| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.000846935469869212, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.693870939738424e-05, | |
| "grad_norm": 22.163108825683594, | |
| "learning_rate": 2e-05, | |
| "loss": 21.3169, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.693870939738424e-05, | |
| "eval_loss": 2.290062189102173, | |
| "eval_runtime": 5153.9246, | |
| "eval_samples_per_second": 9.646, | |
| "eval_steps_per_second": 4.823, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.387741879476848e-05, | |
| "grad_norm": 31.874303817749023, | |
| "learning_rate": 4e-05, | |
| "loss": 19.141, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 5.081612819215272e-05, | |
| "grad_norm": 16.09918212890625, | |
| "learning_rate": 6e-05, | |
| "loss": 16.7916, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 6.775483758953696e-05, | |
| "grad_norm": 25.90465545654297, | |
| "learning_rate": 8e-05, | |
| "loss": 19.5082, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 8.46935469869212e-05, | |
| "grad_norm": 19.378210067749023, | |
| "learning_rate": 0.0001, | |
| "loss": 17.8913, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 8.46935469869212e-05, | |
| "eval_loss": 2.2351300716400146, | |
| "eval_runtime": 5147.8027, | |
| "eval_samples_per_second": 9.658, | |
| "eval_steps_per_second": 4.829, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00010163225638430543, | |
| "grad_norm": 19.173131942749023, | |
| "learning_rate": 0.00012, | |
| "loss": 17.3923, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00011857096578168967, | |
| "grad_norm": 32.09618377685547, | |
| "learning_rate": 0.00014, | |
| "loss": 20.1479, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0001355096751790739, | |
| "grad_norm": 21.93677520751953, | |
| "learning_rate": 0.00016, | |
| "loss": 17.782, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00015244838457645815, | |
| "grad_norm": 21.510074615478516, | |
| "learning_rate": 0.00018, | |
| "loss": 17.7573, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0001693870939738424, | |
| "grad_norm": 40.90770721435547, | |
| "learning_rate": 0.0002, | |
| "loss": 20.1925, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0001693870939738424, | |
| "eval_loss": 2.188427686691284, | |
| "eval_runtime": 5141.4887, | |
| "eval_samples_per_second": 9.669, | |
| "eval_steps_per_second": 4.835, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00018632580337122663, | |
| "grad_norm": 23.243389129638672, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 17.2822, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00020326451276861087, | |
| "grad_norm": 23.551620483398438, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 16.8668, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0002202032221659951, | |
| "grad_norm": 17.5562686920166, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 15.4958, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00023714193156337935, | |
| "grad_norm": 25.39556312561035, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 17.6328, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0002540806409607636, | |
| "grad_norm": 17.877641677856445, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 15.1113, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0002540806409607636, | |
| "eval_loss": 2.129117488861084, | |
| "eval_runtime": 5125.7738, | |
| "eval_samples_per_second": 9.699, | |
| "eval_steps_per_second": 4.85, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0002710193503581478, | |
| "grad_norm": 30.318897247314453, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 17.8594, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00028795805975553206, | |
| "grad_norm": 15.739501953125, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 17.0346, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0003048967691529163, | |
| "grad_norm": 24.859498977661133, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 18.724, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00032183547855030054, | |
| "grad_norm": 16.235403060913086, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 15.4015, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0003387741879476848, | |
| "grad_norm": 15.718273162841797, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 17.4672, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0003387741879476848, | |
| "eval_loss": 2.085993766784668, | |
| "eval_runtime": 5126.7679, | |
| "eval_samples_per_second": 9.697, | |
| "eval_steps_per_second": 4.849, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000355712897345069, | |
| "grad_norm": 17.683870315551758, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 17.1021, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00037265160674245326, | |
| "grad_norm": 18.684045791625977, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 15.9178, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0003895903161398375, | |
| "grad_norm": 20.173887252807617, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 17.3176, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.00040652902553722173, | |
| "grad_norm": 19.331239700317383, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 16.0941, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.000423467734934606, | |
| "grad_norm": 19.496095657348633, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 15.8816, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.000423467734934606, | |
| "eval_loss": 2.0629289150238037, | |
| "eval_runtime": 5134.9215, | |
| "eval_samples_per_second": 9.682, | |
| "eval_steps_per_second": 4.841, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0004404064443319902, | |
| "grad_norm": 33.5986213684082, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 15.7073, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.00045734515372937445, | |
| "grad_norm": 16.573434829711914, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 14.8196, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0004742838631267587, | |
| "grad_norm": 20.38245391845703, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 16.6963, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0004912225725241429, | |
| "grad_norm": 16.84417724609375, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 16.4005, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0005081612819215272, | |
| "grad_norm": 20.52333641052246, | |
| "learning_rate": 0.0001, | |
| "loss": 15.0153, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005081612819215272, | |
| "eval_loss": 2.041257858276367, | |
| "eval_runtime": 5138.2338, | |
| "eval_samples_per_second": 9.676, | |
| "eval_steps_per_second": 4.838, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005250999913189114, | |
| "grad_norm": 20.041860580444336, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 15.1095, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0005420387007162956, | |
| "grad_norm": 23.613113403320312, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 15.7084, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0005589774101136799, | |
| "grad_norm": 15.596351623535156, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 15.3402, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0005759161195110641, | |
| "grad_norm": 20.050952911376953, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 18.9168, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0005928548289084484, | |
| "grad_norm": 14.879345893859863, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 14.3502, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0005928548289084484, | |
| "eval_loss": 2.025693655014038, | |
| "eval_runtime": 5133.43, | |
| "eval_samples_per_second": 9.685, | |
| "eval_steps_per_second": 4.842, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0006097935383058326, | |
| "grad_norm": 17.409738540649414, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 15.9409, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0006267322477032168, | |
| "grad_norm": 19.219921112060547, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 18.1686, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0006436709571006011, | |
| "grad_norm": 15.948378562927246, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 15.5429, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0006606096664979853, | |
| "grad_norm": 17.51921844482422, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 14.8223, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0006775483758953696, | |
| "grad_norm": 19.49251937866211, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 16.4808, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0006775483758953696, | |
| "eval_loss": 2.012857675552368, | |
| "eval_runtime": 5129.6105, | |
| "eval_samples_per_second": 9.692, | |
| "eval_steps_per_second": 4.846, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0006944870852927538, | |
| "grad_norm": 17.339757919311523, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 14.7313, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.000711425794690138, | |
| "grad_norm": 15.278409957885742, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 14.637, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0007283645040875223, | |
| "grad_norm": 15.347859382629395, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 16.3984, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0007453032134849065, | |
| "grad_norm": 15.566729545593262, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 15.9272, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0007622419228822908, | |
| "grad_norm": 15.312088012695312, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 14.5249, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0007622419228822908, | |
| "eval_loss": 2.008061170578003, | |
| "eval_runtime": 5125.393, | |
| "eval_samples_per_second": 9.7, | |
| "eval_steps_per_second": 4.85, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.000779180632279675, | |
| "grad_norm": 18.034000396728516, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 16.604, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0007961193416770592, | |
| "grad_norm": 25.57406997680664, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 19.8734, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0008130580510744435, | |
| "grad_norm": 15.575130462646484, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 15.871, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0008299967604718277, | |
| "grad_norm": 21.707000732421875, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 16.6486, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.000846935469869212, | |
| "grad_norm": 19.565614700317383, | |
| "learning_rate": 0.0, | |
| "loss": 14.055, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.000846935469869212, | |
| "eval_loss": 2.0067672729492188, | |
| "eval_runtime": 5122.5703, | |
| "eval_samples_per_second": 9.705, | |
| "eval_steps_per_second": 4.853, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.01067914379264e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |