_norm": 0.07702717930078506, | |
| "learning_rate": 1.5099792636697652e-05, | |
| "loss": 0.1133, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.6832397754611068, | |
| "eval_loss": 0.10605080425739288, | |
| "eval_runtime": 51.6204, | |
| "eval_samples_per_second": 5.095, | |
| "eval_steps_per_second": 2.557, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.6992782678428227, | |
| "grad_norm": 0.10285110771656036, | |
| "learning_rate": 1.3655035007410178e-05, | |
| "loss": 0.1099, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.715316760224539, | |
| "grad_norm": 0.09180232137441635, | |
| "learning_rate": 1.2277846601482823e-05, | |
| "loss": 0.1134, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.731355252606255, | |
| "grad_norm": 0.08292191475629807, | |
| "learning_rate": 1.0969305137240227e-05, | |
| "loss": 0.1063, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.731355252606255, | |
| "eval_loss": 0.10602111369371414, | |
| "eval_runtime": 51.6076, | |
| "eval_samples_per_second": 5.096, | |
| "eval_steps_per_second": 2.558, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.747393744987971, | |
| "grad_norm": 0.07537797093391418, | |
| "learning_rate": 9.730434613365746e-06, | |
| "loss": 0.1113, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.7634322373696873, | |
| "grad_norm": 0.07908269017934799, | |
| "learning_rate": 8.562204507571914e-06, | |
| "loss": 0.1038, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.7794707297514034, | |
| "grad_norm": 0.12484702467918396, | |
| "learning_rate": 7.465529017935658e-06, | |
| "loss": 0.1155, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.7794707297514034, | |
| "eval_loss": 0.10594786703586578, | |
| "eval_runtime": 51.6233, | |
| "eval_samples_per_second": 5.095, | |
| "eval_steps_per_second": 2.557, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.7955092221331195, | |
| "grad_norm": 0.12337016314268112, | |
| "learning_rate": 6.4412663474931575e-06, | |
| "loss": 0.1158, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.8115477145148358, | |
| "grad_norm": 0.0796642079949379, | |
| "learning_rate": 5.490218032653205e-06, | |
| "loss": 0.1062, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.8275862068965516, | |
| "grad_norm": 0.062203481793403625, | |
| "learning_rate": 4.613128315955217e-06, | |
| "loss": 0.1035, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.8275862068965516, | |
| "eval_loss": 0.10584517568349838, | |
| "eval_runtime": 51.5844, | |
| "eval_samples_per_second": 5.098, | |
| "eval_steps_per_second": 2.559, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.8436246992782679, | |
| "grad_norm": 0.07995738089084625, | |
| "learning_rate": 3.810683563662476e-06, | |
| "loss": 0.1067, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.859663191659984, | |
| "grad_norm": 0.09624777734279633, | |
| "learning_rate": 3.0835117286464153e-06, | |
| "loss": 0.1018, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.8757016840417, | |
| "grad_norm": 0.08080890029668808, | |
| "learning_rate": 2.4321818589823518e-06, | |
| "loss": 0.0953, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.8757016840417, | |
| "eval_loss": 0.1057586520910263, | |
| "eval_runtime": 51.6082, | |
| "eval_samples_per_second": 5.096, | |
| "eval_steps_per_second": 2.558, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.8917401764234163, | |
| "grad_norm": 0.07521943002939224, | |
| "learning_rate": 1.8572036526410708e-06, | |
| "loss": 0.1016, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.9077786688051324, | |
| "grad_norm": 0.11767873167991638, | |
| "learning_rate": 1.359027058624862e-06, | |
| "loss": 0.1072, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.9238171611868484, | |
| "grad_norm": 0.09177617728710175, | |
| "learning_rate": 9.38041924860078e-07, | |
| "loss": 0.1093, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9238171611868484, | |
| "eval_loss": 0.10568080097436905, | |
| "eval_runtime": 51.6479, | |
| "eval_samples_per_second": 5.092, | |
| "eval_steps_per_second": 2.556, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9398556535685647, | |
| "grad_norm": 0.1095619797706604, | |
| "learning_rate": 5.945776931217783e-07, | |
| "loss": 0.1099, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.9558941459502805, | |
| "grad_norm": 0.08429524302482605, | |
| "learning_rate": 3.2890314122919764e-07, | |
| "loss": 0.1109, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.9719326383319968, | |
| "grad_norm": 0.0836464986205101, | |
| "learning_rate": 1.412261727136821e-07, | |
| "loss": 0.1046, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.9719326383319968, | |
| "eval_loss": 0.10573633015155792, | |
| "eval_runtime": 51.5768, | |
| "eval_samples_per_second": 5.099, | |
| "eval_steps_per_second": 2.559, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.987971130713713, | |
| "grad_norm": 0.09745202958583832, | |
| "learning_rate": 3.1693654123921e-08, | |
| "loss": 0.1078, | |
| "step": 1240 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1248, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.026272997179556e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |