| { | |
| "best_metric": 0.07266250252723694, | |
| "best_model_checkpoint": "./vit-base-flowers102/checkpoint-1300", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1796, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00019888641425389755, | |
| "loss": 4.5016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019777282850779511, | |
| "loss": 4.311, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019665924276169265, | |
| "loss": 4.1383, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0001955456570155902, | |
| "loss": 3.8337, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019443207126948776, | |
| "loss": 3.6639, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00019331848552338533, | |
| "loss": 3.4038, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019220489977728286, | |
| "loss": 3.2495, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001910913140311804, | |
| "loss": 2.9898, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00018997772828507797, | |
| "loss": 2.8136, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001888641425389755, | |
| "loss": 2.5779, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.7774509803921569, | |
| "eval_loss": 2.8894846439361572, | |
| "eval_runtime": 16.9816, | |
| "eval_samples_per_second": 60.065, | |
| "eval_steps_per_second": 7.538, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00018775055679287305, | |
| "loss": 2.4361, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00018663697104677061, | |
| "loss": 2.2851, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00018552338530066815, | |
| "loss": 2.1522, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00018440979955456572, | |
| "loss": 2.0138, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00018329621380846326, | |
| "loss": 1.878, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00018218262806236082, | |
| "loss": 1.7814, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00018106904231625836, | |
| "loss": 1.5264, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0001799554565701559, | |
| "loss": 1.5172, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00017884187082405347, | |
| "loss": 1.4431, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.000177728285077951, | |
| "loss": 1.2226, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.9254901960784314, | |
| "eval_loss": 1.5941656827926636, | |
| "eval_runtime": 16.7137, | |
| "eval_samples_per_second": 61.028, | |
| "eval_steps_per_second": 7.658, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00017661469933184855, | |
| "loss": 1.2047, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00017550111358574611, | |
| "loss": 1.2643, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00017438752783964368, | |
| "loss": 1.1202, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00017327394209354122, | |
| "loss": 0.9402, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00017216035634743876, | |
| "loss": 0.8425, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00017104677060133632, | |
| "loss": 0.7436, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00016993318485523386, | |
| "loss": 0.6685, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001688195991091314, | |
| "loss": 0.6464, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016770601336302897, | |
| "loss": 0.5846, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001665924276169265, | |
| "loss": 0.606, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.9529411764705882, | |
| "eval_loss": 0.8011782169342041, | |
| "eval_runtime": 17.625, | |
| "eval_samples_per_second": 57.872, | |
| "eval_steps_per_second": 7.262, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00016547884187082405, | |
| "loss": 0.5889, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0001643652561247216, | |
| "loss": 0.5238, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001633630289532294, | |
| "loss": 0.5481, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00016224944320712695, | |
| "loss": 0.5891, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00016113585746102451, | |
| "loss": 0.4343, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00016002227171492205, | |
| "loss": 0.4681, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.0001589086859688196, | |
| "loss": 0.4484, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00015779510022271716, | |
| "loss": 0.3993, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0001566815144766147, | |
| "loss": 0.4087, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00015556792873051224, | |
| "loss": 0.3413, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_accuracy": 0.9705882352941176, | |
| "eval_loss": 0.4845229685306549, | |
| "eval_runtime": 16.4436, | |
| "eval_samples_per_second": 62.03, | |
| "eval_steps_per_second": 7.784, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0001544543429844098, | |
| "loss": 0.3052, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00015334075723830737, | |
| "loss": 0.2869, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0001522271714922049, | |
| "loss": 0.3613, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 0.00015111358574610245, | |
| "loss": 0.3189, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2199, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00014888641425389755, | |
| "loss": 0.1689, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0001477728285077951, | |
| "loss": 0.2321, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00014665924276169266, | |
| "loss": 0.1722, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0001455456570155902, | |
| "loss": 0.1581, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00014443207126948776, | |
| "loss": 0.1571, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.9813725490196078, | |
| "eval_loss": 0.26114311814308167, | |
| "eval_runtime": 18.4125, | |
| "eval_samples_per_second": 55.397, | |
| "eval_steps_per_second": 6.952, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0001433184855233853, | |
| "loss": 0.1276, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00014220489977728287, | |
| "loss": 0.1203, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0001410913140311804, | |
| "loss": 0.1005, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00013997772828507795, | |
| "loss": 0.1351, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0001388641425389755, | |
| "loss": 0.0928, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00013775055679287305, | |
| "loss": 0.0848, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.0001366369710467706, | |
| "loss": 0.1251, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00013552338530066816, | |
| "loss": 0.0774, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00013440979955456572, | |
| "loss": 0.097, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00013329621380846326, | |
| "loss": 0.1237, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.9784313725490196, | |
| "eval_loss": 0.16912123560905457, | |
| "eval_runtime": 16.6514, | |
| "eval_samples_per_second": 61.256, | |
| "eval_steps_per_second": 7.687, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0001321826280623608, | |
| "loss": 0.0724, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00013106904231625837, | |
| "loss": 0.0754, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0001299554565701559, | |
| "loss": 0.065, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00012884187082405345, | |
| "loss": 0.0679, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.000127728285077951, | |
| "loss": 0.0579, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00012661469933184855, | |
| "loss": 0.0524, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 0.0001255011135857461, | |
| "loss": 0.0727, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00012438752783964366, | |
| "loss": 0.0721, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00012327394209354122, | |
| "loss": 0.0665, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00012216035634743876, | |
| "loss": 0.049, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_accuracy": 0.9892156862745098, | |
| "eval_loss": 0.11457555741071701, | |
| "eval_runtime": 16.6002, | |
| "eval_samples_per_second": 61.445, | |
| "eval_steps_per_second": 7.711, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00012104677060133632, | |
| "loss": 0.0695, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00011993318485523385, | |
| "loss": 0.0487, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00011881959910913141, | |
| "loss": 0.0432, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.00011770601336302896, | |
| "loss": 0.0509, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0001165924276169265, | |
| "loss": 0.1105, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00011547884187082405, | |
| "loss": 0.0894, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0001143652561247216, | |
| "loss": 0.0668, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00011325167037861917, | |
| "loss": 0.0663, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00011213808463251671, | |
| "loss": 0.0423, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00011102449888641426, | |
| "loss": 0.0763, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_accuracy": 0.9862745098039216, | |
| "eval_loss": 0.12085805088281631, | |
| "eval_runtime": 17.6825, | |
| "eval_samples_per_second": 57.684, | |
| "eval_steps_per_second": 7.239, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.00010991091314031181, | |
| "loss": 0.0545, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00010879732739420935, | |
| "loss": 0.0563, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001076837416481069, | |
| "loss": 0.0743, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00010657015590200446, | |
| "loss": 0.04, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.000105456570155902, | |
| "loss": 0.1064, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00010434298440979955, | |
| "loss": 0.0389, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00010322939866369712, | |
| "loss": 0.0372, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00010211581291759467, | |
| "loss": 0.0514, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00010100222717149221, | |
| "loss": 0.0406, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.988864142538976e-05, | |
| "loss": 0.0864, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9803921568627451, | |
| "eval_loss": 0.12226687371730804, | |
| "eval_runtime": 16.7815, | |
| "eval_samples_per_second": 60.781, | |
| "eval_steps_per_second": 7.627, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.877505567928731e-05, | |
| "loss": 0.0337, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.766146993318485e-05, | |
| "loss": 0.055, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.65478841870824e-05, | |
| "loss": 0.0329, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.543429844097996e-05, | |
| "loss": 0.028, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.432071269487751e-05, | |
| "loss": 0.0314, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.320712694877506e-05, | |
| "loss": 0.0515, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.20935412026726e-05, | |
| "loss": 0.0294, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 9.097995545657017e-05, | |
| "loss": 0.0318, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.986636971046771e-05, | |
| "loss": 0.0332, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.875278396436526e-05, | |
| "loss": 0.0786, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_accuracy": 0.9833333333333333, | |
| "eval_loss": 0.1075347512960434, | |
| "eval_runtime": 17.0089, | |
| "eval_samples_per_second": 59.969, | |
| "eval_steps_per_second": 7.525, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.763919821826281e-05, | |
| "loss": 0.0269, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8.652561247216035e-05, | |
| "loss": 0.0247, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 8.541202672605792e-05, | |
| "loss": 0.0234, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 8.429844097995546e-05, | |
| "loss": 0.0241, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.318485523385301e-05, | |
| "loss": 0.0227, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.207126948775056e-05, | |
| "loss": 0.0199, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 8.095768374164812e-05, | |
| "loss": 0.023, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 7.984409799554567e-05, | |
| "loss": 0.0254, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 7.873051224944321e-05, | |
| "loss": 0.0209, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 7.761692650334076e-05, | |
| "loss": 0.0269, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_accuracy": 0.984313725490196, | |
| "eval_loss": 0.09193924814462662, | |
| "eval_runtime": 17.7915, | |
| "eval_samples_per_second": 57.331, | |
| "eval_steps_per_second": 7.194, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.650334075723831e-05, | |
| "loss": 0.0207, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 7.538975501113587e-05, | |
| "loss": 0.0189, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.427616926503342e-05, | |
| "loss": 0.0213, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 7.316258351893096e-05, | |
| "loss": 0.02, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.204899777282851e-05, | |
| "loss": 0.0683, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 7.093541202672605e-05, | |
| "loss": 0.0187, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.982182628062362e-05, | |
| "loss": 0.0209, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 6.870824053452117e-05, | |
| "loss": 0.0176, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 6.759465478841871e-05, | |
| "loss": 0.0199, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 6.648106904231626e-05, | |
| "loss": 0.0178, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_accuracy": 0.9872549019607844, | |
| "eval_loss": 0.07947444170713425, | |
| "eval_runtime": 17.2701, | |
| "eval_samples_per_second": 59.062, | |
| "eval_steps_per_second": 7.412, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 6.536748329621381e-05, | |
| "loss": 0.0184, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.425389755011137e-05, | |
| "loss": 0.0182, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.314031180400892e-05, | |
| "loss": 0.0169, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 6.202672605790646e-05, | |
| "loss": 0.0169, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 6.091314031180401e-05, | |
| "loss": 0.0172, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 5.979955456570156e-05, | |
| "loss": 0.0189, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.8685968819599115e-05, | |
| "loss": 0.0194, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 5.757238307349666e-05, | |
| "loss": 0.0167, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 5.6458797327394206e-05, | |
| "loss": 0.0177, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 5.5345211581291766e-05, | |
| "loss": 0.0165, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.9872549019607844, | |
| "eval_loss": 0.07266250252723694, | |
| "eval_runtime": 16.2341, | |
| "eval_samples_per_second": 62.831, | |
| "eval_steps_per_second": 7.885, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.423162583518931e-05, | |
| "loss": 0.0165, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5.3118040089086864e-05, | |
| "loss": 0.0165, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 5.200445434298441e-05, | |
| "loss": 0.0158, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 5.0890868596881956e-05, | |
| "loss": 0.0155, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.977728285077951e-05, | |
| "loss": 0.0143, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.866369710467706e-05, | |
| "loss": 0.0142, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.7550111358574614e-05, | |
| "loss": 0.0144, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.643652561247217e-05, | |
| "loss": 0.0139, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.532293986636971e-05, | |
| "loss": 0.0148, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.420935412026726e-05, | |
| "loss": 0.0144, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_accuracy": 0.9852941176470589, | |
| "eval_loss": 0.07835763692855835, | |
| "eval_runtime": 17.268, | |
| "eval_samples_per_second": 59.069, | |
| "eval_steps_per_second": 7.413, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 4.309576837416481e-05, | |
| "loss": 0.0138, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 4.1982182628062364e-05, | |
| "loss": 0.0146, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.0868596881959917e-05, | |
| "loss": 0.015, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.975501113585746e-05, | |
| "loss": 0.0148, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.8641425389755015e-05, | |
| "loss": 0.0144, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.752783964365256e-05, | |
| "loss": 0.0149, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.6414253897550114e-05, | |
| "loss": 0.0134, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 3.5300668151447666e-05, | |
| "loss": 0.0141, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.418708240534521e-05, | |
| "loss": 0.0488, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.3073496659242765e-05, | |
| "loss": 0.0138, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_accuracy": 0.9852941176470589, | |
| "eval_loss": 0.07589561492204666, | |
| "eval_runtime": 16.5936, | |
| "eval_samples_per_second": 61.47, | |
| "eval_steps_per_second": 7.714, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.195991091314031e-05, | |
| "loss": 0.0142, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.084632516703786e-05, | |
| "loss": 0.013, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.9732739420935413e-05, | |
| "loss": 0.0138, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 2.8619153674832965e-05, | |
| "loss": 0.0141, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 2.7505567928730515e-05, | |
| "loss": 0.014, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.639198218262806e-05, | |
| "loss": 0.0136, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.5278396436525613e-05, | |
| "loss": 0.0133, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 2.4164810690423166e-05, | |
| "loss": 0.015, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.3051224944320715e-05, | |
| "loss": 0.013, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.1937639198218264e-05, | |
| "loss": 0.0135, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_accuracy": 0.9862745098039216, | |
| "eval_loss": 0.07374249398708344, | |
| "eval_runtime": 16.5728, | |
| "eval_samples_per_second": 61.546, | |
| "eval_steps_per_second": 7.723, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.0824053452115813e-05, | |
| "loss": 0.0134, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.9710467706013363e-05, | |
| "loss": 0.0132, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 1.8596881959910915e-05, | |
| "loss": 0.0121, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1.7483296213808465e-05, | |
| "loss": 0.033, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6369710467706014e-05, | |
| "loss": 0.0378, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1.5256124721603565e-05, | |
| "loss": 0.0123, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.4142538975501116e-05, | |
| "loss": 0.0138, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.3028953229398663e-05, | |
| "loss": 0.0124, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 1.1915367483296214e-05, | |
| "loss": 0.0128, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.0801781737193764e-05, | |
| "loss": 0.0123, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.9852941176470589, | |
| "eval_loss": 0.07696886360645294, | |
| "eval_runtime": 17.0943, | |
| "eval_samples_per_second": 59.669, | |
| "eval_steps_per_second": 7.488, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 9.688195991091315e-06, | |
| "loss": 0.0133, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.574610244988866e-06, | |
| "loss": 0.013, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 7.461024498886416e-06, | |
| "loss": 0.0122, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 6.347438752783964e-06, | |
| "loss": 0.0124, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 5.233853006681515e-06, | |
| "loss": 0.0117, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.120267260579064e-06, | |
| "loss": 0.0116, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 3.006681514476615e-06, | |
| "loss": 0.0122, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.893095768374165e-06, | |
| "loss": 0.0123, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 7.79510022271715e-07, | |
| "loss": 0.0132, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1796, | |
| "total_flos": 2.2241519461722194e+18, | |
| "train_loss": 0.4137112680814314, | |
| "train_runtime": 1023.9197, | |
| "train_samples_per_second": 28.006, | |
| "train_steps_per_second": 1.754 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1796, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 2.2241519461722194e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |