| { | |
| "best_metric": 0.7260138988494873, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-600", | |
| "epoch": 4.0, | |
| "global_step": 844, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019763033175355452, | |
| "loss": 1.4193, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.000195260663507109, | |
| "loss": 1.0491, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00019289099526066352, | |
| "loss": 1.1612, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000190521327014218, | |
| "loss": 1.0393, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00018815165876777252, | |
| "loss": 1.0533, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00018578199052132703, | |
| "loss": 0.9506, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00018341232227488152, | |
| "loss": 0.9148, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00018104265402843603, | |
| "loss": 0.9735, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00017890995260663508, | |
| "loss": 0.783, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0001765402843601896, | |
| "loss": 0.9235, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.6492890995260664, | |
| "eval_loss": 0.8972262740135193, | |
| "eval_runtime": 90.3527, | |
| "eval_samples_per_second": 9.341, | |
| "eval_steps_per_second": 1.173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00017417061611374408, | |
| "loss": 0.9931, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0001718009478672986, | |
| "loss": 1.0035, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00016943127962085308, | |
| "loss": 0.9619, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001670616113744076, | |
| "loss": 0.8618, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001646919431279621, | |
| "loss": 0.9862, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0001623222748815166, | |
| "loss": 0.9977, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0001599526066350711, | |
| "loss": 0.8715, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.0001575829383886256, | |
| "loss": 0.9153, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.0001552132701421801, | |
| "loss": 0.784, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00015284360189573462, | |
| "loss": 0.9253, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.6635071090047393, | |
| "eval_loss": 0.9150586128234863, | |
| "eval_runtime": 91.0647, | |
| "eval_samples_per_second": 9.268, | |
| "eval_steps_per_second": 1.164, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0001504739336492891, | |
| "loss": 0.8305, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0001481042654028436, | |
| "loss": 0.8379, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0001457345971563981, | |
| "loss": 0.707, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0001433649289099526, | |
| "loss": 0.8205, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00014099526066350712, | |
| "loss": 0.7565, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001386255924170616, | |
| "loss": 0.7539, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013625592417061612, | |
| "loss": 0.9389, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0001338862559241706, | |
| "loss": 0.8062, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 0.00013151658767772512, | |
| "loss": 0.7702, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00012914691943127963, | |
| "loss": 0.8371, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_accuracy": 0.693127962085308, | |
| "eval_loss": 0.8071145415306091, | |
| "eval_runtime": 87.0572, | |
| "eval_samples_per_second": 9.695, | |
| "eval_steps_per_second": 1.218, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00012677725118483412, | |
| "loss": 0.7698, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00012440758293838863, | |
| "loss": 0.8412, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00012203791469194314, | |
| "loss": 0.8268, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00011966824644549763, | |
| "loss": 0.6359, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00011729857819905214, | |
| "loss": 0.6765, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00011492890995260664, | |
| "loss": 0.7455, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00011255924170616114, | |
| "loss": 0.7108, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.00011018957345971565, | |
| "loss": 0.6456, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00010781990521327015, | |
| "loss": 0.7036, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00010545023696682465, | |
| "loss": 0.7355, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.707345971563981, | |
| "eval_loss": 0.7562642693519592, | |
| "eval_runtime": 87.3813, | |
| "eval_samples_per_second": 9.659, | |
| "eval_steps_per_second": 1.213, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00010308056872037915, | |
| "loss": 0.6315, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010071090047393366, | |
| "loss": 0.6996, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.834123222748816e-05, | |
| "loss": 0.6379, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.597156398104266e-05, | |
| "loss": 0.5765, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 9.360189573459716e-05, | |
| "loss": 0.6309, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.123222748815167e-05, | |
| "loss": 0.5836, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.886255924170617e-05, | |
| "loss": 0.5075, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8.649289099526067e-05, | |
| "loss": 0.5817, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 8.412322274881517e-05, | |
| "loss": 0.6765, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 8.175355450236967e-05, | |
| "loss": 0.6532, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_accuracy": 0.6895734597156398, | |
| "eval_loss": 0.7542913556098938, | |
| "eval_runtime": 87.8093, | |
| "eval_samples_per_second": 9.612, | |
| "eval_steps_per_second": 1.207, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 7.938388625592418e-05, | |
| "loss": 0.5203, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.701421800947868e-05, | |
| "loss": 0.4772, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 7.464454976303318e-05, | |
| "loss": 0.6391, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.227488151658768e-05, | |
| "loss": 0.5908, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.990521327014218e-05, | |
| "loss": 0.5459, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.753554502369669e-05, | |
| "loss": 0.6922, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 6.516587677725119e-05, | |
| "loss": 0.5738, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 6.279620853080569e-05, | |
| "loss": 0.5743, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 6.0426540284360186e-05, | |
| "loss": 0.5733, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 5.829383886255925e-05, | |
| "loss": 0.5982, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.726303317535545, | |
| "eval_loss": 0.7260138988494873, | |
| "eval_runtime": 88.3787, | |
| "eval_samples_per_second": 9.55, | |
| "eval_steps_per_second": 1.199, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 5.592417061611375e-05, | |
| "loss": 0.6154, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5.355450236966825e-05, | |
| "loss": 0.577, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 5.1184834123222746e-05, | |
| "loss": 0.5146, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.881516587677725e-05, | |
| "loss": 0.43, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.644549763033176e-05, | |
| "loss": 0.4428, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.407582938388626e-05, | |
| "loss": 0.4397, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.1706161137440756e-05, | |
| "loss": 0.3049, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 3.933649289099526e-05, | |
| "loss": 0.3993, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.696682464454976e-05, | |
| "loss": 0.4284, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.4597156398104267e-05, | |
| "loss": 0.4276, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "eval_accuracy": 0.7239336492890995, | |
| "eval_loss": 0.7345680594444275, | |
| "eval_runtime": 87.1766, | |
| "eval_samples_per_second": 9.681, | |
| "eval_steps_per_second": 1.216, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.222748815165877e-05, | |
| "loss": 0.3917, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.9857819905213268e-05, | |
| "loss": 0.3747, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 2.7488151658767774e-05, | |
| "loss": 0.3327, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.5118483412322273e-05, | |
| "loss": 0.3647, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 2.274881516587678e-05, | |
| "loss": 0.3785, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 2.037914691943128e-05, | |
| "loss": 0.4262, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.8009478672985784e-05, | |
| "loss": 0.4531, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.5639810426540286e-05, | |
| "loss": 0.3191, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.3270142180094788e-05, | |
| "loss": 0.427, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.0900473933649289e-05, | |
| "loss": 0.4935, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.7132701421800948, | |
| "eval_loss": 0.749038577079773, | |
| "eval_runtime": 86.4859, | |
| "eval_samples_per_second": 9.759, | |
| "eval_steps_per_second": 1.226, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 8.530805687203793e-06, | |
| "loss": 0.3046, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 6.161137440758294e-06, | |
| "loss": 0.3368, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 3.791469194312797e-06, | |
| "loss": 0.3309, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.4218009478672987e-06, | |
| "loss": 0.3304, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 844, | |
| "total_flos": 1.0430702255838413e+18, | |
| "train_loss": 0.674277721140622, | |
| "train_runtime": 2407.857, | |
| "train_samples_per_second": 5.59, | |
| "train_steps_per_second": 0.351 | |
| } | |
| ], | |
| "max_steps": 844, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.0430702255838413e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |