| { |
| "best_global_step": 20000, |
| "best_metric": 5.635913372039795, |
| "best_model_checkpoint": "tinybert_base_train_book_ent_15p_s_init/checkpoint-20000", |
| "epoch": 24.0, |
| "eval_steps": 10000, |
| "global_step": 27456, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.4370629370629371, |
| "grad_norm": 3.084730386734009, |
| "learning_rate": 4.9900000000000005e-06, |
| "loss": 7.2145, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8741258741258742, |
| "grad_norm": 1.6794264316558838, |
| "learning_rate": 9.990000000000001e-06, |
| "loss": 6.6993, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.3111888111888113, |
| "grad_norm": 0.43600839376449585, |
| "learning_rate": 1.499e-05, |
| "loss": 6.4656, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.7482517482517483, |
| "grad_norm": 0.5343518853187561, |
| "learning_rate": 1.999e-05, |
| "loss": 6.0652, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.1853146853146854, |
| "grad_norm": 3.4993603229522705, |
| "learning_rate": 2.4990000000000003e-05, |
| "loss": 4.2034, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.6223776223776225, |
| "grad_norm": 0.8302733898162842, |
| "learning_rate": 2.9990000000000003e-05, |
| "loss": 4.1233, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.0594405594405596, |
| "grad_norm": 1.1415528059005737, |
| "learning_rate": 3.499e-05, |
| "loss": 6.0294, |
| "step": 3500 |
| }, |
| { |
| "epoch": 3.4965034965034967, |
| "grad_norm": 0.3220045864582062, |
| "learning_rate": 3.999e-05, |
| "loss": 7.6921, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.9335664335664333, |
| "grad_norm": 2.5710136890411377, |
| "learning_rate": 4.499e-05, |
| "loss": 7.4362, |
| "step": 4500 |
| }, |
| { |
| "epoch": 4.370629370629371, |
| "grad_norm": 1.61074960231781, |
| "learning_rate": 4.999e-05, |
| "loss": 6.096, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.8076923076923075, |
| "grad_norm": 0.29658418893814087, |
| "learning_rate": 5.499000000000001e-05, |
| "loss": 6.5745, |
| "step": 5500 |
| }, |
| { |
| "epoch": 5.244755244755245, |
| "grad_norm": 0.688091516494751, |
| "learning_rate": 5.999e-05, |
| "loss": 6.444, |
| "step": 6000 |
| }, |
| { |
| "epoch": 5.681818181818182, |
| "grad_norm": 0.6230639219284058, |
| "learning_rate": 6.499000000000001e-05, |
| "loss": 6.9739, |
| "step": 6500 |
| }, |
| { |
| "epoch": 6.118881118881119, |
| "grad_norm": 1.1754754781723022, |
| "learning_rate": 6.999e-05, |
| "loss": 5.7531, |
| "step": 7000 |
| }, |
| { |
| "epoch": 6.555944055944056, |
| "grad_norm": 4.404361248016357, |
| "learning_rate": 7.499e-05, |
| "loss": 5.3091, |
| "step": 7500 |
| }, |
| { |
| "epoch": 6.993006993006993, |
| "grad_norm": 0.781994104385376, |
| "learning_rate": 7.999000000000001e-05, |
| "loss": 5.9669, |
| "step": 8000 |
| }, |
| { |
| "epoch": 7.43006993006993, |
| "grad_norm": 1.3538726568222046, |
| "learning_rate": 8.499e-05, |
| "loss": 6.3625, |
| "step": 8500 |
| }, |
| { |
| "epoch": 7.867132867132867, |
| "grad_norm": 2.3600542545318604, |
| "learning_rate": 8.999000000000001e-05, |
| "loss": 5.153, |
| "step": 9000 |
| }, |
| { |
| "epoch": 8.304195804195805, |
| "grad_norm": 1.3912559747695923, |
| "learning_rate": 9.499e-05, |
| "loss": 6.0561, |
| "step": 9500 |
| }, |
| { |
| "epoch": 8.741258741258742, |
| "grad_norm": 1.6746373176574707, |
| "learning_rate": 9.999000000000001e-05, |
| "loss": 5.3929, |
| "step": 10000 |
| }, |
| { |
| "epoch": 8.741258741258742, |
| "eval_accuracy": 0.1617547379551555, |
| "eval_loss": 5.967167854309082, |
| "eval_runtime": 1.926, |
| "eval_samples_per_second": 248.696, |
| "eval_steps_per_second": 1.558, |
| "step": 10000 |
| }, |
| { |
| "epoch": 9.178321678321678, |
| "grad_norm": 1.5575922727584839, |
| "learning_rate": 9.714138405132907e-05, |
| "loss": 6.1957, |
| "step": 10500 |
| }, |
| { |
| "epoch": 9.615384615384615, |
| "grad_norm": 1.0062355995178223, |
| "learning_rate": 9.427703941338222e-05, |
| "loss": 5.8074, |
| "step": 11000 |
| }, |
| { |
| "epoch": 10.052447552447552, |
| "grad_norm": 0.869584321975708, |
| "learning_rate": 9.141269477543538e-05, |
| "loss": 4.0817, |
| "step": 11500 |
| }, |
| { |
| "epoch": 10.48951048951049, |
| "grad_norm": 1.2013421058654785, |
| "learning_rate": 8.854835013748855e-05, |
| "loss": 3.59, |
| "step": 12000 |
| }, |
| { |
| "epoch": 10.926573426573427, |
| "grad_norm": 0.9844303727149963, |
| "learning_rate": 8.568400549954171e-05, |
| "loss": 4.4935, |
| "step": 12500 |
| }, |
| { |
| "epoch": 11.363636363636363, |
| "grad_norm": 1.0267465114593506, |
| "learning_rate": 8.281966086159487e-05, |
| "loss": 4.2636, |
| "step": 13000 |
| }, |
| { |
| "epoch": 11.8006993006993, |
| "grad_norm": 1.0032545328140259, |
| "learning_rate": 7.995531622364803e-05, |
| "loss": 4.6715, |
| "step": 13500 |
| }, |
| { |
| "epoch": 12.237762237762238, |
| "grad_norm": 1.1699270009994507, |
| "learning_rate": 7.70909715857012e-05, |
| "loss": 5.0537, |
| "step": 14000 |
| }, |
| { |
| "epoch": 12.674825174825175, |
| "grad_norm": 1.651526927947998, |
| "learning_rate": 7.422662694775435e-05, |
| "loss": 5.8745, |
| "step": 14500 |
| }, |
| { |
| "epoch": 13.111888111888112, |
| "grad_norm": 1.1928800344467163, |
| "learning_rate": 7.136228230980752e-05, |
| "loss": 6.2548, |
| "step": 15000 |
| }, |
| { |
| "epoch": 13.548951048951048, |
| "grad_norm": 1.213062047958374, |
| "learning_rate": 6.849793767186068e-05, |
| "loss": 6.3592, |
| "step": 15500 |
| }, |
| { |
| "epoch": 13.986013986013987, |
| "grad_norm": 1.3651458024978638, |
| "learning_rate": 6.563359303391385e-05, |
| "loss": 6.3495, |
| "step": 16000 |
| }, |
| { |
| "epoch": 14.423076923076923, |
| "grad_norm": 1.116390347480774, |
| "learning_rate": 6.276924839596701e-05, |
| "loss": 6.2993, |
| "step": 16500 |
| }, |
| { |
| "epoch": 14.86013986013986, |
| "grad_norm": 1.324367642402649, |
| "learning_rate": 5.990490375802017e-05, |
| "loss": 6.2516, |
| "step": 17000 |
| }, |
| { |
| "epoch": 15.297202797202797, |
| "grad_norm": 1.1858265399932861, |
| "learning_rate": 5.7040559120073336e-05, |
| "loss": 6.1861, |
| "step": 17500 |
| }, |
| { |
| "epoch": 15.734265734265735, |
| "grad_norm": 1.1394177675247192, |
| "learning_rate": 5.4176214482126495e-05, |
| "loss": 6.0894, |
| "step": 18000 |
| }, |
| { |
| "epoch": 16.17132867132867, |
| "grad_norm": 1.223654866218567, |
| "learning_rate": 5.131186984417965e-05, |
| "loss": 6.0153, |
| "step": 18500 |
| }, |
| { |
| "epoch": 16.60839160839161, |
| "grad_norm": 1.0455007553100586, |
| "learning_rate": 4.844752520623282e-05, |
| "loss": 5.9313, |
| "step": 19000 |
| }, |
| { |
| "epoch": 17.045454545454547, |
| "grad_norm": 1.1198850870132446, |
| "learning_rate": 4.558318056828598e-05, |
| "loss": 5.8995, |
| "step": 19500 |
| }, |
| { |
| "epoch": 17.482517482517483, |
| "grad_norm": 1.2449570894241333, |
| "learning_rate": 4.271883593033914e-05, |
| "loss": 5.8297, |
| "step": 20000 |
| }, |
| { |
| "epoch": 17.482517482517483, |
| "eval_accuracy": 0.1853157281132229, |
| "eval_loss": 5.635913372039795, |
| "eval_runtime": 1.9422, |
| "eval_samples_per_second": 246.63, |
| "eval_steps_per_second": 1.545, |
| "step": 20000 |
| }, |
| { |
| "epoch": 17.91958041958042, |
| "grad_norm": 1.0795655250549316, |
| "learning_rate": 3.98544912923923e-05, |
| "loss": 5.7946, |
| "step": 20500 |
| }, |
| { |
| "epoch": 18.356643356643357, |
| "grad_norm": 1.3464816808700562, |
| "learning_rate": 3.699014665444546e-05, |
| "loss": 5.742, |
| "step": 21000 |
| }, |
| { |
| "epoch": 18.793706293706293, |
| "grad_norm": 1.2070534229278564, |
| "learning_rate": 3.4125802016498626e-05, |
| "loss": 5.725, |
| "step": 21500 |
| }, |
| { |
| "epoch": 19.23076923076923, |
| "grad_norm": 0.9267760515213013, |
| "learning_rate": 3.126145737855179e-05, |
| "loss": 5.6568, |
| "step": 22000 |
| }, |
| { |
| "epoch": 19.667832167832167, |
| "grad_norm": 1.0271191596984863, |
| "learning_rate": 2.8397112740604954e-05, |
| "loss": 5.6281, |
| "step": 22500 |
| }, |
| { |
| "epoch": 20.104895104895103, |
| "grad_norm": 1.1037367582321167, |
| "learning_rate": 2.5532768102658113e-05, |
| "loss": 5.6152, |
| "step": 23000 |
| }, |
| { |
| "epoch": 20.541958041958043, |
| "grad_norm": 1.0020090341567993, |
| "learning_rate": 2.2668423464711275e-05, |
| "loss": 5.5562, |
| "step": 23500 |
| }, |
| { |
| "epoch": 20.97902097902098, |
| "grad_norm": 1.1875511407852173, |
| "learning_rate": 1.9804078826764437e-05, |
| "loss": 5.5316, |
| "step": 24000 |
| }, |
| { |
| "epoch": 21.416083916083917, |
| "grad_norm": 1.1084744930267334, |
| "learning_rate": 1.69397341888176e-05, |
| "loss": 5.5283, |
| "step": 24500 |
| }, |
| { |
| "epoch": 21.853146853146853, |
| "grad_norm": 1.079026699066162, |
| "learning_rate": 1.4075389550870763e-05, |
| "loss": 5.5119, |
| "step": 25000 |
| }, |
| { |
| "epoch": 22.29020979020979, |
| "grad_norm": 1.0321869850158691, |
| "learning_rate": 1.1211044912923923e-05, |
| "loss": 5.4713, |
| "step": 25500 |
| }, |
| { |
| "epoch": 22.727272727272727, |
| "grad_norm": 1.1838607788085938, |
| "learning_rate": 8.346700274977086e-06, |
| "loss": 5.4734, |
| "step": 26000 |
| }, |
| { |
| "epoch": 23.164335664335663, |
| "grad_norm": 1.1609679460525513, |
| "learning_rate": 5.482355637030248e-06, |
| "loss": 5.4625, |
| "step": 26500 |
| }, |
| { |
| "epoch": 23.6013986013986, |
| "grad_norm": 0.9143561720848083, |
| "learning_rate": 2.61801099908341e-06, |
| "loss": 5.4393, |
| "step": 27000 |
| }, |
| { |
| "epoch": 24.0, |
| "step": 27456, |
| "total_flos": 2.175368143688663e+17, |
| "train_loss": 5.765684167821925, |
| "train_runtime": 24142.3828, |
| "train_samples_per_second": 227.291, |
| "train_steps_per_second": 1.137 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 27456, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 24, |
| "save_steps": 10000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.175368143688663e+17, |
| "train_batch_size": 200, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
