| { | |
| "best_metric": 0.7981651376146789, | |
| "best_model_checkpoint": "./results\\checkpoint-3000", | |
| "epoch": 0.9501187648456056, | |
| "eval_steps": 50, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.3752969121140145e-07, | |
| "loss": 1.1402, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.750593824228029e-07, | |
| "loss": 1.0834, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 7.125890736342043e-07, | |
| "loss": 0.9403, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.501187648456058e-07, | |
| "loss": 0.8238, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.187648456057007e-06, | |
| "loss": 0.7737, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7308847904205322, | |
| "eval_runtime": 10.532, | |
| "eval_samples_per_second": 82.796, | |
| "eval_steps_per_second": 5.222, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.4251781472684086e-06, | |
| "loss": 0.7321, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.6627078384798101e-06, | |
| "loss": 0.7221, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9002375296912116e-06, | |
| "loss": 0.7081, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.137767220902613e-06, | |
| "loss": 0.7255, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.375296912114014e-06, | |
| "loss": 0.7094, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7017449736595154, | |
| "eval_runtime": 10.4451, | |
| "eval_samples_per_second": 83.484, | |
| "eval_steps_per_second": 5.266, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.612826603325416e-06, | |
| "loss": 0.6745, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.850356294536817e-06, | |
| "loss": 0.674, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.0878859857482185e-06, | |
| "loss": 0.7235, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.3254156769596202e-06, | |
| "loss": 0.7192, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.5629453681710215e-06, | |
| "loss": 0.7033, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.7025471329689026, | |
| "eval_runtime": 10.4492, | |
| "eval_samples_per_second": 83.451, | |
| "eval_steps_per_second": 5.264, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.8004750593824232e-06, | |
| "loss": 0.7037, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.038004750593825e-06, | |
| "loss": 0.6754, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.275534441805226e-06, | |
| "loss": 0.7119, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.513064133016627e-06, | |
| "loss": 0.7077, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.750593824228028e-06, | |
| "loss": 0.698, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.4873853211009174, | |
| "eval_loss": 0.7025485038757324, | |
| "eval_runtime": 10.6349, | |
| "eval_samples_per_second": 81.994, | |
| "eval_steps_per_second": 5.172, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.98812351543943e-06, | |
| "loss": 0.6954, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5.225653206650832e-06, | |
| "loss": 0.7078, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5.4631828978622335e-06, | |
| "loss": 0.6521, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.700712589073634e-06, | |
| "loss": 0.6937, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.938242280285035e-06, | |
| "loss": 0.6965, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.5229357798165137, | |
| "eval_loss": 0.6914423704147339, | |
| "eval_runtime": 10.7015, | |
| "eval_samples_per_second": 81.484, | |
| "eval_steps_per_second": 5.139, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 6.175771971496437e-06, | |
| "loss": 0.679, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 6.4133016627078396e-06, | |
| "loss": 0.68, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.6508313539192404e-06, | |
| "loss": 0.679, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.888361045130641e-06, | |
| "loss": 0.7055, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.125890736342043e-06, | |
| "loss": 0.6866, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.5091743119266054, | |
| "eval_loss": 0.7029032707214355, | |
| "eval_runtime": 10.4489, | |
| "eval_samples_per_second": 83.453, | |
| "eval_steps_per_second": 5.264, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.363420427553444e-06, | |
| "loss": 0.6868, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.6009501187648464e-06, | |
| "loss": 0.6827, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.838479809976247e-06, | |
| "loss": 0.6765, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 8.07600950118765e-06, | |
| "loss": 0.6836, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 8.31353919239905e-06, | |
| "loss": 0.6894, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.5068807339449541, | |
| "eval_loss": 0.6934388875961304, | |
| "eval_runtime": 9.7535, | |
| "eval_samples_per_second": 89.404, | |
| "eval_steps_per_second": 5.639, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.551068883610452e-06, | |
| "loss": 0.6939, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.788598574821852e-06, | |
| "loss": 0.6903, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.026128266033253e-06, | |
| "loss": 0.6508, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.263657957244656e-06, | |
| "loss": 0.7377, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.501187648456057e-06, | |
| "loss": 0.7484, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.4908256880733945, | |
| "eval_loss": 0.7315114140510559, | |
| "eval_runtime": 11.129, | |
| "eval_samples_per_second": 78.354, | |
| "eval_steps_per_second": 4.942, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.73871733966746e-06, | |
| "loss": 0.6887, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.97624703087886e-06, | |
| "loss": 0.6562, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.0213776722090261e-05, | |
| "loss": 0.6696, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.0451306413301664e-05, | |
| "loss": 0.7045, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.0688836104513065e-05, | |
| "loss": 0.6498, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.5997706422018348, | |
| "eval_loss": 0.6686387658119202, | |
| "eval_runtime": 18.4439, | |
| "eval_samples_per_second": 47.278, | |
| "eval_steps_per_second": 2.982, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.0926365795724467e-05, | |
| "loss": 0.67, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.1163895486935868e-05, | |
| "loss": 0.6821, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.1401425178147269e-05, | |
| "loss": 0.6477, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.163895486935867e-05, | |
| "loss": 0.6935, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.187648456057007e-05, | |
| "loss": 0.6687, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.5745412844036697, | |
| "eval_loss": 0.6807542443275452, | |
| "eval_runtime": 24.027, | |
| "eval_samples_per_second": 36.293, | |
| "eval_steps_per_second": 2.289, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.2114014251781473e-05, | |
| "loss": 0.6542, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.2351543942992874e-05, | |
| "loss": 0.7159, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.2589073634204277e-05, | |
| "loss": 0.7332, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.2826603325415679e-05, | |
| "loss": 0.6521, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.3064133016627078e-05, | |
| "loss": 0.673, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_accuracy": 0.6594036697247706, | |
| "eval_loss": 0.6414512395858765, | |
| "eval_runtime": 24.4855, | |
| "eval_samples_per_second": 35.613, | |
| "eval_steps_per_second": 2.246, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.3301662707838481e-05, | |
| "loss": 0.6498, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.3539192399049882e-05, | |
| "loss": 0.6634, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.3776722090261283e-05, | |
| "loss": 0.6557, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.4014251781472683e-05, | |
| "loss": 0.6924, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.4251781472684086e-05, | |
| "loss": 0.629, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.6192660550458715, | |
| "eval_loss": 0.6346095204353333, | |
| "eval_runtime": 21.937, | |
| "eval_samples_per_second": 39.75, | |
| "eval_steps_per_second": 2.507, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.4489311163895489e-05, | |
| "loss": 0.6623, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.4726840855106888e-05, | |
| "loss": 0.6252, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.496437054631829e-05, | |
| "loss": 0.6662, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.5201900237529693e-05, | |
| "loss": 0.6038, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.5439429928741092e-05, | |
| "loss": 0.611, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.6915137614678899, | |
| "eval_loss": 0.5941868424415588, | |
| "eval_runtime": 23.4369, | |
| "eval_samples_per_second": 37.206, | |
| "eval_steps_per_second": 2.347, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.5676959619952495e-05, | |
| "loss": 0.5919, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.5914489311163897e-05, | |
| "loss": 0.6268, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.61520190023753e-05, | |
| "loss": 0.5221, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.63895486935867e-05, | |
| "loss": 0.7314, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.66270783847981e-05, | |
| "loss": 0.7231, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.6410550458715596, | |
| "eval_loss": 0.6234958171844482, | |
| "eval_runtime": 17.243, | |
| "eval_samples_per_second": 50.571, | |
| "eval_steps_per_second": 3.19, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.6864608076009504e-05, | |
| "loss": 0.6203, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.7102137767220903e-05, | |
| "loss": 0.6168, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.7339667458432306e-05, | |
| "loss": 0.6334, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.7577197149643705e-05, | |
| "loss": 0.6718, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.7814726840855108e-05, | |
| "loss": 0.6445, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.7075688073394495, | |
| "eval_loss": 0.5777224898338318, | |
| "eval_runtime": 24.0528, | |
| "eval_samples_per_second": 36.254, | |
| "eval_steps_per_second": 2.287, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.8052256532066507e-05, | |
| "loss": 0.6013, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.828978622327791e-05, | |
| "loss": 0.5739, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.8527315914489312e-05, | |
| "loss": 0.6245, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.876484560570071e-05, | |
| "loss": 0.5358, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.9002375296912114e-05, | |
| "loss": 0.5986, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.6685779816513762, | |
| "eval_loss": 0.6269965767860413, | |
| "eval_runtime": 24.3364, | |
| "eval_samples_per_second": 35.831, | |
| "eval_steps_per_second": 2.26, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.9239904988123516e-05, | |
| "loss": 0.6833, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.947743467933492e-05, | |
| "loss": 0.6864, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9714964370546318e-05, | |
| "loss": 0.5795, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.995249406175772e-05, | |
| "loss": 0.6274, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.0190023752969123e-05, | |
| "loss": 0.663, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.6972477064220184, | |
| "eval_loss": 0.5657657384872437, | |
| "eval_runtime": 26.7355, | |
| "eval_samples_per_second": 32.616, | |
| "eval_steps_per_second": 2.057, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.0427553444180522e-05, | |
| "loss": 0.5579, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.0665083135391925e-05, | |
| "loss": 0.5483, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.0902612826603327e-05, | |
| "loss": 0.5827, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.114014251781473e-05, | |
| "loss": 0.5757, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.137767220902613e-05, | |
| "loss": 0.6553, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_accuracy": 0.6112385321100917, | |
| "eval_loss": 0.639819860458374, | |
| "eval_runtime": 23.3332, | |
| "eval_samples_per_second": 37.372, | |
| "eval_steps_per_second": 2.357, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.161520190023753e-05, | |
| "loss": 0.5793, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.1852731591448934e-05, | |
| "loss": 0.5839, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.2090261282660333e-05, | |
| "loss": 0.5909, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.2327790973871736e-05, | |
| "loss": 0.5239, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.2565320665083135e-05, | |
| "loss": 0.5402, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.6915137614678899, | |
| "eval_loss": 0.5853348970413208, | |
| "eval_runtime": 28.0695, | |
| "eval_samples_per_second": 31.066, | |
| "eval_steps_per_second": 1.959, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.2802850356294538e-05, | |
| "loss": 0.5396, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.3040380047505937e-05, | |
| "loss": 0.5771, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.327790973871734e-05, | |
| "loss": 0.5087, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.3515439429928742e-05, | |
| "loss": 0.5752, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.375296912114014e-05, | |
| "loss": 0.7053, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.5779816513761468, | |
| "eval_loss": 0.7715405225753784, | |
| "eval_runtime": 22.4698, | |
| "eval_samples_per_second": 38.808, | |
| "eval_steps_per_second": 2.448, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.3990498812351544e-05, | |
| "loss": 0.5453, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.4228028503562946e-05, | |
| "loss": 0.5469, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.446555819477435e-05, | |
| "loss": 0.5702, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.4703087885985748e-05, | |
| "loss": 0.519, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.494061757719715e-05, | |
| "loss": 0.613, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.6112385321100917, | |
| "eval_loss": 0.713302493095398, | |
| "eval_runtime": 22.1389, | |
| "eval_samples_per_second": 39.388, | |
| "eval_steps_per_second": 2.484, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.5178147268408553e-05, | |
| "loss": 0.5345, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.5415676959619956e-05, | |
| "loss": 0.5408, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.5653206650831358e-05, | |
| "loss": 0.5472, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.5890736342042754e-05, | |
| "loss": 0.6234, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.6128266033254157e-05, | |
| "loss": 0.4965, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_accuracy": 0.7385321100917431, | |
| "eval_loss": 0.5148417353630066, | |
| "eval_runtime": 11.4511, | |
| "eval_samples_per_second": 76.15, | |
| "eval_steps_per_second": 4.803, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.636579572446556e-05, | |
| "loss": 0.5747, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.6603325415676962e-05, | |
| "loss": 0.8352, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.6840855106888364e-05, | |
| "loss": 0.6406, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.7078384798099763e-05, | |
| "loss": 0.5701, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.7315914489311166e-05, | |
| "loss": 0.5114, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.6467889908256881, | |
| "eval_loss": 0.6622269749641418, | |
| "eval_runtime": 11.4464, | |
| "eval_samples_per_second": 76.181, | |
| "eval_steps_per_second": 4.805, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.7553444180522565e-05, | |
| "loss": 0.6769, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.7790973871733968e-05, | |
| "loss": 0.5533, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.8028503562945367e-05, | |
| "loss": 0.4968, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.826603325415677e-05, | |
| "loss": 0.4944, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.8503562945368172e-05, | |
| "loss": 0.4682, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_accuracy": 0.7213302752293578, | |
| "eval_loss": 0.5471405386924744, | |
| "eval_runtime": 11.4757, | |
| "eval_samples_per_second": 75.987, | |
| "eval_steps_per_second": 4.793, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.8741092636579575e-05, | |
| "loss": 0.5635, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.8978622327790977e-05, | |
| "loss": 0.4861, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.9216152019002373e-05, | |
| "loss": 0.5109, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 2.9453681710213776e-05, | |
| "loss": 0.5228, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.9691211401425178e-05, | |
| "loss": 0.4801, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.7545871559633027, | |
| "eval_loss": 0.4974513649940491, | |
| "eval_runtime": 11.4419, | |
| "eval_samples_per_second": 76.211, | |
| "eval_steps_per_second": 4.807, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.992874109263658e-05, | |
| "loss": 0.3829, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.0166270783847983e-05, | |
| "loss": 0.527, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 3.0403800475059386e-05, | |
| "loss": 0.5883, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.064133016627079e-05, | |
| "loss": 0.5133, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.0878859857482184e-05, | |
| "loss": 0.5178, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.7236238532110092, | |
| "eval_loss": 0.5414833426475525, | |
| "eval_runtime": 11.5239, | |
| "eval_samples_per_second": 75.669, | |
| "eval_steps_per_second": 4.773, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.111638954869359e-05, | |
| "loss": 0.4607, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.135391923990499e-05, | |
| "loss": 0.4809, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.159144893111639e-05, | |
| "loss": 0.5025, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.1828978622327794e-05, | |
| "loss": 0.514, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.20665083135392e-05, | |
| "loss": 0.581, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.7614678899082569, | |
| "eval_loss": 0.5108276009559631, | |
| "eval_runtime": 11.4369, | |
| "eval_samples_per_second": 76.245, | |
| "eval_steps_per_second": 4.809, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.23040380047506e-05, | |
| "loss": 0.5912, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3.2541567695961995e-05, | |
| "loss": 0.5702, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3.27790973871734e-05, | |
| "loss": 0.5757, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3.30166270783848e-05, | |
| "loss": 0.5328, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3.32541567695962e-05, | |
| "loss": 0.5175, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.6651376146788991, | |
| "eval_loss": 0.5825160145759583, | |
| "eval_runtime": 12.4677, | |
| "eval_samples_per_second": 69.941, | |
| "eval_steps_per_second": 4.411, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3.3491686460807606e-05, | |
| "loss": 0.4831, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.372921615201901e-05, | |
| "loss": 0.3822, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.396674584323041e-05, | |
| "loss": 0.4679, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.4204275534441806e-05, | |
| "loss": 0.4313, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.444180522565321e-05, | |
| "loss": 0.4897, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.7362385321100917, | |
| "eval_loss": 0.5593736171722412, | |
| "eval_runtime": 26.9471, | |
| "eval_samples_per_second": 32.36, | |
| "eval_steps_per_second": 2.041, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.467933491686461e-05, | |
| "loss": 0.4222, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.4916864608076014e-05, | |
| "loss": 0.4754, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.515439429928741e-05, | |
| "loss": 0.5092, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.539192399049881e-05, | |
| "loss": 0.4992, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.5629453681710215e-05, | |
| "loss": 0.5653, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.7672018348623854, | |
| "eval_loss": 0.48585110902786255, | |
| "eval_runtime": 24.8178, | |
| "eval_samples_per_second": 35.136, | |
| "eval_steps_per_second": 2.216, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.586698337292162e-05, | |
| "loss": 0.4569, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.6104513064133013e-05, | |
| "loss": 0.4708, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.6342042755344416e-05, | |
| "loss": 0.5629, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.657957244655582e-05, | |
| "loss": 0.4748, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.681710213776722e-05, | |
| "loss": 0.4647, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7511467889908257, | |
| "eval_loss": 0.5034508109092712, | |
| "eval_runtime": 25.1309, | |
| "eval_samples_per_second": 34.698, | |
| "eval_steps_per_second": 2.189, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.7054631828978624e-05, | |
| "loss": 0.3937, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.7292161520190026e-05, | |
| "loss": 0.4446, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.752969121140142e-05, | |
| "loss": 0.4019, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.7767220902612825e-05, | |
| "loss": 0.4452, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.800475059382423e-05, | |
| "loss": 0.5062, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_accuracy": 0.7775229357798165, | |
| "eval_loss": 0.4737890362739563, | |
| "eval_runtime": 28.0162, | |
| "eval_samples_per_second": 31.125, | |
| "eval_steps_per_second": 1.963, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.824228028503563e-05, | |
| "loss": 0.5397, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.847980997624703e-05, | |
| "loss": 0.3397, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.8717339667458435e-05, | |
| "loss": 0.474, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.895486935866984e-05, | |
| "loss": 0.4831, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.919239904988123e-05, | |
| "loss": 0.4812, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.7362385321100917, | |
| "eval_loss": 0.5663760304450989, | |
| "eval_runtime": 23.8733, | |
| "eval_samples_per_second": 36.526, | |
| "eval_steps_per_second": 2.304, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.9429928741092636e-05, | |
| "loss": 0.4661, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.966745843230404e-05, | |
| "loss": 0.5724, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.990498812351544e-05, | |
| "loss": 0.5851, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.0142517814726843e-05, | |
| "loss": 0.5317, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.0380047505938246e-05, | |
| "loss": 0.4891, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.6123853211009175, | |
| "eval_loss": 0.7159540057182312, | |
| "eval_runtime": 21.4544, | |
| "eval_samples_per_second": 40.644, | |
| "eval_steps_per_second": 2.564, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.061757719714965e-05, | |
| "loss": 0.5505, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.0855106888361044e-05, | |
| "loss": 0.4684, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.109263657957245e-05, | |
| "loss": 0.3916, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 4.133016627078385e-05, | |
| "loss": 0.48, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.156769596199525e-05, | |
| "loss": 0.494, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.6961009174311926, | |
| "eval_loss": 0.6904969811439514, | |
| "eval_runtime": 12.0521, | |
| "eval_samples_per_second": 72.352, | |
| "eval_steps_per_second": 4.564, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.1805225653206655e-05, | |
| "loss": 0.4914, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.204275534441806e-05, | |
| "loss": 0.4142, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.228028503562946e-05, | |
| "loss": 0.3954, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.2517814726840856e-05, | |
| "loss": 0.4304, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.275534441805226e-05, | |
| "loss": 0.4446, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.7603211009174312, | |
| "eval_loss": 0.470760315656662, | |
| "eval_runtime": 11.4415, | |
| "eval_samples_per_second": 76.214, | |
| "eval_steps_per_second": 4.807, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.299287410926366e-05, | |
| "loss": 0.4779, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.323040380047506e-05, | |
| "loss": 0.4142, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.3467933491686466e-05, | |
| "loss": 0.4997, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.370546318289787e-05, | |
| "loss": 0.456, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.394299287410927e-05, | |
| "loss": 0.3551, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.6720183486238532, | |
| "eval_loss": 0.9761282801628113, | |
| "eval_runtime": 11.4416, | |
| "eval_samples_per_second": 76.213, | |
| "eval_steps_per_second": 4.807, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.418052256532067e-05, | |
| "loss": 0.5181, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.441805225653207e-05, | |
| "loss": 0.4539, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.465558194774347e-05, | |
| "loss": 0.5408, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.4893111638954874e-05, | |
| "loss": 0.4418, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.513064133016627e-05, | |
| "loss": 0.4393, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_accuracy": 0.7488532110091743, | |
| "eval_loss": 0.5114963054656982, | |
| "eval_runtime": 11.4593, | |
| "eval_samples_per_second": 76.095, | |
| "eval_steps_per_second": 4.8, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.536817102137767e-05, | |
| "loss": 0.3827, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.5605700712589075e-05, | |
| "loss": 0.3863, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.584323040380048e-05, | |
| "loss": 0.3994, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.6080760095011874e-05, | |
| "loss": 0.498, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.6318289786223276e-05, | |
| "loss": 0.4129, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.7924311926605505, | |
| "eval_loss": 0.4416314661502838, | |
| "eval_runtime": 27.8567, | |
| "eval_samples_per_second": 31.303, | |
| "eval_steps_per_second": 1.974, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.655581947743468e-05, | |
| "loss": 0.305, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.679334916864608e-05, | |
| "loss": 0.3489, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.7030878859857484e-05, | |
| "loss": 0.5725, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.7268408551068886e-05, | |
| "loss": 0.3962, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.750593824228028e-05, | |
| "loss": 0.428, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.713302752293578, | |
| "eval_loss": 0.5976735949516296, | |
| "eval_runtime": 26.9577, | |
| "eval_samples_per_second": 32.347, | |
| "eval_steps_per_second": 2.04, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.7743467933491685e-05, | |
| "loss": 0.4047, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.798099762470309e-05, | |
| "loss": 0.4644, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.821852731591449e-05, | |
| "loss": 0.5173, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.845605700712589e-05, | |
| "loss": 0.4207, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.8693586698337295e-05, | |
| "loss": 0.6847, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_accuracy": 0.7522935779816514, | |
| "eval_loss": 0.4740794599056244, | |
| "eval_runtime": 26.5007, | |
| "eval_samples_per_second": 32.905, | |
| "eval_steps_per_second": 2.075, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.89311163895487e-05, | |
| "loss": 0.4262, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.9168646080760093e-05, | |
| "loss": 0.3127, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.9406175771971496e-05, | |
| "loss": 0.4341, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.96437054631829e-05, | |
| "loss": 0.3944, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.98812351543943e-05, | |
| "loss": 0.4921, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.731651376146789, | |
| "eval_loss": 0.5092917680740356, | |
| "eval_runtime": 29.3929, | |
| "eval_samples_per_second": 29.667, | |
| "eval_steps_per_second": 1.871, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.998680390604381e-05, | |
| "loss": 0.4166, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.996041171813143e-05, | |
| "loss": 0.3623, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.9934019530219056e-05, | |
| "loss": 0.4034, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.990762734230668e-05, | |
| "loss": 0.4566, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.98812351543943e-05, | |
| "loss": 0.4414, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.713302752293578, | |
| "eval_loss": 0.6652459502220154, | |
| "eval_runtime": 25.6161, | |
| "eval_samples_per_second": 34.041, | |
| "eval_steps_per_second": 2.147, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.9854842966481924e-05, | |
| "loss": 0.4862, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.9831089997360785e-05, | |
| "loss": 0.5651, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.98046978094484e-05, | |
| "loss": 0.4251, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.9778305621536024e-05, | |
| "loss": 0.514, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.9751913433623646e-05, | |
| "loss": 0.3697, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_accuracy": 0.7350917431192661, | |
| "eval_loss": 0.5495473146438599, | |
| "eval_runtime": 27.2319, | |
| "eval_samples_per_second": 32.021, | |
| "eval_steps_per_second": 2.02, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 4.972552124571127e-05, | |
| "loss": 0.3555, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.969912905779889e-05, | |
| "loss": 0.3627, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.9672736869886514e-05, | |
| "loss": 0.353, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.964634468197414e-05, | |
| "loss": 0.3561, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.961995249406176e-05, | |
| "loss": 0.3599, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.783256880733945, | |
| "eval_loss": 0.4456700384616852, | |
| "eval_runtime": 27.007, | |
| "eval_samples_per_second": 32.288, | |
| "eval_steps_per_second": 2.037, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.959356030614938e-05, | |
| "loss": 0.3917, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.9567168118237005e-05, | |
| "loss": 0.3342, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.954077593032463e-05, | |
| "loss": 0.3964, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.9514383742412244e-05, | |
| "loss": 0.4588, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.9487991554499866e-05, | |
| "loss": 0.4021, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.7924311926605505, | |
| "eval_loss": 0.43415939807891846, | |
| "eval_runtime": 28.1746, | |
| "eval_samples_per_second": 30.95, | |
| "eval_steps_per_second": 1.952, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.946159936658749e-05, | |
| "loss": 0.3698, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.943520717867511e-05, | |
| "loss": 0.4352, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.9408814990762734e-05, | |
| "loss": 0.4253, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.938242280285036e-05, | |
| "loss": 0.3442, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.935603061493798e-05, | |
| "loss": 0.4341, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.7626146788990825, | |
| "eval_loss": 0.4870525598526001, | |
| "eval_runtime": 27.9701, | |
| "eval_samples_per_second": 31.176, | |
| "eval_steps_per_second": 1.966, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.93296384270256e-05, | |
| "loss": 0.3156, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.9303246239113225e-05, | |
| "loss": 0.3465, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.927685405120085e-05, | |
| "loss": 0.3555, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.9250461863288464e-05, | |
| "loss": 0.4034, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.9224069675376086e-05, | |
| "loss": 0.4811, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_accuracy": 0.694954128440367, | |
| "eval_loss": 1.0977351665496826, | |
| "eval_runtime": 25.7159, | |
| "eval_samples_per_second": 33.909, | |
| "eval_steps_per_second": 2.139, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.919767748746371e-05, | |
| "loss": 0.7109, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.917128529955133e-05, | |
| "loss": 0.4311, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.9144893111638955e-05, | |
| "loss": 0.4666, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.911850092372658e-05, | |
| "loss": 0.6718, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.90921087358142e-05, | |
| "loss": 0.417, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_accuracy": 0.7637614678899083, | |
| "eval_loss": 0.4990720748901367, | |
| "eval_runtime": 32.7311, | |
| "eval_samples_per_second": 26.641, | |
| "eval_steps_per_second": 1.68, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.906571654790182e-05, | |
| "loss": 0.4022, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.9039324359989445e-05, | |
| "loss": 0.3948, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.901293217207707e-05, | |
| "loss": 0.4361, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.898653998416469e-05, | |
| "loss": 0.3763, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.8960147796252307e-05, | |
| "loss": 0.4257, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.7626146788990825, | |
| "eval_loss": 0.6092020869255066, | |
| "eval_runtime": 17.7696, | |
| "eval_samples_per_second": 49.073, | |
| "eval_steps_per_second": 3.095, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.893375560833993e-05, | |
| "loss": 0.3702, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.890736342042755e-05, | |
| "loss": 0.3374, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.8880971232515175e-05, | |
| "loss": 0.4812, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.88545790446028e-05, | |
| "loss": 0.3349, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.882818685669042e-05, | |
| "loss": 0.4071, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.8084862385321101, | |
| "eval_loss": 0.44936081767082214, | |
| "eval_runtime": 11.5824, | |
| "eval_samples_per_second": 75.287, | |
| "eval_steps_per_second": 4.749, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.880179466877804e-05, | |
| "loss": 0.4774, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.8775402480865665e-05, | |
| "loss": 0.3806, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.874901029295329e-05, | |
| "loss": 0.38, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.872261810504091e-05, | |
| "loss": 0.4337, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.869622591712853e-05, | |
| "loss": 0.3033, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_accuracy": 0.783256880733945, | |
| "eval_loss": 0.4898684322834015, | |
| "eval_runtime": 28.0155, | |
| "eval_samples_per_second": 31.126, | |
| "eval_steps_per_second": 1.963, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.866983372921615e-05, | |
| "loss": 0.3711, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.864344154130377e-05, | |
| "loss": 0.3867, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.8617049353391395e-05, | |
| "loss": 0.4458, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.859065716547902e-05, | |
| "loss": 0.3669, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.856426497756664e-05, | |
| "loss": 0.4616, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.46433430910110474, | |
| "eval_runtime": 21.9181, | |
| "eval_samples_per_second": 39.784, | |
| "eval_steps_per_second": 2.509, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.853787278965426e-05, | |
| "loss": 0.5175, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.8511480601741886e-05, | |
| "loss": 0.4101, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.848508841382951e-05, | |
| "loss": 0.3916, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.845869622591713e-05, | |
| "loss": 0.3209, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.8432304038004754e-05, | |
| "loss": 0.4432, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.46843382716178894, | |
| "eval_runtime": 21.7437, | |
| "eval_samples_per_second": 40.104, | |
| "eval_steps_per_second": 2.529, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.840591185009237e-05, | |
| "loss": 0.33, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.837951966217999e-05, | |
| "loss": 0.3966, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.8353127474267615e-05, | |
| "loss": 0.312, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.832673528635524e-05, | |
| "loss": 0.3508, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.830034309844286e-05, | |
| "loss": 0.3636, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_accuracy": 0.7694954128440367, | |
| "eval_loss": 0.6283801198005676, | |
| "eval_runtime": 20.1708, | |
| "eval_samples_per_second": 43.231, | |
| "eval_steps_per_second": 2.727, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.827395091053048e-05, | |
| "loss": 0.5102, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.8247558722618106e-05, | |
| "loss": 0.4305, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.822116653470573e-05, | |
| "loss": 0.3684, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.819477434679335e-05, | |
| "loss": 0.3314, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.8168382158880974e-05, | |
| "loss": 0.4871, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.7729357798165137, | |
| "eval_loss": 0.5208825469017029, | |
| "eval_runtime": 21.7303, | |
| "eval_samples_per_second": 40.128, | |
| "eval_steps_per_second": 2.531, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.814198997096859e-05, | |
| "loss": 0.3994, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.811559778305621e-05, | |
| "loss": 0.3938, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.8089205595143835e-05, | |
| "loss": 0.361, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.806281340723146e-05, | |
| "loss": 0.3131, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.803642121931908e-05, | |
| "loss": 0.4091, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.8027522935779816, | |
| "eval_loss": 0.43396520614624023, | |
| "eval_runtime": 21.2308, | |
| "eval_samples_per_second": 41.072, | |
| "eval_steps_per_second": 2.591, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.80100290314067e-05, | |
| "loss": 0.386, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.7983636843494326e-05, | |
| "loss": 0.3953, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.795724465558195e-05, | |
| "loss": 0.4312, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.793085246766957e-05, | |
| "loss": 0.3507, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.7904460279757194e-05, | |
| "loss": 0.2085, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.8004587155963303, | |
| "eval_loss": 0.5883902311325073, | |
| "eval_runtime": 21.6546, | |
| "eval_samples_per_second": 40.269, | |
| "eval_steps_per_second": 2.54, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.7878068091844817e-05, | |
| "loss": 0.5439, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.785167590393243e-05, | |
| "loss": 0.4659, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.7825283716020055e-05, | |
| "loss": 0.495, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.779889152810768e-05, | |
| "loss": 0.4629, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.77724993401953e-05, | |
| "loss": 0.3517, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.5798487067222595, | |
| "eval_runtime": 22.0365, | |
| "eval_samples_per_second": 39.571, | |
| "eval_steps_per_second": 2.496, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.774610715228292e-05, | |
| "loss": 0.5166, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.7719714964370546e-05, | |
| "loss": 0.4298, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.769332277645817e-05, | |
| "loss": 0.3885, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.766693058854579e-05, | |
| "loss": 0.4401, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.7640538400633414e-05, | |
| "loss": 0.37, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.5207229852676392, | |
| "eval_runtime": 11.4929, | |
| "eval_samples_per_second": 75.873, | |
| "eval_steps_per_second": 4.786, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.761414621272104e-05, | |
| "loss": 0.434, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.758775402480866e-05, | |
| "loss": 0.3609, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.756136183689628e-05, | |
| "loss": 0.4148, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.7534969648983905e-05, | |
| "loss": 0.3513, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.750857746107152e-05, | |
| "loss": 0.4267, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.7752293577981652, | |
| "eval_loss": 0.4631665349006653, | |
| "eval_runtime": 11.4748, | |
| "eval_samples_per_second": 75.993, | |
| "eval_steps_per_second": 4.793, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.748218527315914e-05, | |
| "loss": 0.3484, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.7455793085246766e-05, | |
| "loss": 0.2985, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.742940089733439e-05, | |
| "loss": 0.4244, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.740300870942201e-05, | |
| "loss": 0.3836, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.7376616521509634e-05, | |
| "loss": 0.4646, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.7591743119266054, | |
| "eval_loss": 0.5199323296546936, | |
| "eval_runtime": 11.5496, | |
| "eval_samples_per_second": 75.5, | |
| "eval_steps_per_second": 4.762, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.735022433359726e-05, | |
| "loss": 0.3064, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.732383214568488e-05, | |
| "loss": 0.3248, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.72974399577725e-05, | |
| "loss": 0.3718, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.7271047769860125e-05, | |
| "loss": 0.3535, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.724465558194775e-05, | |
| "loss": 0.3569, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_accuracy": 0.7672018348623854, | |
| "eval_loss": 0.4929494559764862, | |
| "eval_runtime": 12.339, | |
| "eval_samples_per_second": 70.67, | |
| "eval_steps_per_second": 4.457, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.721826339403537e-05, | |
| "loss": 0.5133, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.7191871206122986e-05, | |
| "loss": 0.3535, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.716547901821061e-05, | |
| "loss": 0.4051, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.713908683029823e-05, | |
| "loss": 0.341, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.7112694642385854e-05, | |
| "loss": 0.3356, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.4769574701786041, | |
| "eval_runtime": 11.5756, | |
| "eval_samples_per_second": 75.331, | |
| "eval_steps_per_second": 4.751, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.708630245447348e-05, | |
| "loss": 0.3053, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.70599102665611e-05, | |
| "loss": 0.3905, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.703351807864872e-05, | |
| "loss": 0.4397, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.7007125890736345e-05, | |
| "loss": 0.4199, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.698073370282397e-05, | |
| "loss": 0.3777, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_accuracy": 0.786697247706422, | |
| "eval_loss": 0.4674142897129059, | |
| "eval_runtime": 25.3754, | |
| "eval_samples_per_second": 34.364, | |
| "eval_steps_per_second": 2.167, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.695434151491159e-05, | |
| "loss": 0.3039, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.692794932699921e-05, | |
| "loss": 0.3881, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.6901557139086836e-05, | |
| "loss": 0.2602, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.687516495117445e-05, | |
| "loss": 0.3553, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.6848772763262074e-05, | |
| "loss": 0.3472, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.7373853211009175, | |
| "eval_loss": 0.5634092688560486, | |
| "eval_runtime": 26.2811, | |
| "eval_samples_per_second": 33.18, | |
| "eval_steps_per_second": 2.093, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.68223805753497e-05, | |
| "loss": 0.4972, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.679598838743732e-05, | |
| "loss": 0.3577, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.676959619952494e-05, | |
| "loss": 0.3653, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.6743204011612565e-05, | |
| "loss": 0.2476, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.671681182370019e-05, | |
| "loss": 0.4177, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.783256880733945, | |
| "eval_loss": 0.5188720226287842, | |
| "eval_runtime": 24.6069, | |
| "eval_samples_per_second": 35.437, | |
| "eval_steps_per_second": 2.235, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.669041963578781e-05, | |
| "loss": 0.3181, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.666402744787543e-05, | |
| "loss": 0.3579, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.6637635259963056e-05, | |
| "loss": 0.5733, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.661124307205068e-05, | |
| "loss": 0.3851, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.65848508841383e-05, | |
| "loss": 0.4028, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.4956331253051758, | |
| "eval_runtime": 24.5882, | |
| "eval_samples_per_second": 35.464, | |
| "eval_steps_per_second": 2.237, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.6558458696225924e-05, | |
| "loss": 0.3687, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.653206650831354e-05, | |
| "loss": 0.353, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.650567432040116e-05, | |
| "loss": 0.3628, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.6479282132488785e-05, | |
| "loss": 0.5389, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.645288994457641e-05, | |
| "loss": 0.483, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.8084862385321101, | |
| "eval_loss": 0.43807780742645264, | |
| "eval_runtime": 24.4347, | |
| "eval_samples_per_second": 35.687, | |
| "eval_steps_per_second": 2.251, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.642649775666403e-05, | |
| "loss": 0.3414, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.640010556875165e-05, | |
| "loss": 0.4314, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.6373713380839276e-05, | |
| "loss": 0.2962, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.63473211929269e-05, | |
| "loss": 0.2627, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.632092900501452e-05, | |
| "loss": 0.3413, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_accuracy": 0.7935779816513762, | |
| "eval_loss": 0.5697915554046631, | |
| "eval_runtime": 14.9148, | |
| "eval_samples_per_second": 58.465, | |
| "eval_steps_per_second": 3.688, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.6294536817102144e-05, | |
| "loss": 0.4393, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.626814462918977e-05, | |
| "loss": 0.4142, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.624175244127739e-05, | |
| "loss": 0.3803, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.6215360253365005e-05, | |
| "loss": 0.4087, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.618896806545263e-05, | |
| "loss": 0.3966, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.786697247706422, | |
| "eval_loss": 0.47139275074005127, | |
| "eval_runtime": 23.4839, | |
| "eval_samples_per_second": 37.132, | |
| "eval_steps_per_second": 2.342, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.616257587754025e-05, | |
| "loss": 0.3998, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.6136183689627873e-05, | |
| "loss": 0.3903, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.6109791501715496e-05, | |
| "loss": 0.3425, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.608339931380312e-05, | |
| "loss": 0.321, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.605700712589074e-05, | |
| "loss": 0.3299, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_accuracy": 0.783256880733945, | |
| "eval_loss": 0.46382883191108704, | |
| "eval_runtime": 24.48, | |
| "eval_samples_per_second": 35.621, | |
| "eval_steps_per_second": 2.247, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.6030614937978364e-05, | |
| "loss": 0.4465, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.600422275006599e-05, | |
| "loss": 0.3761, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.597783056215361e-05, | |
| "loss": 0.346, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 4.595143837424123e-05, | |
| "loss": 0.3839, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.592504618632885e-05, | |
| "loss": 0.4783, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.7844036697247706, | |
| "eval_loss": 0.49812304973602295, | |
| "eval_runtime": 24.7287, | |
| "eval_samples_per_second": 35.263, | |
| "eval_steps_per_second": 2.224, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.589865399841647e-05, | |
| "loss": 0.3789, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.5872261810504094e-05, | |
| "loss": 0.4411, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.5845869622591716e-05, | |
| "loss": 0.4694, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.581947743467934e-05, | |
| "loss": 0.2994, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.579308524676696e-05, | |
| "loss": 0.4475, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.8027522935779816, | |
| "eval_loss": 0.4598585367202759, | |
| "eval_runtime": 24.9853, | |
| "eval_samples_per_second": 34.901, | |
| "eval_steps_per_second": 2.201, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.5766693058854584e-05, | |
| "loss": 0.3204, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.574030087094221e-05, | |
| "loss": 0.3833, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.571390868302983e-05, | |
| "loss": 0.2844, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.568751649511745e-05, | |
| "loss": 0.3206, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.5661124307205075e-05, | |
| "loss": 0.3527, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.5331198573112488, | |
| "eval_runtime": 24.8534, | |
| "eval_samples_per_second": 35.086, | |
| "eval_steps_per_second": 2.213, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.563473211929269e-05, | |
| "loss": 0.3891, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.5608339931380314e-05, | |
| "loss": 0.324, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.5581947743467936e-05, | |
| "loss": 0.3532, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.3021, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.552916336764318e-05, | |
| "loss": 0.4124, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.7626146788990825, | |
| "eval_loss": 0.5969462394714355, | |
| "eval_runtime": 19.8238, | |
| "eval_samples_per_second": 43.988, | |
| "eval_steps_per_second": 2.774, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 4.5502771179730804e-05, | |
| "loss": 0.5429, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.547637899181843e-05, | |
| "loss": 0.4458, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.544998680390605e-05, | |
| "loss": 0.3272, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.542359461599367e-05, | |
| "loss": 0.3482, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.5397202428081295e-05, | |
| "loss": 0.3683, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.7775229357798165, | |
| "eval_loss": 0.5118904113769531, | |
| "eval_runtime": 11.5129, | |
| "eval_samples_per_second": 75.741, | |
| "eval_steps_per_second": 4.777, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.537081024016891e-05, | |
| "loss": 0.2501, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.5344418052256534e-05, | |
| "loss": 0.3256, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.5318025864344157e-05, | |
| "loss": 0.4049, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.529163367643178e-05, | |
| "loss": 0.2541, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.52652414885194e-05, | |
| "loss": 0.3894, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.8084862385321101, | |
| "eval_loss": 0.5941323041915894, | |
| "eval_runtime": 25.3823, | |
| "eval_samples_per_second": 34.355, | |
| "eval_steps_per_second": 2.167, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.5238849300607025e-05, | |
| "loss": 0.4195, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.521245711269465e-05, | |
| "loss": 0.2915, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.518606492478227e-05, | |
| "loss": 0.3745, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.515967273686989e-05, | |
| "loss": 0.3915, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.5133280548957515e-05, | |
| "loss": 0.4001, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.7717889908256881, | |
| "eval_loss": 0.4977372884750366, | |
| "eval_runtime": 11.5193, | |
| "eval_samples_per_second": 75.699, | |
| "eval_steps_per_second": 4.775, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.510688836104514e-05, | |
| "loss": 0.2936, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.5080496173132754e-05, | |
| "loss": 0.3909, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.505410398522038e-05, | |
| "loss": 0.356, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.5027711797308e-05, | |
| "loss": 0.3305, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.500131960939562e-05, | |
| "loss": 0.3394, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7981651376146789, | |
| "eval_loss": 0.5128748416900635, | |
| "eval_runtime": 11.5097, | |
| "eval_samples_per_second": 75.762, | |
| "eval_steps_per_second": 4.779, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 21050, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.6839258734592e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |