| { | |
| "best_metric": 0.057097259908914566, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-480", | |
| "epoch": 1.9565217391304348, | |
| "global_step": 720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0001972826086956522, | |
| "loss": 0.6016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019456521739130434, | |
| "loss": 0.168, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019184782608695653, | |
| "loss": 0.1568, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0001891304347826087, | |
| "loss": 0.1165, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.10377205908298492, | |
| "eval_runtime": 18.7416, | |
| "eval_samples_per_second": 39.164, | |
| "eval_steps_per_second": 4.909, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0001864130434782609, | |
| "loss": 0.0853, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00018369565217391304, | |
| "loss": 0.179, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00018097826086956522, | |
| "loss": 0.0871, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001782608695652174, | |
| "loss": 0.0596, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.9673024523160763, | |
| "eval_loss": 0.1255735605955124, | |
| "eval_runtime": 19.1765, | |
| "eval_samples_per_second": 38.276, | |
| "eval_steps_per_second": 4.798, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00017554347826086956, | |
| "loss": 0.1279, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00017282608695652174, | |
| "loss": 0.0814, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00017010869565217392, | |
| "loss": 0.0967, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001673913043478261, | |
| "loss": 0.1317, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.9495912806539509, | |
| "eval_loss": 0.18217410147190094, | |
| "eval_runtime": 17.5926, | |
| "eval_samples_per_second": 41.722, | |
| "eval_steps_per_second": 5.229, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00016467391304347828, | |
| "loss": 0.0585, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00016195652173913046, | |
| "loss": 0.078, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00015923913043478262, | |
| "loss": 0.0794, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001565217391304348, | |
| "loss": 0.0731, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.9741144414168937, | |
| "eval_loss": 0.11143568158149719, | |
| "eval_runtime": 18.204, | |
| "eval_samples_per_second": 40.321, | |
| "eval_steps_per_second": 5.054, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00015380434782608698, | |
| "loss": 0.0842, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00015108695652173913, | |
| "loss": 0.0406, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0001483695652173913, | |
| "loss": 0.1171, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001456521739130435, | |
| "loss": 0.1041, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.08514755964279175, | |
| "eval_runtime": 17.638, | |
| "eval_samples_per_second": 41.615, | |
| "eval_steps_per_second": 5.216, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00014293478260869567, | |
| "loss": 0.0823, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00014021739130434783, | |
| "loss": 0.0536, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.1171, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001347826086956522, | |
| "loss": 0.1292, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.9645776566757494, | |
| "eval_loss": 0.13422255218029022, | |
| "eval_runtime": 17.8577, | |
| "eval_samples_per_second": 41.103, | |
| "eval_steps_per_second": 5.152, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013206521739130434, | |
| "loss": 0.1483, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00012934782608695652, | |
| "loss": 0.1082, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001266304347826087, | |
| "loss": 0.1206, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012391304347826086, | |
| "loss": 0.1232, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.9741144414168937, | |
| "eval_loss": 0.08246026188135147, | |
| "eval_runtime": 17.6614, | |
| "eval_samples_per_second": 41.56, | |
| "eval_steps_per_second": 5.209, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00012119565217391304, | |
| "loss": 0.0745, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00011847826086956522, | |
| "loss": 0.1076, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00011576086956521739, | |
| "loss": 0.0651, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00011304347826086956, | |
| "loss": 0.0782, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.9741144414168937, | |
| "eval_loss": 0.09356382489204407, | |
| "eval_runtime": 17.6794, | |
| "eval_samples_per_second": 41.517, | |
| "eval_steps_per_second": 5.204, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00011032608695652174, | |
| "loss": 0.115, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0001076086956521739, | |
| "loss": 0.0635, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0001048913043478261, | |
| "loss": 0.1061, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00010217391304347828, | |
| "loss": 0.1171, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.9768392370572208, | |
| "eval_loss": 0.06765041500329971, | |
| "eval_runtime": 18.1237, | |
| "eval_samples_per_second": 40.499, | |
| "eval_steps_per_second": 5.076, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.945652173913043e-05, | |
| "loss": 0.086, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.673913043478261e-05, | |
| "loss": 0.0685, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.402173913043478e-05, | |
| "loss": 0.052, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.130434782608696e-05, | |
| "loss": 0.0549, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.9727520435967303, | |
| "eval_loss": 0.06371203064918518, | |
| "eval_runtime": 17.6078, | |
| "eval_samples_per_second": 41.686, | |
| "eval_steps_per_second": 5.225, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.858695652173914e-05, | |
| "loss": 0.058, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.586956521739131e-05, | |
| "loss": 0.1215, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.315217391304349e-05, | |
| "loss": 0.1048, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 8.043478260869566e-05, | |
| "loss": 0.0603, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.9768392370572208, | |
| "eval_loss": 0.06182762607932091, | |
| "eval_runtime": 18.5478, | |
| "eval_samples_per_second": 39.574, | |
| "eval_steps_per_second": 4.96, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.771739130434783e-05, | |
| "loss": 0.0648, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0512, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.228260869565217e-05, | |
| "loss": 0.0867, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 6.956521739130436e-05, | |
| "loss": 0.049, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.9768392370572208, | |
| "eval_loss": 0.057097259908914566, | |
| "eval_runtime": 17.4967, | |
| "eval_samples_per_second": 41.951, | |
| "eval_steps_per_second": 5.258, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.684782608695652e-05, | |
| "loss": 0.0389, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 6.413043478260869e-05, | |
| "loss": 0.087, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.141304347826087e-05, | |
| "loss": 0.0733, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.869565217391305e-05, | |
| "loss": 0.0199, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.9727520435967303, | |
| "eval_loss": 0.07243036478757858, | |
| "eval_runtime": 17.9356, | |
| "eval_samples_per_second": 40.924, | |
| "eval_steps_per_second": 5.129, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.5978260869565226e-05, | |
| "loss": 0.0629, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.32608695652174e-05, | |
| "loss": 0.0752, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.054347826086957e-05, | |
| "loss": 0.0894, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 4.782608695652174e-05, | |
| "loss": 0.0787, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.06181642785668373, | |
| "eval_runtime": 17.5051, | |
| "eval_samples_per_second": 41.931, | |
| "eval_steps_per_second": 5.256, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.5108695652173916e-05, | |
| "loss": 0.1088, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.239130434782609e-05, | |
| "loss": 0.0586, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.9673913043478264e-05, | |
| "loss": 0.0972, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.695652173913043e-05, | |
| "loss": 0.049, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.058572325855493546, | |
| "eval_runtime": 19.2109, | |
| "eval_samples_per_second": 38.207, | |
| "eval_steps_per_second": 4.789, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 3.423913043478261e-05, | |
| "loss": 0.0505, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.152173913043479e-05, | |
| "loss": 0.0226, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.8804347826086957e-05, | |
| "loss": 0.0754, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.608695652173913e-05, | |
| "loss": 0.0356, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.05890597403049469, | |
| "eval_runtime": 17.4571, | |
| "eval_samples_per_second": 42.046, | |
| "eval_steps_per_second": 5.27, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.3369565217391306e-05, | |
| "loss": 0.0261, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.065217391304348e-05, | |
| "loss": 0.1149, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.793478260869565e-05, | |
| "loss": 0.0605, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.5217391304347828e-05, | |
| "loss": 0.0761, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.062350884079933167, | |
| "eval_runtime": 17.6913, | |
| "eval_samples_per_second": 41.489, | |
| "eval_steps_per_second": 5.2, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1011, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 9.782608695652175e-06, | |
| "loss": 0.0473, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 7.065217391304347e-06, | |
| "loss": 0.0595, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.347826086956522e-06, | |
| "loss": 0.0566, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.059235844761133194, | |
| "eval_runtime": 17.9656, | |
| "eval_samples_per_second": 40.856, | |
| "eval_steps_per_second": 5.121, | |
| "step": 720 | |
| } | |
| ], | |
| "max_steps": 736, | |
| "num_train_epochs": 2, | |
| "total_flos": 8.915533311547699e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |