| { | |
| "best_metric": 0.03899691626429558, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-720", | |
| "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": 0.5574, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019456521739130434, | |
| "loss": 0.3197, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019184782608695653, | |
| "loss": 0.271, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.0001891304347826087, | |
| "loss": 0.2484, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.8801089918256131, | |
| "eval_loss": 0.21790249645709991, | |
| "eval_runtime": 16.4161, | |
| "eval_samples_per_second": 44.712, | |
| "eval_steps_per_second": 5.604, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0001864130434782609, | |
| "loss": 0.2372, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00018369565217391304, | |
| "loss": 0.1945, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00018097826086956522, | |
| "loss": 0.2186, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001782608695652174, | |
| "loss": 0.3033, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.8678474114441417, | |
| "eval_loss": 0.28713682293891907, | |
| "eval_runtime": 18.6159, | |
| "eval_samples_per_second": 39.429, | |
| "eval_steps_per_second": 4.942, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00017554347826086956, | |
| "loss": 0.2893, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00017282608695652174, | |
| "loss": 0.3135, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00017010869565217392, | |
| "loss": 0.2673, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001673913043478261, | |
| "loss": 0.2709, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.9073569482288828, | |
| "eval_loss": 0.23233647644519806, | |
| "eval_runtime": 16.7819, | |
| "eval_samples_per_second": 43.738, | |
| "eval_steps_per_second": 5.482, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00016467391304347828, | |
| "loss": 0.2171, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00016195652173913046, | |
| "loss": 0.2393, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00015923913043478262, | |
| "loss": 0.1934, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001565217391304348, | |
| "loss": 0.1878, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.9318801089918256, | |
| "eval_loss": 0.1636728048324585, | |
| "eval_runtime": 16.6111, | |
| "eval_samples_per_second": 44.187, | |
| "eval_steps_per_second": 5.538, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00015380434782608698, | |
| "loss": 0.2186, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00015108695652173913, | |
| "loss": 0.1952, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.0001483695652173913, | |
| "loss": 0.1546, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001456521739130435, | |
| "loss": 0.1655, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.8678474114441417, | |
| "eval_loss": 0.24226485192775726, | |
| "eval_runtime": 17.096, | |
| "eval_samples_per_second": 42.934, | |
| "eval_steps_per_second": 5.381, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00014293478260869567, | |
| "loss": 0.1843, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00014021739130434783, | |
| "loss": 0.12, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.1686, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001347826086956522, | |
| "loss": 0.2393, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.94141689373297, | |
| "eval_loss": 0.1487017571926117, | |
| "eval_runtime": 16.7397, | |
| "eval_samples_per_second": 43.848, | |
| "eval_steps_per_second": 5.496, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00013206521739130434, | |
| "loss": 0.2159, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00012934782608695652, | |
| "loss": 0.1608, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001266304347826087, | |
| "loss": 0.1508, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00012391304347826086, | |
| "loss": 0.1432, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.9209809264305178, | |
| "eval_loss": 0.1679653823375702, | |
| "eval_runtime": 19.1578, | |
| "eval_samples_per_second": 38.313, | |
| "eval_steps_per_second": 4.802, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00012119565217391304, | |
| "loss": 0.1852, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00011847826086956522, | |
| "loss": 0.1845, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.00011576086956521739, | |
| "loss": 0.1238, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00011304347826086956, | |
| "loss": 0.0986, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.9604904632152589, | |
| "eval_loss": 0.11554770916700363, | |
| "eval_runtime": 16.799, | |
| "eval_samples_per_second": 43.693, | |
| "eval_steps_per_second": 5.477, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00011032608695652174, | |
| "loss": 0.1781, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0001076086956521739, | |
| "loss": 0.1229, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0001048913043478261, | |
| "loss": 0.1238, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00010217391304347828, | |
| "loss": 0.0729, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.9577656675749319, | |
| "eval_loss": 0.10841722786426544, | |
| "eval_runtime": 17.3667, | |
| "eval_samples_per_second": 42.265, | |
| "eval_steps_per_second": 5.297, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.945652173913043e-05, | |
| "loss": 0.1156, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.673913043478261e-05, | |
| "loss": 0.0934, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 9.402173913043478e-05, | |
| "loss": 0.082, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 9.130434782608696e-05, | |
| "loss": 0.0553, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.9632152588555858, | |
| "eval_loss": 0.09444452077150345, | |
| "eval_runtime": 16.7805, | |
| "eval_samples_per_second": 43.741, | |
| "eval_steps_per_second": 5.483, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 8.858695652173914e-05, | |
| "loss": 0.1067, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 8.586956521739131e-05, | |
| "loss": 0.0544, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 8.315217391304349e-05, | |
| "loss": 0.1047, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 8.043478260869566e-05, | |
| "loss": 0.0395, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.9673024523160763, | |
| "eval_loss": 0.07385484129190445, | |
| "eval_runtime": 17.3745, | |
| "eval_samples_per_second": 42.246, | |
| "eval_steps_per_second": 5.295, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.771739130434783e-05, | |
| "loss": 0.0566, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0625, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.228260869565217e-05, | |
| "loss": 0.0961, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 6.956521739130436e-05, | |
| "loss": 0.0178, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.9673024523160763, | |
| "eval_loss": 0.10838904976844788, | |
| "eval_runtime": 16.8091, | |
| "eval_samples_per_second": 43.667, | |
| "eval_steps_per_second": 5.473, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.684782608695652e-05, | |
| "loss": 0.0641, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 6.413043478260869e-05, | |
| "loss": 0.0497, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 6.141304347826087e-05, | |
| "loss": 0.0364, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 5.869565217391305e-05, | |
| "loss": 0.0273, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.9727520435967303, | |
| "eval_loss": 0.07216904312372208, | |
| "eval_runtime": 17.0234, | |
| "eval_samples_per_second": 43.117, | |
| "eval_steps_per_second": 5.404, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 5.5978260869565226e-05, | |
| "loss": 0.0558, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 5.32608695652174e-05, | |
| "loss": 0.0379, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.054347826086957e-05, | |
| "loss": 0.0569, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 4.782608695652174e-05, | |
| "loss": 0.0273, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.9754768392370572, | |
| "eval_loss": 0.0689365416765213, | |
| "eval_runtime": 16.7035, | |
| "eval_samples_per_second": 43.943, | |
| "eval_steps_per_second": 5.508, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 4.5108695652173916e-05, | |
| "loss": 0.0087, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.239130434782609e-05, | |
| "loss": 0.0347, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.9673913043478264e-05, | |
| "loss": 0.0196, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.695652173913043e-05, | |
| "loss": 0.0271, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.9795640326975477, | |
| "eval_loss": 0.07317620515823364, | |
| "eval_runtime": 16.8063, | |
| "eval_samples_per_second": 43.674, | |
| "eval_steps_per_second": 5.474, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 3.423913043478261e-05, | |
| "loss": 0.0951, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.152173913043479e-05, | |
| "loss": 0.0726, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.8804347826086957e-05, | |
| "loss": 0.01, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.608695652173913e-05, | |
| "loss": 0.0294, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.9822888283378747, | |
| "eval_loss": 0.057586103677749634, | |
| "eval_runtime": 17.373, | |
| "eval_samples_per_second": 42.249, | |
| "eval_steps_per_second": 5.296, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 2.3369565217391306e-05, | |
| "loss": 0.017, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.065217391304348e-05, | |
| "loss": 0.0271, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.793478260869565e-05, | |
| "loss": 0.0572, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.5217391304347828e-05, | |
| "loss": 0.0449, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.9877384196185286, | |
| "eval_loss": 0.04121558368206024, | |
| "eval_runtime": 16.5738, | |
| "eval_samples_per_second": 44.287, | |
| "eval_steps_per_second": 5.551, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0261, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 9.782608695652175e-06, | |
| "loss": 0.0091, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 7.065217391304347e-06, | |
| "loss": 0.0799, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.347826086956522e-06, | |
| "loss": 0.0076, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.9877384196185286, | |
| "eval_loss": 0.03899691626429558, | |
| "eval_runtime": 17.1545, | |
| "eval_samples_per_second": 42.788, | |
| "eval_steps_per_second": 5.363, | |
| "step": 720 | |
| } | |
| ], | |
| "max_steps": 736, | |
| "num_train_epochs": 2, | |
| "total_flos": 8.915533311547699e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |