| { | |
| "best_metric": 0.0030439873225986958, | |
| "best_model_checkpoint": "./Tb_Dataset/checkpoint-1000", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1304, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03067484662576687, | |
| "grad_norm": 1.666305661201477, | |
| "learning_rate": 0.00019846625766871168, | |
| "loss": 0.4847, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06134969325153374, | |
| "grad_norm": 1.4896280765533447, | |
| "learning_rate": 0.00019693251533742332, | |
| "loss": 0.278, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09202453987730061, | |
| "grad_norm": 1.3254238367080688, | |
| "learning_rate": 0.000195398773006135, | |
| "loss": 0.2401, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12269938650306748, | |
| "grad_norm": 1.1399273872375488, | |
| "learning_rate": 0.00019386503067484663, | |
| "loss": 0.188, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15337423312883436, | |
| "grad_norm": 2.320486307144165, | |
| "learning_rate": 0.0001923312883435583, | |
| "loss": 0.1944, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18404907975460122, | |
| "grad_norm": 0.24434129893779755, | |
| "learning_rate": 0.00019079754601226997, | |
| "loss": 0.0738, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2147239263803681, | |
| "grad_norm": 0.7523425221443176, | |
| "learning_rate": 0.0001892638036809816, | |
| "loss": 0.171, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24539877300613497, | |
| "grad_norm": 3.6861462593078613, | |
| "learning_rate": 0.00018773006134969328, | |
| "loss": 0.178, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27607361963190186, | |
| "grad_norm": 2.111417770385742, | |
| "learning_rate": 0.00018619631901840492, | |
| "loss": 0.2488, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3067484662576687, | |
| "grad_norm": 0.8211548328399658, | |
| "learning_rate": 0.00018466257668711656, | |
| "loss": 0.0996, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3067484662576687, | |
| "eval_accuracy": 0.5625, | |
| "eval_loss": 1.0429491996765137, | |
| "eval_runtime": 0.6744, | |
| "eval_samples_per_second": 23.723, | |
| "eval_steps_per_second": 2.965, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3374233128834356, | |
| "grad_norm": 0.12294389307498932, | |
| "learning_rate": 0.00018312883435582823, | |
| "loss": 0.1403, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36809815950920244, | |
| "grad_norm": 1.0773956775665283, | |
| "learning_rate": 0.00018159509202453987, | |
| "loss": 0.0801, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3987730061349693, | |
| "grad_norm": 1.2252161502838135, | |
| "learning_rate": 0.00018006134969325154, | |
| "loss": 0.1667, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4294478527607362, | |
| "grad_norm": 0.3200875520706177, | |
| "learning_rate": 0.00017852760736196318, | |
| "loss": 0.1365, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4601226993865031, | |
| "grad_norm": 0.4385143518447876, | |
| "learning_rate": 0.00017699386503067485, | |
| "loss": 0.1371, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.49079754601226994, | |
| "grad_norm": 0.11226026713848114, | |
| "learning_rate": 0.00017546012269938652, | |
| "loss": 0.042, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5214723926380368, | |
| "grad_norm": 1.9924014806747437, | |
| "learning_rate": 0.00017392638036809816, | |
| "loss": 0.1138, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5521472392638037, | |
| "grad_norm": 2.601227045059204, | |
| "learning_rate": 0.00017239263803680983, | |
| "loss": 0.1729, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5828220858895705, | |
| "grad_norm": 5.278203964233398, | |
| "learning_rate": 0.00017085889570552147, | |
| "loss": 0.1036, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6134969325153374, | |
| "grad_norm": 1.20023512840271, | |
| "learning_rate": 0.00016932515337423314, | |
| "loss": 0.0481, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6134969325153374, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.5665359497070312, | |
| "eval_runtime": 0.4807, | |
| "eval_samples_per_second": 33.285, | |
| "eval_steps_per_second": 4.161, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6441717791411042, | |
| "grad_norm": 0.7395023107528687, | |
| "learning_rate": 0.0001677914110429448, | |
| "loss": 0.1225, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6748466257668712, | |
| "grad_norm": 0.05845380574464798, | |
| "learning_rate": 0.00016625766871165645, | |
| "loss": 0.1034, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7055214723926381, | |
| "grad_norm": 2.7569024562835693, | |
| "learning_rate": 0.00016472392638036812, | |
| "loss": 0.0538, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7361963190184049, | |
| "grad_norm": 0.06222417205572128, | |
| "learning_rate": 0.00016319018404907976, | |
| "loss": 0.0958, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7668711656441718, | |
| "grad_norm": 1.9317396879196167, | |
| "learning_rate": 0.00016165644171779143, | |
| "loss": 0.09, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7975460122699386, | |
| "grad_norm": 3.952806234359741, | |
| "learning_rate": 0.0001601226993865031, | |
| "loss": 0.0878, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8282208588957055, | |
| "grad_norm": 1.5617386102676392, | |
| "learning_rate": 0.00015858895705521474, | |
| "loss": 0.1304, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8588957055214724, | |
| "grad_norm": 0.32101425528526306, | |
| "learning_rate": 0.0001570552147239264, | |
| "loss": 0.0684, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8895705521472392, | |
| "grad_norm": 0.6578021049499512, | |
| "learning_rate": 0.00015552147239263805, | |
| "loss": 0.0917, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9202453987730062, | |
| "grad_norm": 0.03468528762459755, | |
| "learning_rate": 0.0001539877300613497, | |
| "loss": 0.0391, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9202453987730062, | |
| "eval_accuracy": 0.6875, | |
| "eval_loss": 1.0036612749099731, | |
| "eval_runtime": 0.4639, | |
| "eval_samples_per_second": 34.488, | |
| "eval_steps_per_second": 4.311, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.950920245398773, | |
| "grad_norm": 0.04961508885025978, | |
| "learning_rate": 0.00015245398773006136, | |
| "loss": 0.0435, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9815950920245399, | |
| "grad_norm": 0.7167016863822937, | |
| "learning_rate": 0.000150920245398773, | |
| "loss": 0.0667, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0122699386503067, | |
| "grad_norm": 0.3984834551811218, | |
| "learning_rate": 0.00014938650306748467, | |
| "loss": 0.1054, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.0429447852760736, | |
| "grad_norm": 0.09521777182817459, | |
| "learning_rate": 0.0001478527607361963, | |
| "loss": 0.0565, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.0736196319018405, | |
| "grad_norm": 3.836611032485962, | |
| "learning_rate": 0.00014631901840490798, | |
| "loss": 0.0498, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1042944785276074, | |
| "grad_norm": 0.6689714789390564, | |
| "learning_rate": 0.00014478527607361964, | |
| "loss": 0.0693, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1349693251533743, | |
| "grad_norm": 0.42697030305862427, | |
| "learning_rate": 0.00014325153374233129, | |
| "loss": 0.1017, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.165644171779141, | |
| "grad_norm": 0.28322863578796387, | |
| "learning_rate": 0.00014171779141104295, | |
| "loss": 0.0072, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.196319018404908, | |
| "grad_norm": 0.02922147326171398, | |
| "learning_rate": 0.0001401840490797546, | |
| "loss": 0.0355, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2269938650306749, | |
| "grad_norm": 0.4956642687320709, | |
| "learning_rate": 0.00013865030674846626, | |
| "loss": 0.0711, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2269938650306749, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.5200170874595642, | |
| "eval_runtime": 0.4602, | |
| "eval_samples_per_second": 34.769, | |
| "eval_steps_per_second": 4.346, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2576687116564418, | |
| "grad_norm": 0.35631048679351807, | |
| "learning_rate": 0.00013711656441717793, | |
| "loss": 0.018, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2883435582822087, | |
| "grad_norm": 0.053142815828323364, | |
| "learning_rate": 0.00013558282208588957, | |
| "loss": 0.0114, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3190184049079754, | |
| "grad_norm": 3.590878963470459, | |
| "learning_rate": 0.00013404907975460124, | |
| "loss": 0.0385, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3496932515337423, | |
| "grad_norm": 0.0230648685246706, | |
| "learning_rate": 0.00013251533742331288, | |
| "loss": 0.0333, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3803680981595092, | |
| "grad_norm": 3.86190128326416, | |
| "learning_rate": 0.00013098159509202455, | |
| "loss": 0.0393, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4110429447852761, | |
| "grad_norm": 0.9012386798858643, | |
| "learning_rate": 0.00012944785276073622, | |
| "loss": 0.0571, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.441717791411043, | |
| "grad_norm": 0.07151665538549423, | |
| "learning_rate": 0.00012791411042944786, | |
| "loss": 0.0104, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4723926380368098, | |
| "grad_norm": 0.783393919467926, | |
| "learning_rate": 0.00012638036809815953, | |
| "loss": 0.101, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5030674846625767, | |
| "grad_norm": 0.11429934948682785, | |
| "learning_rate": 0.00012484662576687117, | |
| "loss": 0.0861, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5337423312883436, | |
| "grad_norm": 0.22566761076450348, | |
| "learning_rate": 0.00012331288343558281, | |
| "loss": 0.0258, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5337423312883436, | |
| "eval_accuracy": 0.9375, | |
| "eval_loss": 0.38183438777923584, | |
| "eval_runtime": 0.6415, | |
| "eval_samples_per_second": 24.943, | |
| "eval_steps_per_second": 3.118, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5644171779141103, | |
| "grad_norm": 0.0231198500841856, | |
| "learning_rate": 0.0001217791411042945, | |
| "loss": 0.0112, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5950920245398774, | |
| "grad_norm": 0.07680622488260269, | |
| "learning_rate": 0.00012024539877300614, | |
| "loss": 0.0332, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6257668711656441, | |
| "grad_norm": 4.2723798751831055, | |
| "learning_rate": 0.0001187116564417178, | |
| "loss": 0.0615, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.656441717791411, | |
| "grad_norm": 0.020270884037017822, | |
| "learning_rate": 0.00011717791411042945, | |
| "loss": 0.0025, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.687116564417178, | |
| "grad_norm": 0.017487134784460068, | |
| "learning_rate": 0.0001156441717791411, | |
| "loss": 0.0032, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7177914110429446, | |
| "grad_norm": 3.106546401977539, | |
| "learning_rate": 0.00011411042944785277, | |
| "loss": 0.0776, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7484662576687118, | |
| "grad_norm": 0.2955380380153656, | |
| "learning_rate": 0.00011257668711656441, | |
| "loss": 0.046, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.7791411042944785, | |
| "grad_norm": 0.03838299587368965, | |
| "learning_rate": 0.00011104294478527608, | |
| "loss": 0.0039, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8098159509202454, | |
| "grad_norm": 0.02066592499613762, | |
| "learning_rate": 0.00010950920245398772, | |
| "loss": 0.0411, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8404907975460123, | |
| "grad_norm": 0.029740285128355026, | |
| "learning_rate": 0.00010797546012269939, | |
| "loss": 0.0547, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8404907975460123, | |
| "eval_accuracy": 0.9375, | |
| "eval_loss": 0.3414860665798187, | |
| "eval_runtime": 0.6402, | |
| "eval_samples_per_second": 24.991, | |
| "eval_steps_per_second": 3.124, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.871165644171779, | |
| "grad_norm": 0.06896264851093292, | |
| "learning_rate": 0.00010644171779141106, | |
| "loss": 0.0412, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.9018404907975461, | |
| "grad_norm": 0.17898960411548615, | |
| "learning_rate": 0.0001049079754601227, | |
| "loss": 0.0059, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.9325153374233128, | |
| "grad_norm": 0.02462293580174446, | |
| "learning_rate": 0.00010337423312883437, | |
| "loss": 0.0334, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9631901840490797, | |
| "grad_norm": 0.15255212783813477, | |
| "learning_rate": 0.00010184049079754601, | |
| "loss": 0.0484, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.9938650306748467, | |
| "grad_norm": 1.2095363140106201, | |
| "learning_rate": 0.00010030674846625767, | |
| "loss": 0.0409, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0245398773006134, | |
| "grad_norm": 0.05067542567849159, | |
| "learning_rate": 9.877300613496932e-05, | |
| "loss": 0.0239, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.0552147239263805, | |
| "grad_norm": 0.048488594591617584, | |
| "learning_rate": 9.723926380368099e-05, | |
| "loss": 0.0092, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.085889570552147, | |
| "grad_norm": 0.06358503550291061, | |
| "learning_rate": 9.570552147239264e-05, | |
| "loss": 0.0284, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.116564417177914, | |
| "grad_norm": 0.02291274443268776, | |
| "learning_rate": 9.41717791411043e-05, | |
| "loss": 0.0312, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.147239263803681, | |
| "grad_norm": 0.01959369145333767, | |
| "learning_rate": 9.263803680981595e-05, | |
| "loss": 0.0029, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.147239263803681, | |
| "eval_accuracy": 0.9375, | |
| "eval_loss": 0.06373238563537598, | |
| "eval_runtime": 1.4073, | |
| "eval_samples_per_second": 11.369, | |
| "eval_steps_per_second": 1.421, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1779141104294477, | |
| "grad_norm": 0.015473265200853348, | |
| "learning_rate": 9.110429447852761e-05, | |
| "loss": 0.0052, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.208588957055215, | |
| "grad_norm": 6.127484321594238, | |
| "learning_rate": 8.957055214723928e-05, | |
| "loss": 0.0215, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.2392638036809815, | |
| "grad_norm": 0.014825492165982723, | |
| "learning_rate": 8.803680981595093e-05, | |
| "loss": 0.0051, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2699386503067487, | |
| "grad_norm": 0.09054987877607346, | |
| "learning_rate": 8.650306748466259e-05, | |
| "loss": 0.0089, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.3006134969325154, | |
| "grad_norm": 0.012495328672230244, | |
| "learning_rate": 8.496932515337423e-05, | |
| "loss": 0.0035, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.331288343558282, | |
| "grad_norm": 0.012604707852005959, | |
| "learning_rate": 8.343558282208588e-05, | |
| "loss": 0.0046, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.361963190184049, | |
| "grad_norm": 0.012834266759455204, | |
| "learning_rate": 8.190184049079755e-05, | |
| "loss": 0.0685, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.392638036809816, | |
| "grad_norm": 1.8484301567077637, | |
| "learning_rate": 8.036809815950921e-05, | |
| "loss": 0.0058, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.4233128834355826, | |
| "grad_norm": 0.01303753349930048, | |
| "learning_rate": 7.883435582822086e-05, | |
| "loss": 0.0201, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.4539877300613497, | |
| "grad_norm": 0.011059868149459362, | |
| "learning_rate": 7.730061349693252e-05, | |
| "loss": 0.0543, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4539877300613497, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.7362374067306519, | |
| "eval_runtime": 0.6252, | |
| "eval_samples_per_second": 25.594, | |
| "eval_steps_per_second": 3.199, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4846625766871164, | |
| "grad_norm": 0.013651982881128788, | |
| "learning_rate": 7.576687116564417e-05, | |
| "loss": 0.0023, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.5153374233128836, | |
| "grad_norm": 0.013891194015741348, | |
| "learning_rate": 7.423312883435584e-05, | |
| "loss": 0.0035, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.5460122699386503, | |
| "grad_norm": 0.022767823189496994, | |
| "learning_rate": 7.26993865030675e-05, | |
| "loss": 0.0023, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5766871165644174, | |
| "grad_norm": 0.01230280939489603, | |
| "learning_rate": 7.116564417177914e-05, | |
| "loss": 0.0163, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.607361963190184, | |
| "grad_norm": 0.01010560616850853, | |
| "learning_rate": 6.963190184049079e-05, | |
| "loss": 0.0016, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.638036809815951, | |
| "grad_norm": 0.010164570063352585, | |
| "learning_rate": 6.809815950920245e-05, | |
| "loss": 0.0287, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.668711656441718, | |
| "grad_norm": 0.010293642990291119, | |
| "learning_rate": 6.656441717791412e-05, | |
| "loss": 0.0307, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.6993865030674846, | |
| "grad_norm": 0.01078992523252964, | |
| "learning_rate": 6.503067484662577e-05, | |
| "loss": 0.0106, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.7300613496932513, | |
| "grad_norm": 0.01032053492963314, | |
| "learning_rate": 6.349693251533743e-05, | |
| "loss": 0.0014, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7607361963190185, | |
| "grad_norm": 0.009690840728580952, | |
| "learning_rate": 6.196319018404908e-05, | |
| "loss": 0.0265, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.7607361963190185, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 1.0916930437088013, | |
| "eval_runtime": 0.4595, | |
| "eval_samples_per_second": 34.82, | |
| "eval_steps_per_second": 4.352, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.791411042944785, | |
| "grad_norm": 0.014224858023226261, | |
| "learning_rate": 6.0429447852760736e-05, | |
| "loss": 0.0622, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.8220858895705523, | |
| "grad_norm": 0.013891954906284809, | |
| "learning_rate": 5.88957055214724e-05, | |
| "loss": 0.0153, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.852760736196319, | |
| "grad_norm": 0.0167331974953413, | |
| "learning_rate": 5.736196319018405e-05, | |
| "loss": 0.0026, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.883435582822086, | |
| "grad_norm": 0.03341999277472496, | |
| "learning_rate": 5.582822085889571e-05, | |
| "loss": 0.0119, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.914110429447853, | |
| "grad_norm": 0.012064835987985134, | |
| "learning_rate": 5.429447852760736e-05, | |
| "loss": 0.0149, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.9447852760736195, | |
| "grad_norm": 0.012997813522815704, | |
| "learning_rate": 5.276073619631902e-05, | |
| "loss": 0.0034, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.9754601226993866, | |
| "grad_norm": 0.014833126217126846, | |
| "learning_rate": 5.122699386503068e-05, | |
| "loss": 0.0209, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.0061349693251533, | |
| "grad_norm": 0.010445049963891506, | |
| "learning_rate": 4.9693251533742335e-05, | |
| "loss": 0.0434, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.03680981595092, | |
| "grad_norm": 0.012170241214334965, | |
| "learning_rate": 4.815950920245399e-05, | |
| "loss": 0.0058, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.067484662576687, | |
| "grad_norm": 0.010554679669439793, | |
| "learning_rate": 4.6625766871165645e-05, | |
| "loss": 0.0017, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.067484662576687, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.0030439873225986958, | |
| "eval_runtime": 0.4699, | |
| "eval_samples_per_second": 34.049, | |
| "eval_steps_per_second": 4.256, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.098159509202454, | |
| "grad_norm": 0.010963929817080498, | |
| "learning_rate": 4.5092024539877307e-05, | |
| "loss": 0.0241, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.128834355828221, | |
| "grad_norm": 0.010549047961831093, | |
| "learning_rate": 4.3558282208588955e-05, | |
| "loss": 0.0015, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.1595092024539877, | |
| "grad_norm": 0.012895047664642334, | |
| "learning_rate": 4.2024539877300617e-05, | |
| "loss": 0.0021, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.190184049079755, | |
| "grad_norm": 0.009695703163743019, | |
| "learning_rate": 4.049079754601227e-05, | |
| "loss": 0.0013, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.2208588957055215, | |
| "grad_norm": 0.01083831675350666, | |
| "learning_rate": 3.895705521472393e-05, | |
| "loss": 0.0018, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.2515337423312882, | |
| "grad_norm": 0.030923034995794296, | |
| "learning_rate": 3.742331288343559e-05, | |
| "loss": 0.0012, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.2822085889570554, | |
| "grad_norm": 0.009053406305611134, | |
| "learning_rate": 3.5889570552147236e-05, | |
| "loss": 0.0011, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.312883435582822, | |
| "grad_norm": 0.009371621534228325, | |
| "learning_rate": 3.43558282208589e-05, | |
| "loss": 0.0011, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.3435582822085887, | |
| "grad_norm": 0.009119446389377117, | |
| "learning_rate": 3.282208588957055e-05, | |
| "loss": 0.0011, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.374233128834356, | |
| "grad_norm": 0.009182000532746315, | |
| "learning_rate": 3.1288343558282215e-05, | |
| "loss": 0.0054, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.374233128834356, | |
| "eval_accuracy": 1.0, | |
| "eval_loss": 0.036368854343891144, | |
| "eval_runtime": 0.4537, | |
| "eval_samples_per_second": 35.267, | |
| "eval_steps_per_second": 4.408, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.4049079754601226, | |
| "grad_norm": 0.00895740371197462, | |
| "learning_rate": 2.9754601226993867e-05, | |
| "loss": 0.0017, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.4355828220858897, | |
| "grad_norm": 0.20501597225666046, | |
| "learning_rate": 2.822085889570552e-05, | |
| "loss": 0.0013, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.4662576687116564, | |
| "grad_norm": 0.009406461380422115, | |
| "learning_rate": 2.668711656441718e-05, | |
| "loss": 0.0011, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.4969325153374236, | |
| "grad_norm": 0.009192903526127338, | |
| "learning_rate": 2.5153374233128835e-05, | |
| "loss": 0.0012, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.5276073619631902, | |
| "grad_norm": 0.009342888370156288, | |
| "learning_rate": 2.361963190184049e-05, | |
| "loss": 0.0362, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.558282208588957, | |
| "grad_norm": 0.01002499833703041, | |
| "learning_rate": 2.208588957055215e-05, | |
| "loss": 0.0011, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.588957055214724, | |
| "grad_norm": 0.01044855359941721, | |
| "learning_rate": 2.0552147239263807e-05, | |
| "loss": 0.0011, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.6196319018404908, | |
| "grad_norm": 0.030983537435531616, | |
| "learning_rate": 1.9018404907975462e-05, | |
| "loss": 0.0011, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.6503067484662575, | |
| "grad_norm": 0.009104466997087002, | |
| "learning_rate": 1.7484662576687117e-05, | |
| "loss": 0.0012, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.6809815950920246, | |
| "grad_norm": 0.009420313872396946, | |
| "learning_rate": 1.5950920245398772e-05, | |
| "loss": 0.0234, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.6809815950920246, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.23103433847427368, | |
| "eval_runtime": 0.4704, | |
| "eval_samples_per_second": 34.012, | |
| "eval_steps_per_second": 4.252, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.7116564417177913, | |
| "grad_norm": 0.00850239023566246, | |
| "learning_rate": 1.441717791411043e-05, | |
| "loss": 0.001, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.7423312883435584, | |
| "grad_norm": 0.008601406589150429, | |
| "learning_rate": 1.2883435582822087e-05, | |
| "loss": 0.0013, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.773006134969325, | |
| "grad_norm": 0.011089220643043518, | |
| "learning_rate": 1.1349693251533742e-05, | |
| "loss": 0.0011, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.8036809815950923, | |
| "grad_norm": 0.008406179025769234, | |
| "learning_rate": 9.8159509202454e-06, | |
| "loss": 0.001, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.834355828220859, | |
| "grad_norm": 0.009516136720776558, | |
| "learning_rate": 8.282208588957055e-06, | |
| "loss": 0.0012, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.8650306748466257, | |
| "grad_norm": 0.011050822213292122, | |
| "learning_rate": 6.748466257668712e-06, | |
| "loss": 0.0012, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.895705521472393, | |
| "grad_norm": 0.008980349637567997, | |
| "learning_rate": 5.214723926380368e-06, | |
| "loss": 0.0012, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.9263803680981595, | |
| "grad_norm": 0.00903311651200056, | |
| "learning_rate": 3.680981595092025e-06, | |
| "loss": 0.0014, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.957055214723926, | |
| "grad_norm": 0.00838613323867321, | |
| "learning_rate": 2.147239263803681e-06, | |
| "loss": 0.0253, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.9877300613496933, | |
| "grad_norm": 0.008292277343571186, | |
| "learning_rate": 6.134969325153375e-07, | |
| "loss": 0.0076, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.9877300613496933, | |
| "eval_accuracy": 0.875, | |
| "eval_loss": 0.40373101830482483, | |
| "eval_runtime": 0.4908, | |
| "eval_samples_per_second": 32.599, | |
| "eval_steps_per_second": 4.075, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1304, | |
| "total_flos": 1.6167928713188475e+18, | |
| "train_loss": 0.04957773063711799, | |
| "train_runtime": 1223.9715, | |
| "train_samples_per_second": 17.046, | |
| "train_steps_per_second": 1.065 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1304, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.6167928713188475e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |