| { | |
| "best_metric": 0.07230885326862335, | |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-1120", | |
| "epoch": 3.9130434782608696, | |
| "global_step": 1440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0001986413043478261, | |
| "loss": 0.4743, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0001972826086956522, | |
| "loss": 0.1414, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019592391304347828, | |
| "loss": 0.1163, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019456521739130434, | |
| "loss": 0.1002, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.9591280653950953, | |
| "eval_loss": 0.15037867426872253, | |
| "eval_runtime": 17.5557, | |
| "eval_samples_per_second": 41.81, | |
| "eval_steps_per_second": 5.24, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00019320652173913046, | |
| "loss": 0.1429, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019184782608695653, | |
| "loss": 0.1156, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019048913043478262, | |
| "loss": 0.1891, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001891304347826087, | |
| "loss": 0.1599, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.9659400544959128, | |
| "eval_loss": 0.11539561301469803, | |
| "eval_runtime": 17.7744, | |
| "eval_samples_per_second": 41.295, | |
| "eval_steps_per_second": 5.176, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0001877717391304348, | |
| "loss": 0.1202, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0001864130434782609, | |
| "loss": 0.0683, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00018505434782608698, | |
| "loss": 0.1397, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00018369565217391304, | |
| "loss": 0.1965, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.9618528610354223, | |
| "eval_loss": 0.12847301363945007, | |
| "eval_runtime": 17.015, | |
| "eval_samples_per_second": 43.138, | |
| "eval_steps_per_second": 5.407, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00018233695652173913, | |
| "loss": 0.1995, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00018097826086956522, | |
| "loss": 0.0787, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0001796195652173913, | |
| "loss": 0.149, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001782608695652174, | |
| "loss": 0.1032, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.9632152588555858, | |
| "eval_loss": 0.12126161903142929, | |
| "eval_runtime": 17.4318, | |
| "eval_samples_per_second": 42.107, | |
| "eval_steps_per_second": 5.278, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0001769021739130435, | |
| "loss": 0.0208, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00017554347826086956, | |
| "loss": 0.122, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00017418478260869567, | |
| "loss": 0.0703, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00017282608695652174, | |
| "loss": 0.0667, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.10148755460977554, | |
| "eval_runtime": 17.4321, | |
| "eval_samples_per_second": 42.106, | |
| "eval_steps_per_second": 5.278, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00017146739130434783, | |
| "loss": 0.1348, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00017010869565217392, | |
| "loss": 0.0833, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00016875, | |
| "loss": 0.1079, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0001673913043478261, | |
| "loss": 0.0919, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.9659400544959128, | |
| "eval_loss": 0.11204984039068222, | |
| "eval_runtime": 18.116, | |
| "eval_samples_per_second": 40.517, | |
| "eval_steps_per_second": 5.078, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0001660326086956522, | |
| "loss": 0.1255, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00016467391304347828, | |
| "loss": 0.0411, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00016331521739130434, | |
| "loss": 0.1268, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00016195652173913046, | |
| "loss": 0.0758, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.9645776566757494, | |
| "eval_loss": 0.12156815826892853, | |
| "eval_runtime": 17.22, | |
| "eval_samples_per_second": 42.625, | |
| "eval_steps_per_second": 5.343, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00016059782608695652, | |
| "loss": 0.0613, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00015923913043478262, | |
| "loss": 0.0498, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0001578804347826087, | |
| "loss": 0.0627, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0001565217391304348, | |
| "loss": 0.0477, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.9632152588555858, | |
| "eval_loss": 0.13319940865039825, | |
| "eval_runtime": 17.3065, | |
| "eval_samples_per_second": 42.412, | |
| "eval_steps_per_second": 5.316, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00015516304347826086, | |
| "loss": 0.1778, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00015380434782608698, | |
| "loss": 0.199, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00015244565217391304, | |
| "loss": 0.0943, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00015108695652173913, | |
| "loss": 0.0873, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.10057565569877625, | |
| "eval_runtime": 17.221, | |
| "eval_samples_per_second": 42.622, | |
| "eval_steps_per_second": 5.342, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00014972826086956522, | |
| "loss": 0.0584, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.0001483695652173913, | |
| "loss": 0.1169, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0001470108695652174, | |
| "loss": 0.05, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0001456521739130435, | |
| "loss": 0.1141, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.9659400544959128, | |
| "eval_loss": 0.0971706286072731, | |
| "eval_runtime": 17.0669, | |
| "eval_samples_per_second": 43.007, | |
| "eval_steps_per_second": 5.391, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00014429347826086956, | |
| "loss": 0.0666, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00014293478260869567, | |
| "loss": 0.0744, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00014157608695652174, | |
| "loss": 0.0991, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00014021739130434783, | |
| "loss": 0.0581, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.08075320720672607, | |
| "eval_runtime": 17.6899, | |
| "eval_samples_per_second": 41.492, | |
| "eval_steps_per_second": 5.201, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00013885869565217392, | |
| "loss": 0.0953, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.054, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.0001361413043478261, | |
| "loss": 0.0671, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0001347826086956522, | |
| "loss": 0.0721, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.1063413992524147, | |
| "eval_runtime": 17.0062, | |
| "eval_samples_per_second": 43.161, | |
| "eval_steps_per_second": 5.41, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00013342391304347828, | |
| "loss": 0.0161, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00013206521739130434, | |
| "loss": 0.0899, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00013070652173913046, | |
| "loss": 0.2204, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00012934782608695652, | |
| "loss": 0.1076, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.10360449552536011, | |
| "eval_runtime": 17.2631, | |
| "eval_samples_per_second": 42.519, | |
| "eval_steps_per_second": 5.329, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00012798913043478261, | |
| "loss": 0.1218, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0001266304347826087, | |
| "loss": 0.0423, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0001252717391304348, | |
| "loss": 0.0945, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00012391304347826086, | |
| "loss": 0.103, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.0836576595902443, | |
| "eval_runtime": 17.3063, | |
| "eval_samples_per_second": 42.412, | |
| "eval_steps_per_second": 5.316, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.00012255434782608698, | |
| "loss": 0.0906, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00012119565217391304, | |
| "loss": 0.0521, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00011983695652173914, | |
| "loss": 0.0843, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00011847826086956522, | |
| "loss": 0.0865, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.07968933880329132, | |
| "eval_runtime": 17.7434, | |
| "eval_samples_per_second": 41.367, | |
| "eval_steps_per_second": 5.185, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.00011711956521739131, | |
| "loss": 0.0621, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00011576086956521739, | |
| "loss": 0.0528, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00011440217391304349, | |
| "loss": 0.0554, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00011304347826086956, | |
| "loss": 0.035, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.07715282589197159, | |
| "eval_runtime": 16.9577, | |
| "eval_samples_per_second": 43.284, | |
| "eval_steps_per_second": 5.425, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00011168478260869566, | |
| "loss": 0.0544, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00011032608695652174, | |
| "loss": 0.0654, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00010896739130434783, | |
| "loss": 0.1072, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001076086956521739, | |
| "loss": 0.0721, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.9659400544959128, | |
| "eval_loss": 0.11290551722049713, | |
| "eval_runtime": 17.1536, | |
| "eval_samples_per_second": 42.79, | |
| "eval_steps_per_second": 5.363, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00010625000000000001, | |
| "loss": 0.0804, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.0001048913043478261, | |
| "loss": 0.0376, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.00010353260869565218, | |
| "loss": 0.0543, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00010217391304347828, | |
| "loss": 0.0365, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_accuracy": 0.9645776566757494, | |
| "eval_loss": 0.14068667590618134, | |
| "eval_runtime": 17.8845, | |
| "eval_samples_per_second": 41.041, | |
| "eval_steps_per_second": 5.144, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00010081521739130436, | |
| "loss": 0.0898, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.945652173913043e-05, | |
| "loss": 0.0295, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.809782608695652e-05, | |
| "loss": 0.0819, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.673913043478261e-05, | |
| "loss": 0.066, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_accuracy": 0.9618528610354223, | |
| "eval_loss": 0.12197968363761902, | |
| "eval_runtime": 17.0527, | |
| "eval_samples_per_second": 43.043, | |
| "eval_steps_per_second": 5.395, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.538043478260869e-05, | |
| "loss": 0.0357, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.402173913043478e-05, | |
| "loss": 0.1752, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.266304347826087e-05, | |
| "loss": 0.0781, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.130434782608696e-05, | |
| "loss": 0.0515, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.0839233249425888, | |
| "eval_runtime": 17.1241, | |
| "eval_samples_per_second": 42.864, | |
| "eval_steps_per_second": 5.373, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.994565217391305e-05, | |
| "loss": 0.092, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.858695652173914e-05, | |
| "loss": 0.0738, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.722826086956522e-05, | |
| "loss": 0.0888, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 8.586956521739131e-05, | |
| "loss": 0.0223, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.0724789947271347, | |
| "eval_runtime": 17.0736, | |
| "eval_samples_per_second": 42.99, | |
| "eval_steps_per_second": 5.388, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.45108695652174e-05, | |
| "loss": 0.0657, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.315217391304349e-05, | |
| "loss": 0.0554, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.179347826086957e-05, | |
| "loss": 0.038, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 8.043478260869566e-05, | |
| "loss": 0.0131, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.9632152588555858, | |
| "eval_loss": 0.13555656373500824, | |
| "eval_runtime": 17.1277, | |
| "eval_samples_per_second": 42.855, | |
| "eval_steps_per_second": 5.371, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 7.907608695652175e-05, | |
| "loss": 0.0298, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 7.771739130434783e-05, | |
| "loss": 0.1934, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.635869565217392e-05, | |
| "loss": 0.0707, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0701, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_accuracy": 0.9673024523160763, | |
| "eval_loss": 0.09190992265939713, | |
| "eval_runtime": 17.6207, | |
| "eval_samples_per_second": 41.656, | |
| "eval_steps_per_second": 5.221, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.364130434782608e-05, | |
| "loss": 0.055, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 7.228260869565217e-05, | |
| "loss": 0.0723, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 7.092391304347826e-05, | |
| "loss": 0.0969, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.956521739130436e-05, | |
| "loss": 0.0605, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.07686856389045715, | |
| "eval_runtime": 17.3639, | |
| "eval_samples_per_second": 42.272, | |
| "eval_steps_per_second": 5.298, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 6.820652173913043e-05, | |
| "loss": 0.0986, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 6.684782608695652e-05, | |
| "loss": 0.0368, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 6.548913043478261e-05, | |
| "loss": 0.0392, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.413043478260869e-05, | |
| "loss": 0.0382, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.08056215196847916, | |
| "eval_runtime": 17.5095, | |
| "eval_samples_per_second": 41.92, | |
| "eval_steps_per_second": 5.254, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.277173913043478e-05, | |
| "loss": 0.0602, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.141304347826087e-05, | |
| "loss": 0.0264, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 6.0054347826086955e-05, | |
| "loss": 0.0458, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.869565217391305e-05, | |
| "loss": 0.0409, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_accuracy": 0.952316076294278, | |
| "eval_loss": 0.0872640386223793, | |
| "eval_runtime": 17.1006, | |
| "eval_samples_per_second": 42.922, | |
| "eval_steps_per_second": 5.38, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 5.7336956521739136e-05, | |
| "loss": 0.0527, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5.5978260869565226e-05, | |
| "loss": 0.0658, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 5.461956521739131e-05, | |
| "loss": 0.0592, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 5.32608695652174e-05, | |
| "loss": 0.0302, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.13446882367134094, | |
| "eval_runtime": 17.5217, | |
| "eval_samples_per_second": 41.891, | |
| "eval_steps_per_second": 5.251, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 5.1902173913043484e-05, | |
| "loss": 0.1134, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 5.054347826086957e-05, | |
| "loss": 0.0676, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.918478260869566e-05, | |
| "loss": 0.0615, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.782608695652174e-05, | |
| "loss": 0.0528, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.07230885326862335, | |
| "eval_runtime": 16.9893, | |
| "eval_samples_per_second": 43.204, | |
| "eval_steps_per_second": 5.415, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.646739130434783e-05, | |
| "loss": 0.0578, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.5108695652173916e-05, | |
| "loss": 0.0393, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.05, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 4.239130434782609e-05, | |
| "loss": 0.0657, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.09432989358901978, | |
| "eval_runtime": 17.3576, | |
| "eval_samples_per_second": 42.287, | |
| "eval_steps_per_second": 5.3, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.1032608695652174e-05, | |
| "loss": 0.0554, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.9673913043478264e-05, | |
| "loss": 0.0391, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.831521739130435e-05, | |
| "loss": 0.0367, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 3.695652173913043e-05, | |
| "loss": 0.0546, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.1177757978439331, | |
| "eval_runtime": 17.0297, | |
| "eval_samples_per_second": 43.101, | |
| "eval_steps_per_second": 5.402, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 3.559782608695653e-05, | |
| "loss": 0.0321, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.423913043478261e-05, | |
| "loss": 0.0274, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.2880434782608696e-05, | |
| "loss": 0.0554, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 3.152173913043479e-05, | |
| "loss": 0.0307, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.11735932528972626, | |
| "eval_runtime": 17.2834, | |
| "eval_samples_per_second": 42.468, | |
| "eval_steps_per_second": 5.323, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3.016304347826087e-05, | |
| "loss": 0.0863, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 2.8804347826086957e-05, | |
| "loss": 0.066, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.7445652173913044e-05, | |
| "loss": 0.0261, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 2.608695652173913e-05, | |
| "loss": 0.0388, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.07767504453659058, | |
| "eval_runtime": 17.498, | |
| "eval_samples_per_second": 41.948, | |
| "eval_steps_per_second": 5.258, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.472826086956522e-05, | |
| "loss": 0.0422, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.3369565217391306e-05, | |
| "loss": 0.0442, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.2010869565217393e-05, | |
| "loss": 0.0217, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.065217391304348e-05, | |
| "loss": 0.0127, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.09292852133512497, | |
| "eval_runtime": 17.0066, | |
| "eval_samples_per_second": 43.16, | |
| "eval_steps_per_second": 5.41, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.9293478260869567e-05, | |
| "loss": 0.0559, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.793478260869565e-05, | |
| "loss": 0.0453, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6576086956521738e-05, | |
| "loss": 0.0321, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.5217391304347828e-05, | |
| "loss": 0.0408, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.08991661667823792, | |
| "eval_runtime": 16.9692, | |
| "eval_samples_per_second": 43.255, | |
| "eval_steps_per_second": 5.422, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.3858695652173914e-05, | |
| "loss": 0.038, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0416, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.1141304347826088e-05, | |
| "loss": 0.1011, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 9.782608695652175e-06, | |
| "loss": 0.0445, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_accuracy": 0.9686648501362398, | |
| "eval_loss": 0.08311757445335388, | |
| "eval_runtime": 17.0031, | |
| "eval_samples_per_second": 43.169, | |
| "eval_steps_per_second": 5.411, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.423913043478262e-06, | |
| "loss": 0.0526, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 7.065217391304347e-06, | |
| "loss": 0.0458, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.706521739130435e-06, | |
| "loss": 0.0837, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.347826086956522e-06, | |
| "loss": 0.0493, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_accuracy": 0.9632152588555858, | |
| "eval_loss": 0.07967270910739899, | |
| "eval_runtime": 16.9823, | |
| "eval_samples_per_second": 43.222, | |
| "eval_steps_per_second": 5.417, | |
| "step": 1440 | |
| } | |
| ], | |
| "max_steps": 1472, | |
| "num_train_epochs": 4, | |
| "total_flos": 1.781944272047278e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |