| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2518891687657431, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005037783375314861, | |
| "grad_norm": 0.1343812644481659, | |
| "learning_rate": 2e-05, | |
| "loss": 11.7643, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005037783375314861, | |
| "eval_loss": 11.775632858276367, | |
| "eval_runtime": 0.6521, | |
| "eval_samples_per_second": 128.824, | |
| "eval_steps_per_second": 64.412, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010075566750629723, | |
| "grad_norm": 0.17478109896183014, | |
| "learning_rate": 4e-05, | |
| "loss": 11.7898, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.015113350125944584, | |
| "grad_norm": 0.17163191735744476, | |
| "learning_rate": 6e-05, | |
| "loss": 11.7762, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020151133501259445, | |
| "grad_norm": 0.1776304692029953, | |
| "learning_rate": 8e-05, | |
| "loss": 11.7739, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020151133501259445, | |
| "eval_loss": 11.775409698486328, | |
| "eval_runtime": 0.6545, | |
| "eval_samples_per_second": 128.341, | |
| "eval_steps_per_second": 64.17, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02518891687657431, | |
| "grad_norm": 0.20767977833747864, | |
| "learning_rate": 0.0001, | |
| "loss": 11.7652, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.030226700251889168, | |
| "grad_norm": 0.1783723384141922, | |
| "learning_rate": 0.00012, | |
| "loss": 11.7811, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03526448362720403, | |
| "grad_norm": 0.2075798362493515, | |
| "learning_rate": 0.00014, | |
| "loss": 11.7613, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04030226700251889, | |
| "grad_norm": 0.17259609699249268, | |
| "learning_rate": 0.00016, | |
| "loss": 11.7766, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.04030226700251889, | |
| "eval_loss": 11.7745943069458, | |
| "eval_runtime": 0.6545, | |
| "eval_samples_per_second": 128.344, | |
| "eval_steps_per_second": 64.172, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.04534005037783375, | |
| "grad_norm": 0.16603446006774902, | |
| "learning_rate": 0.00018, | |
| "loss": 11.7747, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05037783375314862, | |
| "grad_norm": 0.16154415905475616, | |
| "learning_rate": 0.0002, | |
| "loss": 11.7738, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.055415617128463476, | |
| "grad_norm": 0.21826648712158203, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 11.7761, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.060453400503778336, | |
| "grad_norm": 0.2048770934343338, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 11.7763, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.060453400503778336, | |
| "eval_loss": 11.773273468017578, | |
| "eval_runtime": 0.6781, | |
| "eval_samples_per_second": 123.871, | |
| "eval_steps_per_second": 61.935, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0654911838790932, | |
| "grad_norm": 0.12033865600824356, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 11.7726, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.07052896725440806, | |
| "grad_norm": 0.1583612710237503, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 11.772, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.07556675062972293, | |
| "grad_norm": 0.1649715006351471, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 11.7676, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08060453400503778, | |
| "grad_norm": 0.1502542793750763, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 11.7657, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.08060453400503778, | |
| "eval_loss": 11.771785736083984, | |
| "eval_runtime": 0.6579, | |
| "eval_samples_per_second": 127.688, | |
| "eval_steps_per_second": 63.844, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.08564231738035265, | |
| "grad_norm": 0.15804621577262878, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 11.7648, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0906801007556675, | |
| "grad_norm": 0.1290988177061081, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 11.7736, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09571788413098237, | |
| "grad_norm": 0.15611040592193604, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 11.7751, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.10075566750629723, | |
| "grad_norm": 0.1558377742767334, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 11.7821, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10075566750629723, | |
| "eval_loss": 11.77033805847168, | |
| "eval_runtime": 0.6652, | |
| "eval_samples_per_second": 126.274, | |
| "eval_steps_per_second": 63.137, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10579345088161209, | |
| "grad_norm": 0.14136968553066254, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 11.7598, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11083123425692695, | |
| "grad_norm": 0.1827390193939209, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 11.7637, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.11586901763224182, | |
| "grad_norm": 0.146404430270195, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 11.7784, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.12090680100755667, | |
| "grad_norm": 0.27841717004776, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 11.7707, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12090680100755667, | |
| "eval_loss": 11.768901824951172, | |
| "eval_runtime": 0.6686, | |
| "eval_samples_per_second": 125.642, | |
| "eval_steps_per_second": 62.821, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12594458438287154, | |
| "grad_norm": 0.15636301040649414, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 11.7764, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1309823677581864, | |
| "grad_norm": 0.17114318907260895, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 11.7703, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.13602015113350127, | |
| "grad_norm": 0.20548300445079803, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 11.7773, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.14105793450881612, | |
| "grad_norm": 0.15028426051139832, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 11.7642, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.14105793450881612, | |
| "eval_loss": 11.767555236816406, | |
| "eval_runtime": 0.6574, | |
| "eval_samples_per_second": 127.78, | |
| "eval_steps_per_second": 63.89, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.14609571788413098, | |
| "grad_norm": 0.13849467039108276, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 11.7754, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.15113350125944586, | |
| "grad_norm": 0.24436236917972565, | |
| "learning_rate": 0.0001, | |
| "loss": 11.777, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1561712846347607, | |
| "grad_norm": 0.12146396189928055, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 11.7583, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.16120906801007556, | |
| "grad_norm": 0.13603943586349487, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 11.7767, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.16120906801007556, | |
| "eval_loss": 11.76652717590332, | |
| "eval_runtime": 0.6557, | |
| "eval_samples_per_second": 128.109, | |
| "eval_steps_per_second": 64.054, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.16624685138539042, | |
| "grad_norm": 0.1841493397951126, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 11.7572, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1712846347607053, | |
| "grad_norm": 0.1587410420179367, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 11.7721, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.17632241813602015, | |
| "grad_norm": 0.1748318076133728, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 11.7697, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.181360201511335, | |
| "grad_norm": 0.15539947152137756, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 11.7722, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.181360201511335, | |
| "eval_loss": 11.765742301940918, | |
| "eval_runtime": 0.6575, | |
| "eval_samples_per_second": 127.75, | |
| "eval_steps_per_second": 63.875, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18639798488664988, | |
| "grad_norm": 0.1843356043100357, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 11.7612, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.19143576826196473, | |
| "grad_norm": 0.14512236416339874, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 11.7649, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1964735516372796, | |
| "grad_norm": 0.14143380522727966, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 11.7659, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.20151133501259447, | |
| "grad_norm": 0.15115749835968018, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 11.7692, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20151133501259447, | |
| "eval_loss": 11.76524829864502, | |
| "eval_runtime": 0.6559, | |
| "eval_samples_per_second": 128.077, | |
| "eval_steps_per_second": 64.039, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20654911838790932, | |
| "grad_norm": 0.16189849376678467, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 11.7604, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.21158690176322417, | |
| "grad_norm": 0.15517611801624298, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 11.7661, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.21662468513853905, | |
| "grad_norm": 0.15952840447425842, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 11.7654, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.2216624685138539, | |
| "grad_norm": 0.18206751346588135, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 11.7605, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.2216624685138539, | |
| "eval_loss": 11.764971733093262, | |
| "eval_runtime": 0.6686, | |
| "eval_samples_per_second": 125.638, | |
| "eval_steps_per_second": 62.819, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.22670025188916876, | |
| "grad_norm": 0.14965149760246277, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 11.7675, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.23173803526448364, | |
| "grad_norm": 0.16042684018611908, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 11.7729, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.2367758186397985, | |
| "grad_norm": 0.2848449945449829, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 11.7682, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.24181360201511334, | |
| "grad_norm": 0.16414609551429749, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 11.7582, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.24181360201511334, | |
| "eval_loss": 11.764909744262695, | |
| "eval_runtime": 0.6556, | |
| "eval_samples_per_second": 128.135, | |
| "eval_steps_per_second": 64.068, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.24685138539042822, | |
| "grad_norm": 0.16592900454998016, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 11.7754, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.2518891687657431, | |
| "grad_norm": 0.19010961055755615, | |
| "learning_rate": 0.0, | |
| "loss": 11.7793, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 10185002188800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |