| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0021892057561516682, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.9189410082022242e-05, | |
| "grad_norm": 6.6969146728515625, | |
| "learning_rate": 1e-05, | |
| "loss": 6.6322, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.9189410082022242e-05, | |
| "eval_loss": 4.265627384185791, | |
| "eval_runtime": 5169.4317, | |
| "eval_samples_per_second": 5.581, | |
| "eval_steps_per_second": 0.698, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.8378820164044484e-05, | |
| "grad_norm": 9.290034294128418, | |
| "learning_rate": 2e-05, | |
| "loss": 9.203, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 8.756823024606673e-05, | |
| "grad_norm": 7.84418249130249, | |
| "learning_rate": 3e-05, | |
| "loss": 7.8093, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00011675764032808897, | |
| "grad_norm": 8.62817668914795, | |
| "learning_rate": 4e-05, | |
| "loss": 8.9271, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0001459470504101112, | |
| "grad_norm": 8.213288307189941, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2468, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00017513646049213346, | |
| "grad_norm": 10.377052307128906, | |
| "learning_rate": 6e-05, | |
| "loss": 9.0711, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0002043258705741557, | |
| "grad_norm": 9.143661499023438, | |
| "learning_rate": 7e-05, | |
| "loss": 7.3199, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00023351528065617793, | |
| "grad_norm": 8.905694961547852, | |
| "learning_rate": 8e-05, | |
| "loss": 6.425, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0002627046907382002, | |
| "grad_norm": 7.937708854675293, | |
| "learning_rate": 9e-05, | |
| "loss": 6.1835, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002627046907382002, | |
| "eval_loss": 2.7248966693878174, | |
| "eval_runtime": 5178.8146, | |
| "eval_samples_per_second": 5.571, | |
| "eval_steps_per_second": 0.696, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002918941008202224, | |
| "grad_norm": 7.350879669189453, | |
| "learning_rate": 0.0001, | |
| "loss": 5.3104, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00032108351090224467, | |
| "grad_norm": 7.203503131866455, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 5.7091, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00035027292098426693, | |
| "grad_norm": 6.413120746612549, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.3065, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00037946233106628914, | |
| "grad_norm": 5.061990737915039, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 3.019, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0004086517411483114, | |
| "grad_norm": 5.457662105560303, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 3.4817, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00043784115123033366, | |
| "grad_norm": 4.922332763671875, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.3424, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00046703056131235587, | |
| "grad_norm": 6.326183319091797, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.8058, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0004962199713943781, | |
| "grad_norm": 4.144868850708008, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.8318, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0005254093814764004, | |
| "grad_norm": 5.910154819488525, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.0695, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0005254093814764004, | |
| "eval_loss": 1.414214015007019, | |
| "eval_runtime": 5168.8987, | |
| "eval_samples_per_second": 5.581, | |
| "eval_steps_per_second": 0.698, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0005545987915584227, | |
| "grad_norm": 4.4212799072265625, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.4084, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0005837882016404448, | |
| "grad_norm": 6.185261249542236, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 2.6826, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006129776117224671, | |
| "grad_norm": 4.956446170806885, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.3085, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0006421670218044893, | |
| "grad_norm": 4.5698676109313965, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 2.2213, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0006713564318865116, | |
| "grad_norm": 5.868464469909668, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.9316, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0007005458419685339, | |
| "grad_norm": 4.061339855194092, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 2.3513, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.000729735252050556, | |
| "grad_norm": 5.618566513061523, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.9962, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0007589246621325783, | |
| "grad_norm": 5.740274429321289, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 2.9059, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0007881140722146005, | |
| "grad_norm": 4.832045555114746, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.051, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0007881140722146005, | |
| "eval_loss": 1.29904043674469, | |
| "eval_runtime": 5166.4607, | |
| "eval_samples_per_second": 5.584, | |
| "eval_steps_per_second": 0.698, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0008173034822966228, | |
| "grad_norm": 5.021420955657959, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.949, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0008464928923786451, | |
| "grad_norm": 7.856975555419922, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 2.8508, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0008756823024606673, | |
| "grad_norm": 5.426181316375732, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.5198, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009048717125426895, | |
| "grad_norm": 5.714426040649414, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 2.613, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0009340611226247117, | |
| "grad_norm": 3.9779551029205322, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 1.9224, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.000963250532706734, | |
| "grad_norm": 5.584859371185303, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 2.6696, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0009924399427887563, | |
| "grad_norm": 7.049388885498047, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 2.5729, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0010216293528707784, | |
| "grad_norm": 9.85444450378418, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 2.7488, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0010508187629528008, | |
| "grad_norm": 5.08799934387207, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.9356, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0010508187629528008, | |
| "eval_loss": 1.2538187503814697, | |
| "eval_runtime": 5165.2612, | |
| "eval_samples_per_second": 5.585, | |
| "eval_steps_per_second": 0.698, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.001080008173034823, | |
| "grad_norm": 4.305152893066406, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.1296, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0011091975831168453, | |
| "grad_norm": 4.324735164642334, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 2.434, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0011383869931988675, | |
| "grad_norm": 5.634093761444092, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.0592, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0011675764032808896, | |
| "grad_norm": 3.8521158695220947, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 2.2043, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001196765813362912, | |
| "grad_norm": 4.983729839324951, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 3.2767, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0012259552234449341, | |
| "grad_norm": 3.823223829269409, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.1992, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0012551446335269565, | |
| "grad_norm": 5.168725490570068, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 2.8579, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0012843340436089787, | |
| "grad_norm": 3.6878905296325684, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 1.9881, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0013135234536910008, | |
| "grad_norm": 5.2658467292785645, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8113, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0013135234536910008, | |
| "eval_loss": 1.2026046514511108, | |
| "eval_runtime": 5163.7463, | |
| "eval_samples_per_second": 5.587, | |
| "eval_steps_per_second": 0.699, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0013427128637730232, | |
| "grad_norm": 4.249842166900635, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.1704, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0013719022738550453, | |
| "grad_norm": 4.676008701324463, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 2.5203, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0014010916839370677, | |
| "grad_norm": 5.7045817375183105, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 2.4709, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0014302810940190899, | |
| "grad_norm": 4.275635719299316, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 2.595, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.001459470504101112, | |
| "grad_norm": 5.484550476074219, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 2.4328, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0014886599141831344, | |
| "grad_norm": 4.854852199554443, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.8273, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0015178493242651565, | |
| "grad_norm": 3.8842687606811523, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 2.2676, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.001547038734347179, | |
| "grad_norm": 4.392478942871094, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 2.5662, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.001576228144429201, | |
| "grad_norm": 3.572255849838257, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.027, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.001576228144429201, | |
| "eval_loss": 1.175163984298706, | |
| "eval_runtime": 5163.0006, | |
| "eval_samples_per_second": 5.588, | |
| "eval_steps_per_second": 0.699, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0016054175545112232, | |
| "grad_norm": 7.922635078430176, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0359, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0016346069645932456, | |
| "grad_norm": 5.117091655731201, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 2.586, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.0016637963746752677, | |
| "grad_norm": 4.736616611480713, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.9285, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0016929857847572901, | |
| "grad_norm": 3.8517062664031982, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 2.0243, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.0017221751948393123, | |
| "grad_norm": 4.804389476776123, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 2.3802, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0017513646049213346, | |
| "grad_norm": 3.5435945987701416, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 2.0933, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0017805540150033568, | |
| "grad_norm": 3.101043939590454, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 1.9417, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.001809743425085379, | |
| "grad_norm": 3.5599617958068848, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 2.2412, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.0018389328351674013, | |
| "grad_norm": 5.158842086791992, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.2728, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0018389328351674013, | |
| "eval_loss": 1.149397850036621, | |
| "eval_runtime": 5164.5613, | |
| "eval_samples_per_second": 5.586, | |
| "eval_steps_per_second": 0.698, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0018681222452494235, | |
| "grad_norm": 5.161924839019775, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.4739, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.0018973116553314458, | |
| "grad_norm": 3.7883992195129395, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.6341, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.001926501065413468, | |
| "grad_norm": 6.4142889976501465, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.7093, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.00195569047549549, | |
| "grad_norm": 3.7092437744140625, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 1.8984, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0019848798855775125, | |
| "grad_norm": 3.7973620891571045, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 1.9869, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.002014069295659535, | |
| "grad_norm": 4.184591293334961, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.0964, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.002043258705741557, | |
| "grad_norm": 3.5979950428009033, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.3372, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.002072448115823579, | |
| "grad_norm": 4.4518141746521, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 2.0167, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.0021016375259056016, | |
| "grad_norm": 4.17356538772583, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.1772, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0021016375259056016, | |
| "eval_loss": 1.136897325515747, | |
| "eval_runtime": 5168.1485, | |
| "eval_samples_per_second": 5.582, | |
| "eval_steps_per_second": 0.698, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0021308269359876235, | |
| "grad_norm": 4.173731803894043, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.9867, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.002160016346069646, | |
| "grad_norm": 4.41037654876709, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 2.3544, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.0021892057561516682, | |
| "grad_norm": 3.905989170074463, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.0434, | |
| "step": 75 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1108631969792e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |