| { | |
| "best_metric": 0.20685863494873047, | |
| "best_model_checkpoint": "./dino-base-2023_11_24-unfreeze/checkpoint-16080", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 16080, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.01, | |
| "loss": 0.3961, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.0735977134690961, | |
| "eval_f1_macro": 0.09660814200449142, | |
| "eval_f1_micro": 0.3295561046263951, | |
| "eval_loss": 0.32934051752090454, | |
| "eval_roc_auc": 0.6004986038194676, | |
| "eval_runtime": 666.7423, | |
| "eval_samples_per_second": 4.198, | |
| "eval_steps_per_second": 0.262, | |
| "learning_rate": 0.01, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.01, | |
| "loss": 0.3418, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.04394426580921758, | |
| "eval_f1_macro": 0.12832379577401481, | |
| "eval_f1_micro": 0.33792181231809953, | |
| "eval_loss": 0.3381994962692261, | |
| "eval_roc_auc": 0.6053684490893504, | |
| "eval_runtime": 663.9506, | |
| "eval_samples_per_second": 4.216, | |
| "eval_steps_per_second": 0.264, | |
| "learning_rate": 0.01, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.01, | |
| "loss": 0.3334, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.07859949982136477, | |
| "eval_f1_macro": 0.2420082514645649, | |
| "eval_f1_micro": 0.39049407685639986, | |
| "eval_loss": 0.3550550937652588, | |
| "eval_roc_auc": 0.6319467749490031, | |
| "eval_runtime": 662.2838, | |
| "eval_samples_per_second": 4.226, | |
| "eval_steps_per_second": 0.264, | |
| "learning_rate": 0.01, | |
| "step": 1608 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.01, | |
| "loss": 0.3323, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.06252232940335835, | |
| "eval_f1_macro": 0.10992415783030689, | |
| "eval_f1_micro": 0.2554886521537749, | |
| "eval_loss": 0.32132452726364136, | |
| "eval_roc_auc": 0.5720375077972393, | |
| "eval_runtime": 688.3751, | |
| "eval_samples_per_second": 4.066, | |
| "eval_steps_per_second": 0.254, | |
| "learning_rate": 0.01, | |
| "step": 2144 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.01, | |
| "loss": 0.3248, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.04501607717041801, | |
| "eval_f1_macro": 0.12977783250072622, | |
| "eval_f1_micro": 0.33549517966695885, | |
| "eval_loss": 0.3164198696613312, | |
| "eval_roc_auc": 0.6023923090497274, | |
| "eval_runtime": 684.0693, | |
| "eval_samples_per_second": 4.092, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.01, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.01, | |
| "loss": 0.3235, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.023937120400142908, | |
| "eval_f1_macro": 0.08064530717139505, | |
| "eval_f1_micro": 0.28637243158573666, | |
| "eval_loss": 0.33460524678230286, | |
| "eval_roc_auc": 0.5833928540072826, | |
| "eval_runtime": 694.014, | |
| "eval_samples_per_second": 4.033, | |
| "eval_steps_per_second": 0.252, | |
| "learning_rate": 0.01, | |
| "step": 3216 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.01, | |
| "loss": 0.32, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.06823865666309396, | |
| "eval_f1_macro": 0.19677941241236616, | |
| "eval_f1_micro": 0.4593774264134962, | |
| "eval_loss": 0.3029455542564392, | |
| "eval_roc_auc": 0.6663014350966586, | |
| "eval_runtime": 675.5021, | |
| "eval_samples_per_second": 4.144, | |
| "eval_steps_per_second": 0.259, | |
| "learning_rate": 0.01, | |
| "step": 3752 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 0.01, | |
| "loss": 0.3138, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.05787781350482315, | |
| "eval_f1_macro": 0.29399778254935266, | |
| "eval_f1_micro": 0.5467535723238907, | |
| "eval_loss": 0.2865561246871948, | |
| "eval_roc_auc": 0.7240336144955091, | |
| "eval_runtime": 678.713, | |
| "eval_samples_per_second": 4.124, | |
| "eval_steps_per_second": 0.258, | |
| "learning_rate": 0.01, | |
| "step": 4288 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.01, | |
| "loss": 0.3052, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.12254376563058235, | |
| "eval_f1_macro": 0.29927318211914594, | |
| "eval_f1_micro": 0.4766527970411466, | |
| "eval_loss": 0.2806786894798279, | |
| "eval_roc_auc": 0.6671674748237758, | |
| "eval_runtime": 665.488, | |
| "eval_samples_per_second": 4.206, | |
| "eval_steps_per_second": 0.263, | |
| "learning_rate": 0.01, | |
| "step": 4824 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 0.01, | |
| "loss": 0.3157, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.0707395498392283, | |
| "eval_f1_macro": 0.20914365928952397, | |
| "eval_f1_micro": 0.47518133978097, | |
| "eval_loss": 0.29552116990089417, | |
| "eval_roc_auc": 0.6732798876351146, | |
| "eval_runtime": 670.1708, | |
| "eval_samples_per_second": 4.177, | |
| "eval_steps_per_second": 0.261, | |
| "learning_rate": 0.01, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 0.01, | |
| "loss": 0.3119, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.13612004287245444, | |
| "eval_f1_macro": 0.215957700272248, | |
| "eval_f1_micro": 0.40282685512367494, | |
| "eval_loss": 0.340472012758255, | |
| "eval_roc_auc": 0.6335842288286788, | |
| "eval_runtime": 655.9163, | |
| "eval_samples_per_second": 4.267, | |
| "eval_steps_per_second": 0.267, | |
| "learning_rate": 0.01, | |
| "step": 5896 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.01, | |
| "loss": 0.3162, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.05323329760628796, | |
| "eval_f1_macro": 0.2964933264503757, | |
| "eval_f1_micro": 0.4898614735996788, | |
| "eval_loss": 0.41625335812568665, | |
| "eval_roc_auc": 0.6862559753775964, | |
| "eval_runtime": 663.2841, | |
| "eval_samples_per_second": 4.22, | |
| "eval_steps_per_second": 0.264, | |
| "learning_rate": 0.01, | |
| "step": 6432 | |
| }, | |
| { | |
| "epoch": 12.13, | |
| "learning_rate": 0.01, | |
| "loss": 0.3184, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.10468024294390854, | |
| "eval_f1_macro": 0.32985179099978684, | |
| "eval_f1_micro": 0.5429481760842708, | |
| "eval_loss": 0.29642441868782043, | |
| "eval_roc_auc": 0.7170301433203381, | |
| "eval_runtime": 670.4311, | |
| "eval_samples_per_second": 4.175, | |
| "eval_steps_per_second": 0.261, | |
| "learning_rate": 0.01, | |
| "step": 6968 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 0.01, | |
| "loss": 0.3131, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 0.01, | |
| "loss": 0.3142, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.08324401571989996, | |
| "eval_f1_macro": 0.3153914854710691, | |
| "eval_f1_micro": 0.5253185233583796, | |
| "eval_loss": 0.30047285556793213, | |
| "eval_roc_auc": 0.7071979710639686, | |
| "eval_runtime": 655.5852, | |
| "eval_samples_per_second": 4.269, | |
| "eval_steps_per_second": 0.267, | |
| "learning_rate": 0.01, | |
| "step": 7504 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 0.01, | |
| "loss": 0.3104, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.0, | |
| "eval_f1_macro": 0.06737263276203229, | |
| "eval_f1_micro": 0.16727748691099475, | |
| "eval_loss": 3.1991209983825684, | |
| "eval_roc_auc": 0.48793717022406635, | |
| "eval_runtime": 660.6942, | |
| "eval_samples_per_second": 4.236, | |
| "eval_steps_per_second": 0.265, | |
| "learning_rate": 0.01, | |
| "step": 8040 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.3042, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.15827081100392998, | |
| "eval_f1_macro": 0.2746423799349764, | |
| "eval_f1_micro": 0.45441654850279417, | |
| "eval_loss": 0.2820233702659607, | |
| "eval_roc_auc": 0.6518577907371156, | |
| "eval_runtime": 657.3215, | |
| "eval_samples_per_second": 4.258, | |
| "eval_steps_per_second": 0.266, | |
| "learning_rate": 0.001, | |
| "step": 8576 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.2788, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.1639871382636656, | |
| "eval_f1_macro": 0.3842369260673032, | |
| "eval_f1_micro": 0.5743618993298161, | |
| "eval_loss": 0.2741135358810425, | |
| "eval_roc_auc": 0.7204845597401146, | |
| "eval_runtime": 676.3087, | |
| "eval_samples_per_second": 4.139, | |
| "eval_steps_per_second": 0.259, | |
| "learning_rate": 0.001, | |
| "step": 9112 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.2724, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.2072168631654162, | |
| "eval_f1_macro": 0.39364174946303265, | |
| "eval_f1_micro": 0.5903019744483159, | |
| "eval_loss": 0.2424442023038864, | |
| "eval_roc_auc": 0.725566189869564, | |
| "eval_runtime": 676.7446, | |
| "eval_samples_per_second": 4.136, | |
| "eval_steps_per_second": 0.259, | |
| "learning_rate": 0.001, | |
| "step": 9648 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.2642, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.21864951768488747, | |
| "eval_f1_macro": 0.4095283624260902, | |
| "eval_f1_micro": 0.6020972084455152, | |
| "eval_loss": 0.24139995872974396, | |
| "eval_roc_auc": 0.7346777613512268, | |
| "eval_runtime": 679.8261, | |
| "eval_samples_per_second": 4.117, | |
| "eval_steps_per_second": 0.257, | |
| "learning_rate": 0.001, | |
| "step": 10184 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.2597, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.22508038585209003, | |
| "eval_f1_macro": 0.4156043928581225, | |
| "eval_f1_micro": 0.6078926484380637, | |
| "eval_loss": 0.22694532573223114, | |
| "eval_roc_auc": 0.734743268508761, | |
| "eval_runtime": 658.6647, | |
| "eval_samples_per_second": 4.25, | |
| "eval_steps_per_second": 0.266, | |
| "learning_rate": 0.001, | |
| "step": 10720 | |
| }, | |
| { | |
| "epoch": 20.52, | |
| "learning_rate": 0.001, | |
| "loss": 0.2575, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.23401214719542693, | |
| "eval_f1_macro": 0.42525523109884183, | |
| "eval_f1_micro": 0.6231395675372087, | |
| "eval_loss": 0.22489750385284424, | |
| "eval_roc_auc": 0.7462705119874329, | |
| "eval_runtime": 659.2031, | |
| "eval_samples_per_second": 4.246, | |
| "eval_steps_per_second": 0.265, | |
| "learning_rate": 0.001, | |
| "step": 11256 | |
| }, | |
| { | |
| "epoch": 21.46, | |
| "learning_rate": 0.001, | |
| "loss": 0.253, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.24294390853876385, | |
| "eval_f1_macro": 0.4638642250813612, | |
| "eval_f1_micro": 0.6291112183686134, | |
| "eval_loss": 0.2261122316122055, | |
| "eval_roc_auc": 0.7521458927635197, | |
| "eval_runtime": 657.2132, | |
| "eval_samples_per_second": 4.259, | |
| "eval_steps_per_second": 0.266, | |
| "learning_rate": 0.001, | |
| "step": 11792 | |
| }, | |
| { | |
| "epoch": 22.39, | |
| "learning_rate": 0.001, | |
| "loss": 0.2491, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.2536620221507681, | |
| "eval_f1_macro": 0.4855675958097963, | |
| "eval_f1_micro": 0.6453674121405751, | |
| "eval_loss": 0.21625448763370514, | |
| "eval_roc_auc": 0.7626888699414743, | |
| "eval_runtime": 684.5667, | |
| "eval_samples_per_second": 4.089, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.001, | |
| "step": 12328 | |
| }, | |
| { | |
| "epoch": 23.32, | |
| "learning_rate": 0.001, | |
| "loss": 0.2484, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.2568774562343694, | |
| "eval_f1_macro": 0.4635023998508455, | |
| "eval_f1_micro": 0.6261708770933863, | |
| "eval_loss": 0.22124029695987701, | |
| "eval_roc_auc": 0.746011775710583, | |
| "eval_runtime": 686.8149, | |
| "eval_samples_per_second": 4.075, | |
| "eval_steps_per_second": 0.255, | |
| "learning_rate": 0.001, | |
| "step": 12864 | |
| }, | |
| { | |
| "epoch": 24.25, | |
| "learning_rate": 0.001, | |
| "loss": 0.2465, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.27724187209717754, | |
| "eval_f1_macro": 0.4780125036155298, | |
| "eval_f1_micro": 0.6486374896779521, | |
| "eval_loss": 0.21179212629795074, | |
| "eval_roc_auc": 0.7621535864293458, | |
| "eval_runtime": 682.8306, | |
| "eval_samples_per_second": 4.099, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.001, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 25.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.241, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.255805644873169, | |
| "eval_f1_macro": 0.5159201891126339, | |
| "eval_f1_micro": 0.6601850610164945, | |
| "eval_loss": 0.21056728065013885, | |
| "eval_roc_auc": 0.7726679521523573, | |
| "eval_runtime": 690.8376, | |
| "eval_samples_per_second": 4.052, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 13936 | |
| }, | |
| { | |
| "epoch": 26.12, | |
| "learning_rate": 0.001, | |
| "loss": 0.2413, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.2722400857449089, | |
| "eval_f1_macro": 0.4978739205590764, | |
| "eval_f1_micro": 0.6389924716808555, | |
| "eval_loss": 0.21349740028381348, | |
| "eval_roc_auc": 0.753602824058608, | |
| "eval_runtime": 690.4835, | |
| "eval_samples_per_second": 4.054, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 14472 | |
| }, | |
| { | |
| "epoch": 27.05, | |
| "learning_rate": 0.001, | |
| "loss": 0.2401, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 27.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.2385, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.27724187209717754, | |
| "eval_f1_macro": 0.4596468154208962, | |
| "eval_f1_micro": 0.6102875778238955, | |
| "eval_loss": 0.2182074636220932, | |
| "eval_roc_auc": 0.7318729802955509, | |
| "eval_runtime": 665.6085, | |
| "eval_samples_per_second": 4.205, | |
| "eval_steps_per_second": 0.263, | |
| "learning_rate": 0.001, | |
| "step": 15008 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "learning_rate": 0.001, | |
| "loss": 0.2366, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.270811003929975, | |
| "eval_f1_macro": 0.5354446223521995, | |
| "eval_f1_micro": 0.6614600569197168, | |
| "eval_loss": 0.2131979614496231, | |
| "eval_roc_auc": 0.7757714197210208, | |
| "eval_runtime": 663.3734, | |
| "eval_samples_per_second": 4.219, | |
| "eval_steps_per_second": 0.264, | |
| "learning_rate": 0.001, | |
| "step": 15544 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "learning_rate": 0.001, | |
| "loss": 0.2345, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.2747409789210432, | |
| "eval_f1_macro": 0.5121854108049624, | |
| "eval_f1_micro": 0.6565579984836998, | |
| "eval_loss": 0.20685863494873047, | |
| "eval_roc_auc": 0.7657921716163675, | |
| "eval_runtime": 663.8358, | |
| "eval_samples_per_second": 4.216, | |
| "eval_steps_per_second": 0.264, | |
| "learning_rate": 0.001, | |
| "step": 16080 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.001, | |
| "step": 16080, | |
| "total_flos": 2.04446834386654e+19, | |
| "train_loss": 0.2889823566028728, | |
| "train_runtime": 84191.1285, | |
| "train_samples_per_second": 3.053, | |
| "train_steps_per_second": 0.191 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 16080, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 2.04446834386654e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |