| { | |
| "best_metric": 0.7511737089201878, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-bootcamp/checkpoint-540", | |
| "epoch": 88.88888888888889, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_accuracy": 0.004694835680751174, | |
| "eval_loss": 4.284899711608887, | |
| "eval_runtime": 0.4639, | |
| "eval_samples_per_second": 459.187, | |
| "eval_steps_per_second": 15.091, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4814814814814814, | |
| "grad_norm": 4.8882575035095215, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 4.3139, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.925925925925926, | |
| "eval_accuracy": 0.03286384976525822, | |
| "eval_loss": 4.184599876403809, | |
| "eval_runtime": 0.4473, | |
| "eval_samples_per_second": 476.176, | |
| "eval_steps_per_second": 15.649, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 4.126770496368408, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 4.1651, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "eval_accuracy": 0.056338028169014086, | |
| "eval_loss": 4.058472156524658, | |
| "eval_runtime": 0.4494, | |
| "eval_samples_per_second": 473.953, | |
| "eval_steps_per_second": 15.576, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.06103286384976526, | |
| "eval_loss": 3.952671766281128, | |
| "eval_runtime": 0.4514, | |
| "eval_samples_per_second": 471.822, | |
| "eval_steps_per_second": 15.506, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 4.966624736785889, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.9272, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "eval_accuracy": 0.06103286384976526, | |
| "eval_loss": 3.8813464641571045, | |
| "eval_runtime": 0.456, | |
| "eval_samples_per_second": 467.068, | |
| "eval_steps_per_second": 15.35, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.925925925925926, | |
| "grad_norm": 5.827086448669434, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.7461, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.925925925925926, | |
| "eval_accuracy": 0.08450704225352113, | |
| "eval_loss": 3.7535641193389893, | |
| "eval_runtime": 0.4503, | |
| "eval_samples_per_second": 473.021, | |
| "eval_steps_per_second": 15.545, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.962962962962963, | |
| "eval_accuracy": 0.107981220657277, | |
| "eval_loss": 3.648564100265503, | |
| "eval_runtime": 0.4501, | |
| "eval_samples_per_second": 473.231, | |
| "eval_steps_per_second": 15.552, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.407407407407407, | |
| "grad_norm": 6.3525919914245605, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 3.5254, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.13615023474178403, | |
| "eval_loss": 3.56028151512146, | |
| "eval_runtime": 0.4596, | |
| "eval_samples_per_second": 463.489, | |
| "eval_steps_per_second": 15.232, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 8.264792442321777, | |
| "learning_rate": 5e-05, | |
| "loss": 3.3478, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "eval_accuracy": 0.13615023474178403, | |
| "eval_loss": 3.4565625190734863, | |
| "eval_runtime": 0.456, | |
| "eval_samples_per_second": 467.076, | |
| "eval_steps_per_second": 15.35, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.925925925925926, | |
| "eval_accuracy": 0.15023474178403756, | |
| "eval_loss": 3.2985825538635254, | |
| "eval_runtime": 0.4601, | |
| "eval_samples_per_second": 462.915, | |
| "eval_steps_per_second": 15.213, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 10.37037037037037, | |
| "grad_norm": 10.286266326904297, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 3.0423, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.962962962962964, | |
| "eval_accuracy": 0.15492957746478872, | |
| "eval_loss": 3.2165536880493164, | |
| "eval_runtime": 0.4547, | |
| "eval_samples_per_second": 468.397, | |
| "eval_steps_per_second": 15.393, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 11.851851851851851, | |
| "grad_norm": 9.160584449768066, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 2.7931, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.215962441314554, | |
| "eval_loss": 3.0202882289886475, | |
| "eval_runtime": 0.4563, | |
| "eval_samples_per_second": 466.806, | |
| "eval_steps_per_second": 15.341, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 12.88888888888889, | |
| "eval_accuracy": 0.29107981220657275, | |
| "eval_loss": 2.8990509510040283, | |
| "eval_runtime": 0.4577, | |
| "eval_samples_per_second": 465.39, | |
| "eval_steps_per_second": 15.294, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 9.414092063903809, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.541, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 13.925925925925926, | |
| "eval_accuracy": 0.29107981220657275, | |
| "eval_loss": 2.794092893600464, | |
| "eval_runtime": 0.4564, | |
| "eval_samples_per_second": 466.714, | |
| "eval_steps_per_second": 15.338, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 14.814814814814815, | |
| "grad_norm": 9.282512664794922, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 2.3487, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 14.962962962962964, | |
| "eval_accuracy": 0.29107981220657275, | |
| "eval_loss": 2.733652353286743, | |
| "eval_runtime": 0.4626, | |
| "eval_samples_per_second": 460.475, | |
| "eval_steps_per_second": 15.133, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.36619718309859156, | |
| "eval_loss": 2.540092945098877, | |
| "eval_runtime": 0.4583, | |
| "eval_samples_per_second": 464.804, | |
| "eval_steps_per_second": 15.275, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 16.296296296296298, | |
| "grad_norm": 11.685836791992188, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 2.1043, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 16.88888888888889, | |
| "eval_accuracy": 0.38028169014084506, | |
| "eval_loss": 2.5088114738464355, | |
| "eval_runtime": 0.4588, | |
| "eval_samples_per_second": 464.257, | |
| "eval_steps_per_second": 15.257, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 17.77777777777778, | |
| "grad_norm": 11.694657325744629, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.8892, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 17.925925925925927, | |
| "eval_accuracy": 0.4131455399061033, | |
| "eval_loss": 2.3595798015594482, | |
| "eval_runtime": 0.4625, | |
| "eval_samples_per_second": 460.528, | |
| "eval_steps_per_second": 15.135, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 18.962962962962962, | |
| "eval_accuracy": 0.41784037558685444, | |
| "eval_loss": 2.317976713180542, | |
| "eval_runtime": 0.4557, | |
| "eval_samples_per_second": 467.42, | |
| "eval_steps_per_second": 15.361, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 19.25925925925926, | |
| "grad_norm": 9.399383544921875, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 1.7167, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4272300469483568, | |
| "eval_loss": 2.1819818019866943, | |
| "eval_runtime": 0.4566, | |
| "eval_samples_per_second": 466.514, | |
| "eval_steps_per_second": 15.331, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 20.74074074074074, | |
| "grad_norm": 8.570808410644531, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 1.5748, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 20.88888888888889, | |
| "eval_accuracy": 0.4413145539906103, | |
| "eval_loss": 2.0546934604644775, | |
| "eval_runtime": 0.4517, | |
| "eval_samples_per_second": 471.59, | |
| "eval_steps_per_second": 15.498, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 21.925925925925927, | |
| "eval_accuracy": 0.49295774647887325, | |
| "eval_loss": 1.9472370147705078, | |
| "eval_runtime": 0.4594, | |
| "eval_samples_per_second": 463.678, | |
| "eval_steps_per_second": 15.238, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 11.712135314941406, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.4052, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 22.962962962962962, | |
| "eval_accuracy": 0.48826291079812206, | |
| "eval_loss": 1.9053350687026978, | |
| "eval_runtime": 0.4637, | |
| "eval_samples_per_second": 459.37, | |
| "eval_steps_per_second": 15.097, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 23.703703703703702, | |
| "grad_norm": 11.016504287719727, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 1.2535, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5117370892018779, | |
| "eval_loss": 1.8178737163543701, | |
| "eval_runtime": 0.4639, | |
| "eval_samples_per_second": 459.108, | |
| "eval_steps_per_second": 15.088, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 24.88888888888889, | |
| "eval_accuracy": 0.5305164319248826, | |
| "eval_loss": 1.7599667310714722, | |
| "eval_runtime": 0.4596, | |
| "eval_samples_per_second": 463.449, | |
| "eval_steps_per_second": 15.231, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 25.185185185185187, | |
| "grad_norm": 10.845667839050293, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 1.1687, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 25.925925925925927, | |
| "eval_accuracy": 0.5492957746478874, | |
| "eval_loss": 1.6921813488006592, | |
| "eval_runtime": 0.4593, | |
| "eval_samples_per_second": 463.784, | |
| "eval_steps_per_second": 15.242, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 10.369373321533203, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.0719, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 26.962962962962962, | |
| "eval_accuracy": 0.5586854460093896, | |
| "eval_loss": 1.6075888872146606, | |
| "eval_runtime": 0.4618, | |
| "eval_samples_per_second": 461.281, | |
| "eval_steps_per_second": 15.159, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5586854460093896, | |
| "eval_loss": 1.5316015481948853, | |
| "eval_runtime": 0.4579, | |
| "eval_samples_per_second": 465.205, | |
| "eval_steps_per_second": 15.288, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 28.14814814814815, | |
| "grad_norm": 12.61413860321045, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 1.0577, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 28.88888888888889, | |
| "eval_accuracy": 0.5774647887323944, | |
| "eval_loss": 1.5365443229675293, | |
| "eval_runtime": 0.4629, | |
| "eval_samples_per_second": 460.173, | |
| "eval_steps_per_second": 15.123, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 29.62962962962963, | |
| "grad_norm": 10.047813415527344, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.9558, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 29.925925925925927, | |
| "eval_accuracy": 0.6291079812206573, | |
| "eval_loss": 1.448791742324829, | |
| "eval_runtime": 0.4616, | |
| "eval_samples_per_second": 461.458, | |
| "eval_steps_per_second": 15.165, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 30.962962962962962, | |
| "eval_accuracy": 0.6150234741784038, | |
| "eval_loss": 1.4185121059417725, | |
| "eval_runtime": 0.4583, | |
| "eval_samples_per_second": 464.754, | |
| "eval_steps_per_second": 15.274, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 31.11111111111111, | |
| "grad_norm": 8.121445655822754, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.8771, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6056338028169014, | |
| "eval_loss": 1.3905900716781616, | |
| "eval_runtime": 0.4637, | |
| "eval_samples_per_second": 459.353, | |
| "eval_steps_per_second": 15.096, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 32.592592592592595, | |
| "grad_norm": 9.448586463928223, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.8146, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 32.888888888888886, | |
| "eval_accuracy": 0.6150234741784038, | |
| "eval_loss": 1.3828094005584717, | |
| "eval_runtime": 0.4593, | |
| "eval_samples_per_second": 463.772, | |
| "eval_steps_per_second": 15.241, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 33.925925925925924, | |
| "eval_accuracy": 0.5821596244131455, | |
| "eval_loss": 1.3927448987960815, | |
| "eval_runtime": 0.4708, | |
| "eval_samples_per_second": 452.446, | |
| "eval_steps_per_second": 14.869, | |
| "step": 229 | |
| }, | |
| { | |
| "epoch": 34.074074074074076, | |
| "grad_norm": 11.940274238586426, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 0.8228, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 34.96296296296296, | |
| "eval_accuracy": 0.6384976525821596, | |
| "eval_loss": 1.3035573959350586, | |
| "eval_runtime": 0.4602, | |
| "eval_samples_per_second": 462.817, | |
| "eval_steps_per_second": 15.21, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 35.55555555555556, | |
| "grad_norm": 10.717940330505371, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6878, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 1.2239941358566284, | |
| "eval_runtime": 0.4658, | |
| "eval_samples_per_second": 457.263, | |
| "eval_steps_per_second": 15.027, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 36.888888888888886, | |
| "eval_accuracy": 0.6713615023474179, | |
| "eval_loss": 1.2388455867767334, | |
| "eval_runtime": 0.4563, | |
| "eval_samples_per_second": 466.842, | |
| "eval_steps_per_second": 15.342, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 37.03703703703704, | |
| "grad_norm": 12.011445045471191, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.6471, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 37.925925925925924, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 1.1344724893569946, | |
| "eval_runtime": 0.4612, | |
| "eval_samples_per_second": 461.873, | |
| "eval_steps_per_second": 15.179, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 38.51851851851852, | |
| "grad_norm": 9.849960327148438, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.6102, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 38.96296296296296, | |
| "eval_accuracy": 0.6572769953051644, | |
| "eval_loss": 1.18145751953125, | |
| "eval_runtime": 0.4589, | |
| "eval_samples_per_second": 464.119, | |
| "eval_steps_per_second": 15.253, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 10.17436408996582, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.6599, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.6525821596244131, | |
| "eval_loss": 1.1720404624938965, | |
| "eval_runtime": 0.4628, | |
| "eval_samples_per_second": 460.238, | |
| "eval_steps_per_second": 15.125, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 40.888888888888886, | |
| "eval_accuracy": 0.6525821596244131, | |
| "eval_loss": 1.1335811614990234, | |
| "eval_runtime": 0.4583, | |
| "eval_samples_per_second": 464.802, | |
| "eval_steps_per_second": 15.275, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 41.48148148148148, | |
| "grad_norm": 8.4024019241333, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.5742, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 41.925925925925924, | |
| "eval_accuracy": 0.6713615023474179, | |
| "eval_loss": 1.0863432884216309, | |
| "eval_runtime": 0.4663, | |
| "eval_samples_per_second": 456.75, | |
| "eval_steps_per_second": 15.011, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 42.96296296296296, | |
| "grad_norm": 11.1184720993042, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.5478, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 42.96296296296296, | |
| "eval_accuracy": 0.6713615023474179, | |
| "eval_loss": 1.0909894704818726, | |
| "eval_runtime": 0.4543, | |
| "eval_samples_per_second": 468.842, | |
| "eval_steps_per_second": 15.408, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.6619718309859155, | |
| "eval_loss": 1.074562668800354, | |
| "eval_runtime": 0.4575, | |
| "eval_samples_per_second": 465.544, | |
| "eval_steps_per_second": 15.3, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 44.44444444444444, | |
| "grad_norm": 8.84053897857666, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.557, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 44.888888888888886, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 1.0723693370819092, | |
| "eval_runtime": 0.4539, | |
| "eval_samples_per_second": 469.244, | |
| "eval_steps_per_second": 15.421, | |
| "step": 303 | |
| }, | |
| { | |
| "epoch": 45.925925925925924, | |
| "grad_norm": 10.450238227844238, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.5753, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 45.925925925925924, | |
| "eval_accuracy": 0.7136150234741784, | |
| "eval_loss": 1.0107887983322144, | |
| "eval_runtime": 0.4643, | |
| "eval_samples_per_second": 458.736, | |
| "eval_steps_per_second": 15.076, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 46.96296296296296, | |
| "eval_accuracy": 0.6431924882629108, | |
| "eval_loss": 1.1295782327651978, | |
| "eval_runtime": 0.4606, | |
| "eval_samples_per_second": 462.398, | |
| "eval_steps_per_second": 15.196, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 47.407407407407405, | |
| "grad_norm": 9.998331069946289, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.5325, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 1.0361039638519287, | |
| "eval_runtime": 0.4659, | |
| "eval_samples_per_second": 457.192, | |
| "eval_steps_per_second": 15.025, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 48.888888888888886, | |
| "grad_norm": 7.716311454772949, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4349, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 48.888888888888886, | |
| "eval_accuracy": 0.6995305164319249, | |
| "eval_loss": 1.0237399339675903, | |
| "eval_runtime": 0.4605, | |
| "eval_samples_per_second": 462.536, | |
| "eval_steps_per_second": 15.201, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 49.925925925925924, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.9789619445800781, | |
| "eval_runtime": 0.4655, | |
| "eval_samples_per_second": 457.587, | |
| "eval_steps_per_second": 15.038, | |
| "step": 337 | |
| }, | |
| { | |
| "epoch": 50.370370370370374, | |
| "grad_norm": 11.521199226379395, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.447, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 50.96296296296296, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 1.0409153699874878, | |
| "eval_runtime": 0.4603, | |
| "eval_samples_per_second": 462.789, | |
| "eval_steps_per_second": 15.209, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 51.851851851851855, | |
| "grad_norm": 7.985869884490967, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.4502, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.6713615023474179, | |
| "eval_loss": 1.0466963052749634, | |
| "eval_runtime": 0.4518, | |
| "eval_samples_per_second": 471.396, | |
| "eval_steps_per_second": 15.492, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 52.888888888888886, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.9772961139678955, | |
| "eval_runtime": 0.4537, | |
| "eval_samples_per_second": 469.512, | |
| "eval_steps_per_second": 15.43, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 53.333333333333336, | |
| "grad_norm": 6.707178592681885, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4345, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 53.925925925925924, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 0.9931105971336365, | |
| "eval_runtime": 0.4568, | |
| "eval_samples_per_second": 466.293, | |
| "eval_steps_per_second": 15.324, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 54.81481481481482, | |
| "grad_norm": 7.90395975112915, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.4557, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 54.96296296296296, | |
| "eval_accuracy": 0.7136150234741784, | |
| "eval_loss": 0.9685100317001343, | |
| "eval_runtime": 0.4567, | |
| "eval_samples_per_second": 466.339, | |
| "eval_steps_per_second": 15.326, | |
| "step": 371 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.7370892018779343, | |
| "eval_loss": 0.9546511769294739, | |
| "eval_runtime": 0.4637, | |
| "eval_samples_per_second": 459.34, | |
| "eval_steps_per_second": 15.096, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 56.2962962962963, | |
| "grad_norm": 8.04978084564209, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.4109, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 56.888888888888886, | |
| "eval_accuracy": 0.6948356807511737, | |
| "eval_loss": 1.001466989517212, | |
| "eval_runtime": 0.4572, | |
| "eval_samples_per_second": 465.926, | |
| "eval_steps_per_second": 15.312, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 57.77777777777778, | |
| "grad_norm": 9.716485023498535, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.4406, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 57.925925925925924, | |
| "eval_accuracy": 0.7230046948356808, | |
| "eval_loss": 0.9410375952720642, | |
| "eval_runtime": 0.4627, | |
| "eval_samples_per_second": 460.386, | |
| "eval_steps_per_second": 15.13, | |
| "step": 391 | |
| }, | |
| { | |
| "epoch": 58.96296296296296, | |
| "eval_accuracy": 0.6807511737089202, | |
| "eval_loss": 0.9764938950538635, | |
| "eval_runtime": 0.4625, | |
| "eval_samples_per_second": 460.496, | |
| "eval_steps_per_second": 15.134, | |
| "step": 398 | |
| }, | |
| { | |
| "epoch": 59.25925925925926, | |
| "grad_norm": 7.345524311065674, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.4039, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.7089201877934272, | |
| "eval_loss": 0.9505257606506348, | |
| "eval_runtime": 0.4624, | |
| "eval_samples_per_second": 460.681, | |
| "eval_steps_per_second": 15.14, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 60.74074074074074, | |
| "grad_norm": 9.50730037689209, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.396, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 60.888888888888886, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.9539108872413635, | |
| "eval_runtime": 0.455, | |
| "eval_samples_per_second": 468.095, | |
| "eval_steps_per_second": 15.383, | |
| "step": 411 | |
| }, | |
| { | |
| "epoch": 61.925925925925924, | |
| "eval_accuracy": 0.676056338028169, | |
| "eval_loss": 1.039088487625122, | |
| "eval_runtime": 0.461, | |
| "eval_samples_per_second": 462.085, | |
| "eval_steps_per_second": 15.186, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 62.22222222222222, | |
| "grad_norm": 7.710968971252441, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3958, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 62.96296296296296, | |
| "eval_accuracy": 0.7136150234741784, | |
| "eval_loss": 0.9576095342636108, | |
| "eval_runtime": 0.4566, | |
| "eval_samples_per_second": 466.515, | |
| "eval_steps_per_second": 15.331, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 63.7037037037037, | |
| "grad_norm": 8.0850830078125, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.3763, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7230046948356808, | |
| "eval_loss": 0.937968909740448, | |
| "eval_runtime": 0.4607, | |
| "eval_samples_per_second": 462.317, | |
| "eval_steps_per_second": 15.194, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 64.88888888888889, | |
| "eval_accuracy": 0.7276995305164319, | |
| "eval_loss": 0.9363034963607788, | |
| "eval_runtime": 0.4601, | |
| "eval_samples_per_second": 462.97, | |
| "eval_steps_per_second": 15.215, | |
| "step": 438 | |
| }, | |
| { | |
| "epoch": 65.18518518518519, | |
| "grad_norm": 8.919660568237305, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.3985, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 65.92592592592592, | |
| "eval_accuracy": 0.7089201877934272, | |
| "eval_loss": 0.9399967193603516, | |
| "eval_runtime": 0.4592, | |
| "eval_samples_per_second": 463.874, | |
| "eval_steps_per_second": 15.245, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "grad_norm": 7.8991618156433105, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.3701, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 66.96296296296296, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.9769109487533569, | |
| "eval_runtime": 0.4528, | |
| "eval_samples_per_second": 470.438, | |
| "eval_steps_per_second": 15.46, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7276995305164319, | |
| "eval_loss": 0.9604400992393494, | |
| "eval_runtime": 0.4581, | |
| "eval_samples_per_second": 464.916, | |
| "eval_steps_per_second": 15.279, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 68.14814814814815, | |
| "grad_norm": 8.74885082244873, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.3729, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 68.88888888888889, | |
| "eval_accuracy": 0.7089201877934272, | |
| "eval_loss": 0.9883025884628296, | |
| "eval_runtime": 0.4585, | |
| "eval_samples_per_second": 464.566, | |
| "eval_steps_per_second": 15.267, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 69.62962962962963, | |
| "grad_norm": 7.356982231140137, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.3958, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 69.92592592592592, | |
| "eval_accuracy": 0.7276995305164319, | |
| "eval_loss": 0.9516283273696899, | |
| "eval_runtime": 0.4626, | |
| "eval_samples_per_second": 460.471, | |
| "eval_steps_per_second": 15.133, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 70.96296296296296, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.9252376556396484, | |
| "eval_runtime": 0.4582, | |
| "eval_samples_per_second": 464.872, | |
| "eval_steps_per_second": 15.277, | |
| "step": 479 | |
| }, | |
| { | |
| "epoch": 71.11111111111111, | |
| "grad_norm": 7.72941255569458, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.359, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7136150234741784, | |
| "eval_loss": 0.9196011424064636, | |
| "eval_runtime": 0.4634, | |
| "eval_samples_per_second": 459.681, | |
| "eval_steps_per_second": 15.107, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 72.5925925925926, | |
| "grad_norm": 8.144768714904785, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.362, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 72.88888888888889, | |
| "eval_accuracy": 0.7230046948356808, | |
| "eval_loss": 0.9104124307632446, | |
| "eval_runtime": 0.4551, | |
| "eval_samples_per_second": 468.073, | |
| "eval_steps_per_second": 15.383, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 73.92592592592592, | |
| "eval_accuracy": 0.7136150234741784, | |
| "eval_loss": 0.9254885315895081, | |
| "eval_runtime": 0.458, | |
| "eval_samples_per_second": 465.078, | |
| "eval_steps_per_second": 15.284, | |
| "step": 499 | |
| }, | |
| { | |
| "epoch": 74.07407407407408, | |
| "grad_norm": 8.89719009399414, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.353, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 74.96296296296296, | |
| "eval_accuracy": 0.7089201877934272, | |
| "eval_loss": 0.9358569979667664, | |
| "eval_runtime": 0.4614, | |
| "eval_samples_per_second": 461.641, | |
| "eval_steps_per_second": 15.171, | |
| "step": 506 | |
| }, | |
| { | |
| "epoch": 75.55555555555556, | |
| "grad_norm": 10.696667671203613, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.345, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.7230046948356808, | |
| "eval_loss": 0.9274182319641113, | |
| "eval_runtime": 0.4627, | |
| "eval_samples_per_second": 460.385, | |
| "eval_steps_per_second": 15.13, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 76.88888888888889, | |
| "eval_accuracy": 0.7370892018779343, | |
| "eval_loss": 0.9205775856971741, | |
| "eval_runtime": 0.4607, | |
| "eval_samples_per_second": 462.326, | |
| "eval_steps_per_second": 15.194, | |
| "step": 519 | |
| }, | |
| { | |
| "epoch": 77.03703703703704, | |
| "grad_norm": 7.255312919616699, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.3414, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 77.92592592592592, | |
| "eval_accuracy": 0.7276995305164319, | |
| "eval_loss": 0.922869086265564, | |
| "eval_runtime": 0.4586, | |
| "eval_samples_per_second": 464.472, | |
| "eval_steps_per_second": 15.264, | |
| "step": 526 | |
| }, | |
| { | |
| "epoch": 78.51851851851852, | |
| "grad_norm": 11.26202392578125, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.3298, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 78.96296296296296, | |
| "eval_accuracy": 0.7417840375586855, | |
| "eval_loss": 0.9102315902709961, | |
| "eval_runtime": 0.458, | |
| "eval_samples_per_second": 465.046, | |
| "eval_steps_per_second": 15.283, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 9.204602241516113, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.3394, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7511737089201878, | |
| "eval_loss": 0.8955033421516418, | |
| "eval_runtime": 0.457, | |
| "eval_samples_per_second": 466.115, | |
| "eval_steps_per_second": 15.318, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 80.88888888888889, | |
| "eval_accuracy": 0.7370892018779343, | |
| "eval_loss": 0.8956274390220642, | |
| "eval_runtime": 0.4576, | |
| "eval_samples_per_second": 465.457, | |
| "eval_steps_per_second": 15.297, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 81.48148148148148, | |
| "grad_norm": 5.736772537231445, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.3384, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 81.92592592592592, | |
| "eval_accuracy": 0.7276995305164319, | |
| "eval_loss": 0.8926992416381836, | |
| "eval_runtime": 0.4598, | |
| "eval_samples_per_second": 463.206, | |
| "eval_steps_per_second": 15.223, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 82.96296296296296, | |
| "grad_norm": 5.7758870124816895, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.3164, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 82.96296296296296, | |
| "eval_accuracy": 0.7417840375586855, | |
| "eval_loss": 0.8884522318840027, | |
| "eval_runtime": 0.4568, | |
| "eval_samples_per_second": 466.241, | |
| "eval_steps_per_second": 15.322, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7370892018779343, | |
| "eval_loss": 0.8940874338150024, | |
| "eval_runtime": 0.4584, | |
| "eval_samples_per_second": 464.708, | |
| "eval_steps_per_second": 15.272, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 84.44444444444444, | |
| "grad_norm": 8.919522285461426, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.3055, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 84.88888888888889, | |
| "eval_accuracy": 0.7417840375586855, | |
| "eval_loss": 0.8963488340377808, | |
| "eval_runtime": 0.4576, | |
| "eval_samples_per_second": 465.517, | |
| "eval_steps_per_second": 15.299, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 85.92592592592592, | |
| "grad_norm": 6.437784194946289, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.3355, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 85.92592592592592, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.8992050290107727, | |
| "eval_runtime": 0.452, | |
| "eval_samples_per_second": 471.26, | |
| "eval_steps_per_second": 15.487, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 86.96296296296296, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.8987627625465393, | |
| "eval_runtime": 0.462, | |
| "eval_samples_per_second": 461.085, | |
| "eval_steps_per_second": 15.153, | |
| "step": 587 | |
| }, | |
| { | |
| "epoch": 87.4074074074074, | |
| "grad_norm": 7.563794136047363, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.3101, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.8968843817710876, | |
| "eval_runtime": 0.4577, | |
| "eval_samples_per_second": 465.337, | |
| "eval_steps_per_second": 15.293, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "grad_norm": 6.048923015594482, | |
| "learning_rate": 0.0, | |
| "loss": 0.3218, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.896327793598175, | |
| "eval_runtime": 0.4773, | |
| "eval_samples_per_second": 446.302, | |
| "eval_steps_per_second": 14.667, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 88.88888888888889, | |
| "step": 600, | |
| "total_flos": 1.8862166953495757e+18, | |
| "train_loss": 1.1065828450520834, | |
| "train_runtime": 510.0334, | |
| "train_samples_per_second": 167.048, | |
| "train_steps_per_second": 1.176 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.8862166953495757e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |