| { | |
| "best_metric": 0.23903648555278778, | |
| "best_model_checkpoint": "/root/pretrain_executions/pretrain_utg4java_220m_seq1024/checkpoint-38422", | |
| "epoch": 49.992029332058024, | |
| "eval_steps": 500, | |
| "global_step": 39200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3953451299218875, | |
| "grad_norm": 0.6701709628105164, | |
| "learning_rate": 7.908163265306123e-05, | |
| "loss": 0.5421, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.790690259843775, | |
| "grad_norm": 0.7041985988616943, | |
| "learning_rate": 0.00015816326530612246, | |
| "loss": 0.517, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.9998405866411605, | |
| "eval_loss": 0.4792475402355194, | |
| "eval_runtime": 146.6334, | |
| "eval_samples_per_second": 85.567, | |
| "eval_steps_per_second": 2.68, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 1.1860353897656624, | |
| "grad_norm": 0.7499191164970398, | |
| "learning_rate": 0.00019924510620574761, | |
| "loss": 0.5394, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.5813805196875497, | |
| "grad_norm": 0.6494282484054565, | |
| "learning_rate": 0.00019763119533527698, | |
| "loss": 0.5327, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.9767256496094374, | |
| "grad_norm": 0.5610156059265137, | |
| "learning_rate": 0.00019601728446480633, | |
| "loss": 0.518, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.9996811732823212, | |
| "eval_loss": 0.45565617084503174, | |
| "eval_runtime": 146.6648, | |
| "eval_samples_per_second": 85.549, | |
| "eval_steps_per_second": 2.68, | |
| "step": 1568 | |
| }, | |
| { | |
| "epoch": 2.3720707795313247, | |
| "grad_norm": 0.6261674761772156, | |
| "learning_rate": 0.00019440337359433573, | |
| "loss": 0.5076, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 2.767415909453212, | |
| "grad_norm": 0.5601200461387634, | |
| "learning_rate": 0.00019278946272386507, | |
| "loss": 0.4972, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 2.9995217599234816, | |
| "eval_loss": 0.4429556429386139, | |
| "eval_runtime": 146.6472, | |
| "eval_samples_per_second": 85.559, | |
| "eval_steps_per_second": 2.68, | |
| "step": 2352 | |
| }, | |
| { | |
| "epoch": 3.1627610393751, | |
| "grad_norm": 0.6102643609046936, | |
| "learning_rate": 0.00019117555185339441, | |
| "loss": 0.4957, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 3.558106169296987, | |
| "grad_norm": 0.6246281266212463, | |
| "learning_rate": 0.00018956164098292379, | |
| "loss": 0.4827, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 3.9534512992188744, | |
| "grad_norm": 0.6781056523323059, | |
| "learning_rate": 0.00018794773011245316, | |
| "loss": 0.4736, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.999362346564642, | |
| "eval_loss": 0.42357587814331055, | |
| "eval_runtime": 146.7171, | |
| "eval_samples_per_second": 85.518, | |
| "eval_steps_per_second": 2.679, | |
| "step": 3136 | |
| }, | |
| { | |
| "epoch": 4.348796429140762, | |
| "grad_norm": 0.5225201845169067, | |
| "learning_rate": 0.00018633381924198253, | |
| "loss": 0.4686, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 4.7441415590626494, | |
| "grad_norm": 0.5367516279220581, | |
| "learning_rate": 0.00018471990837151187, | |
| "loss": 0.4575, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 4.999202933205803, | |
| "eval_loss": 0.4168592095375061, | |
| "eval_runtime": 146.7474, | |
| "eval_samples_per_second": 85.501, | |
| "eval_steps_per_second": 2.678, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 5.139486688984537, | |
| "grad_norm": 0.5979415774345398, | |
| "learning_rate": 0.00018310599750104124, | |
| "loss": 0.4591, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 5.534831818906424, | |
| "grad_norm": 0.6041168570518494, | |
| "learning_rate": 0.0001814920866305706, | |
| "loss": 0.4506, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 5.930176948828311, | |
| "grad_norm": 0.5398473739624023, | |
| "learning_rate": 0.00017987817576009998, | |
| "loss": 0.4499, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 5.999043519846963, | |
| "eval_loss": 0.3998393714427948, | |
| "eval_runtime": 146.7067, | |
| "eval_samples_per_second": 85.524, | |
| "eval_steps_per_second": 2.679, | |
| "step": 4704 | |
| }, | |
| { | |
| "epoch": 6.3255220787502, | |
| "grad_norm": 0.5446251630783081, | |
| "learning_rate": 0.00017826426488962933, | |
| "loss": 0.4444, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 6.720867208672087, | |
| "grad_norm": 0.564083993434906, | |
| "learning_rate": 0.00017665556018325697, | |
| "loss": 0.4401, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 6.9988841064881235, | |
| "eval_loss": 0.3943786323070526, | |
| "eval_runtime": 146.7223, | |
| "eval_samples_per_second": 85.515, | |
| "eval_steps_per_second": 2.679, | |
| "step": 5488 | |
| }, | |
| { | |
| "epoch": 7.116212338593974, | |
| "grad_norm": 0.5197238326072693, | |
| "learning_rate": 0.00017504164931278634, | |
| "loss": 0.4349, | |
| "step": 5580 | |
| }, | |
| { | |
| "epoch": 7.5115574685158615, | |
| "grad_norm": 0.5063862204551697, | |
| "learning_rate": 0.0001734277384423157, | |
| "loss": 0.4274, | |
| "step": 5890 | |
| }, | |
| { | |
| "epoch": 7.906902598437749, | |
| "grad_norm": 0.8238950371742249, | |
| "learning_rate": 0.00017181382757184508, | |
| "loss": 0.4275, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.38801178336143494, | |
| "eval_runtime": 146.7154, | |
| "eval_samples_per_second": 85.519, | |
| "eval_steps_per_second": 2.679, | |
| "step": 6273 | |
| }, | |
| { | |
| "epoch": 8.302247728359637, | |
| "grad_norm": 0.4785802662372589, | |
| "learning_rate": 0.00017019991670137442, | |
| "loss": 0.4218, | |
| "step": 6510 | |
| }, | |
| { | |
| "epoch": 8.697592858281524, | |
| "grad_norm": 0.5460196137428284, | |
| "learning_rate": 0.0001685860058309038, | |
| "loss": 0.4165, | |
| "step": 6820 | |
| }, | |
| { | |
| "epoch": 8.999840586641161, | |
| "eval_loss": 0.3786679804325104, | |
| "eval_runtime": 146.6889, | |
| "eval_samples_per_second": 85.535, | |
| "eval_steps_per_second": 2.679, | |
| "step": 7057 | |
| }, | |
| { | |
| "epoch": 9.092937988203412, | |
| "grad_norm": 0.5532106161117554, | |
| "learning_rate": 0.00016697209496043317, | |
| "loss": 0.4147, | |
| "step": 7130 | |
| }, | |
| { | |
| "epoch": 9.488283118125299, | |
| "grad_norm": 0.5270036458969116, | |
| "learning_rate": 0.00016535818408996254, | |
| "loss": 0.4054, | |
| "step": 7440 | |
| }, | |
| { | |
| "epoch": 9.883628248047186, | |
| "grad_norm": 0.5107512474060059, | |
| "learning_rate": 0.00016374427321949188, | |
| "loss": 0.407, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 9.99968117328232, | |
| "eval_loss": 0.3678111732006073, | |
| "eval_runtime": 146.7168, | |
| "eval_samples_per_second": 85.519, | |
| "eval_steps_per_second": 2.679, | |
| "step": 7841 | |
| }, | |
| { | |
| "epoch": 10.278973377969074, | |
| "grad_norm": 0.4663056433200836, | |
| "learning_rate": 0.00016213036234902125, | |
| "loss": 0.4001, | |
| "step": 8060 | |
| }, | |
| { | |
| "epoch": 10.67431850789096, | |
| "grad_norm": 0.5166866183280945, | |
| "learning_rate": 0.00016051645147855062, | |
| "loss": 0.4012, | |
| "step": 8370 | |
| }, | |
| { | |
| "epoch": 10.999521759923482, | |
| "eval_loss": 0.36857831478118896, | |
| "eval_runtime": 146.752, | |
| "eval_samples_per_second": 85.498, | |
| "eval_steps_per_second": 2.678, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 11.069663637812848, | |
| "grad_norm": 0.5623896718025208, | |
| "learning_rate": 0.00015890254060807997, | |
| "loss": 0.3967, | |
| "step": 8680 | |
| }, | |
| { | |
| "epoch": 11.465008767734735, | |
| "grad_norm": 0.4826233983039856, | |
| "learning_rate": 0.00015728862973760934, | |
| "loss": 0.3902, | |
| "step": 8990 | |
| }, | |
| { | |
| "epoch": 11.860353897656623, | |
| "grad_norm": 0.5024587512016296, | |
| "learning_rate": 0.00015567471886713868, | |
| "loss": 0.3889, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 11.999362346564642, | |
| "eval_loss": 0.35674843192100525, | |
| "eval_runtime": 146.7093, | |
| "eval_samples_per_second": 85.523, | |
| "eval_steps_per_second": 2.679, | |
| "step": 9409 | |
| }, | |
| { | |
| "epoch": 12.255699027578512, | |
| "grad_norm": 0.4992258846759796, | |
| "learning_rate": 0.00015406080799666805, | |
| "loss": 0.3838, | |
| "step": 9610 | |
| }, | |
| { | |
| "epoch": 12.6510441575004, | |
| "grad_norm": 0.4781612455844879, | |
| "learning_rate": 0.00015244689712619742, | |
| "loss": 0.3789, | |
| "step": 9920 | |
| }, | |
| { | |
| "epoch": 12.999202933205803, | |
| "eval_loss": 0.35254529118537903, | |
| "eval_runtime": 146.7424, | |
| "eval_samples_per_second": 85.504, | |
| "eval_steps_per_second": 2.678, | |
| "step": 10193 | |
| }, | |
| { | |
| "epoch": 13.046389287422286, | |
| "grad_norm": 0.49535173177719116, | |
| "learning_rate": 0.0001508329862557268, | |
| "loss": 0.3775, | |
| "step": 10230 | |
| }, | |
| { | |
| "epoch": 13.441734417344174, | |
| "grad_norm": 0.5237115621566772, | |
| "learning_rate": 0.00014922428154935443, | |
| "loss": 0.3734, | |
| "step": 10540 | |
| }, | |
| { | |
| "epoch": 13.837079547266061, | |
| "grad_norm": 0.4549529552459717, | |
| "learning_rate": 0.0001476155768429821, | |
| "loss": 0.37, | |
| "step": 10850 | |
| }, | |
| { | |
| "epoch": 13.999043519846964, | |
| "eval_loss": 0.3443816006183624, | |
| "eval_runtime": 146.6954, | |
| "eval_samples_per_second": 85.531, | |
| "eval_steps_per_second": 2.679, | |
| "step": 10977 | |
| }, | |
| { | |
| "epoch": 14.232424677187948, | |
| "grad_norm": 0.520125150680542, | |
| "learning_rate": 0.00014600166597251147, | |
| "loss": 0.3647, | |
| "step": 11160 | |
| }, | |
| { | |
| "epoch": 14.627769807109836, | |
| "grad_norm": 0.5332316160202026, | |
| "learning_rate": 0.00014438775510204084, | |
| "loss": 0.3678, | |
| "step": 11470 | |
| }, | |
| { | |
| "epoch": 14.998884106488124, | |
| "eval_loss": 0.3436979055404663, | |
| "eval_runtime": 146.7179, | |
| "eval_samples_per_second": 85.518, | |
| "eval_steps_per_second": 2.679, | |
| "step": 11761 | |
| }, | |
| { | |
| "epoch": 15.023114937031723, | |
| "grad_norm": 0.47955256700515747, | |
| "learning_rate": 0.0001427738442315702, | |
| "loss": 0.3664, | |
| "step": 11780 | |
| }, | |
| { | |
| "epoch": 15.41846006695361, | |
| "grad_norm": 0.48371678590774536, | |
| "learning_rate": 0.00014115993336109953, | |
| "loss": 0.36, | |
| "step": 12090 | |
| }, | |
| { | |
| "epoch": 15.813805196875498, | |
| "grad_norm": 0.4756961166858673, | |
| "learning_rate": 0.0001395460224906289, | |
| "loss": 0.3577, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.3342459499835968, | |
| "eval_runtime": 146.7158, | |
| "eval_samples_per_second": 85.519, | |
| "eval_steps_per_second": 2.679, | |
| "step": 12546 | |
| }, | |
| { | |
| "epoch": 16.209150326797385, | |
| "grad_norm": 0.5593659281730652, | |
| "learning_rate": 0.00013793211162015827, | |
| "loss": 0.3536, | |
| "step": 12710 | |
| }, | |
| { | |
| "epoch": 16.604495456719274, | |
| "grad_norm": 0.6009001135826111, | |
| "learning_rate": 0.00013631820074968764, | |
| "loss": 0.3539, | |
| "step": 13020 | |
| }, | |
| { | |
| "epoch": 16.99984058664116, | |
| "grad_norm": 0.48707565665245056, | |
| "learning_rate": 0.000134704289879217, | |
| "loss": 0.3522, | |
| "step": 13330 | |
| }, | |
| { | |
| "epoch": 16.99984058664116, | |
| "eval_loss": 0.33101820945739746, | |
| "eval_runtime": 146.73, | |
| "eval_samples_per_second": 85.511, | |
| "eval_steps_per_second": 2.678, | |
| "step": 13330 | |
| }, | |
| { | |
| "epoch": 17.39518571656305, | |
| "grad_norm": 0.4787095785140991, | |
| "learning_rate": 0.00013309037900874636, | |
| "loss": 0.343, | |
| "step": 13640 | |
| }, | |
| { | |
| "epoch": 17.790530846484934, | |
| "grad_norm": 0.4643840789794922, | |
| "learning_rate": 0.00013147646813827573, | |
| "loss": 0.3466, | |
| "step": 13950 | |
| }, | |
| { | |
| "epoch": 17.999681173282323, | |
| "eval_loss": 0.3281005620956421, | |
| "eval_runtime": 146.7108, | |
| "eval_samples_per_second": 85.522, | |
| "eval_steps_per_second": 2.679, | |
| "step": 14114 | |
| }, | |
| { | |
| "epoch": 18.185875976406823, | |
| "grad_norm": 0.4819445312023163, | |
| "learning_rate": 0.0001298625572678051, | |
| "loss": 0.3415, | |
| "step": 14260 | |
| }, | |
| { | |
| "epoch": 18.58122110632871, | |
| "grad_norm": 0.46530964970588684, | |
| "learning_rate": 0.00012824864639733444, | |
| "loss": 0.3393, | |
| "step": 14570 | |
| }, | |
| { | |
| "epoch": 18.976566236250598, | |
| "grad_norm": 0.5159475207328796, | |
| "learning_rate": 0.00012663473552686382, | |
| "loss": 0.3377, | |
| "step": 14880 | |
| }, | |
| { | |
| "epoch": 18.999521759923482, | |
| "eval_loss": 0.32132235169410706, | |
| "eval_runtime": 146.7396, | |
| "eval_samples_per_second": 85.505, | |
| "eval_steps_per_second": 2.678, | |
| "step": 14898 | |
| }, | |
| { | |
| "epoch": 19.371911366172487, | |
| "grad_norm": 0.45964986085891724, | |
| "learning_rate": 0.00012502082465639319, | |
| "loss": 0.3348, | |
| "step": 15190 | |
| }, | |
| { | |
| "epoch": 19.767256496094372, | |
| "grad_norm": 0.49627387523651123, | |
| "learning_rate": 0.00012340691378592253, | |
| "loss": 0.3316, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 19.99936234656464, | |
| "eval_loss": 0.31626757979393005, | |
| "eval_runtime": 146.7396, | |
| "eval_samples_per_second": 85.505, | |
| "eval_steps_per_second": 2.678, | |
| "step": 15682 | |
| }, | |
| { | |
| "epoch": 20.16260162601626, | |
| "grad_norm": 0.48719242215156555, | |
| "learning_rate": 0.0001217930029154519, | |
| "loss": 0.3294, | |
| "step": 15810 | |
| }, | |
| { | |
| "epoch": 20.557946755938147, | |
| "grad_norm": 0.5443927049636841, | |
| "learning_rate": 0.00012017909204498126, | |
| "loss": 0.3261, | |
| "step": 16120 | |
| }, | |
| { | |
| "epoch": 20.953291885860036, | |
| "grad_norm": 0.4637634754180908, | |
| "learning_rate": 0.00011856518117451063, | |
| "loss": 0.3255, | |
| "step": 16430 | |
| }, | |
| { | |
| "epoch": 20.9992029332058, | |
| "eval_loss": 0.31501948833465576, | |
| "eval_runtime": 146.7591, | |
| "eval_samples_per_second": 85.494, | |
| "eval_steps_per_second": 2.678, | |
| "step": 16466 | |
| }, | |
| { | |
| "epoch": 21.34863701578192, | |
| "grad_norm": 0.46018585562705994, | |
| "learning_rate": 0.00011695127030403999, | |
| "loss": 0.3198, | |
| "step": 16740 | |
| }, | |
| { | |
| "epoch": 21.74398214570381, | |
| "grad_norm": 0.5096014738082886, | |
| "learning_rate": 0.00011533735943356936, | |
| "loss": 0.3226, | |
| "step": 17050 | |
| }, | |
| { | |
| "epoch": 21.999043519846964, | |
| "eval_loss": 0.30657365918159485, | |
| "eval_runtime": 146.7538, | |
| "eval_samples_per_second": 85.497, | |
| "eval_steps_per_second": 2.678, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 22.139327275625696, | |
| "grad_norm": 0.44816407561302185, | |
| "learning_rate": 0.00011372344856309872, | |
| "loss": 0.3178, | |
| "step": 17360 | |
| }, | |
| { | |
| "epoch": 22.534672405547585, | |
| "grad_norm": 0.437168151140213, | |
| "learning_rate": 0.00011211474385672638, | |
| "loss": 0.3172, | |
| "step": 17670 | |
| }, | |
| { | |
| "epoch": 22.93001753546947, | |
| "grad_norm": 0.5836613774299622, | |
| "learning_rate": 0.00011050083298625573, | |
| "loss": 0.3121, | |
| "step": 17980 | |
| }, | |
| { | |
| "epoch": 22.998884106488124, | |
| "eval_loss": 0.30263882875442505, | |
| "eval_runtime": 146.7108, | |
| "eval_samples_per_second": 85.522, | |
| "eval_steps_per_second": 2.679, | |
| "step": 18034 | |
| }, | |
| { | |
| "epoch": 23.32536266539136, | |
| "grad_norm": 0.4829230308532715, | |
| "learning_rate": 0.00010888692211578508, | |
| "loss": 0.3079, | |
| "step": 18290 | |
| }, | |
| { | |
| "epoch": 23.720707795313245, | |
| "grad_norm": 0.4485584497451782, | |
| "learning_rate": 0.00010727821740941275, | |
| "loss": 0.3105, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.3048921227455139, | |
| "eval_runtime": 146.8103, | |
| "eval_samples_per_second": 85.464, | |
| "eval_steps_per_second": 2.677, | |
| "step": 18819 | |
| }, | |
| { | |
| "epoch": 24.116052925235135, | |
| "grad_norm": 0.5251662135124207, | |
| "learning_rate": 0.00010566430653894211, | |
| "loss": 0.3052, | |
| "step": 18910 | |
| }, | |
| { | |
| "epoch": 24.511398055157024, | |
| "grad_norm": 0.4876725971698761, | |
| "learning_rate": 0.00010405039566847148, | |
| "loss": 0.3045, | |
| "step": 19220 | |
| }, | |
| { | |
| "epoch": 24.90674318507891, | |
| "grad_norm": 0.5600521564483643, | |
| "learning_rate": 0.00010243648479800084, | |
| "loss": 0.3048, | |
| "step": 19530 | |
| }, | |
| { | |
| "epoch": 24.99984058664116, | |
| "eval_loss": 0.2986990809440613, | |
| "eval_runtime": 146.734, | |
| "eval_samples_per_second": 85.508, | |
| "eval_steps_per_second": 2.678, | |
| "step": 19603 | |
| }, | |
| { | |
| "epoch": 25.3020883150008, | |
| "grad_norm": 0.5170055627822876, | |
| "learning_rate": 0.00010082257392753021, | |
| "loss": 0.3003, | |
| "step": 19840 | |
| }, | |
| { | |
| "epoch": 25.697433444922684, | |
| "grad_norm": 0.48347124457359314, | |
| "learning_rate": 9.920866305705956e-05, | |
| "loss": 0.2983, | |
| "step": 20150 | |
| }, | |
| { | |
| "epoch": 25.999681173282323, | |
| "eval_loss": 0.2916134297847748, | |
| "eval_runtime": 146.7271, | |
| "eval_samples_per_second": 85.512, | |
| "eval_steps_per_second": 2.678, | |
| "step": 20387 | |
| }, | |
| { | |
| "epoch": 26.092778574844573, | |
| "grad_norm": 0.48907041549682617, | |
| "learning_rate": 9.759475218658892e-05, | |
| "loss": 0.2959, | |
| "step": 20460 | |
| }, | |
| { | |
| "epoch": 26.48812370476646, | |
| "grad_norm": 0.5060804486274719, | |
| "learning_rate": 9.598084131611829e-05, | |
| "loss": 0.2923, | |
| "step": 20770 | |
| }, | |
| { | |
| "epoch": 26.883468834688347, | |
| "grad_norm": 0.4843296706676483, | |
| "learning_rate": 9.436693044564765e-05, | |
| "loss": 0.2918, | |
| "step": 21080 | |
| }, | |
| { | |
| "epoch": 26.999521759923482, | |
| "eval_loss": 0.29019656777381897, | |
| "eval_runtime": 146.6934, | |
| "eval_samples_per_second": 85.532, | |
| "eval_steps_per_second": 2.679, | |
| "step": 21171 | |
| }, | |
| { | |
| "epoch": 27.278813964610233, | |
| "grad_norm": 0.42266514897346497, | |
| "learning_rate": 9.275301957517701e-05, | |
| "loss": 0.2901, | |
| "step": 21390 | |
| }, | |
| { | |
| "epoch": 27.674159094532122, | |
| "grad_norm": 0.5161967873573303, | |
| "learning_rate": 9.113910870470638e-05, | |
| "loss": 0.2889, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 27.99936234656464, | |
| "eval_loss": 0.2833983302116394, | |
| "eval_runtime": 146.7193, | |
| "eval_samples_per_second": 85.517, | |
| "eval_steps_per_second": 2.679, | |
| "step": 21955 | |
| }, | |
| { | |
| "epoch": 28.069504224454008, | |
| "grad_norm": 0.4523755609989166, | |
| "learning_rate": 8.952519783423574e-05, | |
| "loss": 0.2871, | |
| "step": 22010 | |
| }, | |
| { | |
| "epoch": 28.464849354375897, | |
| "grad_norm": 0.44348961114883423, | |
| "learning_rate": 8.791128696376511e-05, | |
| "loss": 0.2847, | |
| "step": 22320 | |
| }, | |
| { | |
| "epoch": 28.860194484297786, | |
| "grad_norm": 0.6467667818069458, | |
| "learning_rate": 8.630258225739276e-05, | |
| "loss": 0.2844, | |
| "step": 22630 | |
| }, | |
| { | |
| "epoch": 28.9992029332058, | |
| "eval_loss": 0.28629302978515625, | |
| "eval_runtime": 146.7547, | |
| "eval_samples_per_second": 85.496, | |
| "eval_steps_per_second": 2.678, | |
| "step": 22739 | |
| }, | |
| { | |
| "epoch": 29.25553961421967, | |
| "grad_norm": 0.4734992980957031, | |
| "learning_rate": 8.468867138692213e-05, | |
| "loss": 0.2787, | |
| "step": 22940 | |
| }, | |
| { | |
| "epoch": 29.65088474414156, | |
| "grad_norm": 0.4827498495578766, | |
| "learning_rate": 8.307476051645148e-05, | |
| "loss": 0.2787, | |
| "step": 23250 | |
| }, | |
| { | |
| "epoch": 29.999043519846964, | |
| "eval_loss": 0.2794826626777649, | |
| "eval_runtime": 146.9198, | |
| "eval_samples_per_second": 85.4, | |
| "eval_steps_per_second": 2.675, | |
| "step": 23523 | |
| }, | |
| { | |
| "epoch": 30.046229874063446, | |
| "grad_norm": 0.5005486607551575, | |
| "learning_rate": 8.146084964598085e-05, | |
| "loss": 0.2758, | |
| "step": 23560 | |
| }, | |
| { | |
| "epoch": 30.441575003985335, | |
| "grad_norm": 0.5253671407699585, | |
| "learning_rate": 7.98469387755102e-05, | |
| "loss": 0.2761, | |
| "step": 23870 | |
| }, | |
| { | |
| "epoch": 30.83692013390722, | |
| "grad_norm": 0.472740113735199, | |
| "learning_rate": 7.823302790503957e-05, | |
| "loss": 0.2726, | |
| "step": 24180 | |
| }, | |
| { | |
| "epoch": 30.998884106488124, | |
| "eval_loss": 0.2779182493686676, | |
| "eval_runtime": 146.7777, | |
| "eval_samples_per_second": 85.483, | |
| "eval_steps_per_second": 2.678, | |
| "step": 24307 | |
| }, | |
| { | |
| "epoch": 31.23226526382911, | |
| "grad_norm": 0.5228144526481628, | |
| "learning_rate": 7.661911703456893e-05, | |
| "loss": 0.2717, | |
| "step": 24490 | |
| }, | |
| { | |
| "epoch": 31.627610393750995, | |
| "grad_norm": 0.47681719064712524, | |
| "learning_rate": 7.501041232819659e-05, | |
| "loss": 0.2664, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.27039337158203125, | |
| "eval_runtime": 146.805, | |
| "eval_samples_per_second": 85.467, | |
| "eval_steps_per_second": 2.677, | |
| "step": 25092 | |
| }, | |
| { | |
| "epoch": 32.022955523672884, | |
| "grad_norm": 0.4973162114620209, | |
| "learning_rate": 7.339650145772596e-05, | |
| "loss": 0.268, | |
| "step": 25110 | |
| }, | |
| { | |
| "epoch": 32.41830065359477, | |
| "grad_norm": 0.5740240216255188, | |
| "learning_rate": 7.178259058725531e-05, | |
| "loss": 0.2668, | |
| "step": 25420 | |
| }, | |
| { | |
| "epoch": 32.813645783516655, | |
| "grad_norm": 0.4842962622642517, | |
| "learning_rate": 7.016867971678468e-05, | |
| "loss": 0.2631, | |
| "step": 25730 | |
| }, | |
| { | |
| "epoch": 32.99984058664116, | |
| "eval_loss": 0.2733234763145447, | |
| "eval_runtime": 146.7109, | |
| "eval_samples_per_second": 85.522, | |
| "eval_steps_per_second": 2.679, | |
| "step": 25876 | |
| }, | |
| { | |
| "epoch": 33.20899091343855, | |
| "grad_norm": 0.499452143907547, | |
| "learning_rate": 6.855476884631404e-05, | |
| "loss": 0.263, | |
| "step": 26040 | |
| }, | |
| { | |
| "epoch": 33.60433604336043, | |
| "grad_norm": 0.4541178345680237, | |
| "learning_rate": 6.69408579758434e-05, | |
| "loss": 0.2603, | |
| "step": 26350 | |
| }, | |
| { | |
| "epoch": 33.99968117328232, | |
| "grad_norm": 0.5029833912849426, | |
| "learning_rate": 6.532694710537276e-05, | |
| "loss": 0.258, | |
| "step": 26660 | |
| }, | |
| { | |
| "epoch": 33.99968117328232, | |
| "eval_loss": 0.26625362038612366, | |
| "eval_runtime": 146.7319, | |
| "eval_samples_per_second": 85.51, | |
| "eval_steps_per_second": 2.678, | |
| "step": 26660 | |
| }, | |
| { | |
| "epoch": 34.39502630320421, | |
| "grad_norm": 0.5090352892875671, | |
| "learning_rate": 6.371303623490213e-05, | |
| "loss": 0.2544, | |
| "step": 26970 | |
| }, | |
| { | |
| "epoch": 34.7903714331261, | |
| "grad_norm": 0.4605717360973358, | |
| "learning_rate": 6.209912536443149e-05, | |
| "loss": 0.254, | |
| "step": 27280 | |
| }, | |
| { | |
| "epoch": 34.99952175992348, | |
| "eval_loss": 0.26669949293136597, | |
| "eval_runtime": 146.7117, | |
| "eval_samples_per_second": 85.521, | |
| "eval_steps_per_second": 2.679, | |
| "step": 27444 | |
| }, | |
| { | |
| "epoch": 35.18571656304798, | |
| "grad_norm": 0.46216222643852234, | |
| "learning_rate": 6.048521449396085e-05, | |
| "loss": 0.254, | |
| "step": 27590 | |
| }, | |
| { | |
| "epoch": 35.58106169296987, | |
| "grad_norm": 0.49629315733909607, | |
| "learning_rate": 5.8871303623490214e-05, | |
| "loss": 0.2521, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 35.97640682289176, | |
| "grad_norm": 0.48311081528663635, | |
| "learning_rate": 5.725739275301958e-05, | |
| "loss": 0.2493, | |
| "step": 28210 | |
| }, | |
| { | |
| "epoch": 35.999362346564645, | |
| "eval_loss": 0.26483407616615295, | |
| "eval_runtime": 146.7384, | |
| "eval_samples_per_second": 85.506, | |
| "eval_steps_per_second": 2.678, | |
| "step": 28228 | |
| }, | |
| { | |
| "epoch": 36.371751952813646, | |
| "grad_norm": 0.43428850173950195, | |
| "learning_rate": 5.564348188254894e-05, | |
| "loss": 0.2455, | |
| "step": 28520 | |
| }, | |
| { | |
| "epoch": 36.76709708273553, | |
| "grad_norm": 0.4786287844181061, | |
| "learning_rate": 5.4029571012078306e-05, | |
| "loss": 0.2454, | |
| "step": 28830 | |
| }, | |
| { | |
| "epoch": 36.9992029332058, | |
| "eval_loss": 0.26446378231048584, | |
| "eval_runtime": 146.73, | |
| "eval_samples_per_second": 85.511, | |
| "eval_steps_per_second": 2.678, | |
| "step": 29012 | |
| }, | |
| { | |
| "epoch": 37.16244221265742, | |
| "grad_norm": 0.5931326746940613, | |
| "learning_rate": 5.241566014160767e-05, | |
| "loss": 0.247, | |
| "step": 29140 | |
| }, | |
| { | |
| "epoch": 37.55778734257931, | |
| "grad_norm": 0.5031745433807373, | |
| "learning_rate": 5.0801749271137035e-05, | |
| "loss": 0.2425, | |
| "step": 29450 | |
| }, | |
| { | |
| "epoch": 37.953132472501196, | |
| "grad_norm": 0.5432093739509583, | |
| "learning_rate": 4.918783840066639e-05, | |
| "loss": 0.2416, | |
| "step": 29760 | |
| }, | |
| { | |
| "epoch": 37.999043519846964, | |
| "eval_loss": 0.2601180672645569, | |
| "eval_runtime": 146.6811, | |
| "eval_samples_per_second": 85.539, | |
| "eval_steps_per_second": 2.679, | |
| "step": 29796 | |
| }, | |
| { | |
| "epoch": 38.34847760242308, | |
| "grad_norm": 0.5319362878799438, | |
| "learning_rate": 4.7573927530195756e-05, | |
| "loss": 0.2392, | |
| "step": 30070 | |
| }, | |
| { | |
| "epoch": 38.743822732344974, | |
| "grad_norm": 0.5319586396217346, | |
| "learning_rate": 4.596001665972512e-05, | |
| "loss": 0.2368, | |
| "step": 30380 | |
| }, | |
| { | |
| "epoch": 38.99888410648813, | |
| "eval_loss": 0.25446435809135437, | |
| "eval_runtime": 146.6972, | |
| "eval_samples_per_second": 85.53, | |
| "eval_steps_per_second": 2.679, | |
| "step": 30580 | |
| }, | |
| { | |
| "epoch": 39.13916786226686, | |
| "grad_norm": 0.4489250183105469, | |
| "learning_rate": 4.434610578925448e-05, | |
| "loss": 0.2368, | |
| "step": 30690 | |
| }, | |
| { | |
| "epoch": 39.534512992188745, | |
| "grad_norm": 0.48287880420684814, | |
| "learning_rate": 4.273740108288213e-05, | |
| "loss": 0.2353, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 39.92985812211063, | |
| "grad_norm": 0.49850553274154663, | |
| "learning_rate": 4.1123490212411495e-05, | |
| "loss": 0.2321, | |
| "step": 31310 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.24883659183979034, | |
| "eval_runtime": 146.7363, | |
| "eval_samples_per_second": 85.507, | |
| "eval_steps_per_second": 2.678, | |
| "step": 31365 | |
| }, | |
| { | |
| "epoch": 40.32520325203252, | |
| "grad_norm": 0.4667394161224365, | |
| "learning_rate": 3.9514785506039155e-05, | |
| "loss": 0.2337, | |
| "step": 31620 | |
| }, | |
| { | |
| "epoch": 40.72054838195441, | |
| "grad_norm": 0.5053902864456177, | |
| "learning_rate": 3.790087463556852e-05, | |
| "loss": 0.2284, | |
| "step": 31930 | |
| }, | |
| { | |
| "epoch": 40.99984058664116, | |
| "eval_loss": 0.2544113099575043, | |
| "eval_runtime": 146.7257, | |
| "eval_samples_per_second": 85.513, | |
| "eval_steps_per_second": 2.678, | |
| "step": 32149 | |
| }, | |
| { | |
| "epoch": 41.115893511876294, | |
| "grad_norm": 0.47476327419281006, | |
| "learning_rate": 3.628696376509788e-05, | |
| "loss": 0.2286, | |
| "step": 32240 | |
| }, | |
| { | |
| "epoch": 41.51123864179818, | |
| "grad_norm": 0.5025794506072998, | |
| "learning_rate": 3.467305289462724e-05, | |
| "loss": 0.2292, | |
| "step": 32550 | |
| }, | |
| { | |
| "epoch": 41.90658377172007, | |
| "grad_norm": 0.4553293287754059, | |
| "learning_rate": 3.3059142024156605e-05, | |
| "loss": 0.225, | |
| "step": 32860 | |
| }, | |
| { | |
| "epoch": 41.99968117328232, | |
| "eval_loss": 0.24568869173526764, | |
| "eval_runtime": 146.7316, | |
| "eval_samples_per_second": 85.51, | |
| "eval_steps_per_second": 2.678, | |
| "step": 32933 | |
| }, | |
| { | |
| "epoch": 42.30192890164196, | |
| "grad_norm": 0.4845215678215027, | |
| "learning_rate": 3.144523115368597e-05, | |
| "loss": 0.2236, | |
| "step": 33170 | |
| }, | |
| { | |
| "epoch": 42.69727403156384, | |
| "grad_norm": 0.5739601850509644, | |
| "learning_rate": 2.983132028321533e-05, | |
| "loss": 0.2234, | |
| "step": 33480 | |
| }, | |
| { | |
| "epoch": 42.99952175992348, | |
| "eval_loss": 0.24620206654071808, | |
| "eval_runtime": 146.7264, | |
| "eval_samples_per_second": 85.513, | |
| "eval_steps_per_second": 2.678, | |
| "step": 33717 | |
| }, | |
| { | |
| "epoch": 43.092619161485736, | |
| "grad_norm": 0.4569677412509918, | |
| "learning_rate": 2.8217409412744688e-05, | |
| "loss": 0.2213, | |
| "step": 33790 | |
| }, | |
| { | |
| "epoch": 43.48796429140762, | |
| "grad_norm": 0.5146024227142334, | |
| "learning_rate": 2.6603498542274052e-05, | |
| "loss": 0.2188, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 43.88330942132951, | |
| "grad_norm": 0.47475871443748474, | |
| "learning_rate": 2.4989587671803416e-05, | |
| "loss": 0.2206, | |
| "step": 34410 | |
| }, | |
| { | |
| "epoch": 43.999362346564645, | |
| "eval_loss": 0.2445935159921646, | |
| "eval_runtime": 146.7897, | |
| "eval_samples_per_second": 85.476, | |
| "eval_steps_per_second": 2.677, | |
| "step": 34501 | |
| }, | |
| { | |
| "epoch": 44.27865455125139, | |
| "grad_norm": 0.45915085077285767, | |
| "learning_rate": 2.337567680133278e-05, | |
| "loss": 0.217, | |
| "step": 34720 | |
| }, | |
| { | |
| "epoch": 44.673999681173285, | |
| "grad_norm": 0.4429190456867218, | |
| "learning_rate": 2.176176593086214e-05, | |
| "loss": 0.2165, | |
| "step": 35030 | |
| }, | |
| { | |
| "epoch": 44.9992029332058, | |
| "eval_loss": 0.24302400648593903, | |
| "eval_runtime": 146.7631, | |
| "eval_samples_per_second": 85.492, | |
| "eval_steps_per_second": 2.678, | |
| "step": 35285 | |
| }, | |
| { | |
| "epoch": 45.06934481109517, | |
| "grad_norm": 0.5038246512413025, | |
| "learning_rate": 2.0147855060391505e-05, | |
| "loss": 0.217, | |
| "step": 35340 | |
| }, | |
| { | |
| "epoch": 45.464689941017056, | |
| "grad_norm": 0.4302615523338318, | |
| "learning_rate": 1.8539150354019162e-05, | |
| "loss": 0.2137, | |
| "step": 35650 | |
| }, | |
| { | |
| "epoch": 45.86003507093894, | |
| "grad_norm": 0.5075607299804688, | |
| "learning_rate": 1.6925239483548523e-05, | |
| "loss": 0.2145, | |
| "step": 35960 | |
| }, | |
| { | |
| "epoch": 45.999043519846964, | |
| "eval_loss": 0.24222899973392487, | |
| "eval_runtime": 146.735, | |
| "eval_samples_per_second": 85.508, | |
| "eval_steps_per_second": 2.678, | |
| "step": 36069 | |
| }, | |
| { | |
| "epoch": 46.255380200860834, | |
| "grad_norm": 0.4777955114841461, | |
| "learning_rate": 1.531653477717618e-05, | |
| "loss": 0.2126, | |
| "step": 36270 | |
| }, | |
| { | |
| "epoch": 46.65072533078272, | |
| "grad_norm": 0.48974084854125977, | |
| "learning_rate": 1.3702623906705539e-05, | |
| "loss": 0.2112, | |
| "step": 36580 | |
| }, | |
| { | |
| "epoch": 46.99888410648813, | |
| "eval_loss": 0.2432757019996643, | |
| "eval_runtime": 146.7494, | |
| "eval_samples_per_second": 85.499, | |
| "eval_steps_per_second": 2.678, | |
| "step": 36853 | |
| }, | |
| { | |
| "epoch": 47.046070460704605, | |
| "grad_norm": 0.46624037623405457, | |
| "learning_rate": 1.2088713036234903e-05, | |
| "loss": 0.2089, | |
| "step": 36890 | |
| }, | |
| { | |
| "epoch": 47.44141559062649, | |
| "grad_norm": 0.4808659553527832, | |
| "learning_rate": 1.0474802165764265e-05, | |
| "loss": 0.2085, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 47.836760720548384, | |
| "grad_norm": 0.4421006143093109, | |
| "learning_rate": 8.86089129529363e-06, | |
| "loss": 0.2087, | |
| "step": 37510 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.24061799049377441, | |
| "eval_runtime": 146.7785, | |
| "eval_samples_per_second": 85.483, | |
| "eval_steps_per_second": 2.678, | |
| "step": 37638 | |
| }, | |
| { | |
| "epoch": 48.23210585047027, | |
| "grad_norm": 0.4642196297645569, | |
| "learning_rate": 7.246980424822991e-06, | |
| "loss": 0.208, | |
| "step": 37820 | |
| }, | |
| { | |
| "epoch": 48.627450980392155, | |
| "grad_norm": 0.47141027450561523, | |
| "learning_rate": 5.633069554352354e-06, | |
| "loss": 0.2067, | |
| "step": 38130 | |
| }, | |
| { | |
| "epoch": 48.99984058664116, | |
| "eval_loss": 0.23903648555278778, | |
| "eval_runtime": 146.7908, | |
| "eval_samples_per_second": 85.475, | |
| "eval_steps_per_second": 2.677, | |
| "step": 38422 | |
| }, | |
| { | |
| "epoch": 49.02279611031405, | |
| "grad_norm": 0.45030030608177185, | |
| "learning_rate": 4.019158683881716e-06, | |
| "loss": 0.2062, | |
| "step": 38440 | |
| }, | |
| { | |
| "epoch": 49.41814124023593, | |
| "grad_norm": 0.48792940378189087, | |
| "learning_rate": 2.4052478134110786e-06, | |
| "loss": 0.2062, | |
| "step": 38750 | |
| }, | |
| { | |
| "epoch": 49.81348637015782, | |
| "grad_norm": 0.40084025263786316, | |
| "learning_rate": 7.913369429404415e-07, | |
| "loss": 0.2055, | |
| "step": 39060 | |
| }, | |
| { | |
| "epoch": 49.992029332058024, | |
| "eval_loss": 0.23992499709129333, | |
| "eval_runtime": 146.7718, | |
| "eval_samples_per_second": 85.486, | |
| "eval_steps_per_second": 2.678, | |
| "step": 39200 | |
| }, | |
| { | |
| "epoch": 49.992029332058024, | |
| "step": 39200, | |
| "total_flos": 6.111014223347712e+18, | |
| "train_loss": 0.31974333125717785, | |
| "train_runtime": 145458.5548, | |
| "train_samples_per_second": 34.503, | |
| "train_steps_per_second": 0.269 | |
| } | |
| ], | |
| "logging_steps": 310, | |
| "max_steps": 39200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 6.111014223347712e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |