| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.832340815391663, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01832340815391663, | |
| "grad_norm": 0.06011037901043892, | |
| "learning_rate": 4e-05, | |
| "loss": 1.296, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03664681630783326, | |
| "grad_norm": 0.05856110155582428, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3316, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.054970224461749886, | |
| "grad_norm": 0.0607464499771595, | |
| "learning_rate": 0.00012, | |
| "loss": 1.2794, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07329363261566652, | |
| "grad_norm": 0.06632011383771896, | |
| "learning_rate": 0.00016, | |
| "loss": 1.3129, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09161704076958314, | |
| "grad_norm": 0.06631691753864288, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2741, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10994044892349977, | |
| "grad_norm": 0.056466877460479736, | |
| "learning_rate": 0.00019998035748930052, | |
| "loss": 1.2717, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1282638570774164, | |
| "grad_norm": 0.05860245227813721, | |
| "learning_rate": 0.00019992143767376668, | |
| "loss": 1.2091, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14658726523133303, | |
| "grad_norm": 0.06553175300359726, | |
| "learning_rate": 0.00019982326370006058, | |
| "loss": 1.1926, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16491067338524965, | |
| "grad_norm": 0.07061401754617691, | |
| "learning_rate": 0.00019968587413584876, | |
| "loss": 1.1767, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1832340815391663, | |
| "grad_norm": 0.07183243334293365, | |
| "learning_rate": 0.000199509322954651, | |
| "loss": 1.1183, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2015574896930829, | |
| "grad_norm": 0.06944898515939713, | |
| "learning_rate": 0.00019929367951463655, | |
| "loss": 1.0868, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.21988089784699955, | |
| "grad_norm": 0.06642703711986542, | |
| "learning_rate": 0.00019903902853137703, | |
| "loss": 1.048, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.23820430600091616, | |
| "grad_norm": 0.06603793054819107, | |
| "learning_rate": 0.00019874547004456562, | |
| "loss": 1.0195, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.2565277141548328, | |
| "grad_norm": 0.06488285213708878, | |
| "learning_rate": 0.00019841311937871675, | |
| "loss": 1.0014, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2748511223087494, | |
| "grad_norm": 0.05940372124314308, | |
| "learning_rate": 0.0001980421070978606, | |
| "loss": 0.9943, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.29317453046266606, | |
| "grad_norm": 0.059967171400785446, | |
| "learning_rate": 0.00019763257895425113, | |
| "loss": 0.9349, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3114979386165827, | |
| "grad_norm": 0.0554397851228714, | |
| "learning_rate": 0.0001971846958311071, | |
| "loss": 0.9045, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3298213467704993, | |
| "grad_norm": 0.055131904780864716, | |
| "learning_rate": 0.00019669863367940935, | |
| "loss": 0.8799, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.34814475492441593, | |
| "grad_norm": 0.04358826205134392, | |
| "learning_rate": 0.00019617458344877816, | |
| "loss": 0.8504, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3664681630783326, | |
| "grad_norm": 0.04535752162337303, | |
| "learning_rate": 0.00019561275101245883, | |
| "loss": 0.828, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3847915712322492, | |
| "grad_norm": 0.04672062397003174, | |
| "learning_rate": 0.00019501335708644414, | |
| "loss": 0.8114, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.4031149793861658, | |
| "grad_norm": 0.04161343351006508, | |
| "learning_rate": 0.00019437663714276618, | |
| "loss": 0.846, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.42143838754008245, | |
| "grad_norm": 0.03887801244854927, | |
| "learning_rate": 0.0001937028413169911, | |
| "loss": 0.7911, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4397617956939991, | |
| "grad_norm": 0.03659196197986603, | |
| "learning_rate": 0.00019299223430995323, | |
| "loss": 0.7669, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.45808520384791573, | |
| "grad_norm": 0.03447382524609566, | |
| "learning_rate": 0.00019224509528376738, | |
| "loss": 0.782, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4764086120018323, | |
| "grad_norm": 0.028725607320666313, | |
| "learning_rate": 0.00019146171775215982, | |
| "loss": 0.7183, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.49473202015574896, | |
| "grad_norm": 0.027673941105604172, | |
| "learning_rate": 0.0001906424094651615, | |
| "loss": 0.7018, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5130554283096656, | |
| "grad_norm": 0.10227353870868683, | |
| "learning_rate": 0.00018978749228820826, | |
| "loss": 0.72, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5313788364635822, | |
| "grad_norm": 0.022650673985481262, | |
| "learning_rate": 0.00018889730207569607, | |
| "loss": 0.6936, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5497022446174988, | |
| "grad_norm": 0.023469725623726845, | |
| "learning_rate": 0.00018797218853904037, | |
| "loss": 0.6765, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5680256527714155, | |
| "grad_norm": 0.018101360648870468, | |
| "learning_rate": 0.000187012515109292, | |
| "loss": 0.6799, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5863490609253321, | |
| "grad_norm": 0.016794538125395775, | |
| "learning_rate": 0.00018601865879436317, | |
| "loss": 0.6732, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6046724690792488, | |
| "grad_norm": 0.017263714224100113, | |
| "learning_rate": 0.00018499101003091993, | |
| "loss": 0.6695, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6229958772331654, | |
| "grad_norm": 0.016381224617362022, | |
| "learning_rate": 0.0001839299725309989, | |
| "loss": 0.6928, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.641319285387082, | |
| "grad_norm": 0.015325487591326237, | |
| "learning_rate": 0.00018283596312340891, | |
| "loss": 0.6622, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6596426935409986, | |
| "grad_norm": 0.014056784100830555, | |
| "learning_rate": 0.0001817094115899799, | |
| "loss": 0.7612, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 0.015031951479613781, | |
| "learning_rate": 0.00018055076049672283, | |
| "loss": 0.6596, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6962895098488319, | |
| "grad_norm": 0.01640532910823822, | |
| "learning_rate": 0.00017936046501996762, | |
| "loss": 0.6837, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.7146129180027485, | |
| "grad_norm": 0.01830482669174671, | |
| "learning_rate": 0.000178138992767547, | |
| "loss": 0.6812, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7329363261566652, | |
| "grad_norm": 0.0472831092774868, | |
| "learning_rate": 0.00017688682359509678, | |
| "loss": 0.674, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7512597343105818, | |
| "grad_norm": 0.012456170283257961, | |
| "learning_rate": 0.00017560444941754427, | |
| "loss": 0.6518, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7695831424644984, | |
| "grad_norm": 0.01401186641305685, | |
| "learning_rate": 0.0001742923740158595, | |
| "loss": 0.6418, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.7879065506184151, | |
| "grad_norm": 0.015530922450125217, | |
| "learning_rate": 0.00017295111283914487, | |
| "loss": 0.6465, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.8062299587723316, | |
| "grad_norm": 0.01402275450527668, | |
| "learning_rate": 0.0001715811928021406, | |
| "loss": 0.6642, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8245533669262483, | |
| "grad_norm": 0.01176263578236103, | |
| "learning_rate": 0.0001701831520782264, | |
| "loss": 0.6336, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8428767750801649, | |
| "grad_norm": 0.013003438711166382, | |
| "learning_rate": 0.00016875753988799982, | |
| "loss": 0.6469, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.8612001832340815, | |
| "grad_norm": 0.011523702181875706, | |
| "learning_rate": 0.00016730491628351487, | |
| "loss": 0.6434, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.8795235913879982, | |
| "grad_norm": 0.011919384822249413, | |
| "learning_rate": 0.00016582585192826543, | |
| "loss": 0.6588, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8978469995419148, | |
| "grad_norm": 0.013994649983942509, | |
| "learning_rate": 0.00016432092787299992, | |
| "loss": 0.6315, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.9161704076958315, | |
| "grad_norm": 0.013580686412751675, | |
| "learning_rate": 0.00016279073532745553, | |
| "loss": 0.6782, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.934493815849748, | |
| "grad_norm": 0.01364163402467966, | |
| "learning_rate": 0.00016123587542810118, | |
| "loss": 0.6334, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.9528172240036646, | |
| "grad_norm": 0.013080372475087643, | |
| "learning_rate": 0.0001596569590019811, | |
| "loss": 0.6233, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.9711406321575813, | |
| "grad_norm": 0.056398555636405945, | |
| "learning_rate": 0.00015805460632675112, | |
| "loss": 0.6557, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.9894640403114979, | |
| "grad_norm": 0.012467793188989162, | |
| "learning_rate": 0.00015642944688700264, | |
| "loss": 0.6315, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.0077874484654146, | |
| "grad_norm": 0.012495579198002815, | |
| "learning_rate": 0.00015478211912696929, | |
| "loss": 0.6177, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.026110856619331, | |
| "grad_norm": 0.010633349418640137, | |
| "learning_rate": 0.00015311327019971413, | |
| "loss": 0.644, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.0444342647732479, | |
| "grad_norm": 0.012223353609442711, | |
| "learning_rate": 0.00015142355571289533, | |
| "loss": 0.6502, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.0627576729271644, | |
| "grad_norm": 0.012305443175137043, | |
| "learning_rate": 0.00014971363947121065, | |
| "loss": 0.6185, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 0.017129750922322273, | |
| "learning_rate": 0.0001479841932156215, | |
| "loss": 0.6154, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.0994044892349977, | |
| "grad_norm": 0.013129614293575287, | |
| "learning_rate": 0.0001462358963594595, | |
| "loss": 0.614, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.1177278973889144, | |
| "grad_norm": 0.01199612207710743, | |
| "learning_rate": 0.00014446943572151867, | |
| "loss": 0.6128, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.136051305542831, | |
| "grad_norm": 0.012518757954239845, | |
| "learning_rate": 0.00014268550525623868, | |
| "loss": 0.6169, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.1543747136967477, | |
| "grad_norm": 0.01321893185377121, | |
| "learning_rate": 0.00014088480578108454, | |
| "loss": 0.6402, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.1726981218506642, | |
| "grad_norm": 0.012497123330831528, | |
| "learning_rate": 0.00013906804470123038, | |
| "loss": 0.613, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.1910215300045808, | |
| "grad_norm": 0.01103185210376978, | |
| "learning_rate": 0.00013723593573165523, | |
| "loss": 0.6114, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.2093449381584975, | |
| "grad_norm": 0.012833209708333015, | |
| "learning_rate": 0.00013538919861675979, | |
| "loss": 0.617, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.227668346312414, | |
| "grad_norm": 0.058991171419620514, | |
| "learning_rate": 0.0001335285588476148, | |
| "loss": 0.6298, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.2459917544663308, | |
| "grad_norm": 0.013424506410956383, | |
| "learning_rate": 0.00013165474737695184, | |
| "loss": 0.6488, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.2643151626202473, | |
| "grad_norm": 0.01241598092019558, | |
| "learning_rate": 0.00012976850033200805, | |
| "loss": 0.6088, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.2826385707741639, | |
| "grad_norm": 0.012560844421386719, | |
| "learning_rate": 0.00012787055872533865, | |
| "loss": 0.6032, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3009619789280806, | |
| "grad_norm": 0.010990115813910961, | |
| "learning_rate": 0.00012596166816371005, | |
| "loss": 0.6282, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.3192853870819974, | |
| "grad_norm": 0.01151216309517622, | |
| "learning_rate": 0.00012404257855518782, | |
| "loss": 0.6297, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.337608795235914, | |
| "grad_norm": 0.011924243532121181, | |
| "learning_rate": 0.0001221140438145353, | |
| "loss": 0.6044, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.3559322033898304, | |
| "grad_norm": 0.013133584521710873, | |
| "learning_rate": 0.00012017682156703807, | |
| "loss": 0.6107, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.3742556115437472, | |
| "grad_norm": 0.014631664380431175, | |
| "learning_rate": 0.00011823167285087063, | |
| "loss": 0.6213, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.3925790196976637, | |
| "grad_norm": 0.011716130189597607, | |
| "learning_rate": 0.00011627936181812234, | |
| "loss": 0.6179, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.4109024278515805, | |
| "grad_norm": 0.013568080961704254, | |
| "learning_rate": 0.00011432065543460015, | |
| "loss": 0.5965, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.429225836005497, | |
| "grad_norm": 0.012749516405165195, | |
| "learning_rate": 0.00011235632317852605, | |
| "loss": 0.6128, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.4475492441594136, | |
| "grad_norm": 0.011930575594305992, | |
| "learning_rate": 0.00011038713673824715, | |
| "loss": 0.6117, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.4658726523133303, | |
| "grad_norm": 0.013386845588684082, | |
| "learning_rate": 0.00010841386970907785, | |
| "loss": 0.6186, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.4841960604672468, | |
| "grad_norm": 0.012542261742055416, | |
| "learning_rate": 0.00010643729728939292, | |
| "loss": 0.5909, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.5025194686211636, | |
| "grad_norm": 0.010874781757593155, | |
| "learning_rate": 0.0001044581959760903, | |
| "loss": 0.5903, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.5208428767750801, | |
| "grad_norm": 0.010801080614328384, | |
| "learning_rate": 0.00010247734325954447, | |
| "loss": 0.5929, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.5391662849289967, | |
| "grad_norm": 0.012022151611745358, | |
| "learning_rate": 0.00010049551731816902, | |
| "loss": 0.6117, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.5574896930829134, | |
| "grad_norm": 0.011683526448905468, | |
| "learning_rate": 9.851349671270909e-05, | |
| "loss": 0.6283, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.5758131012368302, | |
| "grad_norm": 0.01242094673216343, | |
| "learning_rate": 9.653206008038364e-05, | |
| "loss": 0.5901, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.5941365093907467, | |
| "grad_norm": 0.011935061775147915, | |
| "learning_rate": 9.455198582899774e-05, | |
| "loss": 0.5848, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.6124599175446632, | |
| "grad_norm": 0.01208607666194439, | |
| "learning_rate": 9.257405183114473e-05, | |
| "loss": 0.5912, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.63078332569858, | |
| "grad_norm": 0.01227467879652977, | |
| "learning_rate": 9.059903511861891e-05, | |
| "loss": 0.5859, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.6491067338524965, | |
| "grad_norm": 0.0517101027071476, | |
| "learning_rate": 8.862771157715847e-05, | |
| "loss": 0.607, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.6674301420064133, | |
| "grad_norm": 0.010776874609291553, | |
| "learning_rate": 8.666085564163852e-05, | |
| "loss": 0.5788, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.6857535501603298, | |
| "grad_norm": 0.013487796299159527, | |
| "learning_rate": 8.469923999183411e-05, | |
| "loss": 0.5766, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.7040769583142463, | |
| "grad_norm": 0.011671481654047966, | |
| "learning_rate": 8.274363524887315e-05, | |
| "loss": 0.5976, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.722400366468163, | |
| "grad_norm": 0.01118433102965355, | |
| "learning_rate": 8.079480967249737e-05, | |
| "loss": 0.6021, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.7407237746220798, | |
| "grad_norm": 0.013788875192403793, | |
| "learning_rate": 7.88535288592514e-05, | |
| "loss": 0.579, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.7590471827759964, | |
| "grad_norm": 0.013310333713889122, | |
| "learning_rate": 7.692055544171823e-05, | |
| "loss": 0.5979, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.777370590929913, | |
| "grad_norm": 0.053230684250593185, | |
| "learning_rate": 7.49966487889185e-05, | |
| "loss": 0.5906, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.7956939990838294, | |
| "grad_norm": 0.013503102585673332, | |
| "learning_rate": 7.308256470799256e-05, | |
| "loss": 0.6061, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.8140174072377462, | |
| "grad_norm": 0.01030020508915186, | |
| "learning_rate": 7.117905514728107e-05, | |
| "loss": 0.5776, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.832340815391663, | |
| "grad_norm": 0.01204043161123991, | |
| "learning_rate": 6.928686790092235e-05, | |
| "loss": 0.584, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1635, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.3780357496832e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |