| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0020927507115352417, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.1855014230704836e-05, | |
| "grad_norm": 0.23623579740524292, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1918, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.1855014230704836e-05, | |
| "eval_loss": 0.16839955747127533, | |
| "eval_runtime": 950.9919, | |
| "eval_samples_per_second": 10.578, | |
| "eval_steps_per_second": 5.289, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.371002846140967e-05, | |
| "grad_norm": 0.2619614005088806, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1554, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00012556504269211453, | |
| "grad_norm": 0.21573609113693237, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1409, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00016742005692281934, | |
| "grad_norm": 0.20257234573364258, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1308, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00020927507115352419, | |
| "grad_norm": 0.22933052480220795, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1471, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00020927507115352419, | |
| "eval_loss": 0.15534359216690063, | |
| "eval_runtime": 956.5429, | |
| "eval_samples_per_second": 10.517, | |
| "eval_steps_per_second": 5.259, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00025113008538422905, | |
| "grad_norm": 0.14353002607822418, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0659, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00029298509961493387, | |
| "grad_norm": 0.18972477316856384, | |
| "learning_rate": 0.00014, | |
| "loss": 0.1075, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003348401138456387, | |
| "grad_norm": 0.21828801929950714, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1016, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00037669512807634355, | |
| "grad_norm": 0.26345500349998474, | |
| "learning_rate": 0.00018, | |
| "loss": 0.1297, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00041855014230704837, | |
| "grad_norm": 0.2362617701292038, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0807, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00041855014230704837, | |
| "eval_loss": 0.07172638177871704, | |
| "eval_runtime": 956.2922, | |
| "eval_samples_per_second": 10.52, | |
| "eval_steps_per_second": 5.26, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00046040515653775324, | |
| "grad_norm": 0.1236543208360672, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0325, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0005022601707684581, | |
| "grad_norm": 0.23662090301513672, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0623, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0005441151849991629, | |
| "grad_norm": 0.17430146038532257, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0378, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0005859701992298677, | |
| "grad_norm": 0.18858376145362854, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.031, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0006278252134605726, | |
| "grad_norm": 0.25688692927360535, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0241, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006278252134605726, | |
| "eval_loss": 0.028141211718320847, | |
| "eval_runtime": 955.7785, | |
| "eval_samples_per_second": 10.525, | |
| "eval_steps_per_second": 5.263, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006696802276912774, | |
| "grad_norm": 0.17680053412914276, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.021, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0007115352419219822, | |
| "grad_norm": 0.13639086484909058, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0156, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0007533902561526871, | |
| "grad_norm": 0.23436982929706573, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0199, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.000795245270383392, | |
| "grad_norm": 0.16494698822498322, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0099, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0008371002846140967, | |
| "grad_norm": 0.1418473720550537, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0136, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008371002846140967, | |
| "eval_loss": 0.011286907829344273, | |
| "eval_runtime": 955.9815, | |
| "eval_samples_per_second": 10.523, | |
| "eval_steps_per_second": 5.262, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008789552988448016, | |
| "grad_norm": 0.06463932991027832, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.004, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0009208103130755065, | |
| "grad_norm": 0.10179118812084198, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0088, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0009626653273062112, | |
| "grad_norm": 0.07703794538974762, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0042, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0010045203415369162, | |
| "grad_norm": 0.214707612991333, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0082, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0010463753557676209, | |
| "grad_norm": 0.10525104403495789, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0064, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0010463753557676209, | |
| "eval_loss": 0.007706103380769491, | |
| "eval_runtime": 955.9544, | |
| "eval_samples_per_second": 10.524, | |
| "eval_steps_per_second": 5.262, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0010882303699983257, | |
| "grad_norm": 0.054454121738672256, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0045, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0011300853842290306, | |
| "grad_norm": 0.17281416058540344, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0074, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0011719403984597355, | |
| "grad_norm": 0.12564797699451447, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0037, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0012137954126904403, | |
| "grad_norm": 0.13228414952754974, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0029, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0012556504269211452, | |
| "grad_norm": 0.20889438688755035, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0114, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0012556504269211452, | |
| "eval_loss": 0.00621896144002676, | |
| "eval_runtime": 955.609, | |
| "eval_samples_per_second": 10.527, | |
| "eval_steps_per_second": 5.264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00129750544115185, | |
| "grad_norm": 0.07152388989925385, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.005, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0013393604553825547, | |
| "grad_norm": 0.2723340392112732, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0044, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0013812154696132596, | |
| "grad_norm": 0.04145659878849983, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0016, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0014230704838439645, | |
| "grad_norm": 0.06280238926410675, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0019, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0014649254980746694, | |
| "grad_norm": 0.16065062582492828, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0046, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0014649254980746694, | |
| "eval_loss": 0.004886394366621971, | |
| "eval_runtime": 955.9011, | |
| "eval_samples_per_second": 10.524, | |
| "eval_steps_per_second": 5.262, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0015067805123053742, | |
| "grad_norm": 0.18025602400302887, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0063, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.001548635526536079, | |
| "grad_norm": 0.12437538802623749, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0032, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.001590490540766784, | |
| "grad_norm": 0.08849278837442398, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0021, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0016323455549974886, | |
| "grad_norm": 0.034282781183719635, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0013, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0016742005692281935, | |
| "grad_norm": 0.06674959510564804, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0017, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0016742005692281935, | |
| "eval_loss": 0.004585847724229097, | |
| "eval_runtime": 956.0546, | |
| "eval_samples_per_second": 10.522, | |
| "eval_steps_per_second": 5.261, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0017160555834588984, | |
| "grad_norm": 0.028878623619675636, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0012, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0017579105976896032, | |
| "grad_norm": 0.15476588904857635, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.006, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.001799765611920308, | |
| "grad_norm": 0.06052372604608536, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0022, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.001841620626151013, | |
| "grad_norm": 0.01776609756052494, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0009, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0018834756403817178, | |
| "grad_norm": 0.1711350679397583, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0118, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0018834756403817178, | |
| "eval_loss": 0.004424721002578735, | |
| "eval_runtime": 956.2757, | |
| "eval_samples_per_second": 10.52, | |
| "eval_steps_per_second": 5.26, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0019253306546124225, | |
| "grad_norm": 0.12957611680030823, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0053, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0019671856688431274, | |
| "grad_norm": 0.05963761731982231, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.004, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0020090406830738324, | |
| "grad_norm": 0.18093325197696686, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.008, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.002050895697304537, | |
| "grad_norm": 0.24867264926433563, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0088, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0020927507115352417, | |
| "grad_norm": 0.02437353879213333, | |
| "learning_rate": 0.0, | |
| "loss": 0.0016, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0020927507115352417, | |
| "eval_loss": 0.004385404288768768, | |
| "eval_runtime": 956.438, | |
| "eval_samples_per_second": 10.518, | |
| "eval_steps_per_second": 5.259, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8856525581844480.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |