| { | |
| "best_metric": 0.07279229909181595, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-500", | |
| "epoch": 4.0, | |
| "global_step": 664, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019698795180722893, | |
| "loss": 1.6355, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019397590361445782, | |
| "loss": 1.0697, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019096385542168677, | |
| "loss": 0.6398, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00018795180722891569, | |
| "loss": 0.4757, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00018493975903614458, | |
| "loss": 0.3393, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0001819277108433735, | |
| "loss": 0.297, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00017891566265060242, | |
| "loss": 0.2623, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00017590361445783134, | |
| "loss": 0.2053, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00017289156626506026, | |
| "loss": 0.1578, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00016987951807228917, | |
| "loss": 0.1579, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.967479674796748, | |
| "eval_loss": 0.16562165319919586, | |
| "eval_runtime": 5.2378, | |
| "eval_samples_per_second": 46.966, | |
| "eval_steps_per_second": 5.919, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001668674698795181, | |
| "loss": 0.1362, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00016385542168674699, | |
| "loss": 0.1524, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001608433734939759, | |
| "loss": 0.0724, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00015783132530120482, | |
| "loss": 0.2021, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00015481927710843374, | |
| "loss": 0.0813, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.00015180722891566266, | |
| "loss": 0.0626, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00014879518072289158, | |
| "loss": 0.0959, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00014578313253012047, | |
| "loss": 0.0884, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0001427710843373494, | |
| "loss": 0.0713, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00013975903614457834, | |
| "loss": 0.0604, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.983739837398374, | |
| "eval_loss": 0.11851482838392258, | |
| "eval_runtime": 4.2166, | |
| "eval_samples_per_second": 58.341, | |
| "eval_steps_per_second": 7.352, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00013674698795180723, | |
| "loss": 0.0534, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00013373493975903615, | |
| "loss": 0.0292, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00013072289156626507, | |
| "loss": 0.0395, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00012771084337349396, | |
| "loss": 0.0325, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00012469879518072288, | |
| "loss": 0.0298, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00012168674698795181, | |
| "loss": 0.0211, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00011867469879518073, | |
| "loss": 0.07, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00011566265060240964, | |
| "loss": 0.0411, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00011265060240963856, | |
| "loss": 0.0201, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00010963855421686749, | |
| "loss": 0.0178, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_accuracy": 0.9634146341463414, | |
| "eval_loss": 0.1352052390575409, | |
| "eval_runtime": 4.4847, | |
| "eval_samples_per_second": 54.853, | |
| "eval_steps_per_second": 6.912, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00010662650602409638, | |
| "loss": 0.0229, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0001036144578313253, | |
| "loss": 0.0155, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010060240963855423, | |
| "loss": 0.0148, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.759036144578314e-05, | |
| "loss": 0.0156, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.457831325301205e-05, | |
| "loss": 0.0135, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.156626506024096e-05, | |
| "loss": 0.0139, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.855421686746988e-05, | |
| "loss": 0.0229, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 8.55421686746988e-05, | |
| "loss": 0.0122, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 8.253012048192772e-05, | |
| "loss": 0.0121, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 7.951807228915663e-05, | |
| "loss": 0.0112, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_accuracy": 0.975609756097561, | |
| "eval_loss": 0.10706935822963715, | |
| "eval_runtime": 4.4577, | |
| "eval_samples_per_second": 55.186, | |
| "eval_steps_per_second": 6.954, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.650602409638555e-05, | |
| "loss": 0.0269, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.349397590361447e-05, | |
| "loss": 0.0119, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 7.048192771084337e-05, | |
| "loss": 0.0346, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.746987951807229e-05, | |
| "loss": 0.0104, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 6.445783132530121e-05, | |
| "loss": 0.0101, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.144578313253012e-05, | |
| "loss": 0.0114, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.843373493975904e-05, | |
| "loss": 0.0098, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 5.5421686746987955e-05, | |
| "loss": 0.0094, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 5.240963855421687e-05, | |
| "loss": 0.0099, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.9397590361445786e-05, | |
| "loss": 0.0092, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_accuracy": 0.983739837398374, | |
| "eval_loss": 0.07279229909181595, | |
| "eval_runtime": 4.7057, | |
| "eval_samples_per_second": 52.277, | |
| "eval_steps_per_second": 6.588, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.63855421686747e-05, | |
| "loss": 0.009, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.337349397590362e-05, | |
| "loss": 0.0089, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.036144578313254e-05, | |
| "loss": 0.0087, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.734939759036144e-05, | |
| "loss": 0.0085, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 3.433734939759036e-05, | |
| "loss": 0.0083, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.132530120481928e-05, | |
| "loss": 0.0082, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.8313253012048197e-05, | |
| "loss": 0.0083, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.530120481927711e-05, | |
| "loss": 0.0317, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 2.2289156626506025e-05, | |
| "loss": 0.0079, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.927710843373494e-05, | |
| "loss": 0.008, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "eval_accuracy": 0.983739837398374, | |
| "eval_loss": 0.08243442326784134, | |
| "eval_runtime": 5.1012, | |
| "eval_samples_per_second": 48.224, | |
| "eval_steps_per_second": 6.077, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6265060240963857e-05, | |
| "loss": 0.0079, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1.3253012048192772e-05, | |
| "loss": 0.0079, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.0240963855421687e-05, | |
| "loss": 0.0078, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 7.228915662650602e-06, | |
| "loss": 0.0078, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.216867469879518e-06, | |
| "loss": 0.0079, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.2048192771084338e-06, | |
| "loss": 0.0078, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 664, | |
| "total_flos": 8.220718640312156e+17, | |
| "train_loss": 0.1065347370164222, | |
| "train_runtime": 522.7169, | |
| "train_samples_per_second": 20.294, | |
| "train_steps_per_second": 1.27 | |
| } | |
| ], | |
| "max_steps": 664, | |
| "num_train_epochs": 4, | |
| "total_flos": 8.220718640312156e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |