| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.16953941791466515, | |
| "eval_steps": 13, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011302627860977678, | |
| "eval_loss": 0.7253943681716919, | |
| "eval_runtime": 222.5674, | |
| "eval_samples_per_second": 6.695, | |
| "eval_steps_per_second": 0.84, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003390788358293303, | |
| "grad_norm": 6.4600725173950195, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.9057, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006781576716586606, | |
| "grad_norm": 4.287564277648926, | |
| "learning_rate": 3e-05, | |
| "loss": 2.7228, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010172365074879909, | |
| "grad_norm": 4.434852600097656, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.2412, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013563153433173212, | |
| "grad_norm": 4.096804618835449, | |
| "learning_rate": 4.997482666353287e-05, | |
| "loss": 1.6976, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01469341621927098, | |
| "eval_loss": 0.281572163105011, | |
| "eval_runtime": 224.6256, | |
| "eval_samples_per_second": 6.633, | |
| "eval_steps_per_second": 0.832, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.016953941791466517, | |
| "grad_norm": 3.934027910232544, | |
| "learning_rate": 4.984280524733107e-05, | |
| "loss": 1.2107, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.020344730149759818, | |
| "grad_norm": 2.3636763095855713, | |
| "learning_rate": 4.959823971496574e-05, | |
| "loss": 0.8397, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02373551850805312, | |
| "grad_norm": 1.3671884536743164, | |
| "learning_rate": 4.9242238009417175e-05, | |
| "loss": 0.7578, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.027126306866346424, | |
| "grad_norm": 1.3218061923980713, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 0.6645, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02938683243854196, | |
| "eval_loss": 0.16004851460456848, | |
| "eval_runtime": 224.8305, | |
| "eval_samples_per_second": 6.627, | |
| "eval_steps_per_second": 0.832, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.030517095224639728, | |
| "grad_norm": 1.757283091545105, | |
| "learning_rate": 4.820287471297598e-05, | |
| "loss": 0.6531, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.033907883582933035, | |
| "grad_norm": 1.2630192041397095, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.616, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03729867194122634, | |
| "grad_norm": 1.2325754165649414, | |
| "learning_rate": 4.674352832889239e-05, | |
| "loss": 0.6141, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.040689460299519635, | |
| "grad_norm": 1.5199618339538574, | |
| "learning_rate": 4.586433134303257e-05, | |
| "loss": 0.6026, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04408024865781294, | |
| "grad_norm": 1.2060375213623047, | |
| "learning_rate": 4.489061372204453e-05, | |
| "loss": 0.6114, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04408024865781294, | |
| "eval_loss": 0.14285726845264435, | |
| "eval_runtime": 224.8051, | |
| "eval_samples_per_second": 6.628, | |
| "eval_steps_per_second": 0.832, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04747103701610624, | |
| "grad_norm": 1.2959660291671753, | |
| "learning_rate": 4.382678665009028e-05, | |
| "loss": 0.576, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.050861825374399545, | |
| "grad_norm": 1.1517972946166992, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.5597, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05425261373269285, | |
| "grad_norm": 1.2231593132019043, | |
| "learning_rate": 4.144846814849282e-05, | |
| "loss": 0.5803, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05764340209098615, | |
| "grad_norm": 1.1372201442718506, | |
| "learning_rate": 4.01447510960205e-05, | |
| "loss": 0.5299, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.05877366487708392, | |
| "eval_loss": 0.13713335990905762, | |
| "eval_runtime": 224.8683, | |
| "eval_samples_per_second": 6.626, | |
| "eval_steps_per_second": 0.832, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.061034190449279456, | |
| "grad_norm": 1.0566143989562988, | |
| "learning_rate": 3.8772424536302564e-05, | |
| "loss": 0.555, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.06442497880757277, | |
| "grad_norm": 1.0105335712432861, | |
| "learning_rate": 3.7337705451608674e-05, | |
| "loss": 0.5633, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06781576716586607, | |
| "grad_norm": 1.0776220560073853, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 0.5251, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07120655552415937, | |
| "grad_norm": 1.2005013227462769, | |
| "learning_rate": 3.4307341460048633e-05, | |
| "loss": 0.5479, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0734670810963549, | |
| "eval_loss": 0.13397091627120972, | |
| "eval_runtime": 224.8, | |
| "eval_samples_per_second": 6.628, | |
| "eval_steps_per_second": 0.832, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07459734388245268, | |
| "grad_norm": 0.9956369996070862, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.5392, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.07798813224074597, | |
| "grad_norm": 1.1790428161621094, | |
| "learning_rate": 3.110851015344735e-05, | |
| "loss": 0.5354, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.08137892059903927, | |
| "grad_norm": 1.041911005973816, | |
| "learning_rate": 2.9463922369965917e-05, | |
| "loss": 0.5366, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.08476970895733257, | |
| "grad_norm": 1.101559042930603, | |
| "learning_rate": 2.7799111902582696e-05, | |
| "loss": 0.5158, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08816049731562588, | |
| "grad_norm": 0.9680448174476624, | |
| "learning_rate": 2.6121620758762877e-05, | |
| "loss": 0.4675, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.08816049731562588, | |
| "eval_loss": 0.13185884058475494, | |
| "eval_runtime": 224.7208, | |
| "eval_samples_per_second": 6.63, | |
| "eval_steps_per_second": 0.832, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.09155128567391918, | |
| "grad_norm": 1.1247650384902954, | |
| "learning_rate": 2.443904839260488e-05, | |
| "loss": 0.5192, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.09494207403221248, | |
| "grad_norm": 1.0689295530319214, | |
| "learning_rate": 2.2759017277414166e-05, | |
| "loss": 0.5174, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.09833286239050579, | |
| "grad_norm": 1.2236093282699585, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 0.5135, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.10172365074879909, | |
| "grad_norm": 1.1371599435806274, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 0.5161, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10285391353489687, | |
| "eval_loss": 0.13008707761764526, | |
| "eval_runtime": 224.8789, | |
| "eval_samples_per_second": 6.626, | |
| "eval_steps_per_second": 0.832, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1051144391070924, | |
| "grad_norm": 0.9614408612251282, | |
| "learning_rate": 1.781001681419957e-05, | |
| "loss": 0.5149, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.1085052274653857, | |
| "grad_norm": 0.8851366639137268, | |
| "learning_rate": 1.621562939796643e-05, | |
| "loss": 0.5307, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.111896015823679, | |
| "grad_norm": 1.0845979452133179, | |
| "learning_rate": 1.466103737583699e-05, | |
| "loss": 0.5183, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.1152868041819723, | |
| "grad_norm": 1.0732598304748535, | |
| "learning_rate": 1.3153283438175034e-05, | |
| "loss": 0.5277, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.11754732975416785, | |
| "eval_loss": 0.12890119850635529, | |
| "eval_runtime": 225.0833, | |
| "eval_samples_per_second": 6.62, | |
| "eval_steps_per_second": 0.831, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.11867759254026561, | |
| "grad_norm": 1.0611317157745361, | |
| "learning_rate": 1.1699198087116589e-05, | |
| "loss": 0.5579, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.12206838089855891, | |
| "grad_norm": 1.7777539491653442, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 0.5391, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.12545916925685222, | |
| "grad_norm": 1.0344315767288208, | |
| "learning_rate": 8.978109650374397e-06, | |
| "loss": 0.4931, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.12884995761514553, | |
| "grad_norm": 1.0285758972167969, | |
| "learning_rate": 7.723433775328384e-06, | |
| "loss": 0.5289, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.13224074597343882, | |
| "grad_norm": 0.9961230158805847, | |
| "learning_rate": 6.547025062816486e-06, | |
| "loss": 0.5086, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.13224074597343882, | |
| "eval_loss": 0.12818647921085358, | |
| "eval_runtime": 224.9913, | |
| "eval_samples_per_second": 6.622, | |
| "eval_steps_per_second": 0.831, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.13563153433173214, | |
| "grad_norm": 1.7210478782653809, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 0.5592, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13902232269002543, | |
| "grad_norm": 1.005109429359436, | |
| "learning_rate": 4.4499481138022544e-06, | |
| "loss": 0.4997, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.14241311104831875, | |
| "grad_norm": 1.0120928287506104, | |
| "learning_rate": 3.5387801599533475e-06, | |
| "loss": 0.543, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.14580389940661204, | |
| "grad_norm": 0.9547510743141174, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 0.5363, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.1469341621927098, | |
| "eval_loss": 0.12771442532539368, | |
| "eval_runtime": 224.7527, | |
| "eval_samples_per_second": 6.63, | |
| "eval_steps_per_second": 0.832, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14919468776490535, | |
| "grad_norm": 1.0372916460037231, | |
| "learning_rate": 2.0118056862137357e-06, | |
| "loss": 0.4873, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.15258547612319864, | |
| "grad_norm": 1.11273992061615, | |
| "learning_rate": 1.4029167422908107e-06, | |
| "loss": 0.5262, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.15597626448149193, | |
| "grad_norm": 0.9660788774490356, | |
| "learning_rate": 9.009284826036691e-07, | |
| "loss": 0.5585, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.15936705283978525, | |
| "grad_norm": 1.1789458990097046, | |
| "learning_rate": 5.08115039419113e-07, | |
| "loss": 0.4771, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.16162757841198078, | |
| "eval_loss": 0.127611443400383, | |
| "eval_runtime": 225.1337, | |
| "eval_samples_per_second": 6.618, | |
| "eval_steps_per_second": 0.831, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.16275784119807854, | |
| "grad_norm": 0.9032691717147827, | |
| "learning_rate": 2.262559558016325e-07, | |
| "loss": 0.4993, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.16614862955637186, | |
| "grad_norm": 1.033461332321167, | |
| "learning_rate": 5.662812383859795e-08, | |
| "loss": 0.4605, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.16953941791466515, | |
| "grad_norm": 0.9426572322845459, | |
| "learning_rate": 0.0, | |
| "loss": 0.4887, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.218781823926272e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |