| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.911589008363202, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019115890083632018, | |
| "eval_loss": 4.260312557220459, | |
| "eval_runtime": 20.2891, | |
| "eval_samples_per_second": 8.724, | |
| "eval_steps_per_second": 4.387, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05734767025089606, | |
| "grad_norm": 4.044816970825195, | |
| "learning_rate": 3e-05, | |
| "loss": 4.0773, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11469534050179211, | |
| "grad_norm": 3.713122844696045, | |
| "learning_rate": 6e-05, | |
| "loss": 3.9077, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.17204301075268819, | |
| "grad_norm": 4.895143032073975, | |
| "learning_rate": 9e-05, | |
| "loss": 3.1815, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17204301075268819, | |
| "eval_loss": 2.6312191486358643, | |
| "eval_runtime": 19.7481, | |
| "eval_samples_per_second": 8.963, | |
| "eval_steps_per_second": 4.507, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.22939068100358423, | |
| "grad_norm": 2.3883414268493652, | |
| "learning_rate": 0.00012, | |
| "loss": 2.4895, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2867383512544803, | |
| "grad_norm": 1.8660134077072144, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.1507, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.34408602150537637, | |
| "grad_norm": 2.030590534210205, | |
| "learning_rate": 0.00018, | |
| "loss": 2.0577, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.34408602150537637, | |
| "eval_loss": 1.9757664203643799, | |
| "eval_runtime": 19.7462, | |
| "eval_samples_per_second": 8.964, | |
| "eval_steps_per_second": 4.507, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4014336917562724, | |
| "grad_norm": 1.5562611818313599, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 2.0297, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.45878136200716846, | |
| "grad_norm": 1.7431790828704834, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.861, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "grad_norm": 1.7667114734649658, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 1.9087, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5161290322580645, | |
| "eval_loss": 1.8730906248092651, | |
| "eval_runtime": 19.7448, | |
| "eval_samples_per_second": 8.964, | |
| "eval_steps_per_second": 4.508, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5734767025089605, | |
| "grad_norm": 1.5921707153320312, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 1.8208, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6308243727598566, | |
| "grad_norm": 1.4759997129440308, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 1.9535, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6881720430107527, | |
| "grad_norm": 1.3630574941635132, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.8836, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6881720430107527, | |
| "eval_loss": 1.8447675704956055, | |
| "eval_runtime": 19.7508, | |
| "eval_samples_per_second": 8.962, | |
| "eval_steps_per_second": 4.506, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7455197132616488, | |
| "grad_norm": 1.3893134593963623, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 1.9008, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8028673835125448, | |
| "grad_norm": 1.3943053483963013, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.8282, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "grad_norm": 1.3723570108413696, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 1.8482, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8602150537634409, | |
| "eval_loss": 1.8272418975830078, | |
| "eval_runtime": 19.7474, | |
| "eval_samples_per_second": 8.963, | |
| "eval_steps_per_second": 4.507, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9175627240143369, | |
| "grad_norm": 1.2157355546951294, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 1.7012, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.974910394265233, | |
| "grad_norm": 1.4276927709579468, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 1.8858, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.032258064516129, | |
| "grad_norm": 1.1759012937545776, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 2.0545, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.032258064516129, | |
| "eval_loss": 1.7873010635375977, | |
| "eval_runtime": 19.7531, | |
| "eval_samples_per_second": 8.961, | |
| "eval_steps_per_second": 4.506, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0896057347670252, | |
| "grad_norm": 1.0551518201828003, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 1.4021, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.146953405017921, | |
| "grad_norm": 1.3845959901809692, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3838, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2043010752688172, | |
| "grad_norm": 1.4633151292800903, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.4003, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.2043010752688172, | |
| "eval_loss": 1.873085856437683, | |
| "eval_runtime": 19.7425, | |
| "eval_samples_per_second": 8.965, | |
| "eval_steps_per_second": 4.508, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.2616487455197132, | |
| "grad_norm": 2.040634870529175, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.4048, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.3189964157706093, | |
| "grad_norm": 1.6276676654815674, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 1.3703, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.3763440860215055, | |
| "grad_norm": 1.4569168090820312, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.2632, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.3763440860215055, | |
| "eval_loss": 1.8426142930984497, | |
| "eval_runtime": 19.7468, | |
| "eval_samples_per_second": 8.963, | |
| "eval_steps_per_second": 4.507, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4336917562724014, | |
| "grad_norm": 1.6304519176483154, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.3729, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.4910394265232976, | |
| "grad_norm": 1.690694808959961, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.3019, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.5483870967741935, | |
| "grad_norm": 1.8786835670471191, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.3178, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.5483870967741935, | |
| "eval_loss": 1.8405394554138184, | |
| "eval_runtime": 19.7363, | |
| "eval_samples_per_second": 8.968, | |
| "eval_steps_per_second": 4.509, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.6057347670250897, | |
| "grad_norm": 1.877886176109314, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.3041, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6630824372759858, | |
| "grad_norm": 1.9166302680969238, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.3063, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.7204301075268817, | |
| "grad_norm": 1.6511739492416382, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.2796, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7204301075268817, | |
| "eval_loss": 1.8262274265289307, | |
| "eval_runtime": 19.737, | |
| "eval_samples_per_second": 8.968, | |
| "eval_steps_per_second": 4.509, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 1.6190431118011475, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.2655, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.8351254480286738, | |
| "grad_norm": 1.7480906248092651, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.374, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.89247311827957, | |
| "grad_norm": 1.9425756931304932, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.2715, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.89247311827957, | |
| "eval_loss": 1.8266067504882812, | |
| "eval_runtime": 19.7351, | |
| "eval_samples_per_second": 8.969, | |
| "eval_steps_per_second": 4.51, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.968083617316864e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |