| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.009525327207617857, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.621542633957431e-05, | |
| "eval_loss": 3.1601955890655518, | |
| "eval_runtime": 13080.0915, | |
| "eval_samples_per_second": 2.677, | |
| "eval_steps_per_second": 1.338, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00028864627901872294, | |
| "grad_norm": 88.87578582763672, | |
| "learning_rate": 3e-05, | |
| "loss": 100.51, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005772925580374459, | |
| "grad_norm": 85.21216583251953, | |
| "learning_rate": 6e-05, | |
| "loss": 96.5347, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008659388370561688, | |
| "grad_norm": 77.13072967529297, | |
| "learning_rate": 9e-05, | |
| "loss": 80.7471, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008659388370561688, | |
| "eval_loss": 2.1278481483459473, | |
| "eval_runtime": 13090.1143, | |
| "eval_samples_per_second": 2.675, | |
| "eval_steps_per_second": 1.337, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011545851160748917, | |
| "grad_norm": 68.87947845458984, | |
| "learning_rate": 0.00012, | |
| "loss": 64.7623, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0014432313950936146, | |
| "grad_norm": 43.136959075927734, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 44.3736, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0017318776741123375, | |
| "grad_norm": 41.73539352416992, | |
| "learning_rate": 0.00018, | |
| "loss": 31.1364, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0017318776741123375, | |
| "eval_loss": 0.844988226890564, | |
| "eval_runtime": 13063.393, | |
| "eval_samples_per_second": 2.68, | |
| "eval_steps_per_second": 1.34, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0020205239531310604, | |
| "grad_norm": 37.0665397644043, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 25.9659, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0023091702321497835, | |
| "grad_norm": 32.10556411743164, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 19.8184, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002597816511168506, | |
| "grad_norm": 39.13405990600586, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 15.5343, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002597816511168506, | |
| "eval_loss": 0.4686388671398163, | |
| "eval_runtime": 11313.8601, | |
| "eval_samples_per_second": 3.094, | |
| "eval_steps_per_second": 1.547, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0028864627901872292, | |
| "grad_norm": 29.395793914794922, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 14.8443, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.003175109069205952, | |
| "grad_norm": 27.154170989990234, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 13.3753, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003463755348224675, | |
| "grad_norm": 23.787752151489258, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 10.3941, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003463755348224675, | |
| "eval_loss": 0.3380902111530304, | |
| "eval_runtime": 10045.2845, | |
| "eval_samples_per_second": 3.485, | |
| "eval_steps_per_second": 1.743, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003752401627243398, | |
| "grad_norm": 32.828285217285156, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 11.2033, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.004041047906262121, | |
| "grad_norm": 46.57306671142578, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 10.001, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004329694185280844, | |
| "grad_norm": 33.11753845214844, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 8.5669, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004329694185280844, | |
| "eval_loss": 0.25203973054885864, | |
| "eval_runtime": 11473.7667, | |
| "eval_samples_per_second": 3.051, | |
| "eval_steps_per_second": 1.526, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004618340464299567, | |
| "grad_norm": 21.155969619750977, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 8.103, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.00490698674331829, | |
| "grad_norm": 25.17514419555664, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 7.4075, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.005195633022337012, | |
| "grad_norm": 21.17397117614746, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 6.9151, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005195633022337012, | |
| "eval_loss": 0.21289654076099396, | |
| "eval_runtime": 12548.5942, | |
| "eval_samples_per_second": 2.79, | |
| "eval_steps_per_second": 1.395, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005484279301355735, | |
| "grad_norm": 21.539554595947266, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 6.7601, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0057729255803744585, | |
| "grad_norm": 20.63874626159668, | |
| "learning_rate": 0.0001, | |
| "loss": 6.994, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.006061571859393182, | |
| "grad_norm": 16.417226791381836, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 5.5389, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.006061571859393182, | |
| "eval_loss": 0.17797139286994934, | |
| "eval_runtime": 9174.8, | |
| "eval_samples_per_second": 3.816, | |
| "eval_steps_per_second": 1.908, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.006350218138411904, | |
| "grad_norm": 17.829113006591797, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 5.8211, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.006638864417430627, | |
| "grad_norm": 18.48779296875, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 5.1794, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.00692751069644935, | |
| "grad_norm": 17.563810348510742, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 5.6267, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00692751069644935, | |
| "eval_loss": 0.16187982261180878, | |
| "eval_runtime": 9520.9589, | |
| "eval_samples_per_second": 3.677, | |
| "eval_steps_per_second": 1.839, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.007216156975468073, | |
| "grad_norm": 19.880619049072266, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 5.2772, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.007504803254486796, | |
| "grad_norm": 21.604888916015625, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 4.9052, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0077934495335055184, | |
| "grad_norm": 19.327678680419922, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 5.1425, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0077934495335055184, | |
| "eval_loss": 0.14949661493301392, | |
| "eval_runtime": 12097.8181, | |
| "eval_samples_per_second": 2.894, | |
| "eval_steps_per_second": 1.447, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.008082095812524242, | |
| "grad_norm": 17.062747955322266, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 5.306, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.008370742091542965, | |
| "grad_norm": 15.015271186828613, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 4.3013, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.008659388370561688, | |
| "grad_norm": 12.879692077636719, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 4.3707, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.008659388370561688, | |
| "eval_loss": 0.1438611000776291, | |
| "eval_runtime": 12295.3446, | |
| "eval_samples_per_second": 2.847, | |
| "eval_steps_per_second": 1.424, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00894803464958041, | |
| "grad_norm": 14.83832836151123, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 4.7525, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.009236680928599134, | |
| "grad_norm": 12.520251274108887, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 4.5588, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.009525327207617857, | |
| "grad_norm": 17.722375869750977, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 4.3476, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.009525327207617857, | |
| "eval_loss": 0.14178016781806946, | |
| "eval_runtime": 9365.6187, | |
| "eval_samples_per_second": 3.738, | |
| "eval_steps_per_second": 1.869, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.206564752284713e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |