{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.8379156847342236, "eval_steps": 50, "global_step": 100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.008379156847342237, "grad_norm": 0.5054947733879089, "learning_rate": 1e-05, "loss": 2.9505, "step": 1 }, { "epoch": 0.008379156847342237, "eval_loss": 3.054915428161621, "eval_runtime": 43.2673, "eval_samples_per_second": 4.669, "eval_steps_per_second": 4.669, "step": 1 }, { "epoch": 0.016758313694684474, "grad_norm": 0.3709588348865509, "learning_rate": 2e-05, "loss": 2.8603, "step": 2 }, { "epoch": 0.02513747054202671, "grad_norm": 0.5003033876419067, "learning_rate": 3e-05, "loss": 2.8325, "step": 3 }, { "epoch": 0.03351662738936895, "grad_norm": 0.3898451328277588, "learning_rate": 4e-05, "loss": 2.7888, "step": 4 }, { "epoch": 0.041895784236711184, "grad_norm": 0.4698517918586731, "learning_rate": 5e-05, "loss": 2.8257, "step": 5 }, { "epoch": 0.05027494108405342, "grad_norm": 0.4781450927257538, "learning_rate": 6e-05, "loss": 2.9094, "step": 6 }, { "epoch": 0.05865409793139565, "grad_norm": 0.6419183015823364, "learning_rate": 7e-05, "loss": 2.8045, "step": 7 }, { "epoch": 0.0670332547787379, "grad_norm": 0.7329452633857727, "learning_rate": 8e-05, "loss": 2.7656, "step": 8 }, { "epoch": 0.07541241162608013, "grad_norm": 0.8872014284133911, "learning_rate": 9e-05, "loss": 2.9754, "step": 9 }, { "epoch": 0.08379156847342237, "grad_norm": 1.0049771070480347, "learning_rate": 0.0001, "loss": 2.7885, "step": 10 }, { "epoch": 0.0921707253207646, "grad_norm": 1.288948655128479, "learning_rate": 9.99695413509548e-05, "loss": 2.862, "step": 11 }, { "epoch": 0.10054988216810684, "grad_norm": 1.6090466976165771, "learning_rate": 9.987820251299122e-05, "loss": 2.9898, "step": 12 }, { "epoch": 0.10892903901544906, "grad_norm": 1.5656819343566895, "learning_rate": 9.972609476841367e-05, "loss": 2.8883, "step": 13 }, { "epoch": 0.1173081958627913, "grad_norm": 1.6467982530593872, "learning_rate": 9.951340343707852e-05, "loss": 2.7058, "step": 14 }, { "epoch": 0.12568735271013354, "grad_norm": 2.0958547592163086, "learning_rate": 9.924038765061042e-05, "loss": 2.7129, "step": 15 }, { "epoch": 0.1340665095574758, "grad_norm": 1.0730005502700806, "learning_rate": 9.890738003669029e-05, "loss": 2.7274, "step": 16 }, { "epoch": 0.142445666404818, "grad_norm": 1.0618668794631958, "learning_rate": 9.851478631379982e-05, "loss": 2.5979, "step": 17 }, { "epoch": 0.15082482325216026, "grad_norm": 0.4424041211605072, "learning_rate": 9.806308479691595e-05, "loss": 2.5467, "step": 18 }, { "epoch": 0.15920398009950248, "grad_norm": 0.5938422679901123, "learning_rate": 9.755282581475769e-05, "loss": 2.6563, "step": 19 }, { "epoch": 0.16758313694684474, "grad_norm": 0.5060924291610718, "learning_rate": 9.698463103929542e-05, "loss": 2.6239, "step": 20 }, { "epoch": 0.17596229379418696, "grad_norm": 0.5067716836929321, "learning_rate": 9.635919272833938e-05, "loss": 2.5558, "step": 21 }, { "epoch": 0.1843414506415292, "grad_norm": 0.5283418297767639, "learning_rate": 9.567727288213005e-05, "loss": 2.6283, "step": 22 }, { "epoch": 0.19272060748887143, "grad_norm": 0.5521147847175598, "learning_rate": 9.493970231495835e-05, "loss": 2.5582, "step": 23 }, { "epoch": 0.20109976433621368, "grad_norm": 0.6335211396217346, "learning_rate": 9.414737964294636e-05, "loss": 2.6768, "step": 24 }, { "epoch": 0.2094789211835559, "grad_norm": 1.2375539541244507, "learning_rate": 9.330127018922194e-05, "loss": 2.6038, "step": 25 }, { "epoch": 0.21785807803089813, "grad_norm": 1.3271223306655884, "learning_rate": 9.24024048078213e-05, "loss": 2.5917, "step": 26 }, { "epoch": 0.22623723487824038, "grad_norm": 0.4055769145488739, "learning_rate": 9.145187862775209e-05, "loss": 2.3628, "step": 27 }, { "epoch": 0.2346163917255826, "grad_norm": 0.964474081993103, "learning_rate": 9.045084971874738e-05, "loss": 2.5157, "step": 28 }, { "epoch": 0.24299554857292485, "grad_norm": 0.6849140524864197, "learning_rate": 8.940053768033609e-05, "loss": 2.6839, "step": 29 }, { "epoch": 0.2513747054202671, "grad_norm": 0.4390021562576294, "learning_rate": 8.83022221559489e-05, "loss": 2.5509, "step": 30 }, { "epoch": 0.2597538622676093, "grad_norm": 0.7394634485244751, "learning_rate": 8.715724127386972e-05, "loss": 2.5447, "step": 31 }, { "epoch": 0.2681330191149516, "grad_norm": 0.45920076966285706, "learning_rate": 8.596699001693255e-05, "loss": 2.5397, "step": 32 }, { "epoch": 0.27651217596229377, "grad_norm": 0.43246570229530334, "learning_rate": 8.473291852294987e-05, "loss": 2.6423, "step": 33 }, { "epoch": 0.284891332809636, "grad_norm": 0.38203856348991394, "learning_rate": 8.345653031794292e-05, "loss": 2.4129, "step": 34 }, { "epoch": 0.2932704896569783, "grad_norm": 0.37578803300857544, "learning_rate": 8.213938048432697e-05, "loss": 2.5367, "step": 35 }, { "epoch": 0.3016496465043205, "grad_norm": 0.6231028437614441, "learning_rate": 8.07830737662829e-05, "loss": 2.504, "step": 36 }, { "epoch": 0.3100288033516627, "grad_norm": 0.48215481638908386, "learning_rate": 7.938926261462366e-05, "loss": 2.5128, "step": 37 }, { "epoch": 0.31840796019900497, "grad_norm": 0.8526724576950073, "learning_rate": 7.795964517353735e-05, "loss": 2.4417, "step": 38 }, { "epoch": 0.3267871170463472, "grad_norm": 0.42325401306152344, "learning_rate": 7.649596321166024e-05, "loss": 2.484, "step": 39 }, { "epoch": 0.33516627389368947, "grad_norm": 0.45637720823287964, "learning_rate": 7.500000000000001e-05, "loss": 2.4911, "step": 40 }, { "epoch": 0.34354543074103167, "grad_norm": 0.5629859566688538, "learning_rate": 7.347357813929454e-05, "loss": 2.4716, "step": 41 }, { "epoch": 0.3519245875883739, "grad_norm": 0.6218668222427368, "learning_rate": 7.191855733945387e-05, "loss": 2.3725, "step": 42 }, { "epoch": 0.36030374443571617, "grad_norm": 0.5280422568321228, "learning_rate": 7.033683215379002e-05, "loss": 2.6807, "step": 43 }, { "epoch": 0.3686829012830584, "grad_norm": 0.4046926200389862, "learning_rate": 6.873032967079561e-05, "loss": 2.4806, "step": 44 }, { "epoch": 0.3770620581304006, "grad_norm": 0.3766598701477051, "learning_rate": 6.710100716628344e-05, "loss": 2.2953, "step": 45 }, { "epoch": 0.38544121497774286, "grad_norm": 0.5563957691192627, "learning_rate": 6.545084971874738e-05, "loss": 2.5368, "step": 46 }, { "epoch": 0.3938203718250851, "grad_norm": 0.403656542301178, "learning_rate": 6.378186779084995e-05, "loss": 2.4207, "step": 47 }, { "epoch": 0.40219952867242736, "grad_norm": 0.4429378807544708, "learning_rate": 6.209609477998338e-05, "loss": 2.637, "step": 48 }, { "epoch": 0.41057868551976956, "grad_norm": 0.46513986587524414, "learning_rate": 6.0395584540887963e-05, "loss": 2.3988, "step": 49 }, { "epoch": 0.4189578423671118, "grad_norm": 0.4939236640930176, "learning_rate": 5.868240888334653e-05, "loss": 2.6023, "step": 50 }, { "epoch": 0.4189578423671118, "eval_loss": 2.578394889831543, "eval_runtime": 42.4269, "eval_samples_per_second": 4.761, "eval_steps_per_second": 4.761, "step": 50 }, { "epoch": 0.42733699921445406, "grad_norm": 0.4209323823451996, "learning_rate": 5.695865504800327e-05, "loss": 2.4749, "step": 51 }, { "epoch": 0.43571615606179626, "grad_norm": 0.4266431927680969, "learning_rate": 5.522642316338268e-05, "loss": 2.4692, "step": 52 }, { "epoch": 0.4440953129091385, "grad_norm": 0.3953860402107239, "learning_rate": 5.348782368720626e-05, "loss": 2.3527, "step": 53 }, { "epoch": 0.45247446975648076, "grad_norm": 0.4726410210132599, "learning_rate": 5.174497483512506e-05, "loss": 2.4229, "step": 54 }, { "epoch": 0.460853626603823, "grad_norm": 0.4370991289615631, "learning_rate": 5e-05, "loss": 2.5421, "step": 55 }, { "epoch": 0.4692327834511652, "grad_norm": 0.4647808074951172, "learning_rate": 4.825502516487497e-05, "loss": 2.5684, "step": 56 }, { "epoch": 0.47761194029850745, "grad_norm": 0.4097454249858856, "learning_rate": 4.6512176312793736e-05, "loss": 2.4871, "step": 57 }, { "epoch": 0.4859910971458497, "grad_norm": 0.4504952132701874, "learning_rate": 4.477357683661734e-05, "loss": 2.4055, "step": 58 }, { "epoch": 0.49437025399319195, "grad_norm": 0.461972177028656, "learning_rate": 4.3041344951996746e-05, "loss": 2.5632, "step": 59 }, { "epoch": 0.5027494108405341, "grad_norm": 0.42533305287361145, "learning_rate": 4.131759111665349e-05, "loss": 2.4914, "step": 60 }, { "epoch": 0.5111285676878764, "grad_norm": 0.4510684311389923, "learning_rate": 3.960441545911204e-05, "loss": 2.5091, "step": 61 }, { "epoch": 0.5195077245352187, "grad_norm": 0.42407482862472534, "learning_rate": 3.790390522001662e-05, "loss": 2.3668, "step": 62 }, { "epoch": 0.5278868813825609, "grad_norm": 0.431485116481781, "learning_rate": 3.6218132209150045e-05, "loss": 2.4541, "step": 63 }, { "epoch": 0.5362660382299032, "grad_norm": 0.4215572774410248, "learning_rate": 3.4549150281252636e-05, "loss": 2.3595, "step": 64 }, { "epoch": 0.5446451950772454, "grad_norm": 0.5403610467910767, "learning_rate": 3.289899283371657e-05, "loss": 2.7011, "step": 65 }, { "epoch": 0.5530243519245875, "grad_norm": 0.43399372696876526, "learning_rate": 3.12696703292044e-05, "loss": 2.367, "step": 66 }, { "epoch": 0.5614035087719298, "grad_norm": 0.49265211820602417, "learning_rate": 2.9663167846209998e-05, "loss": 2.3943, "step": 67 }, { "epoch": 0.569782665619272, "grad_norm": 0.4823525547981262, "learning_rate": 2.8081442660546125e-05, "loss": 2.3512, "step": 68 }, { "epoch": 0.5781618224666143, "grad_norm": 0.4148399829864502, "learning_rate": 2.6526421860705473e-05, "loss": 2.4461, "step": 69 }, { "epoch": 0.5865409793139565, "grad_norm": 0.45690131187438965, "learning_rate": 2.500000000000001e-05, "loss": 2.3753, "step": 70 }, { "epoch": 0.5949201361612988, "grad_norm": 0.3864487111568451, "learning_rate": 2.350403678833976e-05, "loss": 2.3544, "step": 71 }, { "epoch": 0.603299293008641, "grad_norm": 0.4677547812461853, "learning_rate": 2.2040354826462668e-05, "loss": 2.5031, "step": 72 }, { "epoch": 0.6116784498559832, "grad_norm": 0.37837573885917664, "learning_rate": 2.061073738537635e-05, "loss": 2.3215, "step": 73 }, { "epoch": 0.6200576067033254, "grad_norm": 0.47641924023628235, "learning_rate": 1.9216926233717085e-05, "loss": 2.4183, "step": 74 }, { "epoch": 0.6284367635506677, "grad_norm": 0.5523366928100586, "learning_rate": 1.7860619515673033e-05, "loss": 2.5932, "step": 75 }, { "epoch": 0.6368159203980099, "grad_norm": 0.42463499307632446, "learning_rate": 1.6543469682057106e-05, "loss": 2.4165, "step": 76 }, { "epoch": 0.6451950772453522, "grad_norm": 0.5442836284637451, "learning_rate": 1.526708147705013e-05, "loss": 2.5349, "step": 77 }, { "epoch": 0.6535742340926944, "grad_norm": 0.40604889392852783, "learning_rate": 1.4033009983067452e-05, "loss": 2.4106, "step": 78 }, { "epoch": 0.6619533909400367, "grad_norm": 0.4163293242454529, "learning_rate": 1.2842758726130283e-05, "loss": 2.3373, "step": 79 }, { "epoch": 0.6703325477873789, "grad_norm": 0.48656710982322693, "learning_rate": 1.1697777844051105e-05, "loss": 2.5337, "step": 80 }, { "epoch": 0.6787117046347211, "grad_norm": 0.4434641897678375, "learning_rate": 1.0599462319663905e-05, "loss": 2.4307, "step": 81 }, { "epoch": 0.6870908614820633, "grad_norm": 0.4347304105758667, "learning_rate": 9.549150281252633e-06, "loss": 2.3949, "step": 82 }, { "epoch": 0.6954700183294056, "grad_norm": 0.4212183952331543, "learning_rate": 8.548121372247918e-06, "loss": 2.3767, "step": 83 }, { "epoch": 0.7038491751767478, "grad_norm": 0.44238415360450745, "learning_rate": 7.597595192178702e-06, "loss": 2.612, "step": 84 }, { "epoch": 0.7122283320240901, "grad_norm": 0.44845736026763916, "learning_rate": 6.698729810778065e-06, "loss": 2.2322, "step": 85 }, { "epoch": 0.7206074888714323, "grad_norm": 0.4716636836528778, "learning_rate": 5.852620357053651e-06, "loss": 2.5347, "step": 86 }, { "epoch": 0.7289866457187746, "grad_norm": 0.5085580945014954, "learning_rate": 5.060297685041659e-06, "loss": 2.4772, "step": 87 }, { "epoch": 0.7373658025661168, "grad_norm": 0.455152302980423, "learning_rate": 4.322727117869951e-06, "loss": 2.3143, "step": 88 }, { "epoch": 0.745744959413459, "grad_norm": 0.48934170603752136, "learning_rate": 3.6408072716606346e-06, "loss": 2.3811, "step": 89 }, { "epoch": 0.7541241162608012, "grad_norm": 0.4595421552658081, "learning_rate": 3.0153689607045845e-06, "loss": 2.4414, "step": 90 }, { "epoch": 0.7625032731081435, "grad_norm": 0.5183727741241455, "learning_rate": 2.4471741852423237e-06, "loss": 2.5257, "step": 91 }, { "epoch": 0.7708824299554857, "grad_norm": 0.42374536395072937, "learning_rate": 1.9369152030840556e-06, "loss": 2.4218, "step": 92 }, { "epoch": 0.779261586802828, "grad_norm": 0.45261675119400024, "learning_rate": 1.4852136862001764e-06, "loss": 2.4601, "step": 93 }, { "epoch": 0.7876407436501702, "grad_norm": 0.4387091100215912, "learning_rate": 1.0926199633097157e-06, "loss": 2.5303, "step": 94 }, { "epoch": 0.7960199004975125, "grad_norm": 0.42841577529907227, "learning_rate": 7.596123493895991e-07, "loss": 2.3097, "step": 95 }, { "epoch": 0.8043990573448547, "grad_norm": 0.47183293104171753, "learning_rate": 4.865965629214819e-07, "loss": 2.4747, "step": 96 }, { "epoch": 0.8127782141921969, "grad_norm": 0.5696967840194702, "learning_rate": 2.7390523158633554e-07, "loss": 2.4984, "step": 97 }, { "epoch": 0.8211573710395391, "grad_norm": 0.4509386718273163, "learning_rate": 1.2179748700879012e-07, "loss": 2.2953, "step": 98 }, { "epoch": 0.8295365278868814, "grad_norm": 0.5444111227989197, "learning_rate": 3.04586490452119e-08, "loss": 2.5414, "step": 99 }, { "epoch": 0.8379156847342236, "grad_norm": 0.4556747078895569, "learning_rate": 0.0, "loss": 2.4252, "step": 100 }, { "epoch": 0.8379156847342236, "eval_loss": 2.5602805614471436, "eval_runtime": 45.2034, "eval_samples_per_second": 4.469, "eval_steps_per_second": 4.469, "step": 100 } ], "logging_steps": 1, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 25, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 6.51499696816128e+16, "train_batch_size": 1, "trial_name": null, "trial_params": null }